DEMI 2011 Proceedings

University of Banja Luka
Faculty of Mechanical Engineering

26th - 28th May 2011
DEMI 2011
10th Anniversary International Conference

on Accomplishments in
Electrical and Mechanical Engineering and
Information Technology
PROCEEDINGS
ZBORNIK RADOVA
BANJA LUKA, May 2011.
PROCEEDINGS
ZBORNIK RADOVA
University of Banja Luka

Faculty of Mechanical Engineering
BANJA LUKA, May 2011.
ii
PROCEEDINGS OF THE 10th ANNIVERSARY INTERNATIONAL CONFERENCE ON

ACCOMPLISMENTS IN ELECTRICAL AND MECHANICAL ENGINEERING AND
INFORMATION TECHNOLOGY
Under patronage of:
Ministry of Science and Technology of the Republic of Srpska,
Ministry of Industry, Energy and Mining of the Republic of Srpska
and
City of Banjaluka
Publisher:
Faculty of Mechanical Engineering Banja Luka
For publisher:
PhD. Miroslav Rogi, Full Professor
Editor in Chief:
PhD. Gordana Globoki-Laki, Associate Professor
Organizational board:
Gordana Globoki-Laki, PhD. Associate Professor, Chairman
Miroslav Rogi, PhD. Professor
Sneana Petkovi, PhD. Associate Professor
Zdravko Milovanovi, PhD. Associate Professor
Petar Gvero, PhD. Associate Professor
Strain Posavljak, PhD. Assistant Professor
Darko Kneevi, PhD. Assistant Professor
Tihomir Latinovi, PhD. Assistant Professor
Valentina Golubovi-Bugarski, PhD. Assistant Professor
Milan Tica, MSc. Mechanical Engineering
Stevo Borojevi, MSc. Mechanical Engineering
Bojan Kneevi, MSc. Electrical Engineering
Branislav Sredanovi, BSc. Mechanical Engineering
Branislav Jovkovi, BSc. Mechanical Engineering
Milivoj Stipanovi
and
Ljubo Glamoi, PhD. Ministry of Industry, Energy and Mining of the Republic of
Srpska
Technical processing:
Milivoj Stipanovi
Circulation:
190
iii
iv
SCIENTIFIC BOARD
Blagojevi Aleksa, Faculty of Mechanical Engineering Banjaluka, B&H
Blagojevi Drago, Faculty of Mechanical Engineering Banjaluka, B&H
Bobrek Miroslav, Faculty of Mechanical Engineering Banjaluka, B&H
Bojani Pavao, Faculty of Mechanical Engineering Beograd, Serbia
Bulatovi Miodrag, Faculty of Mechanical Engineering Podgorica, Montenegro
osi Ilija, FTN Novi Sad, Serbia
Daki Pantelija, Faculty of Mechanical Engineering Banjaluka, B&H
Doloek Vlatko, University of Sarajevo, B&H
urikovi Veljko, Faculty of Mechanical Engineering Banjaluka, B&H
Filipovi Ivan, Faculty of Mechanical Engineering Sarajevo, B&H
Gerasimuk G. Vasilj, KPI, Ukrainia
Gruden Duan, TU Wien, Austria
Ivkovi Branko, Faculty of Mechanical Engineering Kragujevac, Serbia
Jokanovi Simo, Faculty of Mechanical Engineering Banjaluka, B&H
Jovievi Vid, Faculty of Mechanical Engineering Banjaluka, B&H
Koji Milo, Harvard University, USA
Kostolansky Eduard, University of Cyril and Metodius Trnava, Slovakia
Kozi ore, Faculty of Mechanical Engineering Beograd, Serbia
Luka Duko, University of Applied Science, Germany
Maksimovi Stevan, Aeronautical Institute, Serbia
Mileti Ostoja, Faculty of Mechanical Engineering Banjaluka, B&H
Milii Dragomir, Faculty of Mechanical Engineering Banjaluka, B&H
Milutinovi Dragan, Faculty of Mechanical Engineering Beograd, Serbia
Nedi Bogdan, Faculty of Mechanical Engineering Kragujevac, Serbia
Ninkovi Dobrivoje, ABB Turbo-Systems AG, Switzerland
Ognjanovi Milosav, Faculty of Mechanical Engineering Beograd, Serbia
Pei Radivoje, Faculty of Mechanical Engineering Kragujevac, Serbia
Planak Miroslav, FTN Novi Sad, Serbia
Pop Nicolae, North University of Baia Mare, Romania
Radovanovi Milan, Faculty of Mechanical Engineering Beograd, Serbia
Ravano Giambattista, University SUPSI, Switzerland
Savi Vladimir, FTN Novi Sad, Serbia
Schmied Joachim, Delta JS, Switzerland
Seok Park Hong, University of Ulsan, Korea
ljivi Milan, Faculty of Mechanical Engineering Banjaluka, B&H
Sokovi Mirko, Faculty of Mechanical Engineering Ljubljana, Slovenia
Stefanovi Milentije, Faculty of Mechanical Engineering Kragujevac, Serbia
Stegi Milenko, FSB Zagreb, Croatia
Thomeensen Trygve, Norwegian University of Science and Technology, Norway
Todi Velimir, FTN Novi Sad, Serbia
Tufeki emo, Faculty of Mechanical Engineering Tuzla, B&H
Veinovi Stevan, Faculty of Mechanical Engineering Kragujevac, Serbia
Vere Miroslav, Faculty of Mechanical Engineering, Bratislava, Slovakia
Zeljkovi Milan, FTN Novi Sad, Serbia
Zrili Ranko, Faculty of Mechanical Engineering Banjaluka, B&H
Zrni Nenad, Faculty of Mechanical Engineering Beograd, Serbia
v
vi
CONTENT
KEYNOTE LECTURES ................................................................................................... 1
1.
2.
3.
4.
5.
6.
Claudio R. Bor
SUSTAINABLE INNOVATION AND INNOVATION FOR SUSTAINABILITY:
THE EVOLUTION AND INVOLUTION OF PRODUCTION PARADIGMS ........... 3
Giambattista Ravano
COOPERATION MODELS BETWEEN UNIVERSITIES AND INDUSTRY IN
APPLIED RESEARCH, SWITZERLAND CASE STUDY AND SOME
PRACTICAL EXAMPLES ...................................................................................... 5
Radivoje Pei, Stevan Veinovi
TRANSPORT ECOLOGY AND GLOBAL CLIMATE CHANGE ............................ 7
Milosav Ognjanovi
DESIGN CONSTRAINTS AND ROBUST DESIGN AS THE MODERN
APPROACH TO MECHANICAL STRUCTURE DEVELOPMENT DESIGN
CONSTRAINTS AND ROBUST DESIGN AS THE MODERN APPROACH
TO MECHANICAL STRUCTURE DEVELOPMENT ........................................... 21
Neven Dui
RENEWABLE ENERGY AS A DRIVER OF ECONOMIC GROWTH ................. 35
Stojan Petrovi, Boidar Nikoli, Emil Hnatko, Jovo Mra, Stevan Veinovi
MALICIOUS ECOLOGY ON VEHICLES AND TRAFFIC EXAMPLE ................. 55
A. MECHANICS AND DESIGN ................................................................................... 57

7.
8.
9.
10.
11.
12.
13.
14.
Michail Leparov, Georgi Dinev, Marieta Jancheva

ABOUT RECEIPTE OF VARIANTS OF TECHNICAL OBJECTS ....................... 59
Leparov M., Yancheva M.
ABOUT THE INTEGRATION OF ASSEMBLY UNITS ........................................ 65
Darko Kneevi, Aleksandar Milainovi, Zdravko Milovanovi
ANALYSIS OF INFLUENCE OF LENGTH OF DEVELOPMENT OF
BOUNDARY LAYER ON FLOW RATE THROUGH RADIAL CLEARANCE
WITHIN HYDRAULIC CONTROL COMPONENTS ........................................... 71
Neboja Radi, Goran Sekuli, Dejan Jeremi
ANALYTICAL-NUMERICAL STRESS ANALYSIS OF SPUR GEARS WITH
STRAIGHT TEETH ............................................................................................. 77
Georgy Dinev, Marieta Yancheva
CAD DESIGN OF FLEXIBLE FRICTION COUPLING ....................................... 83
Dragan Lianin, Marinko Petrovi, Nenad Grujovi, Jelena Borota
COMPARATIVE ANALYSIS OF THE FORMATION OF SMALL GRAIN
GUIDANCE ......................................................................................................... 87
Pavle Stepani, eljko urovi, Aleksa Kronjar, Aleksandra Pavasovi
COMPARISON OF TECHNIQUES FOR DETECTION OF FAILURE
ROLLING ELEMENT BEARINGS ....................................................................... 93
Strain Posavljak, Miodrag Jankovic, Katarina Maksimovic
CRACK INITIATION LIFE OF NOTCHED METALLIC PARTS EXPOSED TO
LOW CYCLE FATIGUE ....................................................................................... 99
vii
15
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
viii
Sran Bonjak, Zoran Petkovi, Milo orevi, Neboja Gnjatovi,

Nenad Zrni
DESIGN IMPROVEMENTS OF THE BUCKET WHEEL WITH DRIVE ............111
Aleksandar Marinkovi, Aleksandar oi, Bratislav Stojiljkovi,
Milan Vulievi
DESIGN OF TESLA-TIFFANY CASCADE FOUNTAIN AS A SAMPLE OF
TESLA`S RESEARCH CREATIVITY IN FIELD OF MECHANICAL
ENGINEERING .................................................................................................117
Svetislav Lj. Markovi
DESIGN SEALS FOR REAL CONNECTIONS .................................................123
Valentina Golubovi-Bugarski, Drago Blagojevi, ore ia, Branislav
Sredanovi
DETECTION OF STRUCTURAL DAMAGE LOCATION USING
FREQUENCY RESPONSE FUNCTION DATA .................................................129
Dragi Stamenkovi, Mato Peri
DETERMINATION OF RESIDUAL STRESSES IN TUBULAR WELDED
STRUCTURAL COMPONENTS ........................................................................135
ivko Pejainovi, Zorana Tanasi, Goran Janji
EFFECT OF MATERIAL PROPERTIES OF MEASURING FORCE
TRANSDUCER ELASTIC ELEMENTS TO METROLOGIY
CHARACTERISTICS .........................................................................................145
Sinia Kuzmanovi, Milan Rackov
EVALUATION OF CONCEPTUAL SOLUTIONS OF UNIVERSAL HELICAL
TWO STAGE GEAR UNITS ..............................................................................151
Ivica amagi, Nemanja Vasi, Zijah Burzi
FATIGUE ANALYSIS FROM FRACTURE MECHANICS ANGLE ....................159
Slobodanka Boljanovi, Stevan Maksimovi, Strain Posavljak
FATIGUE LIFE ESTIMATION OF CRACKED STRUCTURAL
COMPONENTS .................................................................................................165
Ibrahim Badak, Remzo Dedic, Mersida Manjgo
HYDRAULIC INSTALLATION OF EKO CONTAINER ......................................173
Vesna Rankovi, Nenad Grujovi, Goran Milovanovi, Dejan Divac, Nikola
Milivojevi
PREDICTION OF DAM BEHAVIOUR USING MULTIPLE LINEAR
REGRESSION AND RADIAL BASIS FUNCTION NEURAL NETWORK .........179
Nenad Zrni, Sran Bonjak, Vlada Gai, Miodrag Arsi
SOME ASPECTS IN FAILURE ANALYSIS OF CRANES ................................185
Stevan Maksimovi, Ivana Vasovi, Mirko Maksimovi
SOME ASPECTS TO DESIGN OF AIRCRAFT STRUCTURES WITH
RESPECTS TO FATIGUE AND FRACTURE MECHANICS ............................191
Andrija Vujii, Nenad Zrni
STATE-OF-THE-ART IN LIFE CYCLE ASSESSEMENT AS A CORE OF
LIFE CYCLE DESIGN .......................................................................................203
Mersida Manjgo, Ljubica Milovi, Zijah Burzi
STRESS INTENSITY FACTOR AND ITS EFFECT OF STRUCTURAL
INTEGRITY ........................................................................................................209
Vukojevi Nedeljko, Hadikaduni Fuad, Pavi Mate
VIBRATORY STRESS RELIEVING OF TANK FLANGS ..................................215
31.
Ranko Antunovic
VIBRODIAGNOSTICS OF ROTATION MACHINES ........................................ 221
B. PRODUCTION TECHNOLOGIES AND ENGINEERING ...................................... 229

32. Kramar D., Sokovi M., Kopa J.
ADVANCED CUTTING TECHNOLOGY HIGH-PRESSURE JET ASSISTED
MACHINING ....................................................................................................... 231
33. Milan Milovanovi, Milentije Stefanovi
ANALYSIS OF THE EFFECTS OF APPLYING NEW MATERIALS .................. 241
34. Tomasz Kudasik, Tadeusz Markowski, Olimpia Markowska,
Sawomir Miechowicz
APPLICATION OF RAPID PROTOTYPING RESINS FOR PHOTOELASTIC
TESTING ............................................................................................................ 247
35. Slavica Cvetkovi
AUDITING PROCESS DESIGN COMPANY LOGISTICS SYSTEM ................. 253
36. Andonovic Vladan, Vrtanoski Gligorce
CAD/CAM TECHNOLOGY IN DENTAL MEDICINE .......................................... 259
37. Zoran Janju, Aleksandar Petrovi, Aleksandar Jovovi,
Radica Proki-Cvetkovi, Predrag Ili
CHANGES VOLTAGE COMPACTION POLYPROPYLENE FILLED WITH
GLASS POWDER .............................................................................................. 265
38. Miletic Ostoja, Todic Mladen
CHANGING THE WALL THICKNESS PROFILE IN THE PROCESS OF
PROFILING ........................................................................................................ 271
39. Plavka Skakun, Miroslav Planak, Dragia Viloti, Mladomir Milutinovi, Dejan
Movrin, Ognjan Luanin
COMPARATIVE INVESTIGATION OF DIFFERENT LUBRICANTS FOR BULK
METAL FORMING OPERATIONS ..................................................................... 275
40. Borislav Kovljeni
COMPLEXITY ANALYSIS OF CAD/CAM SYSTEMS INTEGRATION IN ERP
BUSINESS ENVIRONMENT ............................................................................. 281
41. S. Aleksandrovi, T. Vujinovi, M. Stefanovi, V. Lazi, D. Adamovi
COMPUTER CONTROLLED EXPERIMENTAL DEVICE FOR
INVESTIGATIONS OF TRIBOLOGICAL INFLUENCES IN SHEET METAL
FORMING .......................................................................................................... 285
42. Radu Alexandru Rou, Viorel Aurel erban, Mihaela Popescu, Uu Drago
Cosmin Locovei
DEPOSITION OF TITANIUM NITRIDE LAYERS BY REACTIVE PLASMA
SPRAYING ......................................................................................................... 291
43. Sawomir Miechowicz, Tadeusz Markowski, Tomasz Kudasik, Olimpia Markowska
DESIGN AND FABRICATION OF MEDICAL MODELS WITH RAPID
PROTOTYPING TECHNIQUES AND VACUUM CASTING .............................. 297
44. Djordje Vukelic, Branko Tadic, Janko Hodolic, Igor Budak, Milovan Lazarevic
DEVELOPMENT AN EXPERT SYSTEM FOR MACHINING FIXTURE
DESIGN .............................................................................................................. 303
45. Bogdan Nedi, Gordana Globoki-Laki
DEVELOPMENT MODEL FOR CONTROL METAL CUTTING PROCESS .......... 309
ix
46. Aurel Prsti, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi

DEVELOPMENT OF EPC PROCESS FOR MANUFACTURING PARTS IN
AUTOMOTIVE INDUSTRY ................................................................................315
47. Obuina Muro, kalji Nedim, Smaji Selver
EFFECT OF SURFACE ROUGHNESS ON WOOD ADHESION ......................321
48. M. Stefanovi, D. Viloti, M. Planak, S. Aleksandrovi, Z.Gulisija, D. Adamovi
FORMING LIMIT INDICATORS IN METAL FORMING .....................................327
49. Runev Dobre, Gligore Vrtanoski, Ljupo Trpkovski
HEATED TOOL BUTT WELDING OF POLYETHYLEN PIPES .........................337
50. Marija Mihailovi, Aleksandra Patari, Zvonko Guliija, Miroslav Soki
INCREASING PRODUCTION EFFICIENCY THROUGH CASTING QUALITY
IMPROVING BY ELECTROMAGNETIC FIELD APPLYING ..............................343
51. Zorana Tanasi, Goran Janji, Bobrek Miroslav, ivko Pejainovi
INFLUENCE OF ORGANIZATIONAL CULTURE ON BUSINESS
PERFORMANCE ................................................................................................349
52. Robert Molnar, Drago Soldat
INNOVATION-THE KEY FACTOR IN ENTREPRENEURIAL CYCLES ............355
53. Vid Jovievi, Stevo Borojevi, Gordana Globoki-Laki, Branislav Sredanovi
LABORATORIES UNDER REQUIREMENTS OF DIRECTIVES AND
STANDARDS OF EUROPEAN UNION .............................................................361
54. Sanja Petronic, Andjelka Milosavljevic, Biljana Grujic, Radovan Radovanovic
Radmila Pljakic
LASER SHOCK PEENING OF N-155 SUPERALLOY EXPOSED TO
AGGRESSIVE MEDIUM ....................................................................................367
55. Bogdan Mari, Ranko Boikovi
LEAN CONCEPT TOOLS IN PROCESS OF TECHNICAL SYSTEMS
OVERHAUL ........................................................................................................373
56. Ranko Radonji, Milan ljivi, ivko Babi, Milentije Stefanovi
NUMERICAL SIMULATION OF HOLE FLANGING OF CIRCULAR SHEETS ..379
57. Dejan Luki, Velimir Todi, Mijodrag Miloevi, Goran Jovii
ONE APPROACH TO THE DEVELOPMENT AND IMPLEMENTATION OF
FLEXIBLE MANUFACTURING SYSTEMS ........................................................385
58. Milentije Stefanovi, Srbislav Aleksandrovi, Dragan Adamovi
PAPER ABOUT PAPERS IN THE AREA OF METAL FORMING
PRESENTED AT DEMI CONFERENCES HELD SO FAR ................................391
59. Todic Mladen, Miletic Ostoja
POSITION OF THE NEUTRAL SURFACE DEFORMATION AT BENDING
TWO LAYER COMPOSITES .............................................................................399
60. Milena Cosi, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi,
Zoran Janjuevi
POSSIBILITY TO USE RHEOCASTING PROCESS FOR MANUFACTURING
PARTS IN AUTOMOTIVE INDUSTRY ...............................................................405
61. Zvonko Guliija, Marija Mihailovi, Aleksandra Patari, Zoran Janjuevi
PROPERTIES OF THE WROUGHT Al ALLOY 7075 OBTAINED BY
ELECTROMAGNETIC CASTING PROCESS ....................................................409
62. Vrtanoski Gligorce, Andonovic Vladan
RAPID TECHNOLOGY IN DENTAL BIOMECHANICS ......................................413
x
63. Stevo Borojevi, Vid Jovievi, Gordana Globoki-Laki, ore ia,

Branislav Sredanovi, Marko Radisavljevi
SELECTION OF VARIANT FOR MATERIAL FLOW TYPE IN CONDITIONS
OF GROUP APPROACH USING THE SOFTWARE SYSTEM TECNOMATIX
PLANT SIMULATION ......................................................................................... 419
64. Velimir Todi, Dejan Luki, Mijodrag Miloevi, Jovan Vukman
TECHNOLOGICAL BASIS FOR THE DEVELOPMENT AND
IMPLEMENTATION OF FLEXIBLE MANUFACTURING SYSTEMS ................ 427
65. Goran Janji, Predrag Nagraisalovi, Zorana Tanasi, Miroslav Bobrek, ivko
Pejainovi
THE PROCESS OF MEASURING EQUIPMENT MANAGEMENT AND ITS
AUTOMATIZATION ............................................................................................ 433
66. S. Mazzola, P. Pedrazzoli, G. Dal Maso, C. R. Bor
VIRTUAL ENVIRONMENT PLATFORM FOR INDUSTRIAL OPERATION
AND MAINTENANCE ......................................................................................... 441
C. THERMOTECHNIQUE AND ENERGETICS ......................................................... 447
67. Nataa Soldat, Mirjana Radii
BASIC ASPECTS OF DEFINING MECHANICAL-TECHNOLOGICAL
SOLUTIONS FOR THE PRODUCTION OF BIOGAS FROM LIQUID MANURE 449
68. Crnojevi C., Lei M.
DETERMINATION OF PRESSURE DROP TWO-PHASE FLOW OIL AND
GAS FOR ISOTHERMAL FLOW IN HORIZONTAL PIPELINE ......................... 453
69. Igor Andreevski, Gligor Kaneve, Ljubica Kaneve, Aleksandar Markoski,
Sevde Stavreva
DEVELOPMENT AND APPLICATION OF REGULATORY DISPERSION
MODEL FOR AIR POLLUTION ASSESSMENT ................................................ 459
70. Gordana Tica, Veljko urikovi, Petar Gvero
DIMENSIONING OBJECT'S COOLING SYSTEM FOR PREDETERMINED
KNOWN RELIABILITY ....................................................................................... 465
71. Mia Vuki, Velimir Stefanovi, Predrag ivkovi, Mirko Dobrnjac
EXPERIMENTAL INVESTIGATION OF THERMAL AND FLOW
PROCESSES IN SHELL AND TUBE HEAT EHCHANGERS ........................... 475
72. Popov G., Klimentov Kl., Kostov B.
INVESTIGATION OF THE ENERGY CONSUMPTION IN REGULATING THE
FLOW RATE OF PUMP SYSTEMS ................................................................... 481
73. ore S. antrak, Slavica S. Risti, Novica Z. Jankovi
LDA, CLASSICAL PROBES AND FLOW VISUALIZATION IN EXPERIMENTAL
INVESTIGATION OF TURBULENT SWIRL FLOW ........................................... 489
74. Popov G., Klimentov Kl., Kostov B.
METHODS TO ESTIMATE THE ENERGY CONSUMPTION IN REGULATING
THE FLOW RATE OF PUMP SYSTEMS .......................................................... 495
75. Majid Soleimaninia
NUMERICAL INVESTIGATION OF HEAT TRANSFER ENHANCEMENT IN
NATURAL CONVECTION AND FORCE CONVECTION IN A FLUID
SATURATED VARIABLE POROSITY MEDIUM ............................................ 501
xi
76. Milica Grahovac

OBJECTIVE FUNCTION DEFINITION FOR PRIMARY HVAC SYSTEM
TOTAL COSTS MINIMIZATION .........................................................................515
77. Diana Alina Bistrian, Manuela Pnoiu, Tihomir Latinovi, Marcel Topor
PARALLEL SOLUTIONS TO ACCELERATE MATHEMATICAL
ALGORITHMS IN HYDRODYNAMIC STABILITY PROBLEMS ........................523
78. Sevde Stavreva, Marko Serafimov, Igor Andreevski
REDUCING CONCUMPTION OF ENERGY OF DATA CENTERS ...................533
79. N. Mani, V. Jovanovi, D. Stojiljkovi
RESULTS OF EXPERIMENTAL INVESTIGATION OF PELLET STOVE
ACCORDING TO EN 14875 ...............................................................................539
80. Mirjana Radii, Nataa Soldat
SOME EXPERIENCES IN THE PRODUCTION OF BIOGAS FROM LIQUID
MANURE ............................................................................................................549
81. Aleksandar Stjepanovi, Slaana Stjepanovi, Ferid Softi, Zlatko Bundalo
TEMPERATURE CHARACTERISTICS OF PHOTOVOLTAIC MODULE .........555
82. Predrag ivkovi, Gradimir Ili, Mirko Dobrnjac, Mladen Tomi,
ana Stevanovi,
WIND POTENTIALS ASSESMENT IN COMPLEX TERRAIN ...........................561
83. Ljubo Glamoi
WIND POWER RESOURCES IN THE REPUBLIC OF SRPSKA ......................567
84. Milovan Kotur, Gostimir Radi
COVENANT OF MAYORS FORESEEN ACTIVITIES AT DISTRICT
HEATING COMPANY TOPLANA A.D. BANJA LUKA ....................................575
85. Milovan Kotur, Branko Usorac, Petar Gvero, Gordana Tica
PARTIAL REPLACING HEAVY FUEL OIL WITH BIOMASS IN THE
DISTRICT HEATING COMPANY IN GRADIKA ..............................................581
86. Milovan Kotur, Zoran Kneevi, Petar Gvero, Gordana Tica
BIOMASS PROJECT IN DISTRICT HEATING COMPANY (DHC) IN
PRIJEDOR, BIH ..................................................................................................587
D. TRAFFIC MEANS ..................................................................................................593
87.
88.
89.
90.
91.
xii
MALICIOUS ECOLOGY ON VEHICLES AND TRAFFIC EXAMPLE ................595
Zlatomir ivanovi, Zoran Jovanovi, eljko akota
A COMPARATIVE ANALYSIS OF CNG AND HYBRID BUSES VS DIESEL
BUSES ...............................................................................................................607
Milan Milovanovi, Dragoljub Radonji, Saa Jovanovi
ADJUSTMENTS OF VEHICLES WITH GAS DRIVE ........................................613
Dalibor Jajcevic, Raimund Almbauer
APPLICATION OF A CYCLIC BOUNDARY CONDITION FOR CFD
SIMULATIONS OF A 2-CYLINDER IC-ENGINE ...............................................619
Melisa Velic, Semir Mulalic, Adnan Pecar
CALCULATING THERMODYNAMIC PROPERTIES BY CREATING AND
USING MODEL OF A DIESEL ENGINE WITH SIX CYLINDERS ....................627
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
Mile Raievi, Miroslav Demi, Neboja Rako, Predrag Milenkovi

DETERMINING THE DURATION OF VIBRATION MEASUREMENTS OF
HUMAN BODY IN LABORATORY CONDITIONS BY SUBJECTIVE
METHODS ........................................................................................................ 633
Miroljub Tomi, Stojan Petrovi, Slobodan Popovi, Nenad Milji
DUAL PORT INDUCTION SYSTEM FOR DMB 1.4 MPI ENGINE ................... 651
Blaevi A., Bibi D., Filipovi I.
FUNCTION AND ADOPTION OF IC ENGINES DUAL MASS FLYWHEEL ..... 661
Jelena Eric Obucina, Jovanka Lukic
HYDRAULIC PUMP IN THE OF VEHICLE STEERING SYSTEM ................... 667
Aleksandar Davini, Radivoje Pei, Dragan Taranovi, Miroslav Ravli
IGNITION SYSTEM OF MULTIPROCESSING OTTO/DIESEL ENGINE ......... 673
Filipovi I., Milainovi A., Blaevi A., Pecar A.
IMPACT OF THE SPECIFIC ABSORBERS ON THE DYNAMIC LOAD OF
THE IC ENGINES CRANKSHAFT ................................................................... 681
Jasna Gliovi, Jovanka Luki, Danijela Miloradovi
IMPROVEMENTS OF GROUND VEHICLES FUEL ECONOMY USING
REGENERATIVE BRAKING ............................................................................. 687
Predrag ivkovi, Mladen Tomi, Gradimir Ili, Mirko Dobrnjac,
Vladimir Lazovi,
INFLUENCE OF TRAFFIC ON AIR QUALITY IN NI ...................................... 693
Boran Pikula, Ivan Filipovi, Mirsad Trobradovi
INVESTIGATION OF DYNAMICS CHARACTERISTICS OF HYBRID
VEHICLES ......................................................................................................... 699
Vladan Ivanovic, Decan Ivanovic, Vladimir Pajkovic
LANDFILL GAS AS A FUEL FOR A VEHICLE FLEET FOR THE CITY
LANDFILL .......................................................................................................... 705
Jovanka Luki, Radivoje Pei, Dragan Taranovi
NVH INVESTIGATION OF POWER STEERING SYSTEM HYDRAULIC
PUMP ................................................................................................................ 711
Vojislav B. Krsti, Boidar V. Krsti, Vuki N. Lazi
POSIBILITY OF DETERMINATION THE ROUTES FOR TRANSPORTATION
OF HAZARDOUS GOODS ON THE BASIS OF THE RISK LEVEL ................. 717
Boidar V. Krsti, Vojislav B. Krsti, Ivan B. Krsti
POSSIBILITIES DETERMINATION OF THE OPTIMAL STRATEGY FOR
PREVENTIVE MAINTENANCE OF THE CARDAN SHIFT VEHICLE USING
POLYCRITERION OPTIMIZATION .................................................................. 723
Pikula Boran, Filipovic Ivan, Kepnik Goran
RESEARCH OF THE EXTERNAL AERODYNAMICS OF THE VEHICLE
MODEL .............................................................................................................. 731
Vladimir R. Pajkovi
ROAD TRAFFIC SAFETY PERFORMANCE IN MONTENEGRO .................... 739
Dobrivoje Ninkovic
SURVEY OF METHODS FOR CALCULATING THE WAVE ACTION IN THE
MANIFOLDS OF INTERNAL COMBUSTION ENGINES .................................. 745
Dragan Taranovic, Radivoje Pesic, Jovanka Lukic, Aleksandar Davinic
TEST BENCH FOR NON-STANDARD MEASUREMENT CHARACTERISTICS
OF RECIPROCATING COMPRESSOR ........................................................... 759
xiii
109. Branislav Aleksandrovi, Rajko Radonji, Marko apan, Aleksandra Jankovi

THE RESEARCH OF CORRELATIONS BETWEEN MOTORCYCLE
OSCILLATORY PROCESSES DURING THE NONSTEADY MODES OF
MOTION ............................................................................................................765
110. Izudin Deli, Izet Ali
TRENDS IN DEVELOPMENT OF CATALYTIC CONVERTER OF
INTERNAL COMBUSTION ENGINES (ICE) .....................................................771
111. Pantelija Daki, Sreten Peri
MONITORING OIL FOR LUBRICATION OF TRIBOMECHANICAL ENGINE
ASSEMBLIES ....................................................................................................777
E. MECHATRONICS ..................................................................................................793
112. Milan Paripovi
ANALYSES THE DAMAGE CAUSED BY ATMOSPERIC DISCHARGE AND
OVERVOLTAGE PROTECTION ........................................................................795
113. Slavia Todorovi, Miroslav Rogi
AUTOMATION AND OPTIMIZATION OF PROJECT OPERATIONS IN THE
BRIDGE CRANE DESIGN PROCESS ...............................................................801
114. Corina Daniela Cunan, Ioan Baciu, Loredana Ghiorghioni
DC STABILIZER WITH DIGITAL CONTROL .....................................................807
115. Marija Milievi, Vladimir Kaplarevic, Zoran Dimi, Vojkan Cvijanovi, Mirko
Buan
DEVELOPMENT OF DISTRIBUTED CONTROL SYSTEM FOR ROBOTS
CONTROL BASED ON REAL-TIME LINUX PLATFORM ..................................813
116. Miroslav Rogi, Bojan Kneevi, Branislav Risti
DEVELOPMENT OF THE CONCEPT OF INTERACTIVE EDUCATION IN
MECHATRONICS ..............................................................................................819
117. Ivan B. Krsti, Boidar V. Krsti, Dragan I. Milosavljevi
EFFECTIVENESS DETERMINATION OF ELECTRONIC DEVICES
PREVENTIVE MAINTENANCE ..........................................................................825
118. Mihailo P. Lazarevi, Vasilije Vasi, Ale Hace, Karel Jezernik
FURTHER RESULTS ON MODELING, INTEGRATED DESIGN AND
SIMULATION OF A MECHATRONIC SYSTEM WITH FPGA ...........................831
119. Miroslav Grubii, Snjeana Rezi
IMPACT OF SENSOR FAILURE ON WORK OF ELECTRONICALLY
CONTROLLED DIESEL ENGINES ....................................................................837
120. Miroslav Kostadinovi, Zlatko Bundalo, Duanka Bundalo
IMPLEMENTATION OF PLANTWEB ALERTS IN A DELTAV SYSTEM ...........843
121. Vahid Bagher Poor, Majid Hashemipour
IMPLEMENTATION OF RFID TECHNOLOGY AND SMART PARTS IN
WIRELESS MANUFACTURING SYSTEMS ......................................................849
122. Tihomir Latinovic, Sorin I Deaconu, Remiquez Labudski, Marcel Topor
INTELLIGENT APPROACH FOR MOBILE ROBOT SIMULATOR WITH
ROBOSIM SOFTWARE .....................................................................................857
xiv
123. Milutinovic D., Glavonjic M., Slavkovic N., Kokotovic B., Milutinovic M.,
Zivanovic S., Dimic Z.
MACHINING ROBOT CONTROLED AND PROGRAMMED AS A MACHINE
TOOL .................................................................................................................. 863
124. Vladimir Kaplarevi, Marija Milievi, Jelena Vidakovi, Vladimir Kvrgi
NEW APPROACH FOR DESSIGNING ROBOT PROGRAMING SYSTEM
BASED ON L-IRL PROGRAMING LANGUAGE ................................................ 873
125. Kostic Aleksandra, Velic Melisa, Bektesevic Jasmin
PRACTICAL STRATEGIES FOR STABILISATION OF ALGORITHMS
BASED ON SECULAR EQUATIONS OF RSPDTM .......................................... 877
126. Platon Sovilj, Nenad abrilo, Vladimir Vujii, Ivan upunski
REMOTE MEASUREMENTS BY ZIGBIT WIRELESS MODULE ...................... 885
127. K. Abhary, D. Djukic, H-Y. Hsu, Z. Kovacic, D. Mulcahy, S. Spuzic, F. Uzunovic
SOME ASPECTS OF KNOWLEDGE ENGINEERING ...................................... 893
128. Nenad Miloradovi, Rodoljub Vujanac, Blaa Stojanovi
STACKING AISLE WIDTH FOR FORKLIFT TRUCKS IN PALLETIZED
STORAGE AND HANDLING SYSTEMS ........................................................... 899
129. Mihajlo J. Stoji, Bojan Kneevi
THE CONTROLLER DESIGN FOR TRACKING TRAJECTORY WITH
CONTROLLED JERK ......................................................................................... 905
130. Deaconu, S. I., Opria N, Popa, G. N., Latinovic T.
ULTRASONIC WELDING SYSTEM FOR AUTOMOTIVE WIRINGS
INDUSTRY ......................................................................................................... 911
131. Draen Paali, Zlatko Bundalo, Duanka Bundalo, Miroslav Kostadinovi
WIRELESS SENSOR NETWORKS IN HOME AUTOMATION ......................... 917
132. Mihailo Lazarevi, Petar Mandi, Vasilije Vasi
SOME APPLICATIONS OF NEUROARM INTERACTIVE ROBOT AND
WEBOTS ROBOT SIMULATION TOOL ............................................................ 923
F. MAINTENANCE OF TECHNICAL SYSTEMS ....................................................... 929
133. Rusmir Bajri, Enver Omazi, Fehmo Mrkaljevi
AVAILABILITY ANALYSIS OF IRREDUNDANT TECHNICAL SYSTEMS ........ 931
134. Aleksandar ivkovi, Milan Zeljkovi, Milorad Rodi, Milivoje Mijukovi
COMPUTER AND EXPERIMENTAL DETERMINATION OF THE HUB UNIT
LIFE .................................................................................................................... 937
135. Danijela Nikolic, Vanja Sustersic, Jasmina Skerlic
DECENTRALIZED WASTEWATER TREATMENT SYSTEMS IN LARGE
SETTLEMENTS ................................................................................................. 943
136. Ivan B. Krsti, Dragan I. Milosavljevi, Boidar V. Krsti
DETERMINATION THE PERIODICITY OF MANAGING OF PREVENTIVE
MAINTENANCE OF TECHNICAL SYSTEMS ................................................... 949
137. Mihaela Popescu, Radu Alexandru Rou, Carmen Opri, Ibolyka Bran
ENVIRONMENT PROTECTION FOR WELDING AND ALLIED TECHNIQUES . 955
138. Milomir upovi, Desimir Jovanovi, Bogdan Nedi
FTA AND FMEA IN PREDICTING INCIDENTAL CONDITIONS IN CABLE
CARS AND SKI LIFTS ....................................................................................... 961
xv
139. Milorad Panteli, Sran Bonjak

MAINTENANCE AND LIFECYCLE OF THE EXCAVATION UNITS ..................967
140. Dusan Jovanic, Drago Soldat
MODELING MAINTENANCE ACTIVITIES ON A WELDED CONSTRUCTION
USING IDEF0 METHODOLOGY ........................................................................973
141. Jasmina Skerlic, Vanja Sustersic, Danijela Nikolic
NATURAL SYSTEMS OF WASTEWATER TREATMENT IN SMALL
SETTLEMENTS ................................................................................................. 979
142. Miodrag Milutinovi, Vladimir Popovi
PROCEDURES FOR RISK BASED MANAGEMENT AND MAINTENANCE
RIMAP PRINCIPLE ............................................................................................985
143. Dragoslav Dobra, Sead Avdi
PURPOSE AND METHOD EDUCATION OF INTERNATIONAL
RECOGNIZED PERSONNEL FOR WELDING ..................................................997
144. Aleksandar Majstorovi
SAFETY PARTS OF BODY TO RESPIRATION WITH BREATHING
APPARATUS IN DANGEROUS ZONE ............................................................1003
145. Stojan Simi
STATUS OF MAINTENANCE SERVICE IN COMPANIES IN REGION IN
TIMES OF RECESSION ..................................................................................1009
146. Remigiusz LABUDZKI
IDENTIFY CHARACTERISTICS OF OBJECTS IN MACHINE VISION ............1015
xvi
PREFACE
This year, the 10th Anniversary International Conference on Accomplishments in
Electrical, Mechanical Engineering and Information Technology DEMI is going to be
held. The number of countries taking part in DEMI is increasing so this year scientists
and researchers from 17 countries are going to take part. This modest jubilee, but with
a considerable number of participants, points to the fact that that DEMI Conference is
becoming an internationally recognized conference with respectable participants from
technologically developed countries.
The first DEMI Conference was held in 1998 and was organized by the Faculty of
Mechanical Engineering in Banja Luka having aimed to draw together university and
institute scientists and professionals as well as experts from a very weakened, postwar economy. Such mission and role of the DEMI Conference has remained one of its
priorities to this day. From 1998-2003, the DEMI Conference was being held every
year, and after that time, it started to be held every two years. The DEMI became a
traditional conference and it took over a significant but demanding role of gathering
researchers and scientists as well as economy experts with the aim to foster faster
implementation of contemporary research and new technologies in production
processes in order to ensure better competitiveness of our industry. Definitely, this task
has not been an easy one to achieve.
th
This 10 anniversary DEMI Conference has seen a significant progress. For the first
time, the official language of the Conference is English. The number of papers to be
presented at the Conference is 145 from 17 countries (former Yugoslav countries,
Austria, Germany, Switzerland, Italy, Norway, Poland, Romania, Bulgaria, Turkey,
Australia, New Zealand). The previous DEMI 2009 Conference brought together
researchers and scientists from 11 countries which indicates a greater interest in this
conference. The Conference activities will be realized in six sections which encompass
the following thematic fields: Production Technologies and Engineering, Mechanics
and Design, Traffic Means, Thermotechnique and Energetics, Maintenance of
Technical Systems and Mechatronics. Key note lectures will be delivered by scientists
from universities from Switzerland, Norway, Serbia and Croatia who are involved in
researches of the highest scientific level in the fields of: mechatronics, energetics,
modern approaches to machine structure and product design, the ecology of
transportation and global warming, paradigms of contemporary production and a
companys adjustment to the current market requirements, etc.
Therefore, we hope that the papers to be presented will contribute in considering and
reflection on the present situation of research in mechanical and electrical engineering
as well as information and communications technologies in the regional conditions,
enable comparison with developed European countries and offer new models for faster
implementation of contemporary research as to encourage production in the country.
As it is the most important scientific and professional conference in the fields of
mechanical engineering, information and communications technologies in our country,
this is expected from the DEMI Conference with good reason.
xvii
On behalf of the Organizational Committee of the DEMI 2011Conference, I would like

to thank all authors, members of the review team as well as institutions, companies
and individuals who contributed to the realization of the Conference program.
The International Conference on Accomplishments in Electrical and Mechanical
Engineering and Information Technology DEMI 2011 will be held at the Faculty of
Mechanical Engineering, University of Banja Luka. We are looking forward to
welcoming you as our dear guests. Welcome to the 10th anniversary DEMI 2011
Conference.
Banja Luka, 12 May 2011
President of the Organizational Committee

of the 10th International DEMI 2011 Conference
Gordana Globoki-Laki, PhD. Associate Professor
xviii
KEYNOTE LECTURES

SUSTAINABLE INNOVATION AND INNOVATION FOR

SUSTAINABILITY:
THE EVOLUTION AND INVOLUTION OF PRODUCTION PARADIGMS
Claudio R. Bor1
Summary: The consumer market is continuosly evolving requiring new products with
more functions, better design, more personalized, and lately, more sustainable. The
evolution of products has been followed closely by the evolution of the means to
manufacturing them, to assemble them and to delivering them to the market. Studying
the evolution of production, it is possible to see that different paradigms have appeared
from pre-industrialization, to industrialization, to mass production and then mass
customization. Innovation has always been the driver of the evolution of products and
related process and the present keynote will present an overview of the past evolution
and the future trends. In particular it will be shown how important is that innovation is
sustainable in time for a company to be successfull and how innovation is a key for the
future sustainable production paradigm.
Key words: Production paradigm, sustainble production, mass customization,
innovation.
Prof. Dr. Ing.Claudio R. Bor, Director ICIMSI-SUPSI, Manno-Lugano, Switzerland
COOPERATION MODELS BETWEEN UNIVERSITIES AND

INDUSTRY IN APPLIED RESEARCH, SWITZERLAND CASE STUDY
AND SOME PRACTICAL EXAMPLES
Giambattista Ravano1
Summary: Switzerland has a very well established economic environment of small
and medium size industries (SMI). Thanks to a positive financial situation and a longterm policy of development an interesting system of cooperation between Universities
(particularly Universities of Applied Sciences) and SMI could be established.
Some key successful factors are described:
Direct financing of industrial projects with proved business plan
Participation of main key player in strategic decision (industrial organizations,
research partners, Federal and Cantonal government)
Balanced mix of cooperation and concurrence between research partners.
Thanks to this approach a positive trend to innovation could be reached.
Three examples of applied research projects in the fields of materials for precision
mechanics, traffic systems, biotech precision systems which turned into innovation
and new business are presented.

1
Prof. Giambattista Ravano, Director Department of Innovative Technology, University of Applied Sciences
and Arts of Southern Switzerland (SUPSI), CH -- 6928 Manno
TRANSPORT ECOLOGY AND GLOBAL CLIMATE CHANGE

Radivoje Pei1, Stevan Veinovi2
Summary: Heavy critics claim that a group of Western scientists has put together the
prepositions for Kyoto protocol. The Kyoto protocol limits the production of the
following six components with anthropogenic activities: carbon dioxide, methane,
nitrous oxide, per fluorocarbons, sulphur hexafluoride and hydro fluorocarbons. The
most dangerous role is assigned to carbon dioxide, so, on that basis, the elimination of
carbon dioxide production is proclaimed as ecological success. The escalation is
continued in such a pace, that clean technologies (?), clean fuels (?) and clean
vehicles (?) and similar are defined. The controversy over global warming gets even
more complicated when you include politics, economics, greed, and the self interests
of the various governments, NGOs and companies. Astronomer Milutin Milankovic
studied changes in the orbital eccentricity, obliquity, and precession of Earth's
movements. Such changes in movement and orientation change the amount and
location of solar radiation reaching the Earth. He hypothesized that when some parts of
the cyclic variations are combined and occur at the same time, they are responsible for
major changes to the earth's climate (even ice ages). The next stage of power train
and fuel strategy involves using new high economy combustion engines that can be
run with partially renewable fuels and used worldwide. Researchers are hard at work
exploring new fuels, engines and vehicle technologies- but there are not clean cars,
clean energy or full renewable fuels. Contrary to popular belief electric vehicles dont
hold all the emission answers, until the electric power comes from coal-fired power
stations or part renewable sources (hydro, wind and sun).
Key words: global warming, IC engine, renewable fuels, transport ecology
1. INTRODUCTION
Rational and controlled vehicle use in transport and traffic is an obligation for
the future. Any activation of energy bears a burden to the environment in full amount. A
warm-up lifts light gases like oxygen and nitrogen, while steam and carbon-dioxide
choke the environment. Such are the signs of nature: there are no pure energy forms,
no clean fuels, and no clean engine, neither clean vehicle. Actual reserves and natural
1
PhD Radivoje Pei, professor, Kragujevac, Faculty of Mechanical Engineering in Kragujevac, Serbia, Email: pesicr@kg.ac.rs
2
PhD Stevan Veinovi, retired professor, Kragujevac, Faculty of Mechanical Engineering in Kragujevac,
Serbia, E-mail: vpst@kg.ac.rs
Pei R., Veinovi S.
gifts coal, petroleum and gas should be used less as power sources and more as
raw materials. The greatest contribution to ecological prolongation of life on our planet
is given by rational, economical technologies and products that carefully engage the
gifts of nature.
Political leaders gathered in Kyoto, Japan, in December 1997 to consider a
world treaty restricting anthropogenic production of greenhouse gases, chiefly carbon
dioxide (CO2). They feared that CO2 would result in anthropogenic - caused global
warming hypothetical severe increases in Earths temperatures, with disastrous
environmental consequences. During the past 10 years, many political efforts have
been made to force worldwide agreement to the Kyoto treaty [1].
There is no convincing scientific evidence that anthropogenic release of carbon
dioxide, methane, or other greenhouse gasses is causing or will, in the foreseeable
future, cause catastrophic heating of the Earth's atmosphere and disruption of the
Earth's climate. Moreover, there is substantial scientific evidence that increases in
atmospheric carbon dioxide produce many beneficial effects upon the natural plant and
animal environments of the Earth. Oregon Petition, from the Oregon Institute of
Science and Medicine, signed by over 17,000 international scientists including more
than 2000 of the world's leading climatologists, meteorologists and planetary/
atmospheric scientists [2].
Carbon Dioxide currently at 370 ppm, for it to be dangerous it would have to be
at 15,000 ppm. This could not be reached even if every fossil fuel was burned.
Thousands and thousands of studies show that higher levels of CO2 are good for
plants. Many scientists believe plants still are not getting enough CO2. Tomato farmers
using exhaust from electricity to grow their tomatoes. Vegetation looses less water
under higher CO2 levels, meaning vegetation in drought prone areas will live longer
and produce more. Rice (the most eaten food in the world) was shown to increase
mass and use less water with higher CO2 levels. Meaning the most important food in
the world highly benefits from CO2 increase.
Fig. 1 Global temperatures

8
Ecology of transport and global change earths climate
The eruption of volcano Tambora in Indonesia in 1815 caused the Year without
summer because of global cooling. Due to the destruction of crops, disease,
contamination of water, etc., tens of thousands more died in the next few following
years. In 1991 Volcano Pinatubo (Philippines) caused the entire earth to cool by 0.5 F
for over one year, fig. 1.
The misjudgment of Intergovernmental Panel on Climate Change (IPCC) and
Potsdam Institute for Climate Impact Research (PIK) are not overlooked. The 50,000member American Physical Society criticized the CO2 hypothesis as well as many
German scientists, physicists and meteorologists. Even former herald of climate
disasters are now well established that the climate is strongly influenced by nature (sun
+ ocean current) than the CO2 in the atmosphere. But it is also annoying: For about ten
years, directed the global temperature is not more to the global warming prophets, it
has ceased to rise, although more and more diligently CO2 leaves the smokestacks
and exhaust pipes of the people.
Figure 2 shows the petition of 18,000 American geologists against climate
policy USA Government.
Fig. 2 Climate disaster: Cause is the sun!
Fig. 3 Annual mean surface

temperatures in the contiguous
United States between 1880 and
2006. The slope of the least-squares
trend line for this 127-year re cord is
0.5C per century [1].
3 December 1973
9 April 2001
The slope decreasing
The slope increasing
Intermediate Trends
Fig. 4 Is it Global Cooling or Global

Warming?
9
Pei R., Veinovi S.
Surface temperatures in the United States during the past century reflect this
natural warming trend and its correlation with solar activity, as shown in fig. 3.
Compiled U.S. surface temperatures have increased about 0.5C per century, which is
consistent with other historical values of 0.4 to 0.5C per century during the recovery
from the Little Ice Age.
This temperature change is slight as compared with other natural variations.
Three intermediate trends are evident, including the decreasing trend used to justify
fears of global cooling in the 1970s. Between 1900 and 2000, on absolute scales of
solar irradiance and degrees Kelvin, solar activity and temperature increased fig. 4.
Secular Variations of the Planetary Orbits, (French: Variations Sculaires des Orbites
Plantaires, abbreviated as VSOP) allows prediction of past and future orbital parameters
with great accuracy.
is obliquity (axial tilt). e is eccentricity. is longitude of perihelion.
esin( )is the precession index, which together with obliquity, controls the seasonal cycle of
is the calculated daily-averaged insolation at the top of the atmosphere, on
insolation.
the day of the summer solstice at 65 N latitude.
Benthic forams and Vostok ice core show two distinct proxies for past global sea level and
temperature, from ocean sediment and Antarctic ice respectively. Vertical gray line is current
conditions, at 2 kilo years A.D.
Fig. 5 Past and future Milankovic cycles [3,4]
10
The Serbian scientist Milutin Milankovic is best known for developing one of the
most significant theories relating Earth motions and long-term climate change.
Milankovic dedicated his career to developing a mathematical theory of climate based
on the seasonal and latitudinal variations of solar radiation received by the Earth. Now
known as the Milankovic Theory, it states that as the Earth travels through space
around the sun, cyclical variations in three elements of Earth-sun geometry combine to
produce variations in the amount of solar energy that reaches Earth:
1. Variations in the Earth's orbital eccentricity (e) the shape of the orbit around
the sun.
2. Changes in obliquity () changes in the angle that Earth's axis makes with
the plane of Earth's orbit.
3. Precession the change in the direction of the Earth's axis of rotation, i.e., the
axis of rotation behaves like the spin axis of a top that is winding down; hence
it traces a circle on the celestial sphere over a period of time.
Together, the periods of these orbital motions have become known as
Milankovic cycles, fig. 5.
Changes in orbital eccentricity affect the Earth-sun distance. Currently, a
difference of only 3 percent (5 million kilometers) exists between closest approach
(perihelion), which occurs on or about January 3, and furthest departure (aphelion),
which occurs on or about July 4. This difference in distance amounts to about a 6
percent increase in incoming solar radiation (insolation) from July to January. The
shape of the Earths orbit changes from being elliptical (high eccentricity) to being
nearly circular (low eccentricity) in a cycle that takes between 90,000 and 100,000
years. When the orbit is highly elliptical, the amount of insolation received at perihelion
would be on the order of 20 to 30 percent greater than at aphelion, resulting in a
substantially different climate from what we experience today.
As the axial tilt (obliquity) increases, the seasonal contrast increases so that
winters are colder and summers are warmer in both hemispheres. Today, the Earth's
axis is tilted 23.5 degrees from the plane of its orbit around the sun. But this tilt
changes. During a cycle that averages about 40,000 years, the tilt of the axis varies
between 22.1 and 24.5 degrees. Because this tilt changes, the seasons as we know
them can become exaggerated. More tilt means more severe seasonswarmer
summers and colder winters; less tilt means less severe seasonscooler summers and
milder winters. It's the cool summers that are thought to allow snow and ice to last from
year-to-year in high latitudes, eventually building up into massive ice sheets. There are
positive feedbacks in the climate system as well, because an Earth covered with more
snow reflects more of the sun's energy into space, causing additional cooling.
Changes in axial precession alter the dates of perihelion and aphelion, and
therefore increase the seasonal contrast in one hemisphere and decrease the
seasonal contrast in the other hemisphere.
Using these three orbital variations, Milankovic was able to formulate a
comprehensive mathematical model that calculated latitudinal differences in insolation
and the corresponding surface temperature for 600,000 years prior to the year 1800.
He then attempted to correlate these changes with the growth and retreat of the Ice
Ages. To do this, Milankovic assumed that radiation changes in some latitudes and
seasons are more important to ice sheet growth and decay than those in others. Then,
at the suggestion of German Climatologist Vladimir Koppen, he chose summer
insolation at 65 degrees North as the most important latitude and season to model,
11
Pei R., Veinovi S.
reasoning that great ice sheets grew near this latitude and that cooler summers might
reduce summer snowmelt, leading to a positive annual snow budget and ice sheet
growth [3,4].
But, for about 50 years, Milankovic's theory was largely ignored. Then, in 1976,
a study published in the journal Science examined deep-sea sediment cores and found
that Milankovic's theory did in fact correspond to periods of climate change (Hays et al.
1976 [5]). Specifically, the authors were able to extract the record of temperature
change going back 450,000 years and found that major variations in climate were
closely associated with changes in the geometry (eccentricity, obliquity, and
precession) of Earth's orbit. Indeed, ice ages had occurred when the Earth was going
through different stages of orbital variation.
Since this study, the National Research Council of the U.S. National Academy
of Sciences has embraced the Milankovic Cycle model. ...orbital variations remain the
most thoroughly examined mechanism of climatic change on time scales of tens of
thousands of years and are by far the clearest case of a direct effect of changing
insolation on the lower atmosphere of Earth (National Research Council, 1982 [6]).
2. RENEWABLE ENERGY
Renewable Energy Energy derived from sources that are not depleted when
used, therefore their use causes little environmental impact. Examples are wind power,
hydroelectric energy and solar energy.
There are some definitions of Renewable energy resource:
Renewable energy resource is: an energy resource that is replaced rapidly by
natural processes. Some examples of renewable energy resources are sunlight,
hydropower (water falling through a dam), and wood; When you use some sunlight to
warm your back, more is made almost immediately available; Water above the dam is
continually replaced by rainfall. If you chop down a tree and burn its wood in your
campfire, it takes awhile for the forest to grow enough to replace that wood, but it will
happen within your lifetime [7]; and
Any energy resource that is naturally regenerated over a short time scale and
derived directly from the sun (such as thermal, photochemical, and photoelectric),
indirectly from the sun (such as wind, hydropower, and photosynthetic energy stored in
biomass), or from other natural movements and mechanisms of the environment (such
as geothermal and tidal energy); Renewable energy does not include energy resources
derived from fossil fuels, waste products from fossil sources, or waste products from
inorganic sources [8].
To our knowledge, there is no legal definition as to what renewable means and the meanings proffered vary quite a bit. Technically, all sources of power are
renewable, just at different rates - so the primary difference between a renewable and
non-renewable is the rate of replenishment. That, in itself, should be a flag that this
(in science terms) is a rather arbitrary and subjective definition. Who is to say what
replenishment rate is good or bad, and on what basis?
Let we consider this definition: Renewable is an energy resource that is
replaced rapidly by natural processes Non-renewable is any resource that is not
replaced in a reasonable amount of time (our lifetime, our childrens lifetime ) and is
thus considered used up and not available to us again. Such words as rapidly and
12
reasonable are subjective, relative terms - not scientific. Another pivotal aspect ignored
in these definitions is the fact that although a source (e.g., wind) may be quickly
replenishable, it uses up other resources (e.g., land) that are non-replenishable. We
will run out of suitable land for wind power sooner than we will run out of fossil fuels.
Shouldnt the entire package be assessed as a whole?
Considering the variability, inadequacy, and political nature of its current
iteration, there is some merit to just exterminate renewable from our vocabulary. But
nature abhors a vacuum, so for sound bite reasons, if we refuse to use renewable,
then we would be well-advised to come up with a good replacement. Then, our choices
are to redefine renewable so that it makes more scientific sense, or to come up with a
substitute [8]. They shared their conundrum with a group of energy experts.
Interestingly, they were unanimous in their consensus that there was no hope of
salvaging renewable. One environmentalist said: Several years ago, I came to the
conclusion that the word renewable, applied as a source of energy, was a pejorative and I treat it as such today (much as I do terms like windmills and windfarms). These
are all words bowdlerized of any positive meaning, designed by the craven to casually
separate people from the contents of their wallets. And so, in my public comments, I
always connect renewables with fraud. Rather than refine the definition, I move that we
ridicule the very concept. Instead I recommend promoting the principle of our making
decisions based on energy density, or something in that vein.
The other scientist said: When questioned on renewable, it is relatively simple
to explain the First Law of Thermodynamics concerning conservation. Energy cannot
be new, thus cannot be renewed. All we are doing is transforming one manifestation of
energy into another - and we should be doing it in a clean, non-polluting, preferably
non-carbon-based manner. This avoids (most times) the controversial subject of
potential, unquantified global warming versus thermal equilibrium - which is too long
and too complex for most listeners. The authors suggest that we should normally use
the phrases clean energy, and clean sources, because they are more accurate than
renewable [9].
3. PROS AND CONS OF BIOFULES
The production of biofuels for transport faces several challenges.
3.1 Energy balance
There is controversy over the energy balance of biofuels production. The
energy balance is the amount of energy needed over the life-cycle to produce biofuels
(input) versus the amount of energy produced (output). According to studies by
Pimentel and Patzek, it takes more energy to make ethanol than is contained in the
ethanol itself. Other studies (e.g. by the US Department of Agriculture) indicate that the
energy balance is positive. The balance also varies largely according to the crops used
and the transformation process.
3.2 Climate change reduction potential
In principle, biofuels are "carbon neutral": when they are used, no more carbon
dioxide is released than has been absorbed during the growth of the plants used to
13
Pei R., Veinovi S.
make these biofuels. Therefore replacing fossil fuels with biofuels for transport could
help in the fight against climate change, fig. 6.
Fig. 6 Biofuels are carbon neutral

But other studies, including a May 2007 report [10] by the United Nations
Energy division, contest this conclusion, saying that the use of biofuels could actually
increase greenhouse gas emissions as land would be converted from forests, wetland
and reserves for conservation to grow more corn or soya beans. The report notes that
with respect to reducing greenhouse-gas emissions, biofuels would be more
appropriately used for combined heat and power production rather than for transport.
3.3 Land Use
Using agricultural land to grow bio-energy crops would compete with the use of
land for food and animal feed production, driving up the prices of commodities like
cereals. According to the European Environment Agency (EEA), reaching the
initial 5.75% target of the biofuels directive would already require biofuel crops to
take over between 4% and 13% of the total agriculture area of the EU-25.
Nevertheless a July 2007 study by the Commission's DG Agriculture foresees
that reaching the new 10% target for biofuels in transport by 2020 would not overly
stretch the (EU's) land availability, requiring a relatively modest 15% of arable land,
which it claims could be largely covered by set aside land, currently reserved under the
Common Agricultural Policy in order to limit excessive production by farmers [11].
European NGOs say palm oil producers in Indonesia and Malaysia are damaging the
environment by planting crops in natural forests and in the middle of protected animal
habitats, and accuse the industry of illegally logging rainforests and violating the rights
of local communities.
14
3.4 Environment and sustainability

Energy crops generally require more fertilizers and pesticides to grow. They
also require more water, draining the earths already scarce resources. Whats more,
biodiversity loss - especially in developing countries seeking to enter this growing
market - is an important risk as forests and grasslands are cleared to plant the vast
quantities of crops needed to make a significant dent in the use of oil in transport.
Calls for binding sustainability criteria to be introduced in laws promoting
increased biofuel use therefore emerged from all sides. In its proposed Renewables
Directive of 23 January 2008 [12], the Commission proposed to introduce certain
standards, including an obligation for biofuels counting towards the 10% target to
deliver life-cycle CO2 savings of at least 35% compared to fossil fuels and a ban on
biofuels planted in protected areas, forests, wetlands and highly biodiverse
grasslands.
But Members of the European Parliament (MEPs) insisted on tougher
conditions. In September 2008 [13], the Industry and Energy Committee backed a
report demanding that biofuels offer at least 45% carbon emission savings compared
to fossil fuels a figure that would rise to 60% in 2015. They also insisted that
additional social and environmental criteria be included to protect natural resources
from both direct and indirect land use changes and to guarantee respect for human
rights and adequate working conditions in biofuel plants, especially in developing
nations.
In an attempt to reach a compromise between the 27 member states on the
issue, a special ad-hoc working group was set up at the end of February 2008, with the
aim of drafting "core criteria" for biofuels (EurActiv 01/04/08). After months of infighting, EU ambassadors appeared to have reached a consensus in September.
The EU's Renewables Directive, adopted on 26 March 2009, initially requires a
35% CO2 saving, which will then be scaled up to "at least 50% in 2017 and 60% in new
installations thereafter. It stipulates that biofuels and bioliquids taken into account in
the 10% target must not be produced from raw materials from land with "high
biodiversity value", land that has a high carbon stock, or peat lands.
The final life-cycle CO2 reduction requirement will be crucial for the biofuel
industry. Indeed, typically, biodiesel made from European-grown rapeseed results in a
greenhouse gas saving of 44% while the typical figure for ethanol made from EU sugar
beet is 48% [13].
When using fossil fuels we can assume that on the part of the land, that is
planed to grow crops for biofuel, we could plant trees. In this case the trees will be able
to absorb a larger amount of CO2 than the crops for biofuels for the equal amount
energy in the fuels. Thus, the statement that biofuels have a lower CO2 emission than
fossil fuels is questionable.
3.5 Biofuel costs
Biofuels are more expensive than traditional fossil fuels. Therefore tax
exemptions are needed to make them competitive. The second generation biofuels
promise to be cheaper but are still under development. In some countries like Brazil,
biofuels can be produced at cheaper prices.
15
Pei R., Veinovi S.
4.
TRANSPORT ECOLOGY
The term transport ecology or ecology in transport implies a complex balancing

of the impact of every anthropogenic activity on the environment according to the laws
of nature! Unilaterally reduction emissions only one component can not to declare as
environmentally, especially when it leads to even greater negative consequences on
the other side.
An initial environmental requirement is ordered refineries to reduce the amount
of lead and sulphur in gasoline. Products of combustion of such fuels have much more
toxic components, heavy metals and Particulate Matter (PM).
Table 1 presents the international drive cancerous categories of fuel, lubricants
and bitumen.
Table 1 Carcinogenic category (IRAC- International Research Agency for Cancer)
Group
Degree of
Carcinogenic
Fuel and lubricants
Benzene,
lubricants, bitumen
2A
Probable
Gasoline
2B
Possible
Diesel, heating oil
3
Not classified
MTBE, alcohols
All of our refineries used metal additives in the engine fuel in order to increase
of the octane number, fig. 7! There is insisting on the separation of gasoline at the
leaded and lead-free gasoline in our regulations! Someone has forgotten to bring in
the legislation for the prohibition of all metallic additives?! (So says the international
standards and recommendations!). Metal additives, from motor fuels, are leading to
pollution: engines, catalysts, the people, all living creatures, all plants, the land, water,
and air!
1
Proven
Fig. 7 World companies offer other metal additives

The worst feature of particle emissions (largely as a result of metals in fuels
and additives) is to stimulate cancerous diseases.
What is the specific of PM emissions from conventional (and perhaps modern)
Diesel engines? The conventional diesel engines emit large amounts of particles
during starting, cold engine operation at full load and when the post-injection is occur.
16
Particles from diesel engines are dangerous because they behave like hovering
objects with large surfaces. Nevertheless they belong the group of hygroscopically
substances that are beautiful and easy can to penetrate to all parts of the lungs
because they are less (PM2) from human bronchioles (PM5 = 0.006 mm).
Sources of PM emissions in road transport are listed in Table 2.
Table 2 Sources of PM emissions in road transport
Sources of PM
Commercial vehicles with diesel engine
Passenger cars without catalyst
Truck tires
Road surface
PM10, mg/km
380 - 1 000
3 - 900
400 3 000
375 - 11 700
It is absurd to make partial and unrealistic regulations on toxic, PM and CO2

emission quantities, without thorough environmental assessment of the automobile and
oil industries, transport and traffic in general. Fig. 8 shows a simultaneous decrease in
emission of toxic components with advanced catalytic technologies.
Also visible is the increase in the amount of carbon dioxide or in other words,
the fuel consumption. In passenger cars the new ecological regulations result in
increased fuel consumption by about 10% and in the trucks up to 15%. Just such a
conclusion confirmed by fig. 9.
Fig. 8 Unrealized European Forecasts

by 2012 (According to the auto-oil"
program)
Fig. 9 Regulations on reducing toxic

emissions is regularly followed by the
increase in fuel consumption
Just what to expect from modern models of passenger cars with air power
performance over 200 HP and an electronic limited speed (usually 250 kilometres per
hour)? They are regularly equipped with multi-degree mechanical (six to eight degrees)
or automatic transmission, a lot of (useful and useless) electronics and electronic
stability control for wanton entry into curve on the road?
Statistical processing of roll-over vehicle after the collision shows that the SUV
- Sport utility vehicle (because of the high centre of gravity) and speed of vehicles
prone to roll-over and at such time the typical security measures are insufficient.
Braking power must be nearly ten times the engine power. Therefore the wheels is
17
Pei R., Veinovi S.
scratching and scraping the surface of road plucking the pieces of asphalt and rubber.
The largest emissions of PM is from wearing asphalt road surface, tires and all
frictional surface such as brakes and clutches, fig. 10a.
Fig. 10a The high-speed, acceleration and

braking tires fall apart
Fig. 10b EU directives: the road surface asphalt - contain carcinogenic components
Launched particle were dispersing in the surrounding air which breathes all
living organisms later, and all particles contain carcinogenic ingredients, fig. 10b.
1- fuel pump, 2- pressure sensor, 3- a tube for fuel under pressure, 4- pressure regulator,
5- injectors, 6- gasoline tank, 7- LPG tank, 8- ECU for mixing and selecting of fuels
Fig. 11 New motor technology enables the use of a combination of fuel as a

substitute for additives
The large tankers are already used the crude oil which they are transporting as
fuel for their engines. All army in the world in the project tasks prescribe multi-fuel
power for combat vehicles. Aviation fuel is not shared according to value of octane
and/or cetane but they shared according to caloric value of fuel which determines the
18
action radius (Range!). Strategic Development Goals reveal that the future for road
transport -and not only for him- in vehicles which use ecological fuel. The current oil
refinery with its irrational consumption of the oil and the high pollution will accelerate
the disappearance of fossil energy reserves [14,15].
Figure 11 shows a demonstration of modern engine technologies. Liquefied
petroleum gas (LPG) has over 100 octane value and therefore LPG has a role of
additives. An analogue picture we can apply to a combination of diesel fuel and
biodiesel (biodiesel have a Cetane number of 60 to 80) and again the biodiesel have
the role of additives to improve inflammability of diesel fuel.
5. CONCLUSION
A review of the research literature concerning the environmental consequences
of increased levels of atmospheric carbon dioxide leads to the conclusion that
increases during the 20th and early 21st centuries have produced no deleterious effects
upon Earths weather and climate. Increased carbon dioxide has, however, markedly
increased plant growth. Predictions of harmful climatic effects due to future increases
in hydrocarbon use and minor green house gases like CO2 do not conform to current
experimental knowledge.
Human use of coal, oil, and natural gas has not harmfully warmed the Earth,
and the extrapolation of current trends shows that it will not do so in the foreseeable
future. The CO2 produced does, however, accelerate the growth rates of plants and
also permits plants to grow in drier regions. Animal life, which depends upon plants,
also flourishes, and the diversity of plant and animal life is increased.
Human activities are producing part of the rise in CO2 in the atmosphere.
Mankind is moving the carbon in coal, oil, and natural gas from be low ground to the
atmosphere, where it is available for conversion into living things. We are living in an
increasingly lush environment of plants and animals as a result of this CO2 increase.
Our children will therefore enjoy an Earth with far more plant and animal life than that
with which we now are blessed.
Such are the signs of nature: there are no pure energy form, no clean fuel, no
clean engine, and neither clean vehicle. Actual reserves and natural gifts coal,
petroleum and gas should be used less as power sources and more as raw
materials. The greatest contribution to ecological prolongation of life on our planet is
given by rational, economical technologies and products that carefully engage the gifts
of nature.
The term transport ecology or ecology in transport implies a complex balancing
of the impact of every anthropogenic activity on the environment according to the laws
of nature! Unilaterally reduction emissions only one component can not to declare as
environmentally, especially when it leads to even greater negative consequences on
the other side.
ACKNOWLEDGMENTS
The paper is the result of the research within the project TR 35041 financed by
the Ministry of Science and Technological Development of the Republic of Serbia.
19
Pei R., Veinovi S.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
20
Robinson A., et al., (2007.) Environmental Effects of Increased Atmospheric

Carbon Dioxide, Journal of American Physicians and Surgeons, vol.12, no. 3
pp.79-90.
Petition project, - http://www.oism.org/pproject/ , accessed on 2011-04-07
Milutin Milankovitch - http://earthobservatory.nasa.gov/Features/Milankovitch/,
accessed on 2011-04-07.
Spasova D. et al, (2007). Milutin Milankovitch a traveler through distant worlds
and times. Ministry of Environmental Protection of Republic Serbia, Belgrade.
Hays, J.D.; Imbrie, J.; Shackleton, N.J. (1976). "Variations in the Earth's Orbit:
Pacemaker of the Ice Ages". Science 194 (4270): 11211132.
National Research Council, (1982.) Solar Variability, Weather, and Climate,
Washington, D.C.: National Academy Press, 1982, p. 7.
Definition: Renewable Energy Resource - http://www.cpast.org/Articles/
fetch.adp?topicnum=11, accessed on 2011-04-07
TREIA's definition of renewable energy has been adopted by the Texas
legislature- http://www.treia.org/mc/page.do?sitePageId=49495, accessed on
2011-04-07.
"Renewable" Energy: In Search of Definition - http://masterresource.org/?p=1643,
Sustainable bioenergy: a framework for decision makers - http://esa.un.org/unenergy/pdf/susdev. Biofuels.FAO.pdf, accessed on 2010-09-107.
Biofuels: Impact on agriculture 'modest' says Commission http://www.euractiv.com/en/climate-environment/biofuels-impact-agriculturemodest-commission/article-165913, accessed on 2010-12-07.
COM(2008) 19, Directive of the European parliament and of the council on the
promotion of the use of energy from renewable sources, Brussels, 23.1.2008,
COM(2008) 19 final
Biofuel-makers denounce target downgrade - http://www.euractiv.com/en/
transport/biofuel-makers-denounce-target-downgrade/article-175298, accessed
on 2010-12-07.
Pei R., Davini A., Veinovi S., (2005.) One engine for all fuels one fuel for all
engines, Proceedings ISBN 86-80941-30-1 Paper EAEC05YU-EN01, 10th
EAEC European Automotive Congress, Belgrade 2005, pp. 1-10.
Pei R., Davini A., Veinovi S., (2008.) New engine method for biodiesel cetane
number testing, Thermal Science, vol. 12, no. 1, pp. 125-138.
DESIGN CONSTRAINTS AND ROBUST DESIGN AS THE MODERN

APPROACH TO MECHANICAL STRUCTURE DEVELOPMENT
Milosav Ognjanovi1
Summary: Actual approaches in technical systems design understand horizontal
integration of various scientific and technological knowledge and various market and
user needs in harmony with human environment. Robust design provides high level of
design results in the first attempt and design structures insensitive of service conditions
variation. Axiomatic method in connection with robust design gives possibility for
design parameters definition using various design constraints. The article has intention
to present and analysed relations between these methods and approaches. The case
study of power transmission (PT) components is carried out in order to present
efficiency of the new approaches in design parameters selection and calculation.
Reliability, vibrations and noise as design constraints in the stage of the Embodiment
design of PT components make conditions for axiomatic method application and
robustness provision. Reliability as the design constraint is defined and modelled in a
specific way suitable for this purpose and application. Also, the model of gear
vibrations and gear units noise generation is presented in a new way suitable for
applying as the design constraint. Those design constraints provide design parameters
definition in an efficient way, with high-level service quality indicators. The presented
models are based on a great volume of experimental data about service conditions
probability, gear and bearing failure probability, gear units vibration and modal
behaviour etc. Theoretical knowledge and models are insufficient yet to provide the
necessary data. The article contains presentation of testing methods and data
processing oriented to provide data necessary for the application in the suggested
approach.
1. INTRODUCTION
Tendency to achieve better design performance imposes necessity for a
Holistic approach that also considers integration in product development. Recent
research on integrated design includes system integration, requirements integration,
knowledge integration, and method and process integration [1]. System integration has
a motivation to support the concept of whole system design, as opposed to its
separate components design. Requirements integration considers vertical and
1
Prof. dr Milosav Ognjanovi, University of Belgrade, Faculty of Mechanical Engineering, Kraljice Marije 16,
11120 Belgrade, Serbia, mognjanovic@mas.bg.ac.rs
21
M. Ognjanovi
horizontal integration. Vertical integration considers development of product

requirements through different design stages similar to simultaneous (concurrent)
engineering principle. Actual trend of integration is horizontal one which considers
different areas of product development because products are becoming more multidisciplinary and their boundaries are expanding especially in the sense of the
numerous new requirements. In Fig.1 some of the possible fields of product
development are presented in the form of horizontal integration. Except from
robustness, the product needs to satisfy aesthetic, ergonomic, eco and other
requirements. Holistic approach can provide effective results. Also, separate design of
multidisciplinary products is not acceptable, and therefore mechanical and electronic
design has to be integrated. Biomimetric design [2] has an important role in searching
for new principles and solutions in the area of biological systems, in order to apply in
technical systems, but integrated in a holistic design.
Software
design
Biomimetric
design
Robust
design
Mechanical
design
Aesthetic
design
Horizontal
integration
of product
development
and design
Electronic
design
Ergonomic
design
Eco design
Fig. 1 Horizontal integration of product development and design areas

2. ROBUST DESIGN OF PT COMPONENTS
Robust design means the technical products insensitive to variation of
operating conditions and also products which are successfully designed in the first
attempt. Power transmission components operate in extremely random conditions.
Operating regimes (loads, speeds, etc), production conditions and failure processes
are random. It is known that random processes are possible to identify, present and
analyze only using the experimental approach supported by statistical indicators.
22
Design constraints and robust design as the modern appr. to mech. structures development
Power transmission operating regime for a certain machine system (vehicle, dredge,
etc.) is possible to identify by systematic measurement in service conditions. Possibility
of power transmission components (gears, bearings, sealing sets, etc.) failure can be
identified by failure probability which needs extensive laboratory tests of the listed
components. The relation between service regime and failure probability leads to
component reliability. Using these elementary reliabilities as design limitations
(constraints) for design parameters definition, the robustness of power transmission
unit is fulfilled. Similar situation occurs with the vibration and noise of power
transmission components. The level and frequency structure of vibration and noise
produced by these components can be used as design constraints for design
parameters definition and harmonization of their interaction. The relations between
operating conditions and component parameters and dynamic responses have to be
harmonized using experimental approach.
Fig. 2 Power Transmission Systems Design PTSD model

For the purpose of power transmission system design the specific procedure is
established and presented in Fig. 2 in the form of the design model. The PTSD (Power
Transmission Systems Design) model consists of four modules which respect general
design procedure of technical systems and specific needs for the design of PT
systems. The first is Solution Module which is oriented to the creation of conceptual
solutions for certain service conditions. Power transmission systems in conceptual
sense are a variation structure i.e. conceptual solutions are the result of various
combinations of gear pairs, shafts, bearings etc. In interactive communication the
module offers all possible combinations and stores in the Conceptual Base. The next
LAHP-module has the task to adapt limitations and constraints to every conceptual
23
M. Ognjanovi
solution and to every design component. Limitation Analytic Hierarchy Processing

LAHP module divides the conceptual design in sub-conceptions or function carriers,
and for that structure the processing limitations and constraints in hierarchy order.
General transmission ratio of the system is decomposed to the level of every
transmission stage. General value of the system reliability (chosen in advance) is
hierarchically decomposed to the level of possible failure. The LAHP-module is the key
part of that approach which supports reverse calculation in order to fulfilled one of the
features of design robustness i.e. to design the system with a chosen general level of
reliability. The next is Design Parameters Definition DPD module based on the
calculation of design parameters, especially dimensions, using axiomatic approach. By
observing the axiomatic rules and by inclusion of design constrains this module fulfils
the features of robustness. The last module is Priority Module whose task is to check
which design solution satisfies service limitations, such as volume, weight, efficiency,
cost etc. in a better way. The calculation of priority indicators is interactive and gives
possibility for additional corrections and adoptions.
The main feature of design robustness contains DPD module based on
axiomatic rules. These are the two axioms, the axiom of independency and the axiom
of information minimum. In Fig.3 the principle of DPD module based on axiomatic rules
is presented. Functional Requirements FR of every design component are defined by
service conditions, where the system operates. This FR is necessary to transform into
Design Parameters DP of design component. Transformation matrix [A] established
-1
by Suh [3] for the relation in Fig.3 is inverse matrix [A] . Numeric values of matrix
members define the relations in design component, which are constrained by
numerous limitations, such as safety or reliability, stiffness, standards, rules, etc.
These limitations and constraints are the result of service conditions, which is deducted
by LAHP module to the level of design component. For the purpose to present this
relation more clearly, the following example is processed.
Fig. 3 Relations in DPD module

In Fig. 4 the example is presented. The assembly of the gear, shaft and
bearing is defined by a great collection of design parameters, especially dimensions.
The calculation of dimensions is reduced to the three dimensions, gear diameter d,
gear width b and shaft diameter dsh. In this way, the axiom of information minimum is
24
fulfilled. Other dimensions are in relation with those calculated. Matrix [G] (Fig.4) is the
shape vector which defines transformation of parameters in the all shape dimensions.
This is the shape parameterization where varying of the shape parameters varies the
complete shape and dimensions. In Fig.4 are presented the two shapes of the same
assembly obtained in this way. Similar approach is incorporated in CAD tools for the
shape modelling.
Fig. 4 Example of DP minimization and variation

The structure of the matrix [A] according to Suh can be uncoupled, coupled and
decoupled. The ideal situation is with uncoupled matrix where one DP is responsible
for one FR. Real situation is more complex. In order to obtain the decoupled matrix of
-1
transformation, the matrix [A] is presented in the form of matrix [C] in the following
form.
d
d
sh

C
SE
c11 0
0 c
22
0
0
0
0
0
0
c33
0
T 1 3
T
T
c44 1
0
0
0
.(1)
The axiom of independency is conditionally fulfilled. The members of matrix [C]

and design parameters d, dsh, carrying capacity of the bearings C and seal type
indicator SE have to be calculated successively. The member c11 is in relation with
elementary reliability R1 against wear failure of gear pair, c11 f R1 . After the
calculation of gear diameter d, it is possible to calculate the shaft diameter because the
shaft loads depend of the gear diameter, i.e. c22 f d , R2 , and also of the shaft
reliability R2. The next step is calculation of bearing carrying capacity using matrix
member c33 which is in relation with the both diameters d and dsh, the total number of
25
M. Ognjanovi
bearing revolutions along the service life n6and of bearing reliability R3, i.e.
c33 f d , d sh , n6 , R3 . At the end, the matrix member c44 is in relation to the shaft
diameter dsh, to the total number of the shaft revolutions n6and to seal reliability R4, i.e.
c44 f d sh , n6 , R4 . The main feature of robustness is covered by reliability. The
values of calculated parameters have to be insensitive to service conditions varying.
For example, gear diameter calculation is in the form of
k11 3
MV
2
Hdes
c11 V Hdes Tmax
13
. (2)
Design available gear teeth flank stress V Hdes is in direct relation to reliability
against teeth flank failure R1. Unreliability Fp=1-R1 is the complex function of service
conditions probability p and of failure probability PF under these service conditions, i.e.
Fp=p PF. If in the service life the gear pair is exposed to the flank stress VH1 with n61
cycles and to stress VH2 with n62 cycles and to VH3 with n63 cycles (Fig. 5), gear wear
unreliability can be calculated as
3
Fp
pi PFi
i 1
pi
;
n6i
n6
PFi 1 e
V
Hi
Ki
Ei
.. (3)
Teeth wear (failure) probability PFi is presented by Weibulls functions, where

the parameters of those functions Ki and Ei are defined for every VHi and n6i (see Fig.5).
For this purpose, it is necessary to have the area of failure probability for a certain gear
pair, which can be obtained by extensive gear wear testing. If calculated unreliability is
close to unreliability which is defined as design constraint, the maximal stress can be
chosen as design available stress, i.e. VHdes=VH1. If it is not, it is necessary to change
the relations in Fig. 5 and to repeat the calculation. However, that definition of VHdes
includes in this way all randomness and variations of service conditions, the calculated
design parameters are insensitive to service conditions varying [4].
Fig. 5 Relation between service stress varying and gear wear probability
26
3. RELIABILITY AS DESIGN CONSTRAINT

The presented approach in the design of power transmission components
towards the new approaches as axiomatic and robust design can be successful with a
great volume of experimental data only. The new methods can succeed in the design
process [5] but remain to be carried out towards experimental data. However, the data
have to be oriented and adapted to be suitable for application in a new way. Reliability
is the term with a very wide area of applications. For the purpose of applying the new
methods and approaches, the reliability is defined in a specific way (equ.3). The main
features of reliability as design constraint are the following. Firstly, elementary reliability
is connected to possible failure, not to the component of the system. In one possible
failure a few components can participate or one component can be exposed to more
than one failure. Secondly, this elementary reliability is composed of cause
probabilities which produce failure in order to avoid possible failures by design
activities (Design Parameters Definition). In this regard, the elementary reliability is
complex probability composed of operating stress probability and failure probability
under that operating stress. Both of these probabilities are the result of extensive
experimental research. For the purpose of power transmission components design, the
experiments and the data processing contains a few sub-fields: failure probability
testing of power transmission components; then measurement and statistical analysis
of service loads and load (stress) spectrum creation which can represent the whole
service life of every component; reliability testing of gear transmission units or the
entire power transmission system.
logVHN
PF(N)
PF=0.9
PF=0.1
b)
a)
Fig. 6 Gear failure probability testing: a) back-to-back testing rig,
b) the range of gear failure probability distribution
PF(VH
logN
Testing of failure probability of gear transmission components is an extremely

extensive process of laboratory testing. It is necessary to make numerous tests and
then apply statistical data processing in order to obtain the range of failure probability
distribution. In Figure 6a back-to-back testing rig for gear wear testing is presented. In
order to obtain the field of wear probability distribution (Fig.6b), it is necessary to test
the three sets of gear pairs and define the three Weibulls distribution functions. Using
these functions and logarithmic scale, the range of failure probability distribution is
defined. This range is an important source of failure probability data for every stress
level or for every stress cycles number (see Fig.5).
27
M. Ognjanovi
Testing of gear train transmission reliability can be carried out with the kind of
objectives and tasks. Complete reliability test of this system can be extremely
extensive and it is very difficult to accelerate this test. In order to accelerate and to
make efficient experimental approach, a combination of calculation and experiments
can provide good results in a shorter time. Reliability model of gear drives based on
the results of failure probabilities and load spectrums of gear drive components
provides possibility to calculate reliability for any of service conditions. Reliability tests
in this sense are directed to the check of calculated results for chosen conditions. In
Fig.7 the testing of car gearboxes reliabilities is presented. For this purpose back-toback system is used, which provides a long-time testing under full load and speed of
rotation with a small consumption of energy. Testing is carried out according to load
spectrum obtained by service conditions analysis
Fig. 7 Reliability testing of car gearboxes in back-to-back system

4. VIBRATION AND NOISE AS DESIGN CONSTRAINTS
In mechanical systems (especially in gear transmission systems), there exist
vibrations caused by fluctuating forces (forced vibrations) and natural vibrations which
are the result of relation between design parameters and disturbance processes in the
system. Besides, in these systems there is one specific kind of vibrations caused by
successive teeth impacts. The vibration caused by impact becomes free damped after
that impact, and attenuates (disappears) in a short time by internal damping. The gear
meshing process is a very good example in this sense (Fig. 8a). Elastic deformations
of previous teeth in mesh replace position of first teeth contact point from position A
into position A. Besides that, the position A is out of contact line, the teeth begin to
mesh with impact. Every impact generates free vibration with the natural frequency of
the system fn. This vibration attenuates in a short time by internal damping of the
material and by friction. The next impacts which are repeated with the frequency f (Fig.
8b) restore free natural vibrations again and again. These vibrations can be called
restorable free vibrations. Figure 8b shows the measured results of these vibrations for
the gear speed of rotation 250 rpm, the teeth number z=41, i.e. the teeth mesh
(impact) frequency f=nz/60=170Hz, and the time between the impacts 1/f=0.006
seconds. The natural frequency of the gear system is fn=2200Hz and the time period of
natural vibration is 1/fn=0.00045 seconds. The maximal amplitude level for this
example is R1 | 4g (g-earth acceleration). The impact force, Fc=vc(cme)0.5, is
proportional to the collision velocity vc, the teeth stiffness c and the mass me which is
equivalent to the reduced rotating masses mr.
28
1/fn
R1
a)
b)
Fig. 8 Restorable angular free vibrations: a) individual teeth impact and free vibrations,
b) time function of restorable free vibrations for low rotation speeds
Figure 9a presents the results of measured gear vibrations. The total level of
vibrations increases until the resonance is (9000rpm-Fig.9a). In the supercritical teeth
mesh frequency range, the vibration level slightly increases or fluctuates close to the
same level. In the case of forced vibrations the level of vibrations decreases (line 3 Fig.
9b). It can be explained by the domination of restorable free vibrations (Fig. 8).
b)
c)
a)
d)
Fig. 9 Restorable free vibrations caused by gear meshing: a) measured results,

b) comparing of response curves of forced (3) and restorable (2) vibrations,
c) frequency spectrum for critical mesh-frequency,
d) frequency spectrum in supercritical mesh-frequency range [6]
29
M. Ognjanovi
Evidence for the specific nature of gear vibrations is frequency spectrums

obtained by the FFT analyses (Fig 9c and d). The first one (Fig.9c) shows resonant
vibrations, when the teeth mesh frequency f is equal to the natural frequency of the
gear pair, fn. This natural frequency responds with a natural vibration level close to 60g.
For a higher mesh frequency f>>fn (supercritical mesh frequency range) (Fig.9d), the
frequency spectrum is dominated by a set of natural frequencies fni with significantly
lower responses (less than 9g). One can notice that there is no response for the tested
mesh frequency f=17066Hz, but only for natural frequencies. With the variation of
mesh frequency f, the set of responding natural frequencies fni varies, and their
response levels change. This shows that the modal structure of the system is not
stable and that the dissipation of disturbance energy in this domain is significantly
higher. That is the reason why the total vibration level for f>>fn (Fig. 9a) varies.
a)
b)
c)
Fig. 10 Testing rigs for gear vibration measurement and analysis: a) driving subsystem with speed and load varying, b) measurement of angular vibration until to 6000
rpm, c) measurement of linear vibration until to 40000 rpm.
In Figure 10 are presented testing rigs with close power circulation (back-toback system) used for gear vibration testing. Testing rigs are prepared for variation of
gear design parameters, load (torque), speed of rotation until it is 40,000 rpm, and gear
vibration (linear and angular) and noise measurement and frequency analysis.
Vibrations of the system are the result of interaction between disturbance in the
system and sensitivity (response) of the system to disturbance. The interaction is a
complex process and the results cannot be predictable with the exact value. Numeric
analysis using the FEM and modal testing provides results for the purpose of the
mentioned relation prediction. In Fig.11 an example of gear drive housing analysis is
presented. Vibration energy and noise emitted by housing walls are a part of
disturbance energy caused by teeth impacts, load and stiffness fluctuation etc. Energy
absorbed into the elastic structure of machine parts (for example, gears) is transmitted
30
through the structure and a small part is emitted into the surroundings in the form of
vibration energy and noise. The process of absorbing and transmitting this energy
through the elastic structure of a machine system is interesting to be treated on the
presented example of gear transmission covers (housings). The mechanisms of
transmission, attenuation, amplifying, frequency modulation etc of elastic waves and
emission into the surroundings are analyzed using numerical and experimental
methods. The cover of the system (gear transmission housing) is the most effective in
this process. This approach contains the analysis of disturbance energy transmission
through machine parts, attenuation in the contacts, losing inside the parts or amplifying
by excitation of natural vibrations. Modal analysis and modal testing of the gear
transmission housing have enabled identification and proving of these processes.
Numerical and experimental impact energy transmission through housing enabled to
identify the conditions for a certain modal shape excitation. Also, these results enabled
to define mechanism of frequency modulation of emitted noise and vibrations in the
surroundings. The measurement of the noise emitted in the surroundings was also
carried out to prove the starting hypothesis and developed mechanism of disturbance
transmission, and to define the value of energy transmission ratio.
a)
b)
Fig. 11 Example of sensitivity analysis of gear unit housing:
a) modal analysis using FEM, b) modal testing [7]
Analytic hierarchy processing (LAHP module Fig. 2) in the case of vibration
and noise limitations (constraints) is the deduction of the total vibration and noise level
31
M. Ognjanovi
of the system to the vibration levels of the system components. The process of
deduction can be characterised by a number of the following vibration features.
x Numerous disturbance sources and vibration and noise sources are in a very
complex relation. The difference between the total vibration or noise level and an
individual source with maximal effect can be small.
x The level of vibration in the sub-critical frequency range is proportional to the ratio
between disturbance frequency and natural frequency.
x The level of vibrations in the resonant frequency range is proportional to damping
characteristics of the system.
x The level of vibrations in the supercritical frequency range is proportional to
disturbance intensity and frequency.
According to the mentioned problem, the Limitation Analytic Hierarchy
Processing - LAHP consists of disturbance frequency calculation and vibration level
limits definition based on the type of the system. Vibration limits of the system subassemblies are possible to define by varying design parameters in the DPD module.
The vibration level of design solutions can be reduced by a set of design
parameters corrections which have to be harmonised and lead to vibration level
minimisation. The Campbells diagram (Fig.12) shows the relation between the
disturbance frequencies kf (teeth mesh frequency f) and the natural frequencies of the
system fni. The high level of vibration is equality of these frequencies and is marked in
Figure 12 by signed points. With the aim to minimise the level of vibration, two groups
of design rules are possible to define. The first of them contains variations of design
parameters which can vary the disturbance (service) frequencies kf and reduce
disturbance forces intensity. The second group of design actions contains variations of
design parameters which can vary the natural frequencies fni with the aim to avoid
equality with the service frequencies. For both groups there exist numerous research
results which are necessary to apply. When this passive approach to vibration and
noise reduction is not enough, the active approach can be the next step. By generating
opposite vibration and noise, absolutely steady and silent gear transmission units are
possible to obtain.
n
f
f
n5
2f
3f
3f
2f
fn5
n4
fn4
n3
fn3
n2
fn2
fn1
n1
f
Fig. 12 Campbells diagram of relation between disturbance and natural
frequencies in the system
32
5. NEW MATERIALS IN MODERN DESIGN

Choice of material in design process is the top decision which is result of
impact of material characteristics and necessary features of mechanical components.
Designers experience and knowledge about traditional materials give the main support
for these decisions. Contradictions and contradictory effects in choice of materials very
often make them not fully successful. It is no material with characteristics which can
cover necessary features of components. Additionally some of characteristics make
negative effects. Creation of materials with certain necessary characteristics is the task
of designers and material science. The common name for materials of this type of
materials is composite materials. Development of these materials was started for
extreme specific applications. Now it is relative wide area of various materials which
covering many areas of traditional materials application. For example traditional
material for gears is the group of various steels. These materials have good volume
and surface strength and durability but support vibration and noise generation. Plastic
gears have no enough level of strength and durability. Special way of gear production
of composite material can satisfy of all necessary features. Special gear systems in
automotive and similar areas for specific kind of motion (Fig. 13a) have to be very
silent and reliable. The only good solution is special composite material.
a)
b)
Fig. 13 The new design and new kind of material production
Technology of production of new (composite) materials is various, starting from

casting similar to plastic until to very modern rapid prototyping (production)
technologies. Materialisation of 3D CAD models using 3D printers (Fig.13b) is
technology in expansion. This technology start with usage of plastic materials, continue
by using various kinds of powder including metal powder and sintering, until to usage
of malty material printing in order to produce composites. Rapid production of matrices
for composite casting or similar production is good implementation of 3D printing in the
trend of technical system components development with certain characteristics of
material.
Biomimetrics (Bionics) is the field of science oriented to transforming of
biological systems into technical. In the area of materials (composites) this direction
and methodology is very perspective in the future. Exist wide area of various kind of
natural composites in flora as wood, bamboos etc., at animals, insects etc.
33
M. Ognjanovi
6. CONCLUSION
Modern approach to mechanical structure development very often is the
segment of horizontal integrated multidisciplinary work. Power transmission
components in the future have to be of high quality integrated with electronic control
systems, software and intelligent systems, etc. Design methodology has to be efficient
and provide high level of quality indicators. Robust and axiomatic design using
reliability, vibration and noise as design constraints are guaranty for success.
1. The robust design can be provided by precise definition of constraints and by these
constraints application in the design process. For this purpose, the design process
is defined as a set of four modules: Solution module, LAHP module, DPD module
and Priority Rating module.
2. Reliability is defined in a specific way as a complex probability of service stress
probability and failure probability of machine parts exposed to this stress. The
procedure of design parameters definition (DPD) using elementary reliability as a
constraint is established.
3. The design for vibration and noise offers an approach to design parameters
harmonisation aiming at vibration and noise level reduction. For the purpose to
vibration and noise reduction, the mechanism of vibration and noise generation in
power transmission systems is presented.
4. Tendencies of the new materials development are also presented.
ACKNOWLEDGEMENT
This article is a contribution to the Ministry of Science and Education of Serbia
funded projects TR 035006.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
34
Liu, S., Boyle, I.M., (2009). Engineering design: perspectives, challenges, and
recent advances, Journal of Engineering Design, Vol. 20, no.1, p. 719
Lenau, T, (2009). Biomimetrics as a design methodology Possibilities and
challenges, CD Proceedings of International Conference on Engineering Design
ICED09, Stanford 2009, p. 5.121-5.132.
Armen Z, James K, Lusine, B. (2007). Modeling and analysis of system
robustness, Journal of Engineering Design Vol.18, p. 243-263.
Ognjanovic M., Benur M. (2011). Experimental Research for Robust Design of
Power Transmission Components, Meccanica, DOI: 10.1007/s11012-010-9331-y
Liang, J., Mourelatos, Z.P., Nikolaidis, E. (2007). A Single-loop Approach for
System Reliability-based Design Optimisation, - Journal of Mechanical Design,
Vol.129, p.1215-1224
Ognjanovic M., Agemi F. (2010). Gear vibrations in supercritical mesh-frequency
range caused by teeth impacts, Strojniski vestnik Journal of Mechanical
Engineering, Vol. 56, no.10, p. 653-662.
Ciric-Kostic, S., Ognjanovic, M. (2007). The Noise of Gear Transmission Units
and the Role of Gearbox Walls. FME Transactions Vol. 35, p. 105-112.

RENEWABLE ENERGY AS A DRIVER OF ECONOMIC GROWTH
Prof. dr. sc. Neven Dui1
Summary: The new European energy strategy, so called energy-climate package,
sets up ambitious goals of having 20% renewable energy in gross energy demand,
reduce carbon dioxide emissions by 20% and improve energy efficiency by 20% by
2020. One of the goals of this is to use the necessary need to reduce the influence on
environment and increase the security of energy supply for opening new and better
quality jobs, regional development and reindustrialization of the European economy.
This is also a potential for new development of Croatian, and also southeast European
economies, based on future inevitable investments that may be used to spearhead the
local economy growth. Countries should approach this in an integral way, best for their
individual economies. It will be necessary to lay down financial support mechanisms
that will create both, supply and demand for new technologies. On one side, financial
mechanisms that open the way to increased investment in renewables and energy
efficiency are being put in practice, like feed-in tariffs, but others are needed to support
innovative development of new technologies which create jobs in industry and
academia. New financial mechanisms, often used in countries that successfully gain
from increasing both supply and demand of renewables, are suggested. They would
help finance the development of new products, by bringing together the research
capacities of academia and industry, while in the same time avoiding extra budgetary
strains. Most of renewable energy technology, especially those that will stem from
application of Buildings directive, are quite simple and can be produced locally.
Meanwhile, those technologies have high development costs, due to need to
constantly innovate, which can then be supported through the suggested financial
mechanism.

1
Prof. dr. sc. Neven Dui, Faculty of Mechanical Engineering and Naval Architecture (FSB) University of
Zagreb
35
Neven Dui
Renewable energy as a driver of

economic growth
DEMI 2011
Banja Luka, May 26/28, 2011
Prof.dr.sc. Neven Dui, University of Zagreb
Faculty of Mechanical Engineering and Naval Architecture
(UNIZAG FSB)
FSB 2011.
EU energy context
Security of energy supply

Import dependence from 50% to 70% by 2030
Employment and regional development

policies
Deindustrialization and trade liberalization
Boosting growth and jobs by meeting our climate
change commitments
Mitigation of global warming

Environmental protection
Sustainable development
FSB 2011.
Energy - climate package 2008
20-20-20 till 2020

20% RES
10% biofuels
Energy neutral buildings
from 2018
20% CO2 emission

reduction
20% more energy
efficiency
Energy neutral buildings
from 2018
FSB 2011.
36
Renewable Energy as a Driver of Economic Growth
The new renewables Directive

1. Sets mandatory national targets for renewable energy
shares, including 10% biofuels share, in 2020 (Articles 3
and 5)
2. Requires national action plans (Article 4)
3. Standardises guarantees of origin (certifying the
renewable origin of electricity or heat) and enables the
transfer of these to provide flexibility to Member States
(Articles 6, 7, 8, 9 and 10)
4. Requires reduction of administrative and regulatory
barriers to the growth of renewable energy (Article 12),
improvements in provision of information and training
(Article 13) and improves renewables access to the
electricity grid (Article 14)
5. Creates a sustainability regime for biofuels (Articles
15-18)
FSB 2011.
The new renewables Directive
FSB 2011.
Nova Direktiva za obnovljive izvore energije
Obaveze zemalja Zapadnog Balkana u postizanju udjela

obnovljivih izvora energije u ukupnoj potronji do 2020
FSB 2011.
37
Neven Dui
New RES Directive
FSB 2011.
New RES Directive

Plin
Vjetar
Neto promjena instaliranih

kapaciteta za proizvodnju
elektrine energije u EU, 20002010 [MW], EWEA
PV
FSB 2011.
RES-e in EU, 1996-2006
FSB 2011.
38
Production of primary energy, EU
EUROSTAT
FSB 2011.
RES Directive - reforms of administrative and regulatory barriers
simplification and streamlined procedures

planning authorities to consider renewable
energy and district heating and cooling systems
minimum levels of renewable energy in building
codes for new or refurbished buildings
promotion of energy efficient renewable energy
certification regimes for installers; mutual
recognition
priority access to the grid system for electricity
from RES
to develop grid infrastructure
to review cost sharing rules
FSB 2011.
RES Directive - Promotion of biofuels

Establishes a biofuels sustainability scheme
Biofuels not meeting the criteria will not be
eligible for target counting, obligation schemes,
tax exemptions or other support
Member States responsible for verification of
compliance however Commission can decide that
certification schemes give reliable proof of
compliance
10% biofuels by 2020
2nd generation biofuels x 2
RESe + RESH2 x 2,5
FSB 2011.
39
Neven Dui
Energy package ETS post 2012
EU wide ETS auctioning

No free emissions for power generation sector
(except cogeneration) increasing the cost of
electricity from coal
FSB 2011.
Energy package buildings

New and reconstructed buildings zero energy from 2019
Define zero energy by 2010
Quota of old buildings to be reconstructed
National Action Plans by 2011
Strict insulation regulation from 2006 for new and reconstructed
buildings (>1000m2, 2009 - >50m2)
Smart metering
Feasibility of solar heating, heat pumps, biomass heating,
cogeneration
Spain: obligation to supply 25% of heating

New package will include obligation
Phase out incandescent light bulbs by 2011

Passive buildings ...
More efficient appliances, boilers, etc.
FSB 2011.
How to create RES market?

Demand
Supply
Feed-in tariff
Obligation
FSB 2011.
40
RES-E support schemes in EU
FSB 2011.
Financial Support Mechanisms

Feed-in tariff systems are characterized by a specific price normally set for a
period of several years, which must be paid by electricity companies, usually
distributors, to domestic producers of green electricity. A variant of the feed-in tariff
scheme is the premium mechanism.
Green certificate system RES-E is sold at conventional power-market prices. In
order to finance the additional cost, all or some consumers are obliged to
purchase a certain number of green certificates from RES-E producers.
Tendering procedure the state places a series of tenders for the supply of RES-E
which is then supplied on a contract basis at the price resulting from the tender.
Tax incentives are used as an additional policy tool. Tax incentives may be a tax
credit or a cash payment or an exemption from tax obligations or low VAT.
Investment incentives: A common investment subsidy is a grant for the
installation of capacity.
FSB 2011.
Effectiveness of a promotion policy for

onshore wind
Source: OPTRES, 2007

FSB 2011.
41
Neven Dui
Evaluation of different FITs design options
(Source: ISI 2008)

FSB 2011.
International Feed-in Cooperation

Web-Site:
Research
Events
Legislative documents
Links
http://www.feed-in-cooperation.org/
FSB 2011.
Feed-in tariffs (FITs)

Already applied in over 60 countries, states and provinces,
a highly effective tool for boosting the viability, and hence value, of
the renewables industry,
FITs have been empirically proven to generate the fastest, lowestcost deployment of renewable energy.
European Commission, SEC(2008) 57, update on renewable
energy policies in the European Union (EU): well-adapted feed in
tariff regimes are generally the most efficient and effective support
schemes for promoting renewable electricity
85% of all new wind capacity and nearly 100% of the new PV
capacity since 1997 were installed in countries using FIT
FSB 2011.
42
Feed-in Tariffs (FITs)
Fixed
Premium
Net meetring
Progression of the remuneration level for

wind onshore within the market option
FSB 2011.
Feed-in tariff design

One of the most important aspects of a feed-in
tariff design is the determination of the tariff
level and the duration of support.
One possibility is to set the tariff level based
on the electricity generation costs from
renewable energy sources. Alternatively,
the support level of RES-E can be based on
the avoided external costs induced by RES
generation.
FSB 2011.
RES Electricity Cost Supply Curves

ADEG project
calculation for 2010
(Croatia)
FSB 2011.
43
Neven Dui
External cost
External costs, c/kWh
Global warming
Noise
Materials
Crops
Occup. Health
Public health
5
4
3
Source DG TREN
2
1
0
Coal
Oil
Gas
Nuclear
Wind
Biomass
FSB 2011.
Feed-in tariff design

Should provide technology-specific tariff levels.
Investment for the plant
Other costs related to the project, such as expenses
for licensing procedures
Operation and maintenance (O&M) costs
Fuel costs (in the case of biomass and biogas)
Inflation
Interest rates for the invested capital
Profit margins for investors
Rejection or curtailment
FSB 2011.
FITs design
Stepped tariff design: different levels of
remuneration are paid for electricity of the same
RES-E technology.
flat tariff design : the opposite of a stepped tariff
design
44
FSB 2011.
28.
Guarantee of origin
guarantee of origin means an
electronic document which has the sole
function of providing proof to a final
customer that a given share or quantity
of energy was produced from renewable
sources as required by Article 3(6) of
Directive 2003/54/EC;
28
FSB 2011.
Conclusions
RES-E support requires continuity and long term

investment policy
Technology-specific tariff levels should be applied
Energy policy should provide mechanisms to ensure
the penetration and to improve the integration of
RES-E into the grid
A premium tariff option can be applied to increase
market orientation
Tariff degression provides incentives for cost
reductions
FSB 2011.
Conclusions
Stepped tariffs may be applied to reflect different
power generation costs within the same technology:
May lead to higher administrative complexity and to reduced
transparency
If not properly designed overall efficiency of the system may be
decreased
Extra premiums may help to reach policy goals

Monitoring of success of financial schemes and RES
policies should be applied.
Revision and proposal of new tariffs every 4-5 years.
FSB 2011.
45
Neven Dui
Conclusion
An effective scheme is one that:
provides tariffs for all levels, from domestic to largescale developments,
takes account of the level of development of each
technology and benefits like (fuel savings, avoided
emissions, local jobs, etc.)
guarantees long term investment security,
is administratively simple and easy to explain in
order to ensure investors and public acceptance.
FSB 2011.
FSB 2011.
Possible programmes in Croatia

100000 solar roofs hot water 2000 jobs
100000 heat pumps heating and cooling
1200 jobs
Refurbishing buildings to current regulation in
10 years
Demand reduction 23 PJ, 10% less primary
energy in Croatia
Construction and building material industry
more than 25000 new jobs
All can be local jobs and technologies
FSB 2011.
46

Demand
Supply
Project
development
Design
Installing
Equipment
production
FSB 2011.
How to create equipment supply?

Support to innovative R&D
Another 2020 goal, 3% GDP for
innovations and R&D
Source? Emission tax?
FSB 2011.
Industry Academia partnership

Innovation Union 2020
Constant flow of innovation and
human resources is necessary for
competitive production
Supporting development is
investing in future
FSB 2011.
47
Neven Dui
Torwards 2050
Four pillars of Post carbon society
Renewable Energy
Buildings as Positive Power Plants
Energy Storage
Smart grids and Plug-in Vehicles
FSB 2011.
Denmark
SOURCE: STORIES project
Main RES: Wind, solar and biomass

Main support schemes:
1. Wind power
On- and Offshore Committees
Feed-in tariffs
Depend on the year the turbine was connected to the grid
and its age
Range 1,3-8 cent/kWh
Tendering procedure (for offshore wind turbines)

2. PVs
Subsidies withdrawn in 01.01.2002
Large scale demonstration projects carried out between
1998-2007
R&D for new types of PV cells
3. Wave energy
R&D stage, in 2003 the first grid connected wave power
plant went into operation
FSB 2011.
Finland
Main RES: hydro power and biomass
Investment Subsidies: 30-40% depending on the novelty of project
Guaranteed Access to the Grid
Tax Subsidies
Feed-in Tariff or Green Certificates
FSB 2011.
48
France
Main RES: hydro power and biomass

Feed-in tariffs
Depend on the technology
Range: 6,57 -40 cent/kWh
Duration: 15-20 years
Special tariffs for Corsica and overseas Departments
9 (PVs: 40 c/kWh + premium for building integration of 15
c/kWh
9 Geothermal: 10 c/kWh + premium for energy efficiency
between 0 and 3 c/kWh)
Tendering system for large renewable projects
Tax credit: 50%of the equipment cost for household installations
Reduction of VAT-Tax: 5,5% reduction in VAT for residential energy
equipment
FSB 2011.
Germany
Leader in wind energy, photovoltaics, solar thermal and biofuel sectors, both as
RES-E producer and manufacturer
Main core of renewable Energy Act:
Priority access for RES-E to the grid
Priority transmission and distribution
Grid operators must purchase the electricity produced from RES
Equalisation of additional costs between all grid operators and electricity
suppliers
Feed-in tariffs
Higher tariffs for geothermal, PVs and certain types of biomass
Decreased tariffs for onshore wind installations at locations with very good
yield
Annual decrease of tariffs for all technologies except small hydro plants (1%6,5%) to take account of technical development
FSB 2011.
Germany 2/2
No specific support schemes for hybrid plants
Growing interest in exploring the potential of
combined RES and energy storage systems
particularly for hydrogen production applications
First hybrid power plant (wind/biogas - 120MW
and hydrogen-500kW) installed on grid in 2008
FSB 2011.
49
Neven Dui
Spain
Worlds second larger producer of wind energy

Feed-in tariffs: fixed total price+price premiums with
recognising the environmental benefits
Project developers should choose one of the following
options:
1. Transfer electricity to the system through the transport or distribution
grid
2. Sell the electricity on the wholesale electricity market
Feed-in tariff and premium differentiate according to technology

and the age of installation
Feed-in tariffs are revised every four years
Internal rate of return of 7%
FSB 2011.
Ireland
Main support scheme:
Feed-in tariffs
Range between 5.7-7.2 cent/kWh
Adjusted taking into account annual indexation
Tax Relief
Relief capped at 50% of all capital investments (excluding land), net of
grants, on a single project
Extension of the relief until 31 December 2011
Other incentives:
Ocean Energy Strategy, Organization of an all-island market,
Establishment of Renewable Energy Development Group and Irish
Energy Research Council
FSB 2011.
Ireland
Other policy Incentives
Ocean energy strategy

Launched in 2006
Increasing the capacity for research and development,
both within academic institutions and commercial
entities developing devices
All-Island market
Renewable Energy Development Group
Focus on achievement of Irelands 2010 target for electricity
from RES
Considering policy options and support mechanisms available
to Government to stimulate increased use of biomass and CHP
Developing a strategy for the promotion of ocean energy
Irish Energy Research Council

Coordination of energy research and technological
development and innovation (RTDI) activities and deliver an
integrated approach to energy RD&D
FSB 2011.
50
Ireland 2/2
No special regulation for RES storage systems
Study on:
Implementation of a wind energy storage facility at
Sorne Hill Wind Farm
Different electricity storage technologies and their
potential to address wind energy intermittency in Ireland
In operation:
Desalination plant with load management on Inis Mean
FSB 2011.
UK
Renewable Obligation Order: Obligation on electricity
suppliers to source an increasing share of their power sales
from RES
Issue of Renewables Obligation Certificates
Several technologies eligible under RO
Penalty
Climate Change Levy: RES-E is excempted
FSB 2011.
Malta
RES market still at an early stage
Substantial solar and wind potential
Support schemes:
A fixed feed-in tariff of 46.6 /MWh for PV installations
below 3.7 kWp; and
A reduction in value-added tax on solar systems from
15% to 5%
FSB 2011.
51
Neven Dui
Cyprus
The system of support to RES includes the
following components:
Definition of the price of kWh in the market.
Provision of long-term contracts (15 years)
Definition of financial incentives. Feed-in tariff system
is implemented combined in many cases with capital
subsidization.
FSB 2011.
Conclusions
Each country adopts different measures on promoting

RES, depending on the RES type and the year of
implementation
No special legislation or policy framework for hybrid
power plants with the notable exception of Greece
FSB 2011.
Greece
Main RES: hydro power and active solar
Main support scheme:

Feed-in tariffs
Distinction between the interconnected system and the noninterconnected islands (particularly favourable tariffs for solar
power plants)
Photovoltaic Development Plant
Installation of 540MWp in the mainland
Installation of 200MWp on the islands
Geographical distribution
Plants <150 kWp are exempted from the obligation to obtain a
production license
Investment incentives
Subsidies: 30-40% of the installation cost depending on RES
technology, region and type of enterprise
FSB 2011.
52
Greece
Hybrid power plants
Law 3468 sets the fundamentals for the promotion of

hybrid power plants in the non-interconnected islands
Public consultation of a proposed Code by RAE for the
operation of hybrid power plants (July 2008)
Remuneration of hybrid power plants for:
Capacity availability
Energy supplied to the grid from the hybrid stations firm
output units or RES units
Hybrid power plants have to pay for the electricity

absorbed from the grid in order to fill their storage
systems
FSB 2011.
Greece
Hybrid power plants
Remuneration of:
Capacity Availability: The price to be paid to the producer for the availability of
the firm output units of the hybrid station cannot be less than the price paid for the
availability of the units of a new incoming conventional plant of corresponding
capacity
Energy produced from the hybrid stations firm output units, exploiting
the energy stored in the storage system, and supplied to the grid of the
island: Conducted on the basis of the mean marginal cost of the conventional peak
units operating on the island during the last year (annual mean variable cost)
Energy the RES units of the hybrid station supply directly to the Network
of the island: Priced according to the existing feed-in tariff depending on RES
technology (currently 87.42 /MWh for wind energy in non-interconnected islands)
Energy the hybrid station absorbs from the Network for fulfilling its
storage system: Conducted on the basis of the mean variable cost of the
conventional base units operating on the island during the last year (annual mean
variable cost)
Power produced in the RES units of the hybrid station and directly supplied to the
Network of the island may be compensated to the power the hybrid station absorbs
from that network for fulfilling its storage system
FSB 2011.
Greece
Hybrid power plants
Hybrid power plants

PPC has installed two hybrid power plants
Island of Kythnos: PVs+Batteries
Island of Ikaria: Wind+hydro power
RAE received applications for installing hybrid

power plants in Crete and Sikinos. Applications
are currently being reviewed
FSB 2011.
53
Neven Dui
THANK YOU FOR YOUR ATTENTION!
Neven.Duic@fsb.hr
FSB 2011.
54
MALICIOUS ECOLOGY ON VEHICLES AND TRAFFIC EXAMPLE

Dr in. Stojan Petrovi, Dr in. Boidar Nikoli, Dr in. Emil Hnatko, Dr in. Jovo
Mra, dr in. Stevan Veinovi
Summary: During the epoch of human progress and ecology there is a big interest
for malware ecology as well protection against them. Identifying the entire value stream
for each product is the first step in ecological thinking. By improvement a technical
facilities we degrease energetic and ecologic environment overload. For example,
vehicles are the cases where to positive effect of complex energy balance over the
propulsion system in them. Around 60 million vehicles are produced per year and 800
million vehicles are in traffic today in the world. The biggest power has been situated in
transport sector: up to 15 times bigger of all energy resources in the EU and up to 10
times bigger of all energy resources in Serbia.
Such huge number of vehicles has, of course, a certain negative influence on nature
and human environment. The overall pollution of environment air, water and soil
enforcement genetic bases uncontrolled changes wherever the entire world of our
Planet exists. The global conclusion is that there is no chance to change the road fleet
from fossil fuels to electric power.
55
A. MECHANICS AND DESIGN

ABOUT RECEIPTE OF VARIANTS OF TECHNICAL OBJECTS

Michail Leparov1, Georgi Dinev, Marieta Jancheva2
Summary: One of the important problems of the design process of assembly units is
receiving of many variants to select the optimal variant by specific criteria. In this work
are offered some ways for assembled unit. Every modification fulfil the conditions:
Performs the same function as the main original technical product;
Differs from the original product on characteristics of the parts (forms, number,
location) and / or connections between them.
The reports presented several ways for modifications of the article by heuristic
algorithms, and they are illustrated by examples.
Key words: variants assembly unit, change, technical object, geometrical, design,
methods for geometrical design.
1. INTRODUCTION
One of the important problems in the design process of the assembly unit (AU)
has received a number of options to choose according to specific criteria of optimal
options. Some methods for obtaining variants of TO are given in [2-7]. Under the
version of the assembled unit in this work means transformation of the product, where
the new product differs from the source technical object (TO), but performs the same
main function. Through this work offers some new ways of getting the options variants
of given AU. Every variant of the option meets the conditions:
Performs the same function as the main output product;
Differs from the original object on characteristics of the parts (forms,
number, location) and/or connections between them.
The methods are obtained by logical analysis and are represented by heuristic
algorithms i.e set of guiding steps for search of solutions.
Prof. PhD, Michail Leparov,Technical University of Sofia, Faculty of Mechanical Engineering, (e-mail:
mleparov@tu-sofia.bg)
2
Assoc. Prof. PhD Georgi Dinev, Technical University of Sofia, Faculty of Mechanical Engineering, (e-mail:
gdinev@tu-sofia.bg)
2
Asist. Prof. PhD Marieta Jancheva,Technical University of Sofia, Faculty of Mechanical Engineering, (email: mmarieta@)mail.bg)
59
2.
METHODS
2.1 Method of Repair"

Algorithm
1. Specify the problem in the source is TO.
2. It is assumed TO that it must be "repaired" (correct) because it are damaged
(literally or figuratively). Make a list of all possible failures. For this purpose, made
logical analysis, power analysis, polls "repair" firms, surveys among users and others.
3. For each damage:
3.1. Damage analyses, and taking into account what happens when it immediately,
and to bring what changes related damage components.
3.2. It is assumed that it TO begin to function only when executed operation of the
damage.
3.3. Based on the analysis addresses the problem, to some extent resembles the
situation of damage.
If necessary, carry out appropriate changes in TO.
Example 1
1. Let's solve the problem which demands a new look and additional functions, or
other realization of the TO function, depicted in Figure 1.
2. a) button 1 after pressure blocked;
b) spring 5break;
c) the forces applied to the cover 2 screws 3 are developed and damage;
d) efforts in the spring and buttons1 a cover 2 are deformed up in the middle
upwards;
e) efforts in the spring and screw the casing 4 is broken;
3.1. a) button can not move;
b) the button is not return to the starting position;
c) cover 2 to move freely up and down and prevents the operation of the facility;
d) the button 1 can do necessary working stroke;
e) spring fell in the resulting hole;
3.3. a) One possible solution is given in Figure 2; button and cover form a bayonet,
pressure and rotation of 90 projections button (fig.2b) stand outside the holes in the
cover, so the button can not return to the starting position (return is realized in his new
rotation of 90 );
b) fig.2v; object consists only of two-button detail and housing-related thread, in
rewinding button when he moved in either direction (up, down);
c) fig.2g; in rotation of the button it moves from the bottom with threaded body and
thrown up from the lower spring where as sharply circuit, after a new rotation button is
wrapped thread at the top of the bonnet on the development of the upper spring button
moves it down sharply, thus reducing the spark to turn the electric circuit;
d) fig.2d; cover was replaced by an elastic membrane that also serves as a spring;
e) figure 2, moving up is done manually, a return to the starting position, by using
spring.
60
About reseipte of variants of technical objects
Fig. 1 Original object "button
Fig. 2 New implementations of GF to TO "button
61
Example 2
1. Let's solve the problem is demand for a new look and additional functions, or other
realization of the function of TO Door.
2. a) the key is broken and so far has remained in the lock;
b) the spring that returns the lock is broken up;
c) hob of the door is removed from one hinge;
d) spring, which influence on locking plate is broken;
3.1. a) can not put a key lock and unlock on the door;
b) not influence lock on tongue;
c) the door does not fulfil its main function-closing and opening a hole in the wall;
d) the key is not influence on locking plate;
3.3. a) door only locks closing;
b) lock has two positions-the first is the traditional, which has normal function, the
second is obtained after pressing a little lock in which the tongue is locked, and thus is
suspended and open the door without a key;
c) the door can be rotated to its base and becoming part of a window, as is hole in
the wall is partially open to air penetration, while access to space is limited;
d) primary key does not affect the plate, and on the inclusion of a code device
which, after entering the correct code, unlatched the lock and allows the switch to
move the locking plate.
2.2 Method "Paradoxes"
Algorithm
1. Pinpointing the problem.
2. Determination of the action which is necessary to be realization to solve the
problem.
Note: In the presence of TS can be detained and its main functions OFi. The effect
of TO its main function, defined by a verb.
3. The action (and any OFi):
3.1. Searching opposite states, e.g. open-closed, include- exclude, move forward and
backward moves, etc.
3.2. For each pair participating in the opposite state:
3.2.1. Identify basic features. This can be done by defining these parts, e.g. through
the dictionary.
3.2.2. Incorporation is a sign of one with the sign of the second opposite state.
Combinations can be of type "adjective-noun" or of another type.
Combinations are formed properly grammatical terms.
3.2.3. Analyze the combination. Record what is understood as meaning the
combination.
3.2.4. As results of the analysis sought to decision the problem.
Example
1. Search a new realization of the closure of TO " Reservoir ".
2. Open.
3.1. Close-open (closed-open).
62
About reseipte of variants of technical objects
3.2.1. According to [1]:

(Closed) - which is limited, there is no free access;
(Open) - in which is free to go;
- Restricted, proprietary, unavailable;
- Free, passable.
3.2.2. Some possible phrases are:
- Limited freedom (free limitations);
- Limited passable (passable limitations);
- Unavailable freedom (free unavailability );
- Inaccessible passable (passable inaccessibility).
3.2.3. One possible subjective interpretation is:
(Free limitation) the movement of the cover is only one coordinate axis;
(Walked limitation) obstacles, but it is possible to achieve "mechanism" to open;
(Free inaccessibility) is insufficient space to open, it is destructive;
(Walked unavailability) can be done "path" to inaccessibility;
3.2.4. Some possible solutions for closing the tank are given in Fig.3 by letter symbols
in the figures correspond to letter names of the phrases from t.3.2.3. Fig.3 is a baseline
figure of fig.3b-cover slides along grooves on fig.3v-fold is somewhat cut into 4 parts
reservoir can bend to fig.3g-shell is attached to the environmental technical
environment, and fig.3d-fitting cover to the body going through braces.
Fig. 3 Some solutions to close of TO "Tank

63
3.
CONCLUSION
In the work are considered the following methods: a) "Repair", and b)

"Paradoxes". The methods allow a systematic search of variants. They can be used as
a "manual" design, and as a basis for developing of system formalized for obtaining
modifications of random assembly unit.
LITERATURE
[1]
[2]
Interpretative Dictionary of Bulgarian AZ, Team, S., MAG-77, 1997.

.
, . 2009 ,
, 2009.
[3] Dinev G, . Leparov, M. Vicheva About search of variants of product with CAD,
The 5 th Intern.Simp. Shape, Mechanical and Industrial Design of Products in
Mechanical Engineering 2008, the Faculty of Technical Sciences, Novi Sad,
Serbia, 2008.
[4] ..
, . 2008
, , 2008.
[5] ., .
CAD , 17 . -. . .
2008, , 2008.
[6] ., . , .
CAD , . , ,
, , .4-5, 2008.
[7]. ., . . .
, 16 . -. . .
-2007 , ,
, 2007.
64
ABOUT THE INTEGRATION OF ASSEMBLY UNITS

Leparov M.1, Yancheva M.2
Summary: One of problems in design process is integration of two already designed
or purchased assembly units. In this work is propose a method called Surfaces of
details used to classified and lead the actions of searching for integration the two
existing assembly units.The algorithm is illustrated by example. The results of the this
work could be used in the design of mechanical and other products.
Key words: integration of assembly unit, assembly unit, design
1. INTRODUCTION
In the process of design of technical products very often arises problem of
integraion parts of the designed product, incl. assembly units. This process is intuitive
thanks to the engineering experience of design team. In the technical books of design
this matter is not enough illustrated very well [1-4 ect.] The purpose of this work is
revealing of some opportunities to integrate the given assembly units (AU).
2. METHOD OF SURFACE OF DETAILS
Algorithm
Default: to AU
1. Analysis:
1.1. Clasification of requirements for each of the two outputs AU in terms of
their functioning in order to deploy one off so as not to interfere with the functioning of
others AU.
1.2. Clarifying the need for common links between both AU (if any), eg.
Presence of generalized flows, particularly the disposal of a AU (component part)
against the other one AU (component part) ect.
1.3. Choose one AU for primary, when he will join the ect. AU. Baisic is the AU,
which apparently is more massive and which shall be the second AU (support AU).
When difficulties of determination for such is considered any of them AU.
1
Prof. PhD eng. Mihail Leparov, Bulgaria,1000 Sofia, Bvd. Kliment Ochridski-8, Technical University,
Faculty of Mechanical Engineering, e-mail: mleparov@tu-sofia.bg
2
Assistant Prof. PhD eng. Marieta Yancheva-Popova, Bulgaria,1000 Sofia, Bvd. Kliment Ochridski-8,
Technical University, Faculty of Mechanical Engineering, e-mail: mmarieta@mail.bg
65
1.4. Classification of existing types of compounds to ensure the immobility of

both AU. Classification may be more detailed or superficial.
1.5. Preselection ways of assembling both AU, and for this purpose to analyze
the classification of t.1.4 and reported characteristics of boht AU.
2. Systematization
2.1. Determination of potential surfaces of the support AU, which serve to join the
base AU. To this end:
2.1.1. Idetify parts (components) of the support AU.
2.1.2. For each part (components) to determine it's surfaces.
2.1.3. Analyze the possibility of any surface directy or indirecty (through) other
surfaces or details) to serve for attaching the assembly support AU taking into account
the chosen mode of assembly of t. 1.5. If necessary are make any changes in AU, so
as to mainitaain it's principal function.
2.2. Determination of potential surfase of the base AU which will serve to join the
support Au. To this end:
2.2.1. Determination of details (components) of the base AU.
2.2.2. t.2.1.3.
2.3. Refine options (combinations) of the mutual disposition of both AU.
2.4. Selection criteeria adopted by one or more options for further design work.
2.5. Structural arrangement of each variant of t.2.4. If nesessary, point of view of
operation, assembly and disassembly, are made corresponding changes in either or
both.
Fig.1 Assembly drawing ''Elektromagnetic'' Fig. 2 Assembly drawing ''Indicator light''

66
About the Integration of Assembly Units
Fig. 3 Some probabilities inregration assembly units

Example
Let be given AU by fig.1.''Elektromagnetic'' and AU from fig.2 ''Indicator light''.
Components and operation of both are AU: el. magnet : 1-coil, 2- restrictive sleeve,3housing,4axis, 5- anchor,6- screw. Supplying the coil creates an electromagnetic field
which is concentrated by magnetic 3.The field attracts the armature 3, which is secured
to the axle 4 through tight fit. Axis through the hole in its lower end is connected to and
moved undepicted device. When off the coil under the action of gravity axis 4, along
with five anchor occupy it's original position. Indicator Lamp: 1 - building, 2 - contact
plate, 3 - disc insulator (2 pcs.), 4 - washers (2 pcs.) 5 - base 6 - screw, 7 - screw 8 lamp. One end of the lamp 8 is fed through a conductor, whose tip is placed between
the washers 4 and housing 1 and the second end of the lamp by-wire connected to the
base 5 ,the screw 7 and the contact plate 2. Let the indicator light to signal opening
action of the electromagnet (attracted anchor).
1.1. Requirements.
(el.magnet) the space around the bottom of the axle must be free of attachment
transfer device;
- Air gap between the armature 5 and housing 3 must be free to move the anchor;
- Details over 6 and 2 should be open about their position when moving the anchor;
- Space in the prismatic part of det.3 must be free to join the environmental technical
67
environment;
(Indicator lamp) - 8 lamp must be visible to a person standing on one side of the
electromagnet;
- Is there access to the locations of the wires connecting them;
1.2 Require general links:
- Electric wires, power lamp must be connected with the power of the coil;
- The relationship between the two AU to be fixed.
1.3 Electromagnets fig.1 is a massive part, and then it will be the base unit, the
indicator lamp will perform an secondary role.
1.4. The classification of compounds between the two are in first approximation is:
1. Detachable1.1. By assembly 1.2. By bolting the two parts concatenated
1.3. Through bolts (screws,tenon) compound 1.4. By elastic element 1.5. By pin
connector 1.6.By bayonet.
2. Permanent: 2.1. By soldering, 2.2. By gluing, 2.3. By welding,2.4. By riveting, 2.5.
By roll, 2.6 By pressed detail.
Note: Each assembly component can participate actively in those compounds in which
it's a constituent of the compound or passive, in which it's pressed by the other
components, some of which are active.
1.5. Let shortlisting includes the following modes of entry: 1.1-1.4.
2.1.1. Detailsof the indicator light is given at the beginning of examples.
2.1.2. The surfaces of each piece is visible from fig.1 and fig.2.
2.1.1. Details of the indicator light is given at the beginning of examples.
2.1.2. The surfaces of each piece is visible from fig. 1 and fig. 2.
2.1.3. Some possibilities are:
(Det.1) a) in the upper forehead is pierced with a threaded hole axis parallel to the axis
of the AU, the lamp is pressed against the electromagnet by a screw that is screwed
into threaded holes (in a spur is added) b) same but on the lower forehead c) by
threaded (assembly) at the top of the outer cylindrical surface, d) by thread (assembly)
at the bottom of the outer cylindrical surface (the same should be extended), e) lateral
cylindrical surface is drilled threaded hole perpendicular to the axis of the AU, the
attachment will be made by screw, which is wrapped in 1;
(Det.2) a) the plate can be brought out through a crack in 1, thus after el.isolation can
serve as an hanger for attachment of the AU, for this purpose can be used as a contact
as the lower and top side,b) media from the base is pressed between parts 2and 3;
hanger reaches the plase of fastening through a crack in det.1.
(Det.3) a) puck can be extended outward like a case det.2 b) hanger from the base AU
is pressed between parts 3 and 5 or between det. 1 and 3;
(Det.4)a) as b) det.3
(Det.5) a) open the plate (base) mounts the AU, eg. with screw connection, as the
contact surface between both can serve as upper and lower surfaces b) bending of the
two horizontal parts down will change the location of the connecting screw connection
d) under the condition of b) can be used connection type assembly or hinge ( the plate
itself acts as a shaft or hole);
(Det.6) a) hanger of the base is pressed between the parts 6 and 4 or between det.1
and 4 b) under the head of det.6 put elastic piece, which is implemented by mounting
on the base;
(Det.7) a) det.7 associated whit threaded hole in the base;
(Det.8) a) thread the lamp 8 can be used to create a threaded coupling for attachment
68
About the Integration of Assembly Units

to the auxiliary base AU.
Note: In some cases, the main 5 becomes unnecessary.
2.2.1.Details of the electromagnet are given at the beginning of examples.
2.2.2. Some possibilities are:
(Det.1) a) through the slit in the hanger det.3 support AU webs pass over the spool of
the coil (or slit in it) and bend down (groove in the spool), Option: Add this to the
hanger electromagnet, and it is attached to the lamp;
(Det.2) a) joint between 2 and hole in an support AU b) by bolting (added ext into
det.2);
(Det.3) a) in the upper forehead is drill two screw holes with axes parallel to the axis of
the AU, the lamp is pressed against the electromagnet by a screw that is screwed into
threaded holes (Different placements of the holes), b) by concentric to the hub 2 more
holes or a bulge, as is done by joining thread (assembly); c) on the lateral cylindrical
surface is pierced with a threaded hole perpendicular to the axis of the AU, the
attachment will be made by screw (screws), which wrapped in det.3 d) by using open
accession of an electromagnet to environmental protection technical e) with additional
openings on the upper face of the prismatic section (the same may be extended), f)
with additional openings on the lateral surface of the prism division g) using the two
fronts of the prismatic section for attachment of auxiliary through elastic element type
"pinch" (base) or 'clip' is fixed by an additional screw connection;
(Det.4) a) by a pin axis that is perpendicular to the axis of the base is attached
supported AU with threaded or assembly;
(Det.5) a) in the cone surface is pierced threaded hole that mounts the AU, provided
that it does not interfere with the interaction of the electromagnet with environmental
technical environment.
(Det.6) a) hanger of the support AU pressed between parts 2 and 6 or between det.2
and 3 b) under the head of det.1 put elastic piece, which is implemented by the
attachment of the supporting AU;
The total number of combinations K of attachment of both AU is a.v = K = 18.14 = 252,
where a and c is the number of possible anchorages for the AU and accordingly the
base AU. For some combination require additional changes accordingly, and any
combinations there are variations of structural conversions.
Some of these combinations are given in Fig.3, where by a long dashed double- dotted
narrow line shows the silhouette of the support AU.
3. CONCLUSION
In this work the method Surface of details has been reviewed. He has been
presented and illustrated the algorithm by example. The method allows a systematic
search of options for integration of assembly units. It could be used as a "manual"
design, and as a basis for developing a formalized system desing of integration of
products.
LITERATURE
[1]
Pahl G., W. Beitz. Engineering Design. A Systematic Approach. Springer - Verlag,

Berlin, 2001.
69
[2]. Otto K., K. Wood Product Design. Techniques in Engineering and New Product
Development, NJ, Prentice,Hall, 2001.
[3]. Polovinkin A.I. Fundamentals of Engineering Design, 3-e Edition, St. Petersburg,
ed.Lany,2007.
[4]. Leparov, M., M. Vicheva, M. Georgiev, Fundamentals of Engineering Design,
Softtreyd, S., 2008.
70
ANALYSIS OF INFLUENCE OF LENGTH OF DEVELOPMENT OF

BOUNDARY LAYER ON FLOW RATE THROUGH RADIAL
CLEARANCE WITHIN HYDRAULIC CONTROL COMPONENTS
Darko Kneevi 1, Aleksandar Milainovi 2, Zdravko Milovanovi3
Summary: In hydraulic control components (directional control valves, servo valves,
LS regulators, etc.) there is a problem of determining of flow rate through the radial
clearances of different lengths overlap between the control piston and body of
component. As lengths of overlap are relatively small, it is necessary to analyze the
development of a full profile of fluid velocity through radial clearance (in sleep mode
servo valve has length of overlap equal zero; directional control valves, depending on
the diameter, have lengths of overlaps of 0.6 to 0.8 mm; while in work of the LS
regulator, piston is continuously located in areas of small opening, zero overlap and
small length of overlap). Common values of radial clearance is approximate in the
range from 1 m (minimum values for servo valves) to 50 m (maximum value for axial
piston pumps). [1]
In this paper is analysed development of full-length profile of velocity inside radial
clearance (i.e. development of boundary layer).
Key words: hydraulic components, radial clearance, length of overlap, boundary layer
1.
INTRODUCTION
Complexity of mathematical description of the flow of hydraulic fluid through the

radial clearances stems from the fact that the mathematical description can not be the
same for flow of hydraulic fluid through the clearances at zero overlap between the
piston and cylinder, the small lengths of overlap and the long lengths of overlap.
Flow of the hydraulic fluid through orifice at zero overlap (cylindrical ring),
depending on pressure, temperature, fluid type and size of the clearance, can be
laminar or turbulent. Due to small size of clearances in hydraulic components, flow of
hydraulic fluid through orifice at zero overlap is mostly in the laminar or transition zone.
The discharge coefficient of flow through the orifice in the laminar flow (or transition
from laminar flow to turbulent) is a function of Reynolds number. Determination of
Rejnolds number complicates the fact that it depends on the viscosity of fluid. Viscosity
1
Assist. Professor, Darko Kneevi, Banja Luka, Faculty of Mechanical Engineering, darkokn@blic.net
Assist. Professor, Aleksandar Milainovi, Banja Luka, Faculty of Mechanical Engineering, acom@blic.net
3
Assoc. Prof., Zdravko Milovanovi, Banja Luka, Faculty of Mechanical Engineering, mzdravko@urc.rs.ba
2
71
of working fluid is significantly changed because of the sudden transformation of

hydraulic energy into heat (coeficient of dynamic viscosity of working fluid at the exit of
the orifice can be 5-6 times smaller than the entrance to the orifice). [2]
For small length of overlap (ie short path of flow of oil through the radial
clearance), flow takes place relatively quickly, so that the generated heat, for the most
part, is retained in the oil and the process can be treated as adiabatic. Further, the
velocity profile along the clearance is variable and it is necessary to analyze length of
development of the boundary layer between the piston and cylinder (ie, the
determination of length needed for fully development of velocity profile along radial
clearance).
Fig. 1 Flow of hydraulic fluid through radial clearance

For longer lengths of overlap, fluid flow is quite slow, so fluid is exchanged heat
with the environment. The change of state of hydraulic fluid in the flow can be
considered isothermal. Such case flow through the radial clearance is analyzed in the
literature, but with one flaw: it ignores the change of oil viscosity with a change of
pressure. This neglect can be lead to the calculated flow rate is 50% higher than the
actual flow rate (depending on value of of working pressure in the hydraulic system ).
By determining the length of development of the boundary layer in flow of fluid
through the radial clearance can be defined criteria for the application of appropriate
formulas for determining the flow rate. [2]
2.
DEVELOPMENT OF BOUNDARY LAYER AT FLOW OF FLUID THROUGH

RADIAL CLEARANCE
This analysis is based on analysis of development of boundary layer for flowing

incompressible fluid with uniform velocity profile at the entrance. The boundary layer
has grown in thickness to comletely fill the clearance. Viscous effects are of
considerable within the boundary layer. For fluid outside the boundary layer viscous
effects are negligible.
As the radial clearance between the piston and cylinder, cr, small compared to
the radius of the piston, it is acceptable analysis of this flow as the flow between two
parallel plates at a distance cr.
The symbol (x) denotes the thickness of the boundary layer on the surface of
the piston for any x.
72
Analysis of Influence of Length of Development of Boundary Layer on Flow Rate Through Radial
Clearance Within Hydraulic Control Components
Fig. 2 Development of boundary layer inside radial clearance

Conditions to the lack of velocity gradient at the boundary of the boundary layer
and sliding on the surface of the piston, are given folowing expressions
uz
wu
wz
0,
(1)
0.
(2)
z G
Assuming change of velocity profile by parabolic law

2
z
,
G G
(3)
cr 2
G z z 2
2 uc 2 dz uc dz ,
0 G G
G
(4)
u
uc
and applying the continuity equation
u0 cr
we obtain
uc
u0
.
2G
1
3cr
(5)
Karman has developed an approximate method of solving the boundary layer,

which is based on the application of momentum equations in integral form, applied to
an element of boundary layer, restricted areas 1, 2, 3 and 4 (Figure 3). [3]
Isolated element of the boundary layer in Figure 2 is shown in Figure 3.
73
Fig. 3 Isolated element of boundary layer

For steady two-dimensional flow, neglecting the mass forces and applying a
basic property of the boundary layer
wp
|0,
wz
x component of momentum equation, applied to the CV, is
pG p
dG
dp
dG
dx p dx G
dx W w dx
dx
dx
dx
G x
U u u dz
G x
G x
d
2
U
u
dzdx
U uc uc n ds U u 2 dz
dx 0
0
(6)
0.
By applying the continuity equation for incompressible fluid flow
v n dS
0,
(7)
KP
on the element of boundary layer of unit width dS ds 1 or dS

equation (7), applied to the control volume CV, becomes
dQ
Q x u c n ds Q x
dx
dx
2
where:
74
0 ,
dz 1 , the continuity
(8)
Analysis of Influence of Length of Development of Boundary Layer on Flow Rate Through Radial
Clearance Within Hydraulic Control Components
G x
G x
Q x
v n dz
udz .
(9)
We combine eguations (8) i (9) to get
dQ
d
uc n ds dx dx dx
2
G x
udz .
(10)
If for two-dimensional steady flow, velocity and pressure are change in the law
uc x i p
uc
p x ,
applying Euler's equations of motion, gives
uc
duc
dx
1 dp
.
U dx
(11)
Integral over the surface 2 is
dQ
U uc
dx
dx
U uc uc n ds
2
d
U uc
dx
G x
udzdx .
(12)
By applying Newton's law of viscosity, shear stress on the surface of the piston
can be written as
Ww
du
dz
(13)
z 0
By incorporating equations (10), (11), (12) and (13) into equation (6) and
dividing by U dx , is obtained
du
uc c
dx
G x
d
dz uc
dx
G x
d
udz
dx
G x
u 2 dz Q
du
dz
(14)
z 0
By integrating equation (14), we get
G
x
20 cr
.
3Re x
(15)
Here is
Re
2u0 cr
2Q U
d SP
2Q U
,
wP
(16)
Reynolds number.
By putting
75
cr
,
2
we get the length of the full development of the boundary layer, xtl,
xtl
cr
0.0375 Re .
(17)
From equation (17) can be seen that for very low Reynolds number Re 30 ,
which is often the case in the hydraulic control valves, the length of the development of
the boundary layer is less than the size of radial clearance cr.
3.
CONCLUSION
It is obvious that in determining the flow rate through the radial clearance within
the hydraulic component must be used various mathematical expressions in the
function of the length of the overlap. The criterion for the application of the appropriate
expressions (for zero overlap, short or long overlap) can not be determined without
experimental investigations.
The reason for this is that the energy of hidraulic fluid flowing through radial
clearance is transformed into heat energy which can lead to increase of fluid
temperature. As working fluids in hydraulic systems are commonly used mineral oils,
which significantly changes the viscosity with changes of temperature and pressure, it
is the determination of flow rate of fluid through the radial clearance within the
hydraulic components is very complex.
While the mathematical description of fluid flow through long clearances is
relatively simple, real difficulty is a mathematical description of the flow of hydraulic
fluid through the radial clearances of small length of overlap (in the limiting case at
zero overlap), which are of great importance in hydraulic control components. For short
lengths of overlap can not be ignored impact of changes of viscosity of hydraulic fluid
with changing temperatures, and must be taken into account the influence of
development of full profile of velocity within the radial clearance.
It can be concluded that the formula for determining the flow through the radial
clearance at short length of overlap had to be covered: clearance geometry, operating
conditions (pressure and temperature of fluid), the type and physical properties of fluid
and changes of these properties with change of temperature and pressure. Expression
(17) can be used as orientation for defining criteria for dividing radial clearances on
short and long, and applying the appropriate formula for calculating of flow rate.
LITERATURE
[1]
[2]
[3]
76
Savi V., Zirojevi LJ. (2003). Uljna hidraulika 3. IKOS, Novi Sad.
Kneevic D. (2007). Influence Clearances Geometry in Hydraulic Components of
Automation Control on Efficiency of Hydraulic Systems, PhD Thesis, Novi Sad.
Munson B.R., Young D.F., Okiishi T.H. (2002). Fundamentals of Fluid Mechanics,
John Wiley & Sons, New York
ANALYTICAL-NUMERICAL STRESS ANALYSIS OF SPUR GEARS

WITH STRAIGHT TEETH
Neboja Radi1, Goran Sekuli2, Dejan Jeremi3
Summary: The main task of the study is to determine the stress state of spur gears
where the teeth run straight across the wheel, using analytical-numerical analysis. For
the purpose of stress analysis of spur gears, two methods are used: analytical method,
performed in accordance with ISO 6336 standards and numerical method which is
based on ANSYS v12 software package.
Of these two analysis, only the numerical one is used in two cases, that is, when
friction between two gears as their teeth mesh together, appears even when it is
neglected. After this, derived results are compared.
Key words: stress in the flank of the tooth, stress in the root of the tooth, numerical
analysis, analytical analysis
INTRODUCTION
The first step in the procedure is the stress state determination in the flank of
the tooth and in the root of the tooth of gears. This is one of the main tasks and most
important things when checking calculation of load capacity. For the purpose of this
task, a pair of the same spur gears is used. Stress state of these particular gears is
defined through analytical and numerical analysis. Analytical method is proposed by
the standard ISO 6336 while the numerical one is based on ANSYS v12 software
package. The paper briefly describes the process of developing the module, simulation
and analysis of conditions. Derived results are compared and analyzed.
1.
DETERMINATION OF THE STRESS STATE USING ANALYTICAL ANALYSIS
In order perform the analytical analysis following data are required: the module
ofa gear m=3 mm, number teeth z=20, pitch diameter d1=60 mm, vertex of a circle
diameter da=66 mm, shaft opening B=15 mm, gear width b=30 mm, roating moment
T=9.954 Nm, dynamic durability in the flank of the tooth Hlim=480 N/mm2, and dynamic
durability in the rooth of a tooth Flim=262 N/mm2.
1
Doc. dr Neboja Radi, Istono Sarajevo, Mainski fakultet I. Sarajevo, (nesor67@yahoo.com)

Dipl. in. ma, Goran Sekuli, Viegrad, (sekulic.goran@gmail.com)
3
Dipl. In. ma, Dejan Jeremi, Istono Sarajvo, Mainski fakultet I. Sarajevo, (dejan.jeremic@yahoo.com)
2
77
Operating stress in the flank of the tooth

Durability of tooth flank implies their resistance to pitting formation. Considering
that pitting is primarily a consequence of material fatigue due to high surface
pressures, the operating stress is expressed by following formulations:
VH
Z H Z E ZH Z E Z B
Ft u 1
K A K V K HD K HE
d1 b u
(1)
Where is:
Ft - circumferential force
d 1 - pich diameter
u kinematic transfer capacity
b gear width
K A is the working condition factor and includes increasing of load due to variations of
input and output moments. K V is a dynamic factor, keeping in mind load increase due
to internal dynamic effects. K HE is load distribution factor in the flank of tooth. K HD is
the influence of unequal load distribution on the pairs of gears in a constant mesh
Drive gear circumferential force is expressed by the next formula:
T
Ft
r1
(2)
9.954
Ft
331.8 N
0.03
In this case, load factors are: K A 1 , K V 1 , K HD
K FD
1.37 , K FE
1.23 , K HE
1.45 ,
1.304 .
Other important factors that fulfill above requirements for a proper opration of
operating stress of a gear, are: shape factor of a gear Z H 2.49 , material flexibility
N
) , single mesh factor Z B
mm 2
conjugate gear factor Z E 1 .
factor Z E
189.8 (
1 , constrain angle factor Z H
0. 9 ,
Inserting the previously calculated values into the operating stress in the flank
of tooth expression, we get next results:
331.8 1 1
N
V H 2.49 189.8 0.9 1 1
1 1 1.23 1.45 344.9

60 30 1
mm 2
Operating stress in the root of the tooth
According to the ISO standards, as the working stress in the root of the tooth,
bending stress V S is used. Therein, we neglect two things: tangential stress which
appears in a consequence of shearing action and normal stress due to pressure.
Adequate correction factors were introduced in the calculation. Taking into
consideration load factors too, the expression of the working stress becomes:
78
Analytical-Numerical Stress Analysis of Spur Gears with Straight Teeth
VF
Y Fa YSa YH YE
Ft
K A K V K FD K FE
b m mn
(3)
Thus, shape factor of the tooth flank is Y Fa 2.86 , stress concentration Y Sa
constrain angle factor YH 0.73 and chamfered teeth factor Y E 1 .
1.56 ,
Putting all these parameters in the formulation of the operating stress in the root of the
tooth, we get:
331.8
N
1 1 1.37 1.304 21.45
V F 2.86 1.56 0.73 1
30 3
mm 2
2.
DETERMINATION OF THE STRESS STATE USING NUMERICAL ANALYSIS
A numerical method is carried out through ANSYS v12 software package. It is

performed for two cases: in the first case, when there is no friction between meshed
gears and in the second case when the friction coefficient is taken into account. In
order to obtain the results of the stress state using numerical analysis, it was both
necessary to create a module of a gear and modify it into gear set, and then to define
the conditions of analysis.
The model is formed by extrusion from geometry. Evolute of a curve of tooth
profile is got through involute function, calculating coordinates of the points used for its
drawing in ANSYS v12 software package. After that, drawing of a vertex circle, root
circle and shaft opening is done. Removing redundant lines, we formed geometry as in
Figure 1.
Fig. 1 Geometry and model of a gear

Fig 2 shows a pair of gears derived from touching meshed gears only at one
point in no-load.
79
Fig. 2 A pair of gears

Defining terms that simulate normal conditions and limitations, we began by
defining a contact region between the flank bearing surfaces of tooth of gears.
Furthermore, we selected a type of contact which was the contact without friction
(choosing frictionless option). The following step is the formation of finite element
meshes. Optimal mesh is formed by choosing a typical region with coupling of gears
and decreasing size of final element. In this way, we reduced the time of network
forming and with comminuting of mesh in the mentioned region, we increased the
accuracy.
Fig. 3 Optimal mesh of final elements

We complete defining condition analysis introducing two things: the load, that
is, torque of 9954 Nmm in the upper (drive) gear and a fixed support at the top smooth
support at the root fo a gear. After that, derived results are graphically presented by
figures 4 and 5. Figure 4 illustrates the values of the stress state when friction between
meshed gears is neglected, while Figure 5 shows the stress state between meshed
gears where friction is defined through type of contact along with friction coefficient
P 0.15 .
80
Analytical-Numerical Stress Analysis of Spur Gears with Straight Teeth
Fig. 4 The result of the stress state when friction is neglected
Fig. 5 The result of the stress state with friction at the contact
3.
COMPARISON OF THE RESULTS DERIVED THROUGH ANALYTICAL AND

NUMERICAL ANALYSIS
The results of the stress state derived, in one case, through analytical method
and in two cases, using numerical method, are given in the following diagrams.
Diagram a) illustrates the comparative results from which we can conclude that values
of the stress in the flank of tooth in all three cases have minor variations. Maximum
operating stress in the root of the tooth in both cases of numerical method has minor
derivations, while the analytical analysis calculated value is much lower.
500
450
400
350
300
250
200
150
100
50
0
Radni napon na
boku zupca
Radni napon u
podnoju
Numerika Numerika Analitika

metoda metoda (sa metoda
(bez trenja) trenjem)
Radni napon na boku

zupca
500,00
450,00
400,00
350,00
300,00
250,00
200,00
150,00
100,00
50,00
0,00
Radni napon u podnoju
Numerika
metoda (bez
trenja)
a)
Numerika
metoda (sa
trenjem)
Analitika
metoda
b)
Fig. 6 Comparing results
81
The reasons for the higher values of the stress state in the root of the tooth
using analytical method may be due to stress concentration of the radius. According
analytical recommendations, it should be taken into account that the gear is treated as
a cantilever and that the stress distribution is linear. For the results derived by
numerical method, we cannot say that we have a linear distribution as in the case of
bending, where the root of the tooth is loaded according ISO standards. If we read the
results of numerical analysis from the areas without stress concentration, than the
image of the compared results is as in the diagram b). From these results, we are able
to say that deviation for analytical method is less, compared to numerical one.
4.
CONCLUSION
Determination of stress state of spur gears with straight teeth is made in order
to compare the results derived using numerical analysis with the results from analytical
analysis based on ISO standards. Analyzing the results, we can conclude that stress
states in the flank of tooth both for numerical and analytical method, are approximate.
Also, we deduce that in relation to the position of maximum stress in the root of the
tooth, critical section of analytical method does not correspond with the position of
maximum stress gained by numerical method. Numerical analysis shows that the
developed model can be used in a simple way for different cases of analysis. It allows
us to change the parameters of condition analysis, as we have in this case altered the
type of contact of meshed gears.
LITERATURE
[1]
[2]
[3]
[4]
[5]
82
Miltenovi, V.: Mainski elementi. Mainski fakultet, Ni, 2006.

Miltenovi, V.: Mainski elementi tablice i dijagrami. Mainski fakultet, Ni,
2006.
Lee, H.: Finte Element Simulations with ANSYS Workbench 12.Schroff
Developments Corporation, 2010.
*** www.khkgears.com
*** www.ansys.com
CAD DESIGN OF FLEXIBLE FRICTION COUPLING

Georgy Dinev1, Marieta Yancheva2
Summary: In this work are presented approach for optimal design and constructive
documentation of flexible-friction coupling in CAD medium. On base of advance design
calculation are prepared constructive documentation of coupling.
The first design process step is creating 2D geometrical modeling of assembly unit. On
base of specify functional parameters and dimensional chains are create 3D
geometrical model of assembly unit. Finally the design mechanical products are
checked by animation model for collision between component parts. The result can be
use in education the students Bachelor and Master degree in science field
Mechanical Engineering.
Key words: CAD, design, flexibly, function, coupling, dimensional, analysis, education,
control, documentation, modeling.
1
INTRODUCTION
Dimensional analysis is important activity in the process of designing products

and development of design documents.
So far considered some approaches to the design of mechanical products in
terms of dimensional-precision analysis [2],[6],[7].
Modern systems CAD/CAM/CAE enables to improve the quality in the
development of design documentation [8]. As a result, in promote precision by
automated assembly of mechanical product. Furthermore, the software allows to
inspected, by simulation of the created geometric patters and demand for their
collisions in their assembly [5]. Despite the great possibilities of software in use, the
credit lies in the knowledge and skills of the working team.
In this case the basic stones are following:
Dimensional-structure analysis of a mechanical product-elastic frictional
coupling from a methodological viewpoint;
Simulation analysis of a mechanical product and check the parametric
models of the component parts.
1
Assoc. Prof. PhD Georgi Dimitrov Dinev, Bulgaria,1000 Sofia, Bvd. Kliment Ochridski-8, Technical
University, Faculty of mechanical Engineering, e-mail: gdinev@tu-sofia.bg
2
Assistant Prof. PhD eng. Marieta Konstantinova Yancheva-Popova, Bulgaria,1000 Sofia, Bvd. Kliment
Ochridski-8, Technical University, Faculty of mechanical Engineering, e-mail: mmarieta@mail.bg
83
Dinev G., Yancheva M.
These stages are applicable mainly for the purpose of designing course on
''Principles of Mechanical Design''.
The main tasks are solved in this work are:
To create a geometric model of elastic frictional coupling;
To make dimensional structural analysis of choose variant constructive
decision;
Analyzing the results for optimization and of axial parameters-clearances, in
the design of mechanical product;
To propose an algorithm to developed constructive documentation of
assembly unit.
2
INVESTIGATION, RESULTS AND DISCUSSION
2.1 Geometrical modeling of elastic friction coupling-mechanical product

On base the specified output data base calculations is developed 3D
geometrical model of the elastic coupling through the graphical system Solid Works,
given in Fig.1. [7]
Fig. 1 3D geometrical model of the elastic coupling

2.2 Dimensional structual analysis of the assembly unit
For the functional dimensioning mehanical product of design through modern
CAD/CAM/CAE systems are based the following algorithm:
output data;
84
CAD design of flexible friction coupling
Modeling of component parts;

3D/2D models;
Construction of the graph model;
Calculation of axial chains;
On the basie of the developed 3D geometrical model of the assembly unit to
study the collision between the relevant it parts;
From an inspection of 3D geometrical models of component parts to make
corrections of their sizes by removal of material.
Fig. 2 2D drawing with coded contact surfaces

On base of the developed constructive solution are choosed a system of two
dimensionale chains. Values of the axial parameters are selected on base
standardized catalog information. The functional elements of the assembly unit are
gived in Fig. 3.
85
Dinev G., Yancheva M.
Fig. 3 Scheme dimensional chains

2.3 Analysis of results
Since the elastic friction coupling has a flexible element, assembly parameter
P1f, and is selected by catalog material are change from 0,5 to 5 mm during operation
process due to deviation of the shaft [3]. From the examination of the 3D geometric
model despite the changing parametres, there was not collision and a strong influence
on the value of the axial parameter P2f.
3D geometrical model can be use for simulation analysis and verification to
parametric models of the component parts. That the, students will get knowledge and
skills for use creative design methods of mechanical products. Also motivation to work
together in particular and especially to analyze and discussion analytical solutions [4].
1. CONCLUSION
The assembly unit of mechanical product flexibly coupling in CAD medium is
developed. On the basis of 2D geometrical model is made a dimensional analysis of
the product. A methods for constructive documentation of mechanical product is
proposed. 3D geometrical model of this assembly unit can be investigated by graphical
system NX-CAD/CAM The results can be used in design process on Principles of
mechanical design.
LITERATURE
[1]
Dinev G.D., N. Gizdov (2009) About teaching in fundamentals construction and

CAD Bulgarian journal of Engineering design, Vol. 2, p.74-79.
[2] Dinev G.D, (2009) Fundamentals of constructive documentation of gear
transmission with cylindrical wheels, SOFTTREJD, Sofia, p.190, (in Bulgarian).
[3] Dinev G.D. (2011) Course on Principles of Mechanical Design, AVANGARD
PRIMA, p. 98.
[4] Dinev G., Z. Vitlarov (2011) Handbook for laboratory exercises on Principles of
Mechanical Design, AVANGARD PRIMA, p. 40.
[5] Dinev G.D. (2011) An approach for simulation design of mechanical product, 6-th
International conference OPTIROB 2011, Sinaja, Romania, May 2011 (printed).
[6] Sandalski B. P., P. V. Goranov, G. D. Dinev, I. Z. Nikolova (2008) Fundamentals
of design and CAD, SOFTTREJD, Sofia, p.338 (in Bulgarian).
[7] Goranov P.V, G.D. Dinev, V.M. Stancheva, M.K. Jancheva (2005) Acseptual
model of internet oriented system for tolerance analyses of mechanical
assemblies. Proc. of Inter. Conference Challenges on higher Education,
Sozopol, vol.3, p.64-67.
[8] User guide on Solidworks (2010).
86
COMPARATIVE ANALYSIS OF THE FORMATION OF SMALL GRAIN

GUIDANCE
Dragan Lianin1, Marinko Petrovi2, Nenad Grujovi3, Jelena Borota4
Summary: Classical theory of internal ballistics for small-arms represents a very rough
approximation. The assumption is that interior profile of the barrel of the small calibers
can be defined on the basis of theoretical and practical tests primarily (experiment) for
a family of weapons, i.e. all weapons of small caliber (including 12.7 mm). The aim is
to determine the feasibility of standardization of one or the other barrel profile (12" and
7"), based on the advantages of one over the other profile, verification of the modified
internal ballistics model and the confirmation of such a hypotheses.
Key words: small grain, ballistics, small caliber, weapon
1.
INTRODUCTION
The barrel is an essential part of weapons. It represents the part in which there
is twofold energy transformation: chemical energy of combustion of gunpowder is
converted into heat and the heat into mechanical energy. The barrel provides the
necessary initial projectile speed and direction of the boom, and the projectiles that
have no wings to stabilize them, through the helical grooves, it provides high rotation
needed for stability when the projectile is moving towards its aim.
The classical theory based on geometrical law of combustion and adiabatic
processes in small arms barrel, gives a fairly satisfactory solutions for large-caliber
weapons, because the assumptions on which these methods are based do not differ
from the actual firing processes. There are several two-phase models developed in the
last decade that represent a significant step forward in studying the process of firing in
small-arms [1]. To solve the internal ballistics problem the most realistic strategy is to
go from the works by P.S. Gough and that the solution to the model should be sought
by using the finite differences model [2], [3].
Fundamentals of numerical treatment PDJ were given by Richtmayer and
Morton, Roach, Patankar and Smith, and for the numerical solution of two-phase flow
in small arms we used three methods: MacCormack, Lax-Wendroff and Spalding. PS
Gough [3] and others, and Gokhale and Krier used MacCormack two-stage, explicit
1
dipl.ma.in. Dragan Lianin, Kragujevac, Mainski fakultet (lisanin@sbb.rs)

prof.dr Marinko Petrovi, Kragujevac, Zastava namenski proizvodi
3
prof.dr Nenad Grujovi, Kragujevac, Mainski fakultet (gruja@kg.ac.rs)
4
dipl.ma.in. Jelena Borota, Kragujevac, inski fakultet (jborota@gmail.com)
2
87
method with finite differences. Later for numerical modelling of flame propagation
through the granular environment Spalding implicit method is used [4].
The entire survey was realized as a contribution to the study of flame
propagation through a granular environment [4]. The experimental determination,
calculation of pressures at firing weapons, and nonstationary gas dynamics is applied
to the internal ballistics problem of small-arm weapons.
In the US armament the 5.56 mm machine gun M16 with a bullet M193 was
introduced, but other members of NATO did not accept it. Parallel study of several types
of bullets, adopt the same bullet for the all members. For the survey Belgium prepared a
bullet under the designation SS109, which uses the same cartridges and has the same
dimensions as the U.S. M193, but the projectile has significantly enhanced features
(which required pitching twisting grooves of the barrel from 12" to 7").
Inside profile of the barrel for small-arms can be defined in terms of theoretical
and above all practical tests [experiment] to a family of weapons that meet the
following requirements: safety of use and reliability and simplicity of construction and
maintenance. For analysis was taken the internal profile of the barrel of the small firing
arms. As the basis was taken 5.56 mm caliber, and discussed it in two variants of
existing models internal ballistics two-phase flow [5].
2.
COMPARATIVE ANALYSIS
Belgium conducted a parallel investigation with three types of bullets (7.62 mm

x 51 SS77; 5.56 x45 mm M193 and 5.56 mm x 45 TW7" SS109, valid gunpowder
L110, P112 armour). Advantage of the system 5.56 mm was showed and confirmed
the accuracy in operational steps in twisting grooves of grain SS109 and M193.
Thereby, they set the initial velocity of the projectile SS109, 4g weight V0 = 940 m/s of
the tube length of 508 mm and V0 = 925 m/s from the barrel length 465 mm. Sweden
also conducted tests of ballistic projectile stability of 5.56 mm projectiles fired from
weapons of different length and steps twisting grooves (guns: U.S. M16A1 - 12", the
Belgian FNC - 7" and the Swedish FFV 890C - 9").
Based on these findings, for the SS109 projectile the step folding grooves 7"
(177.8 mm) is confirmed. Because the bullet 5.56 mm SS109 has the same capsule as
the 5.56 mm M193 bullet (SS92), two basic parameters for design of internal alignment
tube can be defined: a step twist and propellant chambers.
The increased velocity of the projectile is a function of increased pressure
propellant gases p, which in cal. 5.56 mm exceeds the value of 3600 bar. It also means
that it increases the specific pressure on the working side gutters of the barrel, i.e.
pressure force of projectile shell is also greater.
When moving through the channel of the barrel, the projectile receives a large
radial velocity along the axial rotation axis. The shell is faced with significant centrifugal
forces of inertia that are transmitted to the surface of the guidance since the projectile
is in contact with it. At the exit from the barrel, the centrifugal force cause shear stress
in the shell and that may even harm the strength of shell and proper route of the
projectile. Between the shell and core there is usually soft lead lining - layer that, lining
the engraved projectile and gutters protect the channel of barrel from intense damage.
The most frequently used method of increasing ballistic lifetime of the barrel is
chrome. The reason for this is an excellent reliability of this production in terms of the
88
Comparative analysis of the formation of small grain guidance
achieved level of the barrel performance compared to other technologies.

Classically applied cutting technology did not give many opportunities to
change a rectangular profile of the rifling and its substitute. Rifling forging is a relatively
new method, which is currently brought to a satisfactory level of quality. Its use, unlike
other technologies has two main advantages: projectile deposits are made as
simultaneous continuation of rifling operation and with this technology a more flexible
variation of rifling profile is possible.
Fig. 1 Shapes of the grooves

The shape of the cross-section or the profile of the projectile guidance was for
a long time considered to have very significant effect on weapons ballistics, Fig 1.
Studies of the practical results have shown that it has the largest influence on the wear
and lifetime of the barrel. For selection of a profile, the valid criteria is determined to
define the restrictions such as: a minimum distribution of hits on the target, the
maximum lifetime of barrel and a minimum of cleaning (the low-maintenance) and
minimum cost pipe. When shooting caliber weapons below 12.7 mm, there are 4 or 6
grooves.
Adoption of the four-grooves for the profile of the barrel 5.56 mm SS109 is
accompanied by the satisfaction of demands that the specific pressure on the working
side of the guidance in this new profile does not exceed the value that occurs in the
profile with 6 grooves. Three profiles were tested: profile 1 (with 6 grooves), profile 2
and profile 3 (with 4 grooves), for the purpose of comparative analysis and the general
results, Tab. 1.
Table 1 Profile dimensions
Calibar d [mm]
Groove do [mm]
Change interval
'd and 'do [mm]
Twofold depth of the
groove 2G1 [mm]
Surface Sc [mm2]
Profile 1
5.50 5.59
5.63 5.72
Profile 2
5.44 5.53
5.64 5.73
Profile 3
5.38 5.47
5.61 5.70
0.09
0.09
0.09
0.10 0.16
0.17 0.23
0.20 0.26
24.51 25.31
24.36 25.16
24.23 25.03
89
3.
MATHEMATICAL MODEL FOR THE TWO-PHASE FLOW WITH QUASISTATIC

ENGRAVING THEORY
The two-phase flow model takes the pressure of forcing p0 as developed

propellant gas pressure when the projectile was launched and the engraving happend.
The main drawback of this model is the empirical defining the pressure of
forcing, and then the Drozd method defines the initial values of parameters and the
duration of combustion till the moment of initiation of the projectile.
Basic assumptions of the model would be implemented as follows:
Launching of the projectile at the pressure generated in the cartridge for
overcoming a strong link between the projectile and cartridge (plucking the
projectile force).
The projectile travels to the contact of front grooved barrels and transition
cone. Speed that the projectile reaches on the road is relatively small and
does not affect the process of quasistatic engravement.
Engravement is a time process and is not instantaneous, but a gradual
transition in which the projectile passes the transition cone with its length.
The force between the shell and surface of the transition cone remains
constant and is provided through the yield strength of material under static
conditions.
Engraving force increases with the moving of the projectile, caused by
increasing of the contact surface.
The system of equations of two-phase flow at combustion of gunpowder can be
represented by partial differential equations (PDE) in matrix form:
wY
wY
As
wt
w s
(1)
matrix of partial derivatives depending on the time t and generalised

coordinate s
wY
wt
wY
wu wub wH wU we
and
wt wt wt wt wt
w s
wu wub wH wU we
ws ws ws ws ws (2)
matrix 5x5 of the coefficients with the partial derivatives from dependent
variables by generalised coordinate s
As
> akl @k 1,l 1

5,5
(3)
matrix of the right of the equation
>bk @k 1
5
(4)
or the general form
m
wY
wY
k (a
l) b
kl
k
wt
ws
l 1
(5)
where, u - gunpowder gas velocity, ub - gunpowder grain velocity,

as a measure of space that can be fulfil by gunpowder gases,
U-
H - porosity,
density of
gunpowder gases, e - internal energy per unit volume of gunpowder gases; m

90
Comparative analysis of the formation of small grain guidance
during the combustion of gunpowder, and m

the coefficients akl are equal to 0.
3 after the combustion, and some of
In addition to these factors, the mathematical model also depends on

gunpowder pressure, velocity of grain combustion, mass of grain, parameter of surface
forces, energy per unit volume used for grain heating etc.
Quasistatic engravement model [7], was developed for artillery and applied to a
small-arms, does not take into account the change in cross section due to grooved
barrels [2]. This simplification is possible for the artillery, where the diameter of the
rotating band is larger than the diameter of grooves, which is very often not the case
with the projectile of firing weapons.
4.
RESULTS AND PRECISION
Real experiment was carried out with 59 samples of barrels cal. 5.56 mm TW7
[6]. The aim was to make atmospheric conditions [temperature, humidity, atmospheric
pressure ...) and factors related to the techniques involved in experiment (assembly,
handling, loading, firing ...) as similar as possible for all samples. The average time of
maximum pressure values were measured at four measuring points by piezoelectric
transducers (p1msr, p2msr, p3msr, p4msr). The average values with standard deviation are
given in Table 2.
Table 2 Statistic analysis of pressure
Profile
Profile 1
Profile 2
Profile 3
p1msr
3502
3470
3628
st. dev.
100
322
150
p2msr
2608
2403
2748
st. dev.
110
534
128
p3msr
1042
1075
1088
st. dev.
68
80
42
p4msr
700
708
705
st. dev.
22
18
21
Projectile velocity was measured at 10 m from the mouth tube. The average
values of calculated and measured initial projectiles velocities (Vocal, Vomsr) reduced to
the mouth tube are given Table 3.
Table 3 Statistic analysis of initial velocities
Profile
Profile 1
Profile 2
Profile 3
Vocal
863
865
868
Vomsr
873.2
867.2
868.2
st.dev.
11.0
11.6
7.3
It is obvious that experimental results deviate very little from the calculated,
only by +1.2% for Profile 1, +0.27% for profile 2 and +0.02% for the third profile This
means complete correspondence of theoretical and experimental models. The average
values of Rs and D with the profiles are given in table 3, with measurements in
centimetres.
As with velocities, statistic analysis shows that the accuracy obeys the
Gaussian distribution with standard deviations, and that the results of the measured
91
initial velocities correspond to the results of precision.

Table 4 Statistical analysis of precision
Profile
Profile 1
Profile 2
Profile 3
5.
D40
4.18
3.81
3.62
RS40
1.16
1.03
1.01
st.dev.
0.28
0.14
0.16
CONCLUSION
The most stable results of the initial velocity and precision provides the profile 3
with the lowest standard deviation of 7.3 m / s and the smallest average radial drift of
the hits RS40 = 1.01 cm, and diameter of the distribution of hits D40 = 3.62 cm.
This work has no pretensions to study the mathematical models analysed in
literature and there is absolutely reason to doubt that information. That leaves rooms
for discussion about quality, dimensional accuracy and uniformity of the designed
barrels, the quality and uniformity of ammunition and quality of measurement
equipment.
To completely define the most appropriate form of the projectile guidance it is
necessary to repeat the experimental research several times with barrels and
ammunition of different quality and make measurements with quality measuring
equipment in various weather conditions.
The aim of this paper is to point to one possible way of defining a standardized
projectile guidance, as for calibre 5.56 mm, and for all calibres up to 12.7 mm, which
must meet the following criteria: security Service - the pressures, initial velocity and
accuracy must be in pre-defined borderline, simplicity of construction - different levels
of quality of tools and human resources, ease of maintenance, low cost of making
barrels and certitude.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
92
Cvetkovi, M. (1984). The use of non-stationary gas dynamics for the interior
ballistics problem of weapons of small caliber, Ph.D. thesis, Zagreb.
Tani, LJ. (1997). Numerical solution of unsteady internal ballistics model of
small caliber weapons, Ph.D. thesis, Belgrade.
Gough, P. S., Zwarts, F. J. (1977). Modeling Heterogeneous Two-Phase Reacting
Flow, AIAA Journal, AIAA / SAE 13th Propulsion conference, Orlando, Florida,
July 11-13, Vol. 17, No. 1, pp. 17-25.
Jaramaz, S. (1991). Contribution to the study of flame propagation through a
granular environment. Ph.D. thesis, Faculty of Mechanical Engineering, Belgrade.
Tani, LJ. (1987). Determination of the pressure forcing the firing of weapons.
Master thesis, Zagreb.
Petrovic, M. (1999). Analysis of a new form of projectile guidance for barrel cal.
5.56 mm. Ph.D. thesis, Mechanical Engineering, Belgrade.
Stiffler, K. (1982). Projectile sliding forces in a rifled barrel. Mississippi State
University.
COMPARISON OF TECHNIQUES FOR DETECTION OF FAILURE

ROLLING ELEMENT BEARINGS
Pavle Stepani 1, eljko urovi 2, Aleksa Kronjar 3, Aleksandra Pavasovi 4
Summary: This paper presents advanced techniques for the detection of bearing
failures based on vibration signals. Capability of detection and diagnosis of some
effective techniques are discussed and compared on the basis experimental results. In
particular, we analyzed the ID3 data mining technique, detection using statistical
pattern recognition and application of hidden Markov model (HMM). Comparing the
experimental results showed that the highest accuracy of detection techniques has
been realized using HMM, and for determining the type of bearing damage easier to
use a statistical approach to pattern recognition. It is shown that the presence of faults
can be detected on-line monitoring of the probability of trained HMM for the correct
bearing, which is determined by feature extraction from vibration signals. In addition,
the application of HMM can be extended to consider problems of prognosis and
prediction of bearing condition, which gives as an output time to failure.
Key words: failure detection, data mining, statistical pattern recognition, hidden
Markov model, condition based maintenance
1.
INTRODUCTION
Most modern techniques for the detection of bearing failures is based on the
analysis of vibration signals collected from the bearing housings. The common goal of
all techniques is to detect the presence and type of failure at an early stage and to
monitor its development in order to assess the remaining life of a system and choose
the appropriate maintenance plan. Monitoring and predicting of the condition industrial
systems and processes at running are activities whose importance in recent years
great, as based upon the concept of preventive maintenance. Timely knowledge of a
defect allows the user to correct the problem occurred during a regular repair in order
to avoid additional, unplanned and often very costly delay because of sudden failure.
The basic elements necessary for successful diagnosis, prognosis, and condition
based maintenance (CBM) are shown in figure 1.
Dipl. eng. el., Pavle Stepani, Belgrade, Lola Institute, pavle.stepanic@li.rs

Prof. dr, eljko urovi, Belgrade, Faculty of Electrical Engineering, zdjurovic@etf.rs
Dipl. eng. el., Aleksa Kronjar, Belgrade, Lola Institute, aleksa.krosnjar@li.rs
4
Prof. dr, Aleksandra Pavasovi, Belgrade, Faculty of Electrical Engineering, apavasovic@etf.rs
2
3
93
Fig. 1 The basic elements condition based maintenance

The sensors are mounted on bearings that are monitored that gives the
vibration signals. Then the feature extraction is to convert the signal of vibration in
some type of parametric representation suitable for further analysis. Diagnostics is
engaged in fault detection, isolation and identification of them when it happens.
Detection of fault has a task to indicate that if something is wrong in the observed
system, task of fault isolation is that find the defective component in the system while
identification should determine the nature of errors which is detected. In the block for
prediction is based on prognostic models and algorithms estimated the remaining
lifetime of the observed bearings.
Condition based maintenance is generally limited to the failure diagnosis
performance than the remaining life assessment. Diagnosis of failure is one of the
most important phases in the CBM system. As a first step in diagnosis is to determine
whether the error is present or not, and therefore this paper will present and compare
three different methods for detection of bearing failures. Detection of rolling element
bearing faults actually represents the problem of pattern recognition and classification.
The following section will briefly discuss these methods and then focus more on
implementation techniques and the results obtained in this study. For detailed
descriptions of the techniques recommended listed references.
2.
METHODS
2.1 Decision Tree

Decision tree is one of data mining techniques, and basically represents a
classifier that shows all the possible outcomes of a process. The paths that lead to
these outcomes are in the form of tree structure. The nodes that separate the different
classes of branching based on if-then conditions. Algorithms to build trees are
recursively embedded in the decision tree partitioning the training set of data
successively more complete subsets. Partitioning criteria determine characteristic
differences between decision trees, which are created different algorithms. ID3
algorithm uses entropy as an indicator and from initial training set partitioning it made
less subset [1]. Features with the highest information gain are selected for the branch
nodes based on information theory. Construction of decision tree very much depends
on how the features of the training sample selected.
94
Comparison of Techniques for Detection of Failure Rolling Element Bearings
2.2 Statistical Pattern Recognition

The main goal of pattern recognition is to decide which category the observed
pattern belongs. One of the ways in which the pattern recognition can be performed is
that based on the series of patterns estimated conditional probability density functions,
but such a procedure involves a very large amount of data and is associated with a
number of numerical problems. Therefore, it is often applied classification methods are
not optimal in any sense, but they are very simple. Commonly used parametric
methods are linear and quadratic classifier [2]. In this paper we design a quadratic
classifier based on the desired output.
2.3 Hidden Markov Models
HMM consists of two stochastic processes. The first process is a Markov chain
with finite number state, while another process for the given state generates a series of
observations. The word "hidden" means that the HMM states are not directly
opservable. Since the sequence of states, as the basic stochastic process is not
observed, it can only be estimate through another set of stochastic processes that
generate sequences of observations. Thus defined model is the basis for the
description of a number of different problems. From the point of view of modeling the
vibration signals from rolling element bearings, HMM can be viewed as a combination
of two stochastic processes. These are the hidden Markov chains which are temporary
variability of vibration signals in the hidden state sequence and observable process
that explains the spectral characteristics of vibration signals. So, despite the fact that in
every discrete time instant t system is in state qt , it every step emits a visible symbol,
which in general can be a continuous function (e.g. spectrum). HMM can be described
by the following equations:
X k 1 AX k Vk 1,
(1)
Yk CX k Wk ,
where X k denote hidden process, Yk denote observation process, Vk and Wk are
terms for noise, and A and C are parameters.
In general HMM is represented by triplet A, B and S , i.e. in compact notation
A, B,S ,
where the matrix A is the probability distribution of transition states,
matrix B is the probability distribution of occurrence of certain symbols in certain states,

and S is the initial distribution of state [3]. The problem of training HMM is a subject of
adaptation (or re-estimation) model parameters O A, B,S in an effort to learn and
coding the characteristics of observational sequences that enable the model can
continue to identify similar observation sequences. Optimization of parameters can be
done Baum-Welch's iterative algorithm [3], which in this paper and used.
3.
EXPERIMENTAL RESULTS
Experimental data were collected from the station for testing rolling element
bearings. For the purposes of research and comparison of the above techiques, in this
paper we used vibration signals obtained from accelerometer mounted horizontally on
the bearing housing. Data were collected for the two states of bearings: normal
(without defects) and defective (with some type defect). The activity preceding
95
classification is the determination of which measuring quantities will in the best

possible way characterize the pattern that needs to be classified. For the first two
methods (ID3 and quadratic classifier), extracted by 9 parameters in the time domain
and 9 parameters in the frequency domain for a total of 18-dimensional feature vector,
characterizing each of the tested bearings [2,4]. For HMM failure detection feature
extraction from vibration signals was performed using Linear Predictive (LP) modeling
for the model of 10th order.
3.1 ID3 Decision Tree
ID3 uses information gain as a measure of selection between features. In order
to decrease computational complexity of algorithm for classification, dimension
reduction was performed in 18-dimensional feature vectors in the new 6-dimensional
space of features. The reduction of dimensions was based on KL transformation [4].
The process of discretization of continuous features in the ID3 algorithm is actually a
process of selecting optimal threshold [1]. The classification was performed in two
classes: defective and healthy (i.e. functional) rolling element bearings. Figure 2 shows
the decision tree for the detection of bearing condition using ID3 algorithm. To test the
classifier used is the "leaving-one-out" method which is estimated classification
accuracy of 95.41%.
y 3 d 54.04
y 4 d 0.69
y 3 ! 54.04
y 4 ! 0.69
y 2 d 47.53
y 5 d 2.59
y 5 ! 2.59
y1 d 1333
y 2 ! 47.53
y1 ! 1333
Fig. 2 ID3 decision tree

3.2 Quadratic Classifier
Quadratic classifier is designed using the desired output method. For the
application and control of quadratic classifiers previously performed dimension
reduction of the 18-dimensional feature vectors into two-dimensional space of features.
Representation of patterns in two-dimensional space allows visualization of random
vectors and easily calculate the parameters of quadratic classifiers. Dimension
reduction was carried by using the criterion based on scatter measure [2]. Figure 3
shows the position of two-dimensional vectors that characterize the analyzed bearings
and projected quadratic classifier. Testing the classifier, "leaving-one-out" method was
obtained classification accuracy of 98.98%. In Figure 3 shows the obviously
separability between defective and functional bearings. In addition, within the class of
defective bearings noticeable clustering around the characteristic centers which might
indicate to a particular type of bearing damaged. This is primarily result of intelligent
dimension reduction based on scattering matrices which takes account of class
separability.
96
Comparison of Techniques for Detection of Failure Rolling Element Bearings
y2
y1
Fig. 3 Quadratic clasifier ('' - defective bearings, '' - functional bearings)

3.3 HMM Fault Detection
For detecting presence of faults is enough to train one HMM for functional
rolling element bearings. Sequences of observations to train HMM is obtained by
dividing the vibration signal of length one second division in ten equal segments. For
each of these segments (length 0.1 s) the autocorrelation method of LP coefficients are
calculated for a model of order 10th. Accordingly, observational sequence consists of
10 observations, i.e., O O1O2O3 O10 where each of the observations Ot a 10dimensional vector of features extracted from the LP model. Based on training
observations for defective and healthy bearings codebook is generated using K-means
algorithm [3]. Codebook size is M = 128 and on its basis will be carried out vector
quantization (i.e. discretization) of continuous feature vectors. A structure model of
HMM was chosen left-to-right model with three states and discrete observation
symbols. Calculate the probability of resulting in a estimate to what extent a sequence
corresponding to the adopted model in forward-backward procedure [3]. If the
probability above a predefined threshold, then in bearing is not present defect,
otherwise the bearing is faulty. One way to select an appropriate threshold is the
minimum value of probability calculated for training observation sequences from
normal bearings:
thresh
min P O O
(2)
normal
In Figure 4 a) shows the iterative step in the training of HMM and the
corresponding logarithm of probability at each step. Figure 4 b) shows the logarithms
of probability of vibration signals for functional bearings and bearings with defects,
obtained for the HMM of functional bearings. As the figure shows bearing faults are
detected with accuracy 100%. Probability of functional bearing is separated by a
clearly defined threshold out of probability bearings with defect. On-line monitoring of
probability HMM for functional bearings indicates ability to forecasting the condition of
bearing. Developing fault in the bearing will be reflected as an unexpected change in
probability, i.e. a decrease in the probability. Over time, as failure progresses, the
probability of HMM for a functional bearings has a downward trend. When the bearings
closer to end of useful life, the probability decreases rapidly, which indicates serious
damage.
97
Fig. 4 a) HMM training curve b) Probabilities of normal and defective bearings

4. CONCLUSIONS
This paper compares some of the modern techniques for the detection of
bearing failures from vibration signals based on experimental results. Classifier based
on the ID3 decision tree is clear and easy to understand. Good characteristic of this
classifier is a short period of training. Estimated accuracy of detection bearing faults is
95.41%. One of the disadvantages of training phase, which can lead to less accuracy
of detection is over-fitting.
Quadratic classifier is easy to use and has a high accuracy of fault detection
98.98%. This is primarily result of intelligent dimension reduction based on scattering
matrices which takes account of class separability. As a result of this reduction of
dimension is a clustering of damaged bearings pattern around specific centers, which
can be used to determine the fault type.
Detection of bearing failures using discrete HMM of vibration signals showed
classification accuracy of 100%. Time training of HMM is longer than the previous two
techniques and algorithm used in training do not guarantee reaching a global
maximum criterion function. Monitoring probability of HMM for functional bearings can
be predicted his condition in the future and estimate the remaining useful life.
Acknowledgments
This work was created within the research project "Development of the devices
for pilots training and dynamic flight simulation of modern combat aircraft: 3 DoF
centrifuge and 4 DoF spatial disorientation trainer" that is supported by the Ministry of
Science and Technological Development, Republic of Serbia.
REFERENCES
[1] Quinlan, J.R. (1986). Induction of decision trees. Machine Learning 1, 81-106.
[2] Stepanic, P., Latinovic I.V., Djurovic, Z. (2009). A new approach to detection of
defects in rolling element bearings based on statistical pattern recognition. Int. J.
Adv. Manuf. Technol., Vol. 45, No. 1-2, pp. 91-100.
[3] Rabiner, L.R. (1989). A Tutorial on Hidden Markov Models and Selected
Applications in Speech Recognition. Proceedings of the IEEE 77(2), pp. 257-286.
[4] Stepani, P., Kronjar, A. (2009). Detekcija oteenja kotrljajuih leajeva
primenom ID3 stabla odluivanja. Zbornik radova DEMI 2009, pp. 323-328.
98
CRACK INITIATION LIFE OF NOTCHED METALLIC PARTS

EXPOSED TO LOW CYCLE FATIGUE
Strain Posavljak1, Miodrag Jankovic2, Katarina Maksimovic3
Summary: The flat specimens with a central hole and one turbojet engine compressor
disk have been studied in this paper as representatives of notched metallic parts. In
conditions of loading by blocks of positive variable force, the crack initiation life of
specimens was determined experimentally. Using different approaches based on low
cycle fatigue criteria, crack initiation life of mentioned specimens was estimated also.
Original Neubrs rule and Toppers and Sonsino-Birgers modification of this rule, were
used for determination of stress-strain response at critical point. Estimation of crack
initiation life was carried out using Palmgren-Miners rule of linear damage
accumulation supported by Morrows, Manson-Halfords and Smith-Watson-Toppers
curves of low cycle fatigue. Experimentally obtained results and results of estimation of
crack initiation life were compared and analyzed. It was shown that approach which
include Sonsino-Birgers modification of original Neubers rule and Palmgren-Miners
rule of linear damage accumulation supported by Morrows curves of low cycle fatigue,
was made the best result of estimation.This approach was used for crack initiation life
estimation of the studied turbojet engine compressor disk. For the purpose of
estimation, two blocks of rotation frequency were taken into account. One block that
presents different engine ground controls and one block that presents different types of
flights. Estimated crack initiation life of disk, expressed in flight hours, was compared
with practical crack initiation life.
Key words: flat specimens with central hole, turbojet engine compressor disk,
damages, crack initiation life estimation
1.
INTRODUCTION
Crack initiation life estimation of real notched metallic part exposed to low cycle
fatigue understands knowing of data about blocks of loads, cyclic events in that blocks,
cyclic properties of material used or nominated for workmanship, stress-strain
response at critical point or point of expected crack initiation and damages provoked by
all cyclic events.

1
Ph.D., Strain Posavljak, Banja Luka, Faculty of Mechanical Engineering, (s.posavljak@urc.rs.ba)

Ph.D., Miodrag Jankovic, Belgrade, Faculty of Mechanical Engineering, (mjankovic@mas.bg.ac.rs)
3
Ph.D., Katarina Maksimovic, Belgrade, Office of Water Mangement, (tince@net.rs)
2
99
Stress-strain response at critical point (local stress-strain response) for all

cyclic events and method of identification and counting of those events, have special
importance. Local stress-strain response may be determined by measurement, by the
finite element method and by the methods that relate local stresses and strains with
their nominal values.
Flat specimen with central hole and one turbojet engine compressor disk, as
notched metallic parts, are dicussed in this paper. Methodology of crack initiation life
estimation of these parts, based on low cycle fatigue criteria, is presented.
1. CASE OF FLAT SPECIMENS WITH CENTRAL HOLE
2.1 Some information about specimens
Three flat specimen with central hole was made of stell 13H11N2V2MF in heat
treatment state (Heating at 1000C, Oil quenching, Tempering at 640C, Air cooling).
Experimentally obtained cyclic properties of this steel that marked as quenched and
tempered (QT) steel, are contained in Table 1.
Table 1 Cyclic properties of QT steel 13H11N2V2MF [1]
No
1
2
3
4
5
6
7
Propertiy
Modulus of elasticity, E [MPa]
Cyclic strength coefficient, K [MPa]
Cyclic strain hardening coefficient, n
Fatigue strength coefficient, f [MPa]
Fatigue strength exponent, b
Fatigue ductility coefficient, f
Fatigue ductility exponent, c
Value
229184.6
1140.0
0.0579
1557.3
-0.0851
0.3175
-0.7214
Crack initiation life (CIL) of mentioned specimens marked with SPC1, SPC2
and SPC3, was determined experimentally. Using the universal MTS system, the same
were loaded by blocks of positive variable force. Results of experimentally obtained
CIL are included in Table 2.
Table 2 Results of experimentally obtained CIL of specimens [1]
Specimen
SPC1
SPC2
SPC3
Crack initiation life [Blocks]

4000
3600
4200
Geometry of flat specimens with central hole, original block of positive variable
force and the same block with identified cyclic events, are shown in Fig. 1.
100
Crack Initiation Life of Notched Metallic Parts Exposed to Low Cycle Fatigue
Fig. 1 Geometry of flat specimens with central hole, original block of positive variable
force and the same block with identified cyclic events
Cyclic events within block of positive variable force, defined as X-Y-X force
cycles, were identified using method of reservoir [1]. X-Y-X force cycles, sorted
according to level /i/ and number of appearing /Ni/ within block, are contained in Table 3.
Table 3 Identified X-Y-X force cycles within block of positive variable force
Level
i
1
2
3
4
Xi-Yi-Xi force cycle

[kN]
6-120-6
48-120-48
72-120-72
48-96-48
Nuber of force cycles within block,

Ni
1
3
2
1
2.2 Stress-strain response at critical specimen point

At the beginning, the flat specimen with central hole was observed as ideal
elastic body. Its linear stress response, provoked by maximal force Fmax = 120 kN, was
obtained using the Finite Element Method (FEM) implemented in I-DEAS Master
Series software [2]. Discretization was carried out by plane stress parabolic finite
elements. FEM model with trajectories of principal stresses 1 is shown in Fig. 2.
Principal stress 1 has maximum value at critical specimen point P, 1max = 1,P
= 1789.13 MPa and it with belonging strain is not real.
Stress for point which is satisfying far from stress concentration area, taken as
nominal stress n = 480 MPa, is equal to the ratio of force Fmax and surface of full
specimen section (S = 5 50 = 250 mm2).
Stress concentration factor Kt = 3.727, equal to the ratio of stress 1,P and nominal
stress n, was served for transformation of linear stress-strain response at critical
specimen point into nonlinear.
Respecting memory of metals, nonlinear stress-strain response, at critical
specimen point, was described by stabilized hysteresis loops assigned to all ith force
cycles contained in Table 3. Neubers rule [3], Toppers modification [3,4] and SonsinoBirgers modification of this rule [5,6], were used for determining of nonlinear stress
strain response. The next two systems of equations
101
(1)

L

Q
L
Q
W(
K 2t ni
E
(2)
1
n '
2 '
i 1,2,3,4
E
2K
belonge to Neubers rule. The first equations in these systems are two forms of
Neubers hyperbola [3]. The second equation in (1) is equation of cyclic stress-strain
curve and the second equation in (2) is equation of Masings curve [3] for QT steel
13H11N2V2MF.
The values of nominal stresses ni and their ranges ni that used in (1) and
(2), were calculated using expressions
Yi 103
S
Vni
'Vni
(3)
Yi Xi 103
i 1,2,3,4
Values of cyclic properties, used in (1) and (2), with known stress concentration
factor Kt, were taken from Table 1.
Systems (1) and (2) in wich instead of Kt used effective stress concentration
factor Kf=3.582, have been related to Toppers modification of Neubers rule.
Sonsino-Birger's modification of Neuber's rule was based on the application of
the next systems of equations
1 K t ni K t ni
1
2
E
n'

E K'
(4)
i 1
1 K t ni K t ni
1
2
E

n '
2
'
E
2K
(5)
i 1,2,3,4
The first equations, in systems (4) and (5), present two forms of SonsinoBirgers curve. The second equations correspond to the second equations in systems
(1) and (2). The values of Kt , ni , ni , E, K and n used in solving of systems (1) and
102
(2), used and for solving of systems (4) and (5) also. Numerical results of nonlinear
stress-strain response at critical specimen point, based on the application of Neuber's
rule and Topper's and Sonsino-Birger's modification od this rule, are contained in Table
4, Table 5 and Table 6.
Table 4 Numerical results of nonlinear stress-strain respose at critical specimen point
Neuber's rule
i
1
2
3
4

[kN]
6-120-6
48-120-48
72-120-72
48-96-48
Vmi
[MPa]
125.013
345.916
524.554
167.278
'Vi
[MPa]
1514.669
1072.864
715.588
715.588
'Hi
0.00832044
0.00468568
0.00312232
0.00312232
Table 5 Numerical results of nonlinear stress-strain respose at critical specimen point

Topper's modification of Neuber's rule
Level
i
1
2
3
4

[kN]
6-120-6
48-120-48
72-120-72
48-96-48
Vmi
[MPa]
131.960
361.741
533.549
189.934
'Vi
[MPa]
1490.921
1031.359
687.744
687.744
'Hi
0.00780806
0.00450238
0.00300084
0.00300084
Table 6 Numerical results of nonlinear stress-strain respose at critical specimen point Sonsino-Birger's modification of Neuber's rule
Level
i
1
2
3
4

[kN]
6-120-6
48-120-48
72-120-72
48-96-48
Vmi
[MPa]
115.114
326.998
505.556
148.439
'Vi
[MPa]
1496.466
1072.699
715.582
715.582
'Hi
0.00791852
0.00468495
0.00312229
0.00312229
Nonlinear stress-strain response at critical specimen point, determined by

Sonsino-Birgers modification of Neubers rule, is presented in Fig. 2.
103
Fig. 2 Fflat specimen with central hole: FEM model with linear stress response
described by principal stress 1 trajectories (a) and nonlineal stress-strain response
at critical specimen point determined by Sonsino-Birgers modification of Neubers
rule
2.3 Specimen damages and crack initiatioin life estimation
Estimation of specimen damage D, as damege at critical point, provoked by
block of positive variable force in Fig.1, was carriad out using Palmgrem-Miner's rule of
linear damage accumulation [3,7,8,9] defined on the way
4
i 1
Di
Ni
N
i 1
(6)
fi
In above expression the damage provoked by ith force cycle is marked with Di.
This damage presents the ratio of number of appearing Ni of ith cycle within block of
positive variable force and number Nfi of the same cycle that flat specimen made of QT
steel 13H11N2V2MF can endure up to crack initiation. Numbers Ni are given in Table 1
and numbers Nfi were determined using Morrows, Manson-Halfords and SmithWatson-Toppers curve of low cycle fatigue (LCF). Morrows (M) curve [3,10] is the first
equation of the next system
'H
2
'H
2
V'f Vmi b
Nf H'f Ncf
E
'Hi
i 1,2,3,4
2
The first equation of system
104
(7)
'H
2
'H
2
V'f Vmi b V'f Vmi b ' c

H N
Nf
V'
f f
E
f
'Hi
i 1,2,3,4
2
(8)
is Manson-Halfords (MH) curve [3] and the first equation of system

PSWT
PSWT
Vmax
PSWT, i
'H
E
2
Vmax,i
V N
' 2
f
'Hi
E
2
2b
EV'f H 'f Nf
bc
(9)
i 1,2,3,4
is Smith-Watson-Toppers (SWT) curve [3] of low cycle fatigue, where PSWT presents
Smith-Watson-Toppers perimeter in which max,i = mi + i/2 . Stresses mi in systems
(7) and (8) are mean stresses.
Combining applied approaches of determining of nonlinear stress-strain
response at critical specimen point, with different curves of LCF, using PalmgrenMiners rule, we obtained estimated values of damages D and crack initiation life (CIL =
1/D) expressed in blocks of positive variable force. These estimated values are
contained in Table 7.
Table 7 Estimated values of damages D and CIL for the flat specimen with central hole
Topper's
modification
of Neuber's
rule
Neuber's rule
D
CIL
M curve of
LCF
0.000431586
2317
MH curve of
LCF
0.000858519
1164
SWT curve
of LCF
0.000535187
1868
M curve of
LCF
0.000335017
2984
SonsinoBirger's
modification
of Neuber's
rule
M curve of
LCF
0.000349002
2865
The best result of estimated CIL of the flat specimen with central hole was
obtained using Toppers modification of Neubers rule in combination with Morrows
curve of LCF. This result is rejected because the application of effective stress
concentration factor Kf is debatable. Result of estimated CIL obtained using SonsinoBirgers modification of Neubers rule in combination with Morrows curve of LCF (CIL =
2865 blocks) is accepted as the best estimated result. The same amounts
approximately 80 % of the smollest experimentaly obtained CIL that amounts 3600
blocks (Table 2).
105
2. CASE OF ONE TURBOJET ENGINE COMPRESSOR DISK

3.1 Some information about disk
Here observed disk is the first stage low pressure compressor rotor disk of
R25-300 turbojet engine. It is dominantly loaded by centrifugal forces of blades and
own centrifugal forces. For disks of this kind, engine ground controls and flights,
described by blocks of rotation frequency /n/ in time /t/, have important role. In the case
of observed disk, two blocks of rotation frequency, taken into account. Blocks A and B
that present different engine ground controls and different flights (Fig. 3).
Fig. 3 Blocks of rotation frequency (block A presents different ground

controls and block B presents different flights)
Identified X-Y-X cycles of rotation frequency in blocks A and B are given in
Table 8. Identification was carried out similar to identification of force cycles in block of
positive variable force in Fig. 1.
Table 8 Identified X-Y-X cycles of rotation frequency
Block A
Xi - Yi - Xi
0-100-0
35-100-35
50-100-50
80-100-80
85-100-85
35-85-35
i
1
2
3
4
5
6
Ni
1
3
1
2
1
1
i
1
2
3
4
5
6
7
8
Block B
Xi - Yi - Xi
0-100-0
70-100-70
75-100-75
80-100-80
85-100-85
70-95-70
75-90-75
70-85-70
Ni
1
3
2
1
3
1
1
1
Prescribed service life of observed disk, which because of premature cracks

can not be reached, amounts 1200 flight hours. Practical service life as crack initiation
life expressed in flight hours (CILh), for Weibulls probabilities 0.001 and 0.999,
amounts 28 and 676 flight hours [1]. Material for workmanship of disk is steel
13H11N2V2MF. Achieved quality of this steel is questionable. Because of that, the
question was asked. Whether the disk made of QT steel 13H1N2V2MF with cyclic
properties in Table 1, can have a longer CIL?
106
3.3 Stress-strain response at critical disk point

For determining of stress-strain response at critical disk point, it was enough to
observe one blade and critical region of our disk (at the first as ideal elastic bodies).
Linear stress response and nodal reactions at blade root contact surfaces, in
conditions of maximal rotation frequency n=186 s-1, were obtained using the finite
element method (FEM) [2]. Using FEM and the same rotation frequency, mentioned
reactions in transformed form used as nodal forces for obtaining of linear stress
response of critical region of disk. Axysimmetric linear stress response of disk, when it
observed as blisk (bladed disk), was obtained also (Fig. 4).
Fig. 4 Linear stress response of critical region of disk (left) and linear
axisymmtrical stress response of disk when it observed as blisk (right)
According to Fig. 4, maximal equivalent stress Veq,max at point of expected crack
initiation (critical point P) and belonging strain are unreal. It can see that critical point P
corresponds to real crack initiation. Equivalent stress at point P of blisk, which
corresponds to critical disk point P, taken as nominal stress Vn, was served for
calculation of so-called equivalent stress concentration factor Keq = 7.45 which is
defined as ratio of eq,max and n.
Similar, as in case of the flat specimen with central hole, nonlinear stress-strain
response at critical disk point was determined using Sonsino-Birgers modification of
Neubers rule based on application of the system of equations (4) and (5) in which
instead of Kt used Keq. The values of corresponding cyclic properties of QT steel
13H11N2V2MF were the same. Nominal stresses ni and their ranges ni which are
used in mentioned systems, were obtained using expressions
2
Y
223 i
100
Y 2 X 2
ni 223 i i
100 100
i iA 1,2,...,6/ i iB 1,2,...,8
ni
(10)
Numerical results of nonlinear stress-strain response at critical disk point,

provoked by blocks of rotation frequency A and B, are included in Table 9 and Table 10.
107
Table 9 Numerical results of nonlinear stress-strain response at critical disk point,

provoked by block A
Vmi
Xi-Yi-Xi
[%]
0-100-0
35-100-35
50-100-50
80-100-80
85-100-85
35-85-35
i
1
2
3
4
5
6
[MPa]
112.778
157.173
235.722
555.780
624.315
-42.239
'Vi
[MPa]
1484.231
1395.440
1238.343
598.227
461.157
996.615
'Hi
0.00768147
0.00650403
0.00545603
0.00261026
0.00201214
0.00434976
Table 10 Numerical results of nonlinear stress-strain response at critical disk point,

provoked by block B
Vmi
Xi-Yi-Xi
[MPa]
112.778
431.368
491.333
555.780
624.315
350.542
410.124
200.794
[%]
1
2
3
4
5
6
7
8
0-100-0
70-100-70
75-100-75
80-100-80
85-100-85
70-95-70
75-90-75
70-85-70
'Vi
[MPa]
1484.231
847.050
727.121
598.227
461.157
685.396
564.704
385.902
'Hi
0.00768147
0.00369601
0.00317266
0.00261026
0.00201214
0.00299058
0.00246396
0.00168384
3.4 Disk damages and crack initiation life estimation

Estimation of damages DA and DB provoked by blocks of rotation frequency A
and B, of observed disk, was carried out using Palmgren-Miners rule in the form
6
DA
DB
Ni
fi A
1
D N
i A
i 1
Ni

N
1 fi B
(11)
D
i B
i 1
Numbers Ni in (11) we taken from Table 8 and numbers Nfi determined using
system (7), in which index i taken values i=iA=1,...,6 and i=iB=1,...,8.
Resource between the two overhauls of R25-300 turbojet engine amounts 400
flight hours [1]. The different engine ground controls in that resource is covered with
438 blocks A and the different flights with 685 bocks B of rotation frequency. On the
base of this fact, simple expressions for estimation of damages D400h and D1h were
developed.
108
'

'

K

'
K

K

'

' '
(12)
K

Damage D400h is damage accumulated between two overhauls and D1h is the one flight
hour damage.
From (12), using values for DA and DB, we obtained values for damages D400h,
D1h and estimated crack initiation life expressed in flight hours (CILh). The data about
DA, DB, D400h, D1h and CILh = 1/D1h are contained in Table 11.
Table 11 The data about DA, DB, D400h, D1h and CILh
DA
0,00060924
438DA
0,26684543
DB
0,00025664
685DB
0,175796791
D1h =
CILh =
D400h
0,44264222
0,00110661
903
In Fig. 5, estimated CILh = 903 flight hours is compared with practical CIL.
Fig. 5 Practical crack initiation life (CIL) and estimated crack initiation life
(CILh) of observed disk
Estimated CILh = 903 flight hours is greater than practical CIL = 676 flight hours
that corresponds to Weibull's probability P(t) = 0.999 [1].
109
3. CONCLUSION
Methodology of crack initiation life estimation, applied for specimens with
central hole and for observed disk, can be applied in the similar form, for all notched
metallic part expossed to low cucle fatigue.
Because of the use of efective stress concentration factor Kf in the endurance
limit area and because that the same, in the finite endurance are, depends on the
number of load cycles N, application of Topper's modification of Neuber's rule, for
determination of nonlinear stress-strain response, should be rejected.
On the example of the first stage low pressure compressor rotor disk of R25300 turbojet engine it is shown that with cyclic material properties we can influence on
crack initiation life.
Research in relate to crack initiation life of notched metallic parts are not
finshed. The question of accuracy of crack initiation life estimation and today is open.
There are many unsolved problems. For example, application of linear damage
accumulation is questionable, because it does not take into account interaction
between cycles and order of their appearance. On the estimation accuracy of crack
initiation life, accuracy of nonlinear stress-strain response, at critical points,
significantly influence. Here is accuracy increased by introducing of Sonsino-Birgers
curve that is more accurate than Neubers hyperbola.
LITERATURE
[1]
Posavljak, S. (2008). Fatigue Life Investigation of Aero Engine Rotating Disks,

Doctoral dissertation (in Serbian), University of Belgrade, Faculty of Mechanical
Engineering.
[2] Lawry M. H. (1998).
I-DEAS Master Series, Mechanical CAE/CAD/CAM
Software, Student Guide, Structural Dynamics Research Corporation, SDRC Part
Number P-60002
[3] Bannantine, J.A., Comer, J., Handrock, J. (1990). Fundamentals of Material
Fatigue Analysis, Prentice-Hall, Englewood Clifs, New Jersy.
[4] Topper T. H., Wetzel R. M., Morroe JoDean (1969). Neubers Rule Applied to
Fatigue of Notched Specimens, Journal of Materials, JMLSA, Vol 4, No. 1, pp.
200-209.
[5] Sonsino, C.M. (1993). Zur Bewertung des Schwingfestigkeitsverhaltens von
Bauteilen mit Hilife rtlicher Beanspruchungen, Konstruktion 45, 25-33.
[6] Birger, I. A. (1985). Prognozirovanie resursa pri malociklovoj ustalosti, Problemy
prochnosti, No 10, 39-44.
[7] Jankovic, M. (2001). Low Cucle Fatigue (in Serbian), University of Belgrade,
Faculty of Mechanical Engineering.
[8] Palmgren, A. (1924). Die Lebensdauer von Kugellagern, Verfahenstechnik,
Berlin, 68, 339-341.
[9] Miner, M. A. (1945). Cumulative Damage in Fatigue, Journal of Applied
Mechanics, 76, A159-164.
[10] Morrow, J. (1968). Fatigue Design Handbook, Advances in Fatigue, Vol. 4.
Society of Automotive Engineers, Warrendale, Pa., Sec 3.2, pp. 21-29.
110
DESIGN IMPROVEMENTS OF THE BUCKET WHEEL WITH DRIVE

Sran Bonjak1, Zoran Petkovi2, Milo orevi3,
Neboja Gnjatovi4, Nenad Zrni5
Summary: Generaly, there are three basic reasons for the bucket wheel excavators
subsystems redesign: (1) better customization of the machine versus operating
conditions; (2) easier maintenance and (3) failure of the substructure under
consideration. The improved design solutions, outlined in this paper, were developed
to satisfy the first and second reason. Namely, an obsolete conception as well as
constant perrenial exploitation in heavy duty conditions made it necessary to redesign
the excavating device of the bucket wheel excavator SchRs 350. Several characteristic
problems which occur during exploitation and maintenance are herein discussed/listed.
This paper presents a concept of redesign which includes substitution of the existig
spur - geared bucket wheel gearbox with a planetary one and the redesign of the BW
body and its connection with the output gearbox shaft. It also shows the results of the
comparative load analysis for the original and the redesigned bucket wheel body.
Protection of the original bucket wheel drive of the excavator SchRs 1760 was realised
by magnetic powder clutches. Because of the relatively high price and also the
problems noticed during exploitation (e. g. overheating resulting in fusing of the
powder) this study developed a solution design of the bucket wheel drive with fluid
couplings. Thanks to the mentioned redesign, partial revitalizations of the bucket wheel
excavators are realised, leading to considerable increase of the machine's reliability as
well as the reliability of the complete surface mining system, and, additionally, making
the maintenance process easier.
Key words: bucket wheel, gearbox, overload clutch, external load identification,
redesign, FEM
1. INTRODUCTION
Bucket wheel excavators (BWE) present the backbone of the open pit coal
mining system. Speaking freely, the above class of machines is built up andaround the
bucket wheel (BW), which in turn realizes the basic function of the machine soil
1
Professor Sran Bonjak, Belgrade, Faculty of Mechanical Engineering, sbosnjak@mas.bg.ac.rs

Professor Zoran Petkovi, Belgrade, Faculty of Mechanical Engineering, zpetkovic@mas.bg.ac.rs
3
Researcher Milo orevi, Belgrade, Faculty of Mechanical Engineering , mddjordjevic@mas.bg.ac.rs
4
Teaching Assistant Neboja Gnjatovi, Faculty of Mechanical Engineering, ngnjatovic@mas.bg.ac.rs
5
Associate Professor Nenad Zrni, Faculty of Mechanical Engineering, nzrnic@mas.bg.ac.rs
2
111
Sran Bonjak, Zoran Petkovi, Milo orevi, Neboja Gnjatovi, Nenad Zrni
excavation. The design of the BW and its drive determines basic BWE performances
above all, capacity. As stated in [1] design errors in these areas would be difficult and
expensive to correct, if at all, after the machine has been constructed.
From a historical point of view, BWs have been developed from the cell-type
(which was. up untill the mid-1950s, the only BW design type in use), through the halfcell type (introduced in the mid-1960s), to the present cell-less type which became the
prefered BW design.
During almost a century of BW improvement, the BW gearboxes were
improved as well.. Thanks to their compactness and relatively small weight, planetary
gears pushed out the spur geared gearboxes, which have been used predominantly
until then.
Besides BWs and gearboxes, under permanent improvement are also the
concepts of their bearings as well as the design of supporting elements and power
transmission parts [2].
The natural tendency to constantly improve the performance of the BWE,
especially its capacities, has not been followed by adequate calculation methods and
production technology. Good proof of this is the relatively frequent damage of BWE
subsystems [3-7]. Regardless of the cause, failures of high performance machines, of
which BWE is one, always lead to very high financial losses [8,9].
Generally, there are three basic reasons for the BWE redesign: (1) better
customization of the machine versus operating conditions; (2) easier maintenance and
(3) failure of the studied substructure.
This paper will present several BW with drive redesign solutions developed by
the University of Belgrade - Faculty of Mechanical Engineering.
2. CASE 1 BWE SchRs 350
An obsolete conception and constant perrenial exploitation in heavy duty
conditions made it necessary to redesign the excavating device of the BWE SchRs
350, Fig. 1(a), which was put into service in 1961 [10].
One bedding of the BW shaft (bearing mark 23076K) is placed between BW
and gearbox, Fig. 1(b). In order to enable replacement of the mentioned bearing it is
necessary to dismantle the complete gearbox, including the output shaft gear,
Fig. 1(c), which is a relatively long and complex procedure. First of all, the machine
should be disburdened by reclining the BW boom, counterweight and discharging
boom, Fig. 2(a). The dismantling of the BW gearbox should be done only upon
completing the previous. Because of the structure complexity, it is necessary to open
the gearbox, Fig. 1(c), and do its succesive dissasembling. Only then is it possible to
dismantle the gear from the output shaft, Fig. 1(c). After replacing the bearing , the
gear is to be heated before installation, Fig. 2(b). A minimum of 7 days is needed to
carry out the above procedure.
Low reliability, followed by frequently occurring breakdowns, leads to
considerable decrease of the BWE exploitation effects. Besides that, it was necessary
to do the following: (a) Build in a new BW gearbox of the planetary type, providing axial
dissasembling of the BW and its shaft connection; the existing BW drive enables a
-1
-1
-1
choice of BW shaft speed among the following values: 6 min , 7.5 min and 9 min ,
while the speed of the appropriate planetary gearbox output shaft is 8.2 min-1;
112
Design improvements of the bucket wheel with drive
(b) Redesign and manufacture the new BW, Figs 2(c) and 2(d), with a larger number of
buckets: instead of nB = 8 for the original , nB = 10 for the redesigned BW.
Changeability of number of buckets to catch the soil is the main cause of
changeability of external load caused by the resistance-to-excavation. The in-house
developed software RADBAG [6] was applied for analyzing the external load, Fig. 3,
and determining its dynamic characteristics caused by the change in the total number
of buckets. Based on the research results, it is conclusive that the redesigned
excavating device gives more favorable external loading.
In order to make the substitution of the BW shaft bearing easier and the
downtime of the machine and complete system substantially shorter, the connection
between the BW gearbox and the BW shaft is redesigned, Fig. 4.
Finally, the general view of the redesigned BW boom head is shown in Fig. 5.
Fig. 1 (a) BWE SchRs 350; (b) BW shaft bedding; (c) BW drive dismantling
Fig. 2 (a) BWE temporary supporting; (b) heating the output gear;
(c) 3D model of the redesigned BW body; (d) BW body FEA
Fig. 3 External load coordinates

113
Fig. 4 Redesigned connection of the BW gearbox and BW shaft
Fig. 5 Redesigned BW boom head

3. CASE 2 BWE SchRs 1760
The original solution of the BW drive is shown in Fig. 6. The BW gearbox (5) is
driven by two AC motors (1) of 550 kW each, which are connected to the input stages
by cardan shafts (2) and overload clutches (3) of the magnetic powder type. Because
of the relatively high price of overload clutches and the problems noticed during
perennial exploitation (e.g. overheating resulting in fusing of the powder), it was
necessary to substitute the above mentioned clutches with fluid couplings. The type of
fluid couplings (TE) is determined by the external load characteristics, Fig. 7, which are
identified by applying the procedure described in [6]. The nominal fluid coupling size
(866) is defined by the motor power and motor shaft nominal number of speed. The
drum of the existing double shoe brake is mounted on the output shaft of one of the
fluid couplings, Fig. 8. Installation of the fluid couplings in the BW drive increases
reliability and enables full machine overload protection.
114
Design improvements of the bucket wheel with drive
Fig. 6 BWE SchRs 1760 : 1 AC drive motor; 2 - cardan shaft; 3 - magnetic powder
clutch; 4 - brake; 5 BW gearbox; 6 - auxiliary drive motor; 7 - auxiliary drive brake
-2800
Mzr [kNm]
-2600
-2400
-2200
-2000
-1800
-1600
-1400
-1200
M[ ]
-70
-61
-53
-46
-40
-35
-30
-25
-20
-15
-11
-7
-2
11
15
20
25
30
35
40
46
53
61
70
Fig. 7 Dependence of the BW torque on the slewing platform position (angle )
Fig. 8 Fluid coupling: (a) without brake; (b) with brake

115
4. CONCLUSION
The presented concepts of the BWE working devices' redesign eliminate the
original design drawbacks which deal with the accessibility and possibility of repair or
substitution of the damaged subassemblies as well as with safety against overloading.
Besides, they are in full agreement with contemporary BWE excavating device
conceptions. Finally, thanks to the presented redesign, partial revitalizations of the
bucket wheel excavators were realised, leading to a considerable increase of the
machine's reliability as well as the reliability of the complete surface mining system
while at the same time making the maintenance process easier.
LITERATURE
[1] Durst, W., Vogt, W. (1989). Bucket Wheel Excavator. Trans Tech Publications,
Clausthal-Zellerfeld.
[2] Gnilke, M. Intelligent retrofit solutions for bucket wheel excavators - Review of
recent projects in central - and eastern Europe, from http://www.baumaschine.de,
[3] Rusiski, E., Czmochowski, J., Iluk, A., Kowalczyk, M. (2010). An analysis of the
causes of a BWE counterweight boom support fracture. Engineering Failure
Analysis, vol. 17, no. 1, p. 179-191.
[4] Rusiski, E.; Harnatkiewicz, P.; Kowalczyk, M., Moczko, P. (2010). Examination of
the causes of a bucket wheel fracture in a bucket wheel excavator, Engineering
Failure Analysis, vol. 17, no. 6, p. 1300-1312.
[5] Bonjak, S., Zrni, N., Simonovi, A., Momilovi, D. (2009). Failure analysis of the
end eye connection of the bucket wheel excavator portal tie-rod support.
Engineering Failure Analysis, vol. 16, no. 3, p. 740-750.
[6] Bonjak, S., Petkovi, Z., Zrni, N., Simi, G., Simonovi, A. (2009). Cracks, repair
and reconstruction of bucket wheel excavator slewing platform. Engineering
[7] Bonjak, S.; Petkovi, Z.; Zrni, N.; Panteli, M., Obradovi, A. (2010). Failure
analysis and redesign of the bucket wheel excavator two-wheel bogie, Engineering
[8] Dreyer, E. (1995). Cost-effective prevention of equipment failure in the mining
industry. International Journal of Pressure Vessels and Piping, vol. 61, no. 2 - 3,
p. 329-347.
[9] de Castro, P.M.S.T., Fernandes, A. A. (2004). Methodologies for failure analysis: a
critical survey. Material and Design, vol. 25, no. 2, p. 117-123.
[10] Bonjak, S., Petkovi, Z., Gnjatovi, N., orevi, M. (2009). Redesign of the
Bucket Wheel Excavating Device. MHCL09 Conference Proceedings, p. 123-128.
Acknowledgements
A part of this work is a contribution to the Ministry of Science and Technological
Development of Serbia funded project TR 35006.
116
DESIGN OF TESLA-TIFFANY CASCADE FOUNTAIN AS A SAMPLE

OF TESLA`S RESEARCH CREATIVITY IN FIELD OF MECHANICAL
ENGINEERING
Aleksandar Marinkovi 1, Aleksandar oi 2, Bratislav Stojiljkovi 3, Milan
Vulievi 4
Summary: Nikola Tesla (1856-1943) made numerous inventions and discoveries in
various fields, from transmission of electrical energy, fluorescent lighting, wireless
transmission and remote control to numerous applications of high-frequency currents
in industry and medicine and peculiar inventions in the field of mechanical engineering
and aeronautics. One of his inventions which differs considerably from everything
else this great scientist and inventor dedicated the largest part of his creativity to was
a fountain. The abundant documents in the archives preserved in the Nikola Tesla
Museum in Belgrade demonstrate Teslas great engagement in this at first glance
less significant segment of his creative activity. This paper presents the original
design of Tesla-Tiffany cascade fountain, as a result of Teslas creativity in field of
mechanical engineering. Using the original sketches, technical drawings and some
calculations related to the dimensions of fountain elements, the authors analyze and
describe his designs of fluid propulsion and flow. In addition, the paper gives some
recent calculations as a comment and contribution to Teslas original design, in order
to present some modifications and improvements in his construction. CAD models
based on original variants of fountain and its components are also presented.
Key words: Nikola Tesla, Mechanical Engineering, Design, Cascade Fountain
1.
INTRODUCTION
Teslas abundant and interesting creative activity in the field of mechanical

engineering has only sporadically been quoted in the greatest number of biographical
depictions of his life and opus. Accordingly, such terms as e.g. turbines, pumps,
fountains, speedometers or aeroplanes, often act confusing even on a connoisseur,
1
Assistant Professor, Aleksandar Marinkovi, Belgrade, Mechanical Engineering Faculty University of

Belgrade, Serbia, amarinkovic@mas.bg.ac.rs
Teaching and Research Assistant, Aleksandar oi, Belgrade, Mechanical Engineering Faculty University
of Belgrade, Serbia, acocic@mas.bg.ac.rs
3
Museum Curator, Bratislav Stojiljkovi, Belgrade, Nikola Tesla Museum, Serbia, bratislav.stojiljkovic@teslamuseum.org
4
MSc student, Milan Vulievi, Belgrade, Mechanical Engineering Faculty University of Belgrade, Serbia,
v_milance@hotmail.com
2
117
Aleksandar Marinkovi, Aleksandar oi, Bratislav Stojiljkovi, Milan Vulievi
deemed to be at least averagely informed about Teslas inventions and discoveries.

Nikola Tesla devoted himself to machine engineering at the beginning of the 20th
century, at the time when his grandiose Long Island project failed, due to the lack of
funds. He desperately needed a new invention at the time, similar to the induction
motor, required by the industry, of a simple construction and cheap to produce, in the
order to fund his further research. Next twenty years of his life Tesla dedicated
exclusively to researches in the field of mechanical engineering, making also use of his
inventions in the field of alternating currents (motors and generators) in order to secure
the necessary funds [1].
Thanks to his specialist knowledge and exceptional creativity he was able to
solve some technical problems in mechanical engineering and to realize a wide range
of new inventions and constructive solutions based explicitly on physical laws. One of
Tesla's solutions in the field of mechanical engineering, less known until now, is his
patent for a fountain. This invention differs from customary solutions applied to
fountains and aquariums of the time, where the fluid was sprayed or jetted with
appropriate devices for decorative purposes [2].
2. TESLA'S FOUNTAIN AND COOPERATION WITH TIFFANY STUDIOS
The Tesla fountain, a completely new type for that time, had to introduce a new
type of beauty, which relied principally "on a fascinating spectacle of the large mass of
fluid in motion and the display of seemingly great power". Water or other fluid would be
lifted from the fountain receptacle to the horizontal flared out top of conduit, to overflow
it in the form of a unique circular cascade, making a splendid sight for an observers
eye. By applying only 1/25 of a horsepower (~30W) to the shaft and assuming a lift of
eighteen inches (~45.7cm), more than one hundred gallons (~378,5l) per minute could
be propelled to flow over the flared top of conduit, one foot in diameter with the depth
of the overflowing fluid of approximately one-half inch (~1.27cm). As the circulation
would be extremely rapid the total quantity of liquid required would be comparatively
small [3].
Fig. 1 The circular water cascade at the top of the overflow

The new type of fountain would allow for realization of beautiful and striking
views through additional colourful illumination and the disposition of voluminous
cascades, with relatively low consumption of energy. Teslas application for his model
of fountain was filed at the United States Patent Office on October 28, 1913, and the
letters patent No.1.113.716 was issued one year later, on October 13, 1914. Over
118
Design of Tesla-Tiffany Cascade Fountain as a Sample of Tesla`s Research Creativity in Field of

Mechanical Engineering
more than eight years, with shorter or longer breaks, Tesla developed numerous
models of fountains, but without significant commercial success. The preserved
archive documentation, e.g. general designs, technical drawings, calculations,
correspondence with companies and individuals, magazine articles, various brochures
and the like, provide an insight into the scope of his research in this field of mechanical
engineering.
After submitting his patent application for "certain new and useful
improvements in fountains" at the end of October 1913, Tesla started searching for the
ways and options to realize his invention. He offered the model of his cascade
fountain, based on the new principle, to Louis Tiffany, inventor of a special stained
glass technique, famous designer and the owner of Tiffany Studios. The review of the
preserved archive documents (letters from the correspondence between these two
important men and between Tesla and Tiffany's associates, as well as calculations,
drawings of the fountain and other documents), shows that this cooperation lasted from
December 4, 1913, to January 9, 1916 [4].
At the end of 1913 and the beginning of 1914 Tesla intensively cooperated with
Joseph Briggs, Manager of Tiffany Studios, on the development of particular structural
parts of the cascade fountain and the manner of its illumination. Joint venture was
formalized by signing a contract on business cooperation between Nikola Tesla and
Study Tiffany on February 17, 1914. For the Tesla Tiffany cascade fountain Tesla
elaborated a complete design of the plant and its components. He also specified
characteristics of the propelling motor and proposed a very interesting solution for
complete electric illumination.
Fig. 2 Vertical cross section of Tesla-Tiffany Cascade Fountain,

New York, December 22, 1913
For circulation of water and creation of the fascinating spectacle of a large
mass of fluid in motion, the classic propeller was planned, with blades directly fixed to
the driving shaft of electromotor. The driving assembly was placed vertically along the
central axis of fountain in the interior of the elevated conical overflow. In the design of
119
this fountain Tesla did not make use of his new principle of energy transmission from
the fluid and to the fluid i.e. of his invention of a turbine with bladeless discs, patented
on May 6, 1913.
For the illumination of the fountain and realization of a more beautiful and more
fascinating spectacle Tesla used different automatically controlled coloured lights.
Electric lighting of Tesla Tiffany cascade fountain consisted of two lighting systems
independent from one another, differing in number and power of electric bulbs used.
One system, installed on the outer side of the fountain, was composed of four sets, two
bulbs of different colours in each, distributed at an angle of 90 and placed vertically
under the top of the elevated conical overflow. Two sets had blue bulbs while the
remaining two were equipped with violet and yellow-green bulbs, respectively. The
other lighting system was placed in the interior of the fountain, under the glass bell of a
particular design. This system consisted of three bulbs of greater power, distributed at
an angle of 120 and installed horizontally above the driving assembly of the fountain
in the middle of the elevated conical overflow. A circular glass plate, composed of nine
equal segments, was placed above the horizontally built-in bulbs. Each one of these
segments was of stained glass: yellow-green, blue or violet. The glass segments of the
same colour were fixed at an angle of 120, thus forming a uniformly coloured set.
Alternation of colours of electrical lighting during the work of the fountain, as
requested by Louis Tiffany, had to be realized according to the following sequence:
beginning from the yellow-green the colour had to be changed into blue, from blue to
violet, from violet into blue and finally from blue into yellow-green. The time required for
these alternations had to be about three minutes.
Though the cooperation between NikolaTesla and Louis Tiffany ended without
the realization of planned projects and expected commercial success, it represents a
part of Teslas opus in the field of fountains and makes evident his broad interests,
extreme creativity and ease in solving complex engineering problems in his own,
original way.
3. CALCULATIONS OVERVIEW WITH 3D MODELS
Starting point for hydraulic calculation of fountain will be the weir height,
denoted as h in Fig 3. Flow over weirs is very complicated in nature because various
complex phenomena are present: streamlines curvature, recirculation flows, dominant
effect of inertial forces, etc. In this paper simplified hydraulic calculation is considered,
based on engineering principles which are simplified theory corrected by experimental
data [5]. Based on original drawings and fontain dimensions, we may assume that weir
height is h = 1 cm . Now the volume flow rate over weir is determined by following
equation:
Q Cd D S g h 3 / 2 ,
where Cd is discharge coefficient and g is gravitational acceleration. Discharge
coefficient is complex function and it depends on Reynolds and Webber number and
also on weir shape. It is determined experimentally and in literature values are known
for other types of weirs, like crested weir, V shaped weir, trapezoidal weir, etc. For type
of the weir on the fountain there are no data in literature, but based on other similar
types it can be estimated that its value is Cd = 0.85.
120
Design of Tesla-Tiffany Cascade Fountain as a Sample of Tesla`s Research Creativity in Field of

Mechanical Engineering
Fig. 3 Sketch of the weir of Fountain

Now it possible to calculate volume flow rate using previous equation and its
value is Q = 4.182 lit/s . With this flow rate average fluid velocity over weir can be
calculated using the equation:
Q
,
uav,1 { u1
DS h
and its value is u1 = 0.532m/s . Now the the pump head can be calculated
using famous Bernoulli equation, written for water level in lower reservoir of the
fountain and the weir. The equation is:
Hp
u2
H D 1 Hl
2g
u2
H K* 1 ,
2g
where D is kinetic energy correction factor (Coriolis coefficient) and Hl is sum

of all head losses. Head losses are usually expressed in terms of fluid kinetic energy,
like Ku12/2g, where K is head loss coefficient. Based on geometry of fountain and its
dimensions, and velocity distribution in the weir cross section it can be estimated that
K = D + K | 5, and it follows that value of pump head is Hp = 0.53m. With some
standard value of efficiency factor of pump Kp = 0.8 it can estimated that power needed
for work of the pump is P = U g Hp Q /Kp = 30W. It can be concluded that power
needed for pump drive is relatively low, which is expected if we have in mind that lift of
the fountain is only H = 0.51m. Based on data for flow rate and pump head we can
choose appropriate pump for this fountain from catalogs of pump manufacturers.
Concerning the calculation of transmission mechanism for lightening system,
we can say that main goal was to reduce nominal rotational speed of electric motor
from nem = 1150min-1 to requested nex = 0,28min-1 for driving of lighting system. For this
purpose Tesla had constructed a transmission system that consists of four stage gear
reduction, with 96 and 12 tooth cylindrical gears and gear ratio i = 8 for each stage.
Taking into account a friction torque loss during the rotation of lightening mechanism,
we could calculate electric power for driving this mechanism [6]. Preliminary calculation
of electric power needed for the transmission system gives us insignificant energy loss
(less than a 1 mW) compared with hydraulic part, due to a very small dimensions and
light gears material such as negligible external load.
At the end of this paper, the Fountain CAD model was made using CATIA V5
software tool [7], where complete Assembly such as driving mechanism were designed
as shown in Fig. 4.
121
Fig. 4 CAD Model of the Fountain and driving system with electric motor
4.
CONCLUSION
This paper is an attempt aimed to present the Tesla-Tiffany Cascade Fountain

as an example of Teslas reseach creativity in field of Mechanical Engineering. Based
on materials from Nikola Tesla Museum with original sketches, technical drawings and
some calculations related to the dimensions of fountain elements, authors analyse the
last variant of fountain and its components focused on calculation and energy loss
aspects of its operating. Even it was not the main objective of the paper, here is
presented the CAD model of the Fountain where its assembly with driving and
transmission mechanism could be observed. For further investigation of this Teslas
inventions in field of Mechanical Engineering, authors attend to continue analysis of
few other transmission and shape variations in order to present some modifications
and improvements in his construction and also to made CAD models based on original
variants of the Fountain.
LITERATURE
[1] B. Jovanovic (2001). Tesla Spirit, Work, Vision, Freemental, Belgrade.
[2] Tesla III Millennium, The Fifth International Conference, Proceedings, 1518.10.1996. Belgrade,
[3] Nikola Tesla. (1996). Patent IV, Institution for Textbooks and Teaching
instruments, Belgrade.
[4] Tesla's Fountain. (2007). Nikola Tesla Museum, Belgrade.
[5] White, F.M. (2003), Fluid Mechanics, McGraw Hill.
[6] Decker, K.H. (1995), Maschinen elemente. Hanser Verlag, 677 S.
[7] Hoffmann, M., Hack, O., Eickenberg, S. (2005). CAD/CAM mit CATIA V5. Carl
Hanser Verlag. Mnchen, Wien.
122
DESIGN SEALS FOR REAL CONNECTIONS

Svetislav Lj. Markovi1
Summary: In order to ensure non-leakage of surfaces, the joints are sealed with flat
seals of elastic material. Increase of the reliability of sealing is achieved by using the
seal coated firming. Soft material seals are used for joints made with rivets, hooks ...,
when the seal on connection suffers the additional burden. Setting the seal is based on
fixing the seals in parallel with surfaces that are to be sealed, so that a seal covers the
entire surface.
Key words: gaskets, joints, design
1.
INTRODUCTION
Sealing gaskets of soft material is always associated with relatively bigger or

smaller changes in distance between the parts which are fixed. In mechanical
engineering it is usually needed to seal compounds based on metal on metal
principle, while preserving the correct mutual arrangement of parts circuit.
When installing gaskets it is very important to take into account the proper
setting, centering gaskets. To increase the reliability of sealing, gaskets are lubricated
and sealants are used.
2.
SEALING OF SOLID COMPOUNDS
The task of sealing the solid compounds can be solved in several ways. Non
demountable and rarely separable joints are sealed with mixtures with sealing
capabilities: bakelite, liquid glass (sodium silicate).
Surfaces previously treated by ironing or by scratching, when connecting are
covered with a thin layer of sealant - lubricant. Oilers are usually produced by the color
of molten red hot oil (lead white, lead paint), iron powder or colloidal graphite in oil. It
also applies to suspension of colloidal graphite, covered with dry silver-plated graphite.
Another way of sealing the solid compounds consists of setting up a
submerged elastic seals of rectangular or circular cross-section of joint surfaces.
Gaskets are placed in the channels created along the entire rim of the joint. In the free
state, the gasket gets above the joint appearances of strictly fixed size (Fig. 1, left),
which depends on the material and the desired force of sealing. In the tensile state of
1
Dr Svetislav Lj. Markovi, prof., Visoka kola tehnikih strukovnih studija aak, Srbija,
e-mail: svetom@open.telekom.rs
123
the surface of the joint a contact occurrs, and the gasket material gets plastic or elastic
seal deformation and allows the sealing of the surface of the joint (Fig. 1, right).
Fig. 1 Sealing compound, metal on metal principle, the positions before and after
execution
To increase the tightness of the seal surfaces that are sealed, shallow
channels are carved up (Fig. 2) in which deformable gasket materiasl are set. For this
reason, a seal cam should be ridged (Figure 3). With tightening, the reefs get
compressed and form a red channel that acts like labyrinth seal.
Fig. 2 Sealing compound metal on metal principle, by use of the channels, the
positions before and after execution
124
Design seals for real connections
Fig. 3 Sealing compound metal on metal principle, with the cam seals, the positions
before and after execution
In Figure 4 it is presented the elastic sealing gasket set in an enclosed space,
which consist of the channel on the one side and protrusion on another surface. This
way of presentation is an advantage over the round edges, which consist of channels
and drains that must be made with the necessary degree of accuracy.
Fig. 4 Sealing compound metal on metal with round flanges

Round (circular) rims are sealed with the elastic metal rings (Fig. 5), usually Z type (so-called Gopher rings).
Fig. 5 Sealing compound by Gopher rings, the positions before and after execution
125
3.
TIN GASKETS
To ensure the tightness of the surface, the joints are usually sealed by tin mat
made of elastic material. Parts of the mechanical transmissions are assembled with
soft seals (in those cases where there is no necessity of exact mutual position of
parts). Gaskets of soft material are used for joints realized with the helping hand of
screws, bolt, hooks ... when seal suffers tensile loads.
There is a big application of steel-reinforced seals that include elastic materials
(rubber, plastics, asbestos ...), closed by a coating of soft metal (copper, brass). The
diversity of such seals is shown in Figure 6.
Fig. 6 Reinforced-reinforced seals

Figure 7 shows an example of application of reinforced gasket for sealing
threaded connections. Gaskets of this type can be used multiple times.
Fig. 7 Sealing of threaded continuation by reinforced seals

4.
SEALING COMPOUNDS WHICH CONTAIN LIQUIDS
Figures 8-15 show the sealing possibilities of cylindrical joints which are subject
to the pressure of fluid (a case of wet tubes of an internal combustion engine cooled
with liquid).
126
Design seals for real connections
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
The simplest form of sealing is achieved by placing a rubber ring round section
in the paper tube channel (Fig. 8). In the free state the ring withdraws above the
surface of the paper tube. In introducing paper tube in the water jacket, the ring is
compressed and reinforces the paper tube in the casing (lining). To ensure that sealing
127
compound is placed well, a few rings one after the other are placed (Fig. 9). The
improved construction of such a seal is shown in Figure 10. Here is a channel created
with a chamfered part directed to the side opposite of the of fluid pressure effect.
Under the pressure of fluid the rings are constantly pushing the narrowed part of the
channal and adhere onto the sealing surfaces with the force proportional to the
pressure. Among the rings there is a furrow which has a drainage hole connected to
the atmosphere (air). In case of breach of the first ring, the fluid leaks out through the
drain hole. The second ring which is, in a given case, relieved of the pressure,
prevents further penetration of liquids.
To increase the reliability of the working party of the sealing, several sealing
rings are set (Fig. 11). Different types of channels and rings are shown in Figure 12. In
these cases it is necessary to assure that the intersection channel is higher than the
intersection of the rings, otherwise the rubber as incompressible material can cause a
significant radial force and the occurrence of deformation at the seal. To provide a
compact grip of the rings to the wall jackets, on the inside of the rings wavy two-side
ring springs are set (Figure 13).
In order to avoid overloading of the sealing rings and to maintain a constant
pressure during the exploitation, in the compounds are introduced elastic elements in
the form of annular conical spring (Figure 14) or the cam ring springs of round cross
section (Figure 15).
In mechanical engineering it is often necessary to seal compounds which
contain liquids, and which are connected to each other by round windows and
openings. They are to be sealed by tin seals of elastic material. Compounds that are
exposed to high pressure and high temperatures are tightened by asbestos reinforced
seals (Figure 16).
Fig. 16 Copper-asbestos gasket

5.
CONCLUSION
Rubber is widely used for sealing joints. Mostly used is synthetic rubber which
is has many advantages over natural rubber. Some of the advantages of synthetic
rubber are acceptable resistance to oil, petrol, kerosene, high temperature, high
elasticity. Disadvantages of rubber can be repaired by its reinforcement (by pressing)
or vulcanization.
REFERENCES
[1] P. I. Orlov, Osnovi konstruirovanija 1, Mainostrojenije, Moskva, 1988.
[2] Markovi S.: Technological suitability of machinery parts, Proceedings 12th
International Research/Expert Conference Trends in the Development of
Machinery and Associated Technology TMT 2008, Istanbul, Turkey, 2008.
[3] Markovi S.: Razvoj oblika mainskih proizvoda u zavisnosti od tehnologije izrade,
Zbornik radova sa 32. savetovanja proizvodnog mainstva Srbije sa
meunarodnim ueem, Novi Sad, 2008.
128
DETECTION OF STRUCTURAL DAMAGE LOCATION USING

FREQUENCY RESPONSE FUNCTION DATA
Valentina Golubovi-Bugarski1, Drago Blagojevi2, ore ia3, Branislav
Sredanovi4
Summary: The method presented in this paper is designed to locate the damage in
the structure, using data from the frequency response function. The method is based
on the Generalized Damage Index. Functioning of method is demonstrated by
numerical example, where freely supported beam was analyzed.The experimental
examples were performed for two beam like structures: freely supported beam and
cantilever beam.The main advantages of the method are that it is not necessary to
undertake a modal identification, there is no need for any analytical or numerical model
of the structure, and it uses all measured data, in form of frequency response
functions, without further treatment.
Keywords: mechanical structure, damage location, modal analysis, frequency
response functions.
1
INTRODUCTION
Structural damage can be defined as changes introduced into a system that

adversely affect the current or future performance of that system. If one is focused on
the study of damage identification in structural and mechanical systems, than damage
may also be defined as any deviation in the structures original geometric or material
properties that may cause undesirable stresses, displacements or vibrations on the
structure. These weaknesses and deviations may be due to: cracks, loose bolts,
broken welds, corrosion, fatigue, etc. All of them should cause a decrease in the
structures stiffness, and some will also affect its mass and damping properties.
Therefore, structural damages should always, at a sufficient level of severity, cause a
change in a structures vibration behavior, described by modal properties: natural
frequencies, damping loss factor and mode shapes. Since the changes on the dynamic
characteristics can be measured and studied, it is possible to trace what structural
changes have caused the dynamic characteristic to change, thus identifying damage. It
is obvious that concept of damage detection is based on comparison between two
1
Doc. dr Valentina Golubovi-Bugarski, Banja Luka, Mainski fakultet Banja Luka, (valentina@urc.rs.ba )
2
Prof. dr Drago Blagojevi, Mainski fakultet Banja Luka, (dbl@urc.rs.ba )
3
Doc. dr ore ia, Mainski fakultet Banja Luka, (djordje@urc.rs.ba )
4
Branislav Sredanovi, Mainski fakultet Banja Luka
129
Valentina Golubovi-Bugarski, Drago Blagojevi, ore ia, Branislav Sredanovi
different states of the system, one of which is assumed to represent the initial and
often undamaged state. The structural damage detection can be divided into four
levels, based on the amount of information provided regarding the damage state.
According to Rytter [1] these four levels are: 1) identification of damage existence in a
structure; 2) identification of damage existence in a structure and location of damage;
3) identification of damage existence in a structure, location of damage and
quantification of damage severity; 4) identification of damage existence in a structure,
location of damage and quantification of damage severity and prediction of the
remaining service life of the structure.
2
DETECTION OF DAMAGE LOCATION
In practical damage evaluation it is difficult to excite the structural modes

(frequency) of a higher order, because the excitation of higher modes needs entering a
higher level of excitation energy, which is technically hardly feasible. Therefore, it is of
interest to study the techniques of evaluation of structural damage that require only a
few basic mode shapes and/or modal frequency. Such methods usually do not require
an analytical model of the structure, but only a few modal frequencies and modal
shapes that meet state structures before and after the occurrence of damage, which
can be obtained by dynamic tests. Sampaio and Maia [2] present such method based
on the Detection and Relative Damage Quantification indicator concerning the
detection and the relative severity of damage, as well as the Generalized Damage
Index for localization of damage. This method belongs to the class of methods that use
the change in the frequency response functions to detect, locate and relatively quantify
the damage. The main advantages of the method are: 1) it is not necessary to perform
modal identification; 2) there is no need for any analytical or numerical model of the
structure; 3) it uses all measured data in the form of frequency response functions,
without further treatment.
Theoretical basis of the method is derived from the equation of motion of a
multiple-degree of freedom system with hysteretical damping, which is often used in
describing of complex structures dynamics [3]. For such system, the system
receptance matrix, containing all inforamtion about the dynamic characteristic of the
system is given by:
>D Z @ >K iD @ Z 2 >M @1

Each element
(1)
D jk (Z ) of the matrix corresponds to an individual frequency
response function, FRF, describing the relation between the response at a particular
coordinate j and a single force excitation applied at coordinate k:
D jk (Z )
Xj
Fk
Fi
0, i 1...N ; i z k
The column vector, k, of the receptance matrix,
(2)
^D k Z `, describes the shape
(in space) exhibited by the structure at each excitation frequency Z, given by the
130
Detection of Structural Damage Location Using Frequency Response Function Data
responses normalized by the applied forces. When a structure is damaged its stiffness
and damping change and, in consequence, so does the receptance matrix:
> D (Z )@ > K i D @ Z >M @

d
1
(3)
where the superscript d stands for damaged. It is reasonable to assume that the
smaller the degree of correlation between the column vectors, ^D k (Z )` and
^D
d
(Z ) , the larger the damage.
The Generalized Damage Index, GDI, can be applied to the localization stage
in damage detection, that is level 2 according to Rytter. To localize a damage of the
structure, a comparison between shape vectors should be done. The Generalized
Damage index, based on the coefficient Ei(f) calculated from the curvature of the
frequency response function for each location p, the frequency fi and the level of
damage d, is defined as :
E d p, f
where
N
d
N
D icc f d D icc f d D icc f d D icc f D icc f D icc f
i 1
i1
(4)
N
N
d
d
D icc f D icc f D icc f D icc f D icc f D icc f
i 1
i 1
represents the conjugate operator and N is the total number of co-ordinates
or measuring points. The second derivate of the FRFs can be obtained by a central
difference approximation:
D icc f D i 1 f 2D i f D i 1 f
, i=p=(2... N-1)
(5)
As mentioned above, it is necessary to measure the FRF functions for two

different states of structure, one of which is assumed to represent the initial and often
undamaged state, while theotherstatecorrespondstoalevelofdamaged.
3
NUMERICAL EXAMPLE
To show the effectiveness of the proposed method, the model of a free beam
is used, made by the finite element software package ANSYS [4]. The model is made
of 160 beam elements, with 161 nodes and two degrees of freedom per each node, (y,
Tx). Beam dimensions are 400u10u10 mm (Lubuh), Young 's modulus is 2.1x1010
N/m2, the density is 7820 kg/m3, Poisson's ratio 0.3. The first four modal frequencies
and corresponding mode shapes of beam are obtained by modal analysis in the
frequency range of 1-3000 Hz, with frequency resolution of 1 Hz. The harmonic
analysis is done applying a random excitation of unit force at the node 30, while the
response to the given excitation is measured in every fourth node, that is in a total for
40 nodes the response is determined to a given excitation of structure, figure 1a. The
damage is simulated by reduction of the cross-section of 64-68th beam element, figure
131
1b. Six levels of damage are simulated: do is the undamaged beam, while the levels of
d1-d5 are damaged beam by 10%, 20%, 30%, 40% and 50% reduction of crosssection, respectively. The harmonic analysis procedure is repeated for each level of
damage. Using routines of Matlab software, the Generalized Damage Index is
calculated from the equation (4), for each of the six levels of damage d = (0,1,2,3,4,5),
four modal frequencies f = (f1, f2, f3, f4) and location p = (1,..,N-1). The calculated
values are graphically interpreted at figure 2, overlapping curves of GDI index for all six
level of damage.
Force
Damagedelements
30
d0
64
d1
d2
d3
d4
d5
Fig. 1 a) harmonic analysis of the beam, b) six levels of damage

From the graphic display of GDI it is obvious that the Generalized Damage
Index reliably indicates the location of 16 and 17, which correspond to the damaged
elements 64-68. Also may be noted that value of the GDI increases with increasing the
level of damage for the location of damaged elements, so that the GDI gives
information about the relative increase of damage.
Indikator oteenja GDI za pet nivoa oteenja
GDI
1.03
GDI
1.02
1.01
64
68
1
0.99
GDI_1
GDI_2
GDI_3
GDI_4
GDI_5
0.98
1
10 13 16 19 22 25 28 31 34 37 40
LokacijaLOCATION
(broj vora)
Fig. 2 The GDI indicates the location of damage for each level of damage
4
EXPERIMENTAL EXAMPLES
For the experimental analysis we used two steel beams of dimensions 400 u
10 u 10 mm [4]. One beam was suspended by common strings to simulate free-free
conditions; the other was clamped to simulate the cantilever beam, figure 3. On the
freely supported beam there were 19 uniformly arranged measuring points, while on
the cantilever beam there were 15 measuring points. The damage was simulated
introducing a cut at a certain location of the beam, figure 4. The propagation of
damage was simulated by deepening the cut for 1 mm at each of 7 (5) levels of
damage for each beam.
132
Detection of Structural Damage Location Using Frequency Response Function Data
Fig. 3 Measuring equipment and beam boundary conditions:

a) the freely supported beam, b) the cantilever beam
The test beams were excited by modal hammer (type Endevco 2302-10) in
each of 19 (15) nodes. Accelerometer (Bruel & Kjaer type 4507), which caught the
signal response, was set as showed at figure 4. Excitation and response signals were
measured by the measuring system PULSE (Bruel & Kjaer type 3560 C) in the
frequency range 0 y 3200 Hz, and processed in Pulse Labshop 9.0 software. Modal
test was repeated for each level of damage.
12345678910111213141516
1413121110987654321
Fig. 4 Models of test beams with position of cut, measuring points and accelerometers:
a) freely supported beam, b) cantilever beam
After 10 modal tests of the freely supported beam had been done, all FRFs
measured at location 5 were overlapped, figure 5a. The cantilever beam was modal
tested for 7 times, and the overlapped FRFs measured for location 11 are showed at
figure 5b. It can be noted that there is some frequency shift due to increasing of the
damage, that is frequencies move to the left (decreasing) due to of decreasing of the
stiffness of the beam (when level of damage increasing).
For the purpose to locate the damage, the Cumulative GDI was calculated by
successive adding the values of GDI (eq. 4), for the each level of damage. Figure 6a
shows cumulative GDI indicating the location of damage between measurement
location 14 and 15 of free-free beam. Figure 6b shows cumulative GDI indicating the
location of damage between measurement location 4 and 5 of cantilever beam.
133
[(m/s)/N]
Frequency Response H1(odgovor,pobuda) - File (Magnitude)
D:\Valentina\DOKTORAT\G1\G1 bo1\Frequency Response H1(odgovor,pobuda) - Mark
OVERLOAD
10k
[(m/s)/N] Frequency Response H1(odgovor,pobuda) -50% (Magnitude)

\Valentina\DOKTORAT\U
OVERLOAD
k3\k3 50%\Frequency Response H1(odgovor,pobuda) 1k
1k
100
100
10
10
1
100m
1
0
400
800
1.2k
1.6k
[Hz]
2k
2.4k
2.8k
3.2k
400
800
1.2k
1.6k
[Hz]
2k
2.4k
2.8k
3.2k
11
12
Fig. 5 Overlapped FRFs: a) free-free beam, b) cantilever beam

K3 suma sukcesivnih nivoa oteenja
G2
Indikator GDI-suma sukcesivnih nivoa oteenja
1.06
15
14
1.02
GD I
GDI
1.04
1
0.98
0.96
1
10
Lokacija (broj vora)
11
12
13
14
15
16
1.08
1.06
1.04
1.02
1
0.98
0.96
0.94
0.92
0.9
10
Lokacija
Fig. 6 The GDI indicates location of damage: a) free-free beam, b) cantilever beam
5
CONCLUSION
This paper presents one method for detection of damage location using
frequency response functions data. The results of numerical example and experiments
show that the Generalized Damage Index, GDI, is able to detect damage location on
the structure. As certain measurement inaccuracies happened on the certain locations
during testing it could be averaged by successive adding the values of GDI for the
each level of damage. However, it is supposed that GDI should increase continuously
on the location on damage. So, some improvement of the GDI indicator is proposed
that is the Cumulative GDI. Proposed damage detection method showed good
performance even for the hammer excitation and one response transducer available.
LITERATURE
[1]
[2]
[3]
[4]
134
Rytter, A. (1993). Vibration based inspection of civil engineering structures, Ph.D.

Dissertation, Department of Building Technology and Structural Engineering,
AalborgUniversity, Denamrk.
Sampaio, R.P.C., Maia, N.M.M. (2008). Strategies for an efficient indicator of
structural damage, Mechanical system and Signal processing, Vol.22.
Maia, N.M.M, Silva, J.M.M. et al. (1997). Theoretical and Experimental modal
Analysis, Research studies Press Ltd.
Golubovi-Bugarski, V. (2010). Models of correlation between structural damage
and dynamics response of mechanical structure, Ph.D. Dissertation, University of
Banja Luka, B&H.
DETERMINATION OF RESIDUAL STRESSES IN TUBULAR WELDED

STRUCTURAL COMPONENTS
Dragi Stamenkovi1, Mato Peri2
Summary: Numerical simulation of welding process with an example of two tubes of
both same and different steel materials properties have been carried out in this paper.
It is shown that the residual stress field and temperature field are symmetric in case of
welding same materials, while in case of welding materials with different thermomechanical properties they are asymmetric. So, it can be concluded that beside
thermo-mechanical properties of material, pipe diameter and pipe wall thickness can
also influence the asymmetry of residual stress and temperature field.
Key words: Welding process, Residual stress, Dissimilar butt-weld, Finite element
methods, Abaqus
1.
INTRODUCTION
Residual stresses and distortion are common unavoidable adverse

consequences of welding process that appear due to localized heating and subsequent
rapid cooling material. High residual stresses in regions near the weld can cause
fatigue, stress or fracture corrosion cracking. Therefore, estimating the magnitude and
distribution of welding residual stresses is of great importance.
In the early days of exploration of residual stresses and deformations in the
welded structures, we relied on experimental methods: X-ray diffraction [1, 2], hole
drilling [3], ultrasonic [4] and cracking [5]. In order to estimate residual stresses, in the
last 30 years, finite element method has been used, by which is tried to substitute a
rather expensive experiment [6-15]. Analytical methods remained limited to simpler
problems. On the example of welding two tubes with the same material properties,
Teng [6] proved that pipe diameter and wall thickness influence the size and
distribution of residual stresses in welding of steel pipe.
The welding of pipes and plates of steel with different thermo-mechanical
properties is used in the case of constructing power plants and ships, which can be
very well simulated using numerical methods. In the example of welding two tubes of
size 219 x 8.18 mm, and with the V shape of weld groove, made from carbon and
alloy steel, Akbari [7] showed that the temperature fields and residual stress fields are
asymmetric.
1
2
MSc, Dragi Stamenkovi, Belgrade, Termoelektro d.o.o., dragi33@gmail.com

MSc, Mato Peri, Zagreb, Bureau of Energetics and Mechanical Engineering, mato.peric@bestprojekt.hr
135
This paper presents two numerical simulations of welding two steel pipes of
dimensions 324.0 x 3.96 mm with the constant width of the weld of 3.84 mm
throughout the whole tube wall, model A and model B. In the model A the welding of
two pipes of the same carbonate steel was simulated, whilst in the model B the welding
of tubes of carbon and stainless steel is stimulated.
2.
FINITE ELEMENT ANALYSIS APPROACH
The numerical simulations performed in this work use sequential thermalstress solution procedure in which the transient heat transfer analysis is followed by
the thermal stress analysis. Temperatures predicted by the heat transfer analysis are
used as the loading for thermal stress analysis.
2.1.1. Thermal model
The governing equation for transient non-linear heat transfer analysis is:
w wT w wT w wT
wT
k z
(1)
kx
k y
Q UC
wx wx wy wy wz wz
wt
where, k x , k y , k z thermal conductivity in the x , y and z respectively, T is the
current temperature, Q is the heat generation, U is the density, C is the specific
heat capacity and t is the time, respectively. General solution of equation (1) is
obtained introducing the initial and boundary conditions, as follows:
Initial condition:
T ( x, y, z ,0) T0 ( x, y, z )
(2)
Boundary conditions:
wT
wT
wT
k x
qs hc (T Tf ) hr (T Tr ) 0
Nx ky
N y kz
(3)
wy
wz
wx
where, N x , N y , N z are the direction cosine normal to the boundary, hc , hr are the
convection and radiation heat transfer coefficients respectively, qs is the boundary
heat flux and Tr is the temperature of radiation heat source and Tf are the
surrounding temperature. Radiation heat losses are dominant near the weld and can
be expressed by equations:
VHF (T 2 Tr 2 )(T Tr )
8
where, V 5.67 u 10 J/(m2K4) is the Steffan-Boltzman constant, H
hr
(4)
is the effective
emissivity and F is the configuration factor. Moving away from the weld zone radiation
heat losses decrease and increase the share of convection heat loses hc . The total
heat input is given by:
q
136
KUI
VH
(5)
Determination of residual stresses in tubular welded structural components
where,
the arc efficiency, V is volume of the weld bead,
I is current, U is the arc
voltage and VH is volume of the activated weld bead element.

2.2. Mechanical model
The equilibrium equation can be expressed as:
V ij Ubi
V ij V ji
(6)
(7)
In equations (6) and (7)
V ij
is the stress tensor,
is density and
bi is the
body force. In numerical simulation, it was used thermal-plastic constitutive equations:
>D @>dH @ >C @dT

>dV @ >D @>dH @ >C @dT
where, >D @ is the elastic stiffness matrix, >D @ is the plastic stiffnes matrix, C
>dV @
ep
th
ep
th
> @
thermal stiffness matrix, dH

increment.
3.
(8)
(9)
th
is the
is the strain increment and dT is the temperature
BUTT-WELDED OF TWO PIPES
Two numerical simulations of welding were performed, model A and model B.

In the model A the welding of two pipes of the same carbon steel, A 106-B, is
simulated. In the model B the welding of tubes of carbon steel, A 106-B and stainless
steel, A240 TP304, is stimulated. Diameter of tubes are 324 mm, thickness of wall
3.96 mm and lenght of each tube 200 mm. Mechanical and thermal properties are
given in Figure 1, 2, 3 and 4. The material is modelled as an elastic ideally plastic.
Welding parameters chosen for this analysis are as follows: tungsten inert gas welding,
welding current I=110 A, welding voltage U=20 V and welding speed v=5 mm/s. The
following values are assumed: the convective heat transfer hc=15 W/m2K, the arc
efficiency K =70% and the emissivity = 0.8. Welding of pipes is modelled in single
pass. The specific heat flux input is q 2.076 u 1010 J/m3s. Width of welding bead is
3.84 mm and constant throughout the whole tube wall. The analyses are made with
standard method for modelling of weld. Three-dimensional mesh consists of 14400
elements. The same mesh is used both for thermal and mechanical analysis. In
thermal model DC3D8 elements are used, while for mechanical model C3D8 elements
are used, according to ABAQUS code [15]. The smallest elements whose size is 5 x
1.92 x 1.32 mm are used at the root of the welded joint. Geometry of welded pipes and
mechanical boundary conditions are shown in Figure 5.
137
1,8
1,6
Thermal conductivity
Density
Specific heat
Thermal Properties
1,4
1,2
C (J/g/C)
-2
U (10 g/mm )
0,8
0,6
-1
O (10 J/mm/s/C)
0,4
0,2
0
0
200
400
600
800
1000
Temperature (C)
1200
1400
1600
Fig. 1 Thermal properties of steel A106-B

300
Youngs Modulus (GPa)
Yield Stress (MPa)
Thermal Expansion Coefficient
Poisson ratio
Mechanical Properties
250
200
-7
D(10 m/m/K)
150
100
10-3Q
50
0
0
200
400
600
800
1000
1200
1400
1600
Temperature (C)
Fig. 2 Mechanical properties of steel A106-B

1,2
Thermal Properties
1
-2
3
U (10 g/mm )
0,8
Thermal conductivity
Specific heat
Density
C (J/g/C)
0,6
0,4
-1
O (10 J/mm/s/C)
0,2
0
0
200
400
600
800
1000
Temperature (C)
1200
1400
Fig. 3 Thermal properties of steel A240 TP304
138
1600
300
Youngs Modulus (GPa)

Yield Stress (MPa)
Thermal Expansion Coefficient
Poisson ratio
Mechanical Properties
250
200
-7
D(10 m/m/K)
150
100
10-3Q
50
0
0
200
400
600
800
1000
1200
1400
1600
Temperature (C)
Fig. 4 Mechanical properties of steel A240 TP304

400
200
200
weld
O324 x 3.96
y
z
A240 TP304
A106-B
0
start of welding
E
90
270
180
mechanical restraint in x, y and z direction

mechanical restraint in y and z direction
mechanical restraint in y direction
Fig. 5 Geometry of welded pipes and mechanical boundary conditions
139
4.
RESULTS
Temperature field 50 s after welding for model A and model B are shown in
Figures 6 and 7. Comparison of temperature profiles in z direction for model A and
model B, 50 s after welding is shown in Figures 8. Temperature profiles, shown in
Figures 8, are the same at inner and outer surface of tubes for Model A and Model B.
In Figures 9 and 10 are shown residual stress field , z , in z direction for Model A i
Model B. In Figures 11 and 12 are shown comparison of residual stresses in z
direction at inner and outer surfaces for E =180, Model A and Model B.
Fig. 6 Temperature field 50 s after welding, Model A
Fig. 7 Temperature field 50 s after welding, Model B
140
120
100
Model A
Temperature (C)
80
Model B
60
40
A106-B
A240
20
0
0
50
100
150
200
250
300
350
400
-20
Z - Coordinate (mm)
Fig. 8 Comparison of temperature profiles in z direction, E =180, Model A and Model B
Fig. 9 Residual stress in z in z direction, Model A
Fig. 10 Residual stress in z in z direction, Model B

141
250
200
Model A
Temperature (C)
150
Model B
100
50
A240
0
0
A106-B
50
100
150
200
250
300
350
400
-50
-100
Z - Coordinate (mm)
Fig. 11 Comparison of residual stresses z in z direction, at inner surface for E =180,

Model A and Model B
100
Model A
Temperature (C)
50
Model B
0
0
-50
50
100
150
200
250
A240
300
350
400
A106-B
-100
-150
-200
Z - Coordinate (mm)
Fig. 12 Comparison of residual stresses z in z direction, at outer surface for E =180,

Model A and Model B
5.
DISCUSSION AND CONCLUSION
Numerical simulation of welding two tubes was conducted, carbon steel A106B, A model, and after this the numerical simulation of welding two tubes made from
two different materials carbon steel A106-B and stainless steel A-240 TP304, model B.
The weld of the same width across the whole thickness of the tube wall was assumed.
Weld of temperature field of such a form becomes uniform along the thickness of the
tube at any point of time.
Model A: the symmetry of the temperature field was confirmed (Figures 6 and 8),
after this stress fields (Figures 9, 11 and 12) which coincides with the results in Ref [7].
Model B: the asymmetry of the temperature field was confirmed (Figures 7 i 8),
after this stress fields (Figures 10, 11 and 12).
Maximum tensile and compressive strain in z direction in welded tubes made
of carbon steel and stainless steel, at outer and inner surface of the tube are more
balanced than in Ref [7], which could be attributed to the uniformity of temperature field
along the thickness of the wall because of the constant width of the weld tube wall
thickness.
142
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
Norton JH, Rosenthal D. Stress measurement by X-ray diffraction. Proceedings

of the Society for Experimental Stress Analysis 1944;1(2):73-76.
Norton JH, Rosental D. Recent contribitions to the X-ray method in the field of
stress analysis. Proceedings of the Society for Experimental Stress Analysis
1947:5(1):71-77.
Pang HL, Pukas SR.Residual Stress Measurements in a crusi-form welded joint
using hole drilling and strain gauges. Strain 1989; February:7-14.
Chu SL, Peukhrt H, Schnieder E. Residual stress in a welded steeel plate and
their measurements using ultrasonic techniques. MRL Bull. Res. Dev.,
1987;1(2):45-50.
Masubuchi K, Martin DC. Invesigation of residual stresses by use of hydrogen
cracking. Welding J., 1961;40:553s-563s.
Teng TL, Chang PH. Three-dimensional thermomechanical analysis of
circumferentiallly welded thin-waled pipes. International Journal of Pressure
Vessels and Piping 75 (1998) 237-247.
Akbari D., Sattari-Far I. Effect of welding heat input on residual stresses in buttweld of dissimilar pipe joints. International Journal of Pressure Vessels and
Piping 86 (2009) 769-776.
Deng D. FEM prediction of welding residual stress and distortion in carbon steel
considering phase transformation effects. Materials and Design 30 (2009) 359366.
Deng D, Liang W, Murakawa H. Determination of welding deformation in fillet
welded joint by menas of numerical simulation and comparison with
experimental measurements. Journal of Materials Processing Technology 183
(2007) 219-225.
Deng D, Murakawa H., Liang W., Numerical simulation of welding distortion in
large structures. Computer methods in applied mechanics and engineering 196
(2007) 4613-4627.
Lee CH, Chang KH, Three-dimensional finite element simulation of residual
stresses in circumferential welds of steel pipe diameter effects, Material Science
and Engineering A (2008) 210-218
Peric M., Tonkovic Z., Karsaj I.: Numerical analysis ofresidual stresses using a
shell/3D modeling technique, Proceedings of the International Conference on
Advances in Welding Science and Technology for Construction, Energy and
Transportation AWST-2010, pp.75-80
Stamenkovic D., Vasovic I.: Finite Element Analysis of Residual Stresses in Butt
Welding Two Similar Plates, Scientific Technical Review, ISSN 1820-0206,
2009,Vol. LIX, No.1, pp.57-60.
Peric M., Stamenkovic D., Milkovic V.: Comparison of residual stresses in buttwelded plates using software packages abaqus and ansys, Scientific Technical
Review , ISSN 1820-0206, 2010,Vol.60,No.3-4, pp.22-26
ABAQUS/Standard, User's guide and theoretical manual, Version 6.8, Hibbit,
Karlsson & Serensen, Inc. (2008).
143
EFFECT OF MATERIAL PROPERTIES OF MEASURING FORCE

TRANSDUCER ELASTIC ELEMENTS TO METROLOGIY
CHARACTERISTICS
ivko Pejainovi1, Zorana Tanasi2, Goran Janji3
Abstract: An influence of caracteristics of the material of elastic element on metrologic
caracteristics of measuring force transducer is analysed in this paper. The most
important caracteristics of the materials and these elements are included and their
significance in fulfillment of the general functions and metrologic caracteristics of
measuring transducer are pointed out. Results of the analysis are the basis for choice
of materials and for design of the optimal shape of these elements, and consequently
make this process much faster and more efficient.
Keywords: force tranducer, elastic element, material of elastic element.
1.
INTRODUCTION
Measuring force transducers designed on the principle of strain gauge are the
serial connection of three transducer elements: elastic element, gauge and measuring
bridge.
F
Elastic
element
Strain
gauge
Measuring
bridge
Um
Fig. 1 Structural scheme of measuring force transducer

Elastic elements are first-primar transducing elements and the most important
mechanical parts of measuring force transducer. Input values of elastic elements are
forces (or torque, pressure etc.), and output values are displacements (linear or
angular) or strains.
Elastic elements are suitably shaped rigid bodies (rod, beam, ring etc.) on which
surfaces gauges are applied and connected to the respective bridging configuration.
Their function is to act as a reaction to measured force. In such activities elastic elements
are deformed along with applied gauges, electrical resistance of gauges changes, and in
1
ivko Pejainovi Ph.D. Assistant Prof., Banja Luka, Faculty of Mechanical Engineering, (zivkop@teol.net)
Zorana Tanasi Ph.D., Banja Luka, Faculty of Mechanical Engineering, (tami9997@teol.net)
3
Goran Janji MSc., Banja Luka, Faculty of Mechanical Engineering, (gjanjic@urc.rs.ba)
2
145
measuring diagonal of the bridge occur the changes equivalent to the measuring force.
Tensions, i.e. strains of elastic element must remain purely elastic, that is within the
Hookes law, and strong enaugh to provide an appropriate output signal.
Their main caracteristics are: linearity between forces and strains, small
histeresis, small creep, small tension relaxations etc. Achieving these characteristics
mainly depend on the structural design of elastic elements and properties of the
material from which it is made.
These caracteristics of elastic element which have the greatest influence on
metrologic caracteristics of measuring transducer are analysed in this paper. Analysis
includes mechanical and thermal characteristics and features related to the workability
of the material of elastic element.
2.
MECHANICAL CHARACTERISTICS OF MATERIAL
The most important mechanical caracteristic which predominantly affects the

accuracy or measuring force transducer is linear relationship between tension and
strain in area of material elasticity of elastic element. According to Hooke's law
relationship between tension and strain in this area is determined with expression:
V EH
(1)
where is:
V - tension in material,
E - module of material elasticity,
H - relative strain.
Linearity of expression 1 depends on material elasticity module of elastic

element. For the stress range corresponding to the rated load of elastic element,
material should have perfect linearity between stress and deformation. In addition to
material of elastic element, also other influences can cause the occurrence of
nonlinearity at the output of the transducer. In general, such materials that have stable
characteristics in a wide range of factors influencing changes are needed for elastic
elements construction. Such materials are rare in practice. Their characteristics to a
greater or lesser extent depend on the rate of change and the size of workload, time of
activity, vibration, temperature, etc.
Under the load, elastic element is deformed instantly, and then the
deformations continue to increase under the law of damping (Fig. 2a). After unloading,
the deformations of elastic element disappear, also with a time delay. This
phenomenon is called the material creep (with delayed elastic effect) and depends on
the speed of change of measured values. When loads are changing rapidly creeping is
not so prominent as in a short period of time fails to develop. The highest creep occurs
with medium frequency loads 20-200 Hz [6]. It is usually not manifested individually,
but together with mechanical hysteresis, resulting increase of hysteresis loops. It is
especially expressed at elevated temperatures.
Discrepancy between the curves depending on the stress and strain, obtained
by the loading and unloading elastic elements, creates a hysteresis loop (Fig. 2b).
Width of hysteresis loop of elastic elements for stresses in the range of rated load is
low. It increases with increasing stress so behaviour of material of elastic element
when it is loaded above the rated load must be considered.
146
Effect of material properties of measuring force transducer elastic elements to metrologiy characteristics
b
'H
a
'H
a.
b.
Fig. 2 Diagrams: a) creep, b) elastic hysteresis

The mechanical and thermal treatment of elastic elements should be strictly
adhere to the prescribed treatment regime. Long and over-heating of the elastic
element leads to local changes in mechanical properties as reflected by the increase in
hysteresis. In some cases, because of the importance of hysteresis, size of the
allowable stresses of elastic elements are not determined by the factor of safety
compared to the material yield strength, but based on the size of allowable hysteresis.
This applies to the elastic elements for high-class accuracy, and allowed stresses do
not exceed 0.1 02 [6].
Mechanical characteristics of materials such as the limit of proportionality,
elastic limit, yield strength and tensile strength is not related directly to the
characteristics of elastic elements. As the limits of proportionality and flexibility are
considered as conditional borders, allowed level of work load (and thus the maximum
size of mechanical signals) is determined by the limit of the material flow. Safety
factors for this limit are chosen in the range 0.3 to 0.5 (for elastic elements of the
higher class of accuracy to 0.1) [6]. The nominal load of elastic element that represents
the upper limit of measurement range is defined in this way. Elastic elements for force
measuring are required to withstand much heavier loads than the nominal. These
loads are called safe or secured load limits. The material must not happen any change
(plastic deformation, cracks) to the boundary of the safe load. The value of the safe
load limit depends on the type of elastic element and is 150-200% of rated load, or
300-500% compared to the limits of destruction of materials [2, 3, 4]. In order to
prevent breakage (especially in the areas of stress concentration) the material must
have sufficient flexibility. Therefore, the elastic elements rarely require higher hardness
of 50 HRC, because the materials with higher hardness prone to brittle fracture. If it is
certain that the safe load limit to be exceeded it is necessary to install some protection
against overload.
For elastic elements dynamicaly loaded with a large number of cycles, it is
necessary to take into consideration fatigue strength and regarding it to select the
appropriate safety factor. For most of elastic elements, fatigue strength is of secondary
importance, since the sensitivity of electro-strain gauges to fatigue is limiting constraint.
In some cases of choise of material of elastic element may be limited by
specific requirements for material density, temperature coefficient of linear expansion,
the corresponding relationship between elastic modulus and bulk density, etc.
147
3.
THERMAL PROPERTIES OF MATERIALS
500 0C
400 0C
300 0C
250 0C
200 0C
100 0C
Napon
0
20 C
Tension
Previously analyzed mechanic properties of materials used for making elastic

elements are related to standard operating conditions (room temperature without
aggressive and other influences). As the electro transducers with strain gauges can be
used in a wide temperature range and under different external influences, special
attention should be paid to the thermal properties of materials, since at extreme
temperatures, range of materials to create elastic elements is substantially limited.
Experimental studies indicate that mechanic properties of some metals (yield
strength and tensile strength) significantly depend on temperature. Figure 3 shows
several diagrams obtained during the investigation of carbon steel in tension at
different temperatures [5]. The diagram shows that the yield strength at higher
temperatures is less obvious and above 300C it can not be seen in the diagram.
According to the same diagrams, steel strength rises to 250C and then rapidly
decreases.
Specifino izduenje
specific strain
Fig. 3 Diagrams of tensile of carbon steel at different temperatures

As these sizes do not relate directly to the elastic elements, more interesting
are the first parts of these diagrams. According to these diagrams limit of
proportionality of steel decreases with increasing temperature. At the same time, the
slope of the diagram reduces and thus also the modulus of elasticity for 1-3% at 60 C
[4]. This information is very important because it directly influences to the linearity of
the materials in the field of elasticity according to the expression 1. Keeping this in
mind it is necessary to choose a material for which the elastic modulus in a certain
temperature range is linear and stable enough. In this case, the errors that occur due
to changes in elastic modulus can be taken into account by introducing corrections or
compensations, which is done only for elastic elements of higher class accuracy.
Another important requirement for the material of elastic element is the stability
of modulus in time. Research shows that the instability arising in the material after its
mechanic and thermal treatment is associated with the subsequent action and
relaxation. The lowest time-dependent instability modules of elasticity have materials
with no significant residual stress.
148
Effect of material properties of measuring force transducer elastic elements to metrologiy characteristics
By analyzing the thermal properties of the material of elastic elements, one

should bear in mind that the electro-strain gauge is a source of heat that is transferred
to the elastic element. Temperature gradients in the asymmetrical elements, especially
if they are unbalanced in relation to applied gauge, can cause interference at the
output. To minimize these temperature effects during the reaction of elastic elements,
the elements should be of symmetrical forms with respect to applied gauge.
The transducers in use are exposed to a larger or smaller temperature
changes. The impacts of these changes can be reduced by appropriate constructional
solutions. In this case, one must take into consideration heat flow path between the
casing of the transducer and the elastic element, and within the element. For any
technical solution, the change of temperature gradients within the element is inversely
proportional to the thermal conductivity of the material element. Therefore, the thermal
conductivity of the material of elastic element is important feature that must be taken
into consideration when selecting materials. The effects of thermal expansion of
materials of the first order at exit of elastic element are eliminated with selfcompensating gauges and full-bridge configuration. In addition, there are higher-order
effects arising from changes in the cross section dimensions, torque arm and the other
parameters, and is about 0.1 to 0.2% at 60C [4].
The analysis of thermal effects on the material of elastic element and its behavior
in such conditions shows that they are numerous and significantly affect the final results.
Thermal effects are successfully solved by applying these recommendations and using
some existing practical and very efficient methods of compensation of these effects.
4.
WORKABILITY OF MATERIAL OF ELASTIC ELEMENT
For selecting the material of elastic elements, in addition to mechanical and

thermal properties, an important properties are related to their workability. In the
production of elements the principle of monolithism should be respected. In this way
the equality of all influential factors on the characteristics of elastic elements is
ensured. All connections of any kind applied to the elastic elements, to a greater or
lesser degree cause a slight displacement and friction that can cause certain nonlinearity. Welded joints require special attention because they create residual stresses,
and metallurgical effects contribute to the appearance of micro-plasticity and limiting
endurance to fatigue.
Fig. 4 Elastic element of the two-component force measuring transducer

149
By adjusting the basic, relatively simple shapes of elastic elements to the task
of measuring, very complex configurations of elastic elements are formed (Fig. 4).
Keeping this in mind and respecting the previously mentioned principle, a complicated
and precise machining is a common requirement in the manufacture of elastic
elements. In this case, the workability of the material is an important feature to be
taken into account when selecting materials because it significantly affects the
performance and cost.
More materials suitable for elastic elements have good workability in the state
before heat treatment. The predicted mechanical properties are achieved by
subsequent heat treatment at appropriate temperatures and cooling in oil or water. In
elastic elements of complex shape with segments of various thicknesses, these
processes are accompanied by the occurrence of subsequent deformation of the
designed shape. Materials prone to such changes should be avoided in the selection.
5.
CONCLUSION
In the design of force transducers based on the principle of electro-resistant

strain gauges central place belongs to the design of their elastic elements. For
successful realization of this task, it is necessary to know the material properties of
elastic elements. The choice of material of elastic elements must be compatible with
commercial by type, quantity and usable form, and sometimes price. None of the
materials to create elastic element does not possess all necessary features so that the
choice represents a compromise between the existing properties. In the particular
case, the optimal material should be selected according to the specific priority of
features.
Summarizing the previously performed analysis it should be noted that any kind
of imperfection in the chosen material of elastic element has negative effect on the
metrological characteristics of the transducer in which an elastic element is embedded.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
150
Pejainovi ., Prilog optimalnom oblikovanju elastinih elemenata mjernih

pretvaraa sile u cilju poboljanja metrolokih karakteristika, Doktorska
disertacija, Mainski fakultet Banja Luka, 2005.
Gommola G., The Application and Installation of Load Cells, Spectris Messtechnik
GmbH. Frankfurt/Main, Zarbock, 2000.
Hoffmann K., An Introduction to Measurements using Strain Gages, HottingerBaldwin Messtechnik GmbH, Darmstadt,1989.
Measurements Group, Strain Gage Based Transducers Their Design and
Construction, The Technical Staff of Measurements Group, Inc North Carolina,
Releigh, 1988.
Timoenko S., Otpornost materijala II dio, Graevinska knjiga, Beograd, 1966.
Makarov R. A., Tenzometrija v mainostroenii, Mainostroenie, Moskva, 1975.
EVALUATION OF CONCEPTUAL SOLUTIONS OF UNIVERSAL

HELICAL TWO STAGE GEAR UNITS
Sinia Kuzmanovi1, Milan Rackov2
Summary: Universal gear reducers are very simple and quite improved mechanisms,
but despite of that their further development is constantly made in order to find better
solution. On the basis of the present solutions it can be note that almost every
manufacturer has its own conceptual solution of gear reducer. Different conceptual
solutions of universal gear reducers do not make a problem for customers because it is
not the point of their interest. They are only important to have quality gear unit,
primarily a strong, reliable, durable, with low-vibration and noise and that is cheap, also
they want short delivery time, low maintaining costs and the repair time is short.
However, manufacturers of gear are very important what kind of concept solution is
adopted and therefore it is necessary to evaluate some conceptual solutions during the
time of their development in order to adopt the best. This paper gives the evaluation of
particular solutions in order to realize the degree of goodness of some solutions.
Key words: evaluation, conceptual solution, universal gear unit, reducer
1. INTRODUCTION
Universal helical gear reducers are the mechanisms which have very wide
application in mechanical engineering. They are regularly bought as finalized product
and none of their costumers and users do not think about their manufacturing.
However, manufacturers of gear reducers are continuously working to improve the
technical characteristics of their products in order to achieve certain advantages at the
market. This paper deals with the affect of arrangement of pinions, gear wheels and
bearings inside of gear unit on the technical characteristics of universal helical gear
reducers.
Universal gear reducers are usually made as single, two, three stage and
multistage gear units. Some manufacturers do not produce single stage gear units
because they are not often required, and the same case is with multistage gear
reducers. Multistage gear units are usually made by combining of two and three stage
units, although there are manufacturers which make only multistage gear units in
special large housing with several gear pairs, but it is rare case.
Universal reducers can be made as geared motors with special reducer electric
1
2
Prof. DSc., Sinia Kuzmanovi, Novi Sad, Univ. of Novi Sad, Faculty of Technical Sciences, (kuzman@uns.ac.rs)
Ass. MSc., Milan Rackov, Novi Sad, Univ. of Novi Sad, Faculty of Technical Sciences, (racmil@uns.ac.rs)
151
motor or with standard IEC motor and they can be made without installed motor with
free input shaft or with IEC motors interface.
Gear reducer without installed motor is now very little used because of more
complex structures and difficult installation. It usually need another coupling (for
connecting motor and reducer), larger space for installation (because of those coupling
application), and there are additional difficulties with aligning input shaft of reducer and
motor shaft. However, manufacturers of gear reducers (due to market competition)
must also have those reducers in their offer.
2. CONCEPTUAL SOLUTIONS OF GEARS AND BEARINGS ARRANGEMENT
There are different practical conceptual solutions of gear unit design despite
the small possibilities of variation, arrangement and position of gears and bearings
within the gear reducer housing. These possibilities are the smallest for the single
stage gear reducers and the following conceptual solutions are possible here, which is
shown on Fig.1 and Fig.2. Two stage gear reducers are available as geared motors,
Fig.2-1 (with special reducer electric motor or with standard IEC motor), with an
adapter for IEC motors, Fig.2-2 (with one or two bearings) and without installed motor,
Fig.2-3 (with free input shaft). Geared motors with standard IEC motors are not
considered here, because most of the manufacturers do not offer this solution, and it
would only slightly expand the number of possible conceptual solutions. In this paper
only two stage gear reducer are analyzed.
Fig. 1 Possible conceptual solutions of universal single stage gear reducer

(1 geared motor, 2 adapter for IEC motor, 3 adapter with free input shaft)
3. PROBLEM INTERPRETATIONS
Based on the theory of possible conceptual solutions, it should adopt the best
solution, which is not so simple. In fact, almost every manufacturer of gear reducer has
a different conceptual solution of gear units, so that it is very difficult to determine
which solution is best.
The output shaft is usually designed that only one bearing accepts axial force
in order to make cheaper solution and make possible easier installation and removal of
the output shaft assembly.
Minimum consumption of material must be considered during design, the lower
volume of machining as possible and sufficient strength and stiffness of the housing,
with a simple assembly and removal of gear unit. Additionally, it is necessary to take
into account the possibility of using the same set of gears in the frame of particular size
of gear reducer. For example, gear pair from single stage gear unit is the first gear pair
of two and three stages gear reducer.
152
Evaluation of Conceptual Solutions of Universal Helical Two Stage Gear Units
1. Universal two stage gear reducer with special reducer motor installed
2. Universal two stage gear reducer with adapter for IEC motor or with standard IEC
motor
3. Universal two stage gear reducer with free input shaft (so-called B interface)
Fig. 2 Characteristic conceptual solutions of universal two stage gear reducer
Today there is a tendency to achieve a high gear ratio, so it means to use large
driven gear, which resulted in a need to make the frontal opening of the housing (and/or
the top opening for two and three stages gear reducer). It is necessary because the
standard frontal opening (which size depends on the electric motor flange) cannot be used
for installing so large gears. This concept housing, so called monoblock construction, is
somewhat weaker, but their possibility of simple installation gives them a certain
advantage. Large back cover of single stage gear reducer provides completely installation
of large gears, so there is no need to make opening on the top of gear unit.
Easy assembly and removal of gears must be provided when defining their
arrangement and position with smaller as possible overall dimensions of gear reducer.
Also, minimum use of special tools must be enabled because today fast delivery is
required (usually within 72 hours) and fast repair of gear reducer, which is usually done
in service centers near the major markets since fast delivery, service, maintenance and
repair are often a crucial factors for decision to purchase gear reducer. Generally, fast
and quality service is an essential support for gear reducer sales, and to ensure this
constructive gear reducer must be prepared for it.
4. THE PROPOSAL OF EVALUATION
Evaluation of conceptual solution can be done in many ways by a simple
comparison of so-called partial indicators of quality: the weight of gear reducer (m),
gear ratio (u), output torque (T2), etc. Also, complex indicators can be used: the ratio
153
between gear ratio and weight (u/m), ratio between output torque and weight (T2/m),
etc. In order to obtain more complete assessment of gear reducer, it is necessary to
introduce more complex indicators, such as uT2/m. It should be noted that this method
of assessment can evaluate only gear units with the same shaft height. In order to
obtain a more complete assessment, it is necessary to introduce the shaft height value
(h) and to considerate complex indicator (uT2/h) or more complex indicator (uT2/mh),
where the value of these indicators should be as higher.
Of course, evaluation can be done for gear reducer housings made of same
material (cast iron). If the materials are different, obtained values will not be
comparable, although obtained assessment can refer a lot about the benefits of
individual solutions. Also, it should note that some manufacturers are oriented to high
load capacity (high output torque) and small gear ratio, while others are oriented to
smaller output torque and high gear ratios. Additionally, some manufacturers offer two
gear sets in the frame of one housing (or the same shaft height): with small transmitted
torque and high gear ratio and the other with high transmitted torque and small gear
ratio, so that they can successfully compete with other manufacturers. With this
approach manufacturers slightly complicate their product range, but certainly achieve
great advantage at the market. In a rough assessment of these solutions it cannot
obtain enough accurate evaluation of the goodness of particular solutions, and
therefore it is necessary evaluate assessment for each gear ratio for the same shaft
height and compare it with other manufacturers solutions.
5. ASSESSMENT OF CHARACTERISTIC SOLUTIONS
This paper deals with assessment of technical characteristics of two stage
universal gear reducers (foot mounted type) of several leading manufacturers of gearboxes
which names will not be mentioned, so they will be marked as Manufacturer 1 to 6 (Fig.3).
Geared motors are not observed, but variant without installed motor with the classical solid
input shaft, in order to make accurately comparison and evaluation of gear units.
Man. 1
Man. 2
Man. 3
Man. 4
Man. 5
Man. 6
Fig. 3 Frontal view of universal single stage gear reducer

On the basis of analyzed solutions of these six manufacturers, it can be
concluded that the Manufacturers 1, 3 and 6 offer ten sizes of a two stage gearbox (ten
shaft heights), but for some shaft heights they offer two gear sets in the frame of one
housing: with small transmitted torque and high gear ratio and the other with high
transmitted torque and small gear ratio. It can be concluded the most required shaft
heights are between 90 and 160 mm, so they offer two gear sets and can successfully
compete with other manufacturers. Manufacturer 2 offers eleven sizes of gear units,
Manufacturer 4 offers eight and Manufacturer 5 offers seven shaft heights. The values
154
of output torque (T2), gear ratio (u), weight (m) and housing volume (V) for each shaft
height are shown in Fig.4.
T2 max,
Nm
u max
h, mm
m, kg
V, m
h, mm
h, mm
3
h, mm
Fig. 4 Diagrams of maximal output torque (T2), gear ratio (u), weight of gear reducer
(m) and housing volume (V) of universal two stage gear reducers with free input shaft
for different axis heights (h), Manufacturers 1 - 6
Analyzing obtained curves, it can be noticed that the increase of output torque
is fairly consistent for all manufacturers and it is especially higher for the Manufacturers
1 and 6 for greater shaft heights. Manufacturer 2 and 6 offer the highest gear ratio.
Furthermore, it is obvious that reducers of all manufacturers have almost constant
value of gear ratio, except the Manufacturer 2 which offers high gear ratio for shaft
height until 200 mm and after that it is reduced. Comparing the reducer weight, it can
be observed that the weight is evenly increased for all manufacturers and the reducers
of Manufacturer 2 have the highest weight for largest shaft heights. Increasing the
volume is also the same, but for higher shaft heights reducers of Manufacturers 1 and
6 have the largest overall dimensions. However, Manufacturer 6 reducers are
nonsymmetrical and they have lower volume than its calculated.
Since gearboxes are not standardized, manufacturers offer a variety of reducer
shaft heights. Shaft heights of these six manufacturers are not same, except for the
height of 250 mm, which suggests this is the most required shaft height for two stage
155
gear units. Also, similar shaft height is for 130 mm (125, 130 and 132 mm) and 315
mm (300 and 315 mm).
T2 max, Nm
umax
m, kg
V, m3
Fig. 5 Comparison of output torque (T2), gear ratio (u), weight of gear reducer (m) and
housing volume (V) of universal two stage gear reducers with free input shaft for the
shaft height h = 250 mm, Manufacturers 1 - 6
Therefore, the initial assessment is taken for shaft height of 250 mm which all
six manufacturers can provide. Comparison of output torques, gear ratios, weights and
volumes for shaft height of 250 mm is given in Fig.5.
Comparing the most important technical characteristics of gear reducers with
shaft height 250 mm, it is observed that the biggest output torque is provided by gear
reducer of Manufacturer 5 (T2max = 5000 Nm). The highest gear ratio is achieved by
reducer of Manufacturer 6 (umax = 62.04, three times higher than Manufacturer 3). Gear
unit of Manufacturer 5 has the smallest weight (m = 117.1 kg). Gear reducers of
Manufacturer 1 and 6 have the largest mass and volume. Gear unit of Manufacturer 2
have large weight and the smallest volume, which means the Manufacturer 2 has
better arranged and more compact design, but unfortunately with the lowest output
torque. Therefore, every reducer of these manufacturers has some typical
characteristic, so that according to these basic indicators it is impossible to adopt the
best solution.
Therefore, complex indicators are used for evaluation, such as the ratio of
output torque and weight (T2 max/m) or the ratio of gear ratio and weight (umax/m), Fig.6.
Comparing complex indicators of two stage gear units (Fig.6-1,2) of those six
manufacturers, it can be concluded that Manufacturer 5 has an advantage, because it
has the highest torque per weight unit and gear ratio value per weight unit. After them
reducers of Manuf. 3 and 6 have high torque per weight unit, but reducers of Manuf. 6
and 2 have high gear ratio per weight unit. However, after this comparison it is not
clear which manufacturer has better reducers after Manufacturer 5.
156
It should considerate more complex indicators such as umax T2 max / m to obtain

a more complete assessment of the quality of gear reducers, Fig.6-3. Based on
analysis for this indicator for shaft height 250 mm and consideration of achieving as
highest as possible output torque and gear ratio for the minimum mass of gearbox, it is
confirmed that gear units of Manufacturer 5 are the best, and after them there are gear
units of Manufacturer 6 which have the most rational utilization of material, much better
than other gear reducers. Therefore, it should adopt such kind of gear units with
maximum output torque and gear ratio per mass unit.
T2 max / m
u max / m
u max T2 max / m
Fig. 6 Comparison of complex characteristics of universal two stage gear reducers with
free input shaft for the shaft height h = 250 mm, Manufacturers 1 - 6
This evaluation has been performed from the viewpoint of design optimization,
which means obtained amount of technical parameters based on the designed weight
or volume. If considerate the price of gear unit, it would obtain more complex
assessment of gear reducer, but since the price is a market category, it will not be
discussed here.
u max T2 max /
mh
u max T2 max /
mV
h, mm
h, mm
Fig. 7 Complex indicators of technical characteristics quality of gear reducers for

different shaft heights of universal two stage gear reducers with free input shaft
157
This method of assessment can be used only for gear units with the same shaft
height, so the evaluation for the shaft height of 250 mm cannot guarantee the same
conclusion for other shaft heights. Introducing the shaft height values is necessary to
obtain more complex evaluation that will allow comparison of gear units for different
sizes of gearbox. Therefore, it should considerate an indicator u max T2 max /h or even
more complex u max T2max /mh or u maxT2 max/Vh (Fig. 7).
Analyzing obtained indicators, the gear ratio and output torque are in the most
favorable comparison with the weight and volume for smaller shaft heights (100 - 200
mm). Therefore, the Manufacturers 2, 5 and 6 have the best relations of complex
indicators until the shaft height 315 mm. Manufacturer 3, which provides the lowest
gear ratio, has the lowest values of these complex indicators. After shaft height 355
mm, reducers of Manufacturer 1 are leading and the best solution.
6. CONCLUSION
Based on the implemented evaluation, it can be seen that the weight of
gearbox of certain manufacturers are not uniform. Also, gear ratios and output torque
are different for particular manufacturer (even within the certain sizes of the same
manufacturer). Based on complex indicators, such as the ratio of output torque and
weight (T2 /m), or the gear ratio and weight (u/m), we can see a big difference between
different solutions, but reducers of Manufacturer 5 should be highlighted because they
have the best relations of technical characteristics. In order to obtain a more complete
assessment, a complex indicator (u max T2max /m or u maxT2 max/V) is introduced into
consideration on the basis of which was also confirmed the advantage of Manufacturer
5. It should consider that this method of assessment can evaluate only gearboxes with
the same shaft height. Shaft height is introduced into consideration in order to obtain
more complex assessment, so indicators (u max T2max /mh) or (u maxT2 max/Vh) are used.
Based on these evaluations, it is concluded that Manufacturers 5, but also 2 and 6
have reducers with the best relations of complex indicators, so that this concept should
be adopted in the development of new two stage series of gear units. Therefore, it
should adopt such kind of gear units with maximum output torque and gear ratio per
mass unit.
Of course, the price as market characteristic is an additional argument for the
purchase of gear unit, but also the speed of delivery, provided service, spare parts,
which is all not discussed in this assessment of most favorable and optimal design.
LITERATURE
[1] Kuzmanovi, S. (2009) Universal Helical Gear Reducers, University of Novi Sad,
Faculty of Technical Sciences, Novi Sad
[2] Kuzmanovi, S., Ianici, S., Rackov, M. (2010). Analysis of Typical Method of Connection
of Electric Motor and Gear Unit in the Frame of Universal Motor Gear Reducers,
Machine Design 2010, Faculty of Technical Sciences, Novi Sad, p. 141-146.
[3] Rackov, M., Kuzmanovi, S. (2011). Proposal of Assessment Method for the
th
Conceptual Design of Universal Helical Gear Reducers, Proceedinga of 7
International Scientific Conference IRMES 2011, University of Ni, Mechanical
Engineering Faculty, 27-28 April 2011, Zlatibor, Serbia, p. 469-474.
[4] Catalogues: SEW, Nord, Rossi, Bonfiglioli, Motovario, Radicon
158
FATIGUE ANALYSIS FROM FRACTURE MECHANICS ANGLE

Ivica amagi1, Nemanja Vasi2, Zijah Burzi3
Summary: Crack propagation influenced by variable loads that are lower than the
loading level of quasistatic fracture is called material fatigue. Fatigue is causing the
largest number of damages and failures of parts and structures during exploitation. In a
number of cases damages can be assigned to the material state, but a lot of damages
are consequence of a poor constructive design and more often bad welded joints [1].
Accordingly, fatigue tendency of a part of machine, structure or welded joint, does not
depend merely on fatigue strength of the material used for a part fabrication but its
geometry as well. Therefore, when increasing a machine or structure fatigue
endurance is set as a task, it should be taken into consideration that mere selection of
more durable material for parts fabrication is not sufficient and often is inefficient, and
new constructive solutions presents better alternative. Defects in metals and various
forms of stress concentration (notches, grooves, holes welded and mechanical joints)
in elements cant be avoided. Therefore, period until fatigue crack initiates, Ni, can be
neglected, from angle of total fatigue life until crack appearance, Nu. Considering that
macroscopic crack growth rate cant be affected by structural alternation, it remains to
determine the crack growth rate in laboratory conditions for the given material of
specimen or element, and after the crack size determination during occasional
inspection to estimate the remaining life of machine part or structure. Theses of linear
elastic fracture mechanics provide the possibility for this estimation.
Key words: fatigue, crack, welded joint, fracture mechanics parameters, fracture
1. INTRODUCTION
The basic progress that fracture mechanics made in the material fatigue scope
is in analytical parse of the fatigue fracture phenomenon on the initiation period, in
which fatigue crack appears, and the propagation or the growth period that follows and
in which crack enlarges to the critical size at which sudden fracture occurs. Thus, total
number of cycles, Nt, after which crack occurs, is divided to the number of cycles
needed for fatigue crack initiation, Ni, and the number of cycles need for its
propagation up to the critical size, Np:
M.Sci, Ivica amagi, Blace, Faculty of Technical Sciences K. Mitrovica, ivica.camagic@pr.ac.rs

M.Sc, Nemanja Vasi, Kragujevac, Faculty of Technical Sciences K. Mitrovica, nemanja.vasic@pr.ac.rs
3
PhD, Zijah Burzi, Belgrade, Military Institute of Techniques, zijah_burzic@rvkds.net
2
159
Nt = Ni + Np .
(1)
Parallel introduction of experimental and theoretical approach allowed

development in the study of a material behaviour under variable load influence, since
just theoretical approach cannot fully explain initiation and propagation of the fatigue
crack. Today, all factors that influence the da/dN = f(DK) dependence, in so called lowcycle fatigue when plastic deformation is established in the loop of a single hysteresis,
are intensively researched. Stress and strain state analysis, on a tip of a growing
fatigue crack, by linear-elastic fracture mechanics (LEFM) procedure have lead to the
Paris equation formulation for all metals and alloys, connecting the fatigue crack
growth rate with the stress intensity factor range on a crack tip.
2. LINEAR-ELASTIC FRACTURE MECHANICS CONCEPT
Crack growth during fatigue is a very complex process that depends on a
series of variables such as:
Effective stress field intensity on the crack tip defined by K-factor;
Loading type and form;
Working environment (aggressiveness, temperature, humidity);
Mechanical and metallurgical properties of the metal.
Introduction of the fracture mechanics into fatigue behaviour research
originated from the crack growth analysis during the cyclic loading. Change of the
crack length, a, is shown in Fig. 1 [4], in terms of cycles number, N, on a three levels of
the upper stress Vu ('V1>'V2>'V3) with the lower stress Vl=0, where each specimen
had an equal starting crack length a0.
It is noticeable that with the incensement of cycles number, N, and crack
length, a, crack growth rate, defined with tangent slope, continuously rises. Also,
increase of the stress range 'V, is followed by more rapid velocity gradient increase. In
other words, crack with length a1 shown in Fig. 1 grows more rapidly with stress
amplitude 'V1 compared to 'V2 and 'V3 respectively.
A number of theoretically and empirically defined dependences in following
form da/dN = f(F, a), can be found in literature, and they emphasize the importance of
loading and crack length. First to define the range of stress intensity factor 'K = f(V, a)
as a basic parameter that controls the fatigue crack growth rate as:
'K = Kmax - Kmin = Y(Vmax - Vmin) (Sa)1/2 = Y'V (Sa)1/2,
(2)
were Paris and associates [2, 3,]. If at the same time Vmin < 0 it is accepted that Kmin =
0 because the K-factor does not exist for pressure stresses.
Crack growth rates da/dN in terms of the 'K are determined by graphical or
numerical procedure from an appropriate a-N, t curve. Experimental results presented
on a double-logarithm scale most commonly have specific S-shape, schematically
represented in Fig. 2 [2].
Three regions can be noticed on this curve from the crack growth mechanism
and different influential factors intensity points of view. In region I decreasing of the 'K
is followed by sudden reduction of the crack growth rate. 'K value at which the rates
are at the order of magnitude 10-10 m/cycle or lower defines the sensitivity threshold of
the stress intensity factor range fatigue threshold, 'Kth [5]. Below 'Kth fatigue cracks
160
Fatigue Analysis From Fracture Mechanics Angle
behaves as cracks without growth tendency. In region II the dependence log(da/dN) to

log'K is basically linear and it is represented with straight line described by Paris and
associates [2] with the degree function as:
da
dN
C ( 'K ) m
(3)
'V3
'V1 > 'V2 > 'V3
Crack length, a
'V2
Fracture
'V1
a1
Number of cycles, N
Fig. 1 Fatigue crack length dependence on the number of cycles

Sudden crack growth, before final fracture, appears in region III. This apparent
instability is connected with the K-factor maximum value, Kmax, approaching the critical
value of the fracture toughness, KIc, for given material, which is connected to the early
phase of brittle fracture. This possibility is expressed within high strength materials,
with low fracture toughness, where tensile specimens dimensions allow linear-elastic
behaviour at the K-factor levels close to the fracture toughness at plane deformation,
KIc, as well.
Paris equation application proved to be especially fertile in the domain of weld
metal fatigue and welded structures. Unlike homogenous materials and parts made
from them, where the number of cycles necessary for fatigue crack initiation
dominantly participates in the total number of cycles until fracture, it was noticed soon
after the Paris equation was formulated that with welds the total number of cycles until
fracture is mainly determined by the fatigue crack growth. Following checks confirmed
that the number of cycles needed for the crack initiation within different kinds of welds
and materials does not exceed 25% of the total number of cycles until fracture [5]. The
reason for this lies in geometric unhomogenousity of welds and in a presence of small
enough surface roughness, especially on crossing between raised and base metal, in
form of intrusions or inclusions of slag with depth not higher than 0,02-0,04 mm, on
which initial fatigue crack rapidly forms caused by the stress concentration [5].
Based on the previous analysis of the influence factors on fatigue crack growth
rate it can be seen that beside the 'K range, as the most influential, the crack growth
161
rate is influenced by many mechanical, geometrical and metallurgical factors and

environment properties. However, till now, a series of empiric, semi empiric or purely
theoretical models for prediction of the crack growth rate was developed, but it is
specific that each of them is applicable in certain 'K region and for certain metals and
testing conditions.
log
da
dN
R>0
III
II
R< 0
da
m
= C ('K)
dN
' K th
' K c log ' K
Fig. 2 Principle form of the growth rate change da/dN = f('K) for R = 0 and S-curve
shift directions for Rz0
Paris equation, given in expression 3, where C and m are material constants, is
valid only in the crack growth region II (Fig. 2), and only for the single value of R =
Kmax/Kmin ratio. Since crack growth rate lines are nearly parallel for different values of R
in region II [1], coefficient m values will be the same as well, but coefficients C will
differ, since they depend on the R ratio. That dependence can be simply presented as:
da
dN
Cv
(1 R ) nv
( 'K ) mR ,
(4)
where coefficients Cv and nv are determined for the ratio R = 0.

Expression 4, with substitution:
n
v
K max
'K
(1 R )
nv
(5)
was modified by Walker in the following form:

da
dN
v
C v 'K mR nv K max
n
v ,
C v 'K mv K max
where mv = mR - nv.
Forman and associates suggested expression:
162
(6)
Fatigue Analysis From Fracture Mechanics Angle
da
dN
CF ( 'K ) mF
(1 R ) K Ic 'K
CF ( 'K ) mF 1 K max
,
K Ic K max
(7)
that beside the R ratio, includes influence of rapid fracture as well, when the Kmax is
converging to the fracture toughness KIc. This kind of behaviour can be found within
high strength metals, with low fracture toughness, where sizes of specimens for fatigue
crack growth rate testing allow linear elastic behaviour at the K-factor levels close to
the KIc, as well.
Klesnil and Lukas included the R ratio influence on the crack growth behaviour
close to the threshold 'Kth (region I). That empiric model, valid for the region II as well,
has a following form:
da
dN
( 'K ) mK
m
'K thK
CK
J
(1 R )
(8)
where CK and mK are material constants, and J is a constant that is dependent on the
material-environment system.
The influence of heterogeneity of structure and mechanical properties can be
overviewed more clearly on the example of butt welded joint (2/3 X-weld) of high
strength low alloyed steel Nionikral-70, welded with Tenacito-75 basic low-hydrogen
electrode in diameters of 3,25 and 4 mm, supplied by Acroni Jesenice. The fatigue
crack growth rate diagrams, da/dN, the stress intensity range change, 'K, for the
samples tested at the room temperature are given in Fig. 4. Values of the coefficient C
and m, and the stress factor intensity range on a fatigue threshold Kth, are given in
Table 1.
Fig. 4 da/dN-'K dependency diagram for: a) BM, b) WM and c) HAZ specimens
163
Table 1 Paris equation coefficients

Growth region
BM-1
WM-1
HAZ-1
Coefficient C
-14
3,9810
3,3010-19
2,9010-16
Coefficient m
4,139
8,462
6,403
'Kth, MPa m1/2

10,22
8,45
8,71
3. CONCLUSION
Without a doubt fracture mechanics approaches have large potential for
successful application in all cases of welded structures safety evaluation. Brittle
fracture tendency in presence of a crack and in variable load conditions should be
connected with slope change of the part of the curve in Paris law validity area. In
particular example, slower crack growth is confirmed at specimens with a crack in base
metal and heat affected zone, because for the same crack growth rate higher stress
intensity factor is requested. Maximal crack growth rate can be expected on a level of
stress intensity factor range witch approaches the fracture toughness at plane strain
KIc, since brittle fracture is achieved at such level.
LITERATURE
[1]
[2]
[3]
[4]
[5]
164
Miller, K.J., OcDonnell, W.J. (1999). The fatigue limit and its elimination, Fatigue
Fracture Engineering Materials Structures, Vol. 22, p. 545-557.
Paris, P.C., Erdogan, F. A Critical Analysis of Crack Propagation Laws, Trans.
ASME, Journal Basic Eng., Vol. 85, No. 4, p. 528.
Paris, P.C., Sih, G.C., Stress Analysis of Cracks, Fracture Toughness Testing
and Application, STP 381, p. 30-83.
Burzi, Z. (1997). Ispitivanje promenljivim optereenjem glatkih i zarezanih
epruveta, 7. Tematski zbornik radova, Eksperimentalne i numerike metode u
oceni integriteta konstrukcije, V. Plana, p. 75-92.
Radon, J.C. (1982). Determination of Threshold Stress Intensties, Fatigue of Low
Alloy Steel BS4360-50D, Int. J. Fatigue, p. 225.
FATIGUE LIFE ESTIMATION OF CRACKED STRUCTURAL

COMPONENTS
Slobodanka Boljanovi1, Stevan Maksimovi2, Strain Posavljak3
Summary: This paper presents a computational model/method for calculating the
residual fatigue life for surface-cracked structures. The analysis considers the major
three-dimensional aspects for the semi-elliptic crack problem including crack shape
(planar crack with crack front curvature) and local stress variation in a plate. The plate
is subjected to a tensile load. Stress intensity factor, as an important parameter for
fatigue life estimation is determined by applying analytical and numerical methods. The
fatigue life of surface-cracked problems has been estimated using the Paris equation
by incrementing the crack depth. Calculated residual life correlates well with
experimentally obtained data.
Key words: fatigue; semi-elliptical surface crack, crack growth; life evaluation
1. INTRODUCTION
Structural components are subject to environment during service and in
combination with possible cracks initiated at structural discontinuities such as holes or
material defects this could lead even to failure. Reliable fatigue life prediction is very
important for safe design and maintenance of structural components subjected to
cyclic loading [1,2]. Under cyclic loading or under static loading and due to effects of
differrent environment factors any surface flaw, if it exists, has the potential of
subcritacally growing into a surface crack. Analysis of the structure containing such
flaws is needed for modeling and estimation of the corresponding crack propagation
rate as well as residual fatigue life.
Experimental investigations show that crack shape of propagating surface
cracks in a plate under cyclic loading is approximately semi-elliptical. Due to
complexity of geometry, for fatigue crack growth analysis of semi-elliptical problems, it
is neccessary to examine both the depth and surface directions. Actually, theoretical
model for fatigue crack growth is based on two coupled Paris fatigue crack growth
relations.
For reliable prediction of crack growth rates and fracture strengths of
components accurate stress analysis of surface crack problems is needed. Over the
1
Ph.D, Slobodanka Boljanovi, Belgrade, VTI, Ratka Resanovia 1, (slobodanka.boljanovic@gmail.com)

Ph.D, Stevan Maksimovi, Belgrade, VTI, Ratka Resanovia 1, (s.maksimovic@open.telekom.rs)
3
Ph.D, Strain Posavljak, Banja Luka, Faculty of Mech. Eng., Stepe Stepanovia 75, (s.posavljak@urc.rs.ba)
2
165
S. Boljanovi, S. Maksimovi, S. Posavljak
years, several methods have evolved to compute stress intensity factor for structural
components containing semi-elliptical cracks. In general, for fatigue crack growth
analysis of surface crack problems, two approaches may theoretically be employed to
model three-dimensional cracking problems. The first approach in which surface crack
is analysed as a full three-dimensional problem, and the second, which consists of
replacing the surface crack by an sequivalents two-dimensional or line crack created
mathematically by combining suitable analytical models with correction functions.
According to that, some of numerical methods which can be used to calculate stress
intensity factor for fatigue crack growth analysis are: the finite element method [3], the
boundary-integral equation method [4,5], the finite element alternating method [6] and
the 3-D line method [7-9].
The objective of this paper is to formulate a computational model for fatigue life
estimation of structural components with surface semi-elliptical cracks. In fatigue
analysis, both, crack depth direction and surface direction are investigated. Moreover,
the stress intensity factor is calculated by applying analytical method and numerical
method (FEM).
2. FATIGUE CRACK GROWTH
An important aspect in the crack growth analysis of cracked structural
components is the evaluation of residual life under service conditions. This means the
calculation of the time i.e. number of loading cycles required to grow a crack from a
tolerable size to a critical length.
The fatigue life analysis of a complex surface crack problem requires
considering two crack growth directions. Actually, Newman and Raju [3] intoduced that
the aspect ratio change of surface cracks should be calculated by assuming that a
semi-elliptical profile is always maintained and for fatigue life estimation it is adequate
to use two coupled Paris fatigue laws known as stwo-point plus semi-ellipses method:
da
db
m
m
C A 'K A A ,
C B 'K B B ,
dN
dN
(1)
where 'KA and 'KB are the ranges of stress intensity factor at the depth and surface
points of a surface crack and they can be calculated by applying analytical and/or
numerical methods, CA, CB, mA and mB are material constants experimentally obtained.
In crack growth analysis very important aspect is to evaluate fatigue life up to failure.
Final number of loading cycles for surface cracked problems can be calculated for both
directions if equations for crack growth rate are integrated:
af
N depth
da
a C A 'K A m A , N surface
0
bf
db
C 'K
mB
b0
(2)
Eq. (2) can be used to estimate the number of loading cycles required for the
surface crack to grow through the thickness for the assumed incremental increase in
crack depth. The corresponding incremental increase in the surface direction is
calculated using Eq. (3). Every subsequent crack length is determined as:
ai 1 ai 'a and bi 1 bi 'b .
(3)
This process is repeated until the crack depth reaches the through-thickness.
166
Fatigue life estimation of cracked structural components
The incremental life is estimated corresponding to every incremental growth and the
cumulative total life is evaluated.
3. STRESS INTENSITY FACTOR FOR SEMI-ELLIPTIC CRACK PROBLEM
In the study of fatigue crack growth and failure behaviour of surface cracks, the
essential element is an accurate calculation of the stress intensity factor. Stress
intensity factor solutions are required for the assessment of fracture strength and,
residual fatigue life, or for a damage tolerance analysis to aid structural design.
In fracture mechanics the stress intensity factor can be calculated with
following equation:
YV Sa ,
KI
(4)
where V is the level of external loading, a presents the crack length and Y is the
geometry correction factor. Due to the fact that surface cracked problem is examined in
this paper, Eq. (4) must be adjusted to considered geometry. The complex geometry
of surface cracked problems could be introduced by formulating the adequate
geometry correction factor. First, Irwin [10] proposed an expression for the stress
intensity factor, around an elliptical crack in an infinite elastic solid, subjected to
uniaxial tension and based on an analysis by Green and Sneddon [11]. As a result of
Irwin's considerations, the geometry correction factor along the elliptical crack in an
infinite solid subjected to uniaxial stress V acting normal to the plain of the crack can
be expressed as:
2
2
sin I cos 2 I
E k
0.25
(5)
where a is the depth and 2b is the surface length of a semi-elliptical flaw. The
parameter I is the angle in the parametric equation of ellipse and E(k) is a complete
integral of the second kind [12], i.e.:
0.5
E k
1.64
1 1.47 a .
(6)
Later, the geometry correction factor for a surface crack in an infinite elastic
solid subjected tensile was modified in order to define the stress intensity factor for
surface crack in a finite elastic solid [12-14] (Fig.1). Actually, in Eq. (5), the
magnification factor Mf is included, which clearly shows the effects of a front and back
free surfaces of the plate:
2
2
sin I cos 2 I
Y Mf
E k
0.25
(7)
The factor Mf for finite width w and finite thickness t is formulated as:
167
Mf
2
4
M 1 M 2 a M 3 a g f w ,
t
t
(8)
where M1, M2, M3 present factors depending of depth crack length a and surface crack
length b, i.e.:
a
M 1 1.13 0.09 ,
b
0.89
M 2 0.54
,
a
0.2
b
M3
(9)
(10)
24
a
0.5
14 1 ,
a
b
0.65
b
1
(11)
g is factor depending of depth crack length a, thickness t, as well as angle I:

2
a
2
g 1 0.1 0.35 1 sin I ,
(12)
and factor fw can be expressed like:
Sb a
.
f w sec
2w t
(13)
With defining the geometry correction factor (Eqs. (7)-(13)), it is possible to calculate
the stress intensity factors at any point along a semi-elliptical crack in a finite plate
subjected to tensile load. Additionally, in this paper, numerical approach is built for the
stress intensity factors calculation by applying a finite element method. All calculated
results using analytical and numerical approaches for determination of the stress
intensity factors are presented in the section that follows.
Fig. 1 Geometry of a semi-elliptical surface crack in a finite plate.

168
4. NUMERICAL RESULTS
To illustrate computation model for fatigue life of structural components with a
semi-elliptical crack, a few numerical examples are presented in this section. These
examples examine stress intensity factor calculation as well as fatigue life estimation.
In order to verify the validation of presented model for surface crack growth simulation
obtained results are compared with experimental data.
Example 4.1 Crack growth estimation of semi-elliptical surface crack
This example deals with crack life calculation of structural element with semielliptical surface crack. Analysed external loading is axial with constant amplitude
(Pmax=35 kN, R = 0.1). Geometry characteristics are: a0=2.42 mm, af=9.43 mm,
b0=2.54 mm, bf=12.95 mm, 2w=28 mm, t=11 mm, L=102 mm. Material characteristics
of AISI 4130 alloy steel are: Vy=415 MPa, Vu=559 MPa, E=210 GPa and Paris'
parameters CA=6.35*10-12, mA=3.99, CB=5.61*10-12, mB=3.89.
The first parameter that has to be examined in fatigue crack growth analysis of
surface-cracked structure is the stress intensity factor. This important parameter is
calculated by applying Eq. (4) with defined geometry characteristics as well as external
loading. Due to the fact that surface crack is analysed, it was neccessary to examine
the calculation of stress intensity factors for both surface and depth directions. For
calculation of the stress intensity factors Eq. (4) with Eqs. (6)-(13) were used.
Computed values of stress intensity factors for adequate crack increments are
presented in Fig.2.
AISI 4130 (R=0.1)
0.5
KIA , KIB [MPa m ]
50
40
30
20
10
0
0
6
8 -3 10
a, b [m] (x 10 )
KIa-a (Calculated curve)
KIb-b (Calculated curve)
12
14
Fig. 2 Stress intensity factor versus crack length

(a- Depth growth, b-Surface growth).
Using the fatigue performance data, according to the structural geometry and
defined fatigue model in previous section, it is possible to calculate crack growth rate
and fatigue life up to failure. Obtained results for crack length versus number of loading
cycles up to failure are presented in Fig.3 for both, depth and surface directions. At the
same Figures, all computed results for number of loading cycles up to failure are
compared with experimental data [15].
169
AISI 4130 (R=0.1)
AISI 4130 (R=0.1)
10
14
12
10
-3
a [m] (x10 )
-3
a [m] (x10 )
8
6
4
8
6
4
0
0
6
8 10 12
4
N [cycles] (x10 )
14
a - N (Calculated curve)
a - N (Experiment)
16
18
6
8 10 12
4
N [cycles] (x10 )
14
16
18
b - N (Calculated curve)
b - N (Experiment)
Fig. 3 Crack length versus number of loading cycles for semi-elliptical crack
(a- Depth growth, b-Surface growth).
It is indicated in Fig.3 that the estimated values of number of loading cycles up
to failure are conservative when compared to experimental data. In engineering
practice existance of conservativity in fatigue crack growth analysis is always benefitial
since in this way safe residual service life of structural elements could be determined.
Additionally, conservativity of computed results is often result of defined criteria (often
too strict) which are used to formulate adequate analytical relations for crack growth
analysis.
Example 4.2 Stress analysis of a surface-cracked plate
In this example, stress intensity factor calculation was carried out. The
considered plate with surface crack (a=b) is subjected to tensile load. External loading
is with constant amplitude (P = 50 kN). The plate is made of AISI 4130 steel
0.5
(Vy=415MPa, Vu=559MPa, E=210 GPa, KIC=80 MPam ). Geometry characteristics of
the surface cracked plate are: a0=b0=3 mm, 2w=50 mm, t=10 mm.
In addition to analytical approach for stress intensity factor evaluation used in
the previous example, numerical approach based on finite element method is built in
this paper. For this purpose singular six-node finite elements [16,17] are used.
Actually, step-by-step, for each increment of crack length different meshes are
modeled by using super-elements around crack tip [18], Fig. 4.
Fig. 4 The part of finite element mesh of the surface-cracked plate.

170
The step-by-step procedure is repeated until the computed crack growth is very close
to the final failure of the plate. A representation of the finite element analysis for the
surface-cracked plate (a=b) made of AISI 4130 steel is shown in Fig.4 and Fig.5.
V1
L1
C1
23.39
21.97
20.55
Surface 11
19.14
17.72
16.3
14.89
13.47
12.06
10.64
9.223
7.807
6.391
4.975
Z
Y
X
Output Set: MSC/NASTRAN Case 1
Deformed(0.0366): Total Translation
Contour: Solid VonMises Stress
3.558
2.142
0.726
Fig. 5 Stress distribution of the surface-cracked plate subjected to tensile load.

Computed results are listed in Table 1 for stress intensity factor for different values of
crack length in both directions (surface and depth) by using proposed analytical
method and a finite element method.
Table 1 Comparison of the computed stress intensity factors using analytical and
numerical method (Positions: A - I =900, B - I =00)
KAFEM
KAAnal.
KBFEM
KBAnal.
a=b
-3
0.5
0.5
0.5
[MPam ]
[MPam ]
[MPam0.5]
[MPam ]
Step
10 [m]
1
3
6.74
6.55
7.92
7.41
2
4
7.88
7.68
9.24
8.87
3
5
8.89
8.41
11.18
9.99
4
6
9.99
9.81
12.48
11.41
It can be observed in Table 1 that the analytical method gives almost the same
solutions as finite element method, so both methods for stress intensity factor
calculation can be used in crack growth analysis of surface-cracked plate subjected to
tensile load.
5. CONCLUSION
The fatigue growth model to analyse the semi-elliptical crack in a plate
subjected to tensile load is developed. For the stress intensity factor calculations, both
analytical and numerical approaches are applied. Fatigue life up to failure is estimated
by using stwo-point plus semi-ellipses method. The fatigue lives up to failure are
compared with experimental results available in the literature. Comparisons of
computed number of loading cycles up to failure, point out the fact that the analytical
approach for calculation of stress intensity factor is in a good agreement with those
obtained by applying finite element method. Presented fatigue growth model is
applicable in engineering practice for fatigue life estimation of semi-elliptical crack
problems in a plate subjected to a tensile.
171
ACKNOWLEDGMENT
This research work is financially supported within the project No. OI 174001
(SANU-Mathematical Institute, Belgrade) as well as the project No. TR 35052 (VTIAeronautical Department, Belgrade) by the Ministry of Science and Technological
Development, Serbia.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
172
Boljanovi, S, Maksimovi, S., Djuri, M. (2009). Analysis of Crack Propagation Using the
Strain Energy Density Method. Scientific Technical Review, vol. LIX, No.2, p.12-17.
Boljanovi S., Maksimovi, S. (2009). Fatigue Life Analysis of Cracked Structural
Components using Crack Closure Effects, SEECCM 2009 Conference Proceedings,
Rhodes, Greece.
Newman, J.C.Jr., Raju, I.S. (1981). Stress Intensity Factor Equations for Cracks in ThreeDimensional Finite Bodies Subject to Tension and Bending loads. NASA TM 83200.
Cruse, T.A. (1972). Numerical evaluation of Elastic Stress Intensity Factor by the
Boundary-Integral Equation Method. In: Swedlow, J.L. (editor). The Surface Crack: Phisical
Problems and Computational Solutions. American Society of Mechanical Engineers, New
York, p.153-170.
Cruse, T.A. (1974). An Improved Boundary-Integral Equation Method for Three
Dimensional Elastic Stress Analysis. Computers and Structures, vol. 4, p.741-754.
Smith, F.W., Sorensen, D.R. (1976). The Semi-Elliptical Surface Crack A Solution by the
Alternating Method. International Journal of Fracture, vol. 12, no. 1, p.47-57.
Rice, J.R. (1972). The Line-Spring Model for Surface Flaws. In: Swedlow, J.L. (editor). The
Surface Crack: Physical Problems and Computational Solutions. American Society of
Mechanical Engineers, New York.
Parks, D.M., Lockett, R.R. (1980). Development and Calibration of the Non-Linear LineSpring Model for Analysis of Surface Cracked Plates and Shells. In: EPRI Ductile Fracture
Research Review Document, EPRI Rep. No-80-10-LD WS 80-912, p.9.1-9.12.
Delale, F., Erdogan, F. (1981). Line-Spring model for Surface Cracks in a Reissner Plate.
International Journal Engng. Sci. vol.19. p.1331-1340.
Irwin, G.R. (1962). The Crack Extension Force for a Part Through Crack in a Plate. Trans.
ASME. Journal of Applied Mechanics, vol. 84, p.651-654.
Green, A.E. , Sneddon, I.N. (1950). The Distribution of Stress in the Neighborhood of a Flat
Elliptical Crack in a Elastic Solid. Proc. Cambridge phil. Soc., vol. 46, p.159-164.
Jones, R. et al. (2004). Weight Functions, CTOD, and Related Solutions for Cracks at
Notches. Engineering Failure Analysis, vol. 11, no. 1, p.79-114.
Anderson, T.L. (1995). Fracture Mechanics: Fundamentals and Applications. CRC Press.
Newman, J.C.R.I., Raju, I.S. (1986). Stress-intensity Factor Equations for Cracks in Threedimensional Finite Bodies Subjected to Tension and Bending Loads. In: Atluri Satya N.
(editor). Computational Methods, Elsevier Science, p.312-334.
Song, P.S., Shieh, Y.L. (2004). Crack Growth and Closure Behaviour of Surface Cracks.
International Journal of Fatigue, vol.26, no.4, p.429-436.
Sukumar, N., Kumusa, M. (1992). Application of the Singular Finite Element to Crack and
Sharp Notches in Orthotropic Media. International Journal of Fracture, vol. 58, p.177-192.
Maksimovic, S., Burzic, Z. and Maksimovic, K. (2006). Fatigue Life Estimation of Notched
Structural Components: Computation and Experimental Investigations. ECF16 Conference
Proceedings, E.E. Gdoutos (editor), Alexandroupolis, Greece, Springer.
Msc/NASTRAN, Theoretical Manuels.
HYDRAULIC INSTALLATION OF EKO CONTAINER

Ibrahim Badak1, Ph.D., Remzo Dedic2, Ph.D., Mersida Manjgo3, Ph.D.
Summary: This paper describes the hydraulic installation that is embedded in the
prototype design of ecological containers. Identify the problems that were created to
solve this task, and the ways in which they are resolved.
Keywords: hydraulic installation, ecological containers
1
INTRODUCTION
Ecological containers are a novelty in the way of collecting waste in the narrow
city centers, parks and pedestrian zones, where the rule resides a large number of
people on a relatively small area. In this way, there is a large amount of waste that is
difficult to take. The proposed solution provides a significant contribution to solving this
problem.
2
WORKING PRINCIPLE
Ecological container is a steel frame in which the store an adequate number of

standard containers, which are then lowered into a special mechanism suitable holes
dug in the sidewalk or the street. (Fig. 1)

1
Univerzitet Demal Bijedi Mainski fakultet Mostar, Sjeverni logor bb, BiH, ibrobadzak@hotmail.com
Sveuilite u Mostaru, Fakultet strojarstva i raunarstva , Matice hrvatske bb, BiH, rdedic2001@yahoo.com
3
Univerzitet Demal Bijedi Mainski fakultet Mostar, Sjeverni logor bb, BiH, manjgom@yahoo.com
2
173
Ibrahim Badak, Ph.D., Remzo Dedic, Ph.D., Mersida Manjgo, Ph.D.

Fig. 1 Ecological container
In this way the surface of the pavement or in the plane of the square remains
the only series basket wastes. Each of them indicated that the wastes are injected
(plastic bottles, glass bottles, aluminum bottles, paper, etc.)
Inserted items to cart, the body of the basket, sinking into a container located
under the basket, resulting in a higher capacity, thereby reducing the number of
required discharge. It also prevents animals to approach through garbage and rubber
insulation significantly reduces the intensity of odors.
Emptying the container is achieved in the same way as the classical
discharge: After the arrival of a vehicle to transport waste joins hydraulic
installation by quick-detachable coupling, a shifting lever valve wiring is
accomplished raising support structure with a container to the level of the
pavement (Figure 2).
174
Hydraulic installation of eko container

Fig.2Container Lifting
Containers are then extracted from the supporting structure one by one,
empty the classical way, and then returned to supporting construction
(Figure 3).
After completion of discharge, the action of the valve handle wiring, the
bearing structure of the container is again lowered into the hole
Workers separate hydraulic installation and go.

Fig. 3 Empty containers ecological
3
HYDRAULIC INSTALLATIONS
Hydraulic scheme of hydraulic installation of ecological container is shown in

Figure 4.
175
Executive elements of hydraulic systems are two hydraulic cylinders (1)

connected at the ends of supporting structure.
As the friction in the guides and the possible uneven distribution of loads can
be expected for some delay in one cylinder to the other, so in hydraulic system install
integrated flow regulator (3) which is substantially prevented.
The cylinders are one way to have a smaller number of lines.

Fig.4Hydraulic installation ecological container
The cylinders are one way to have a smaller number of lines.

The control element (6) is a distributor 3 / 3, and the inability to purchase the
prototype installation is embedded distributor 4 / 3. Manage distributor is done by hand,
a return to the neutral position is achieved with springs.
On suporting construction container flow between the regulator and cylinder
are made of metal tubing, the flow regulator valve and a hose with a coupler finally
enabling you to quickly assembling and disassembling hydraulic installations, without
leaking of hydraulic oil.
176
Hydraulic installation of eko container
Driving part of the hydraulic installation is achieved with a vehicle for waste.
This reduces the price of hydraulic installation of eco-container, and avoids potential
problems mounting the drive under the earth (because of the possibility of water
penetration), and bringing power lines to container.
The drive unit is axial - piston pump. Reservoir contained 100 gallons of
hydraulic oil.
4
EXPERIMENTAL TESTING
Experimental testing of ecological containers carried on the polygon in the

factory that produced it. (Figure 5).

Fig.5Experimental testing of a prototype
Experimental data have found certain irregularities in raising the support
structure, and the cause it was too big gap between the sliding surfaces and the air still
present in a hydraulic installation.
After irradiation, reducing the gap and hydraulic installations, these problems
have disappeared.
5
CONCLUSION
Ecological containers represent a new way of waste disposal, which very

efficiently solve several problems (transport, environment, health)
Experiments with the prototype confirmed the hypothesis and the possibility of
using hydraulic installation vehicles for transport of waste. The following steps should
make certain improvements in design, a larger series would reduce the price of these
products and thus make them more accessible to the market, as well as their
application in practice.
177
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
178
Basta, T., (1972). Mainska hidraulika, Nauna knjiga, Beograd

Jovanovic, P.,Ciric, M., (1985) Cilindri u hidraulici i pneumatici, Odravanje
maina i opreme, Beograd
Jovanovic, S., (1981) Uljna hidraulika I, Tehnika knjiga, Beograd,
Jovanovic, S., (1983) Uljna hidraulika II, Tehnika knjiga, Beograd,
Kelic, V., (1985) Hidroprenosnici, Nauna knjiga, Beograd,
Savic, V., (1987) Hidraulina ulja i odravanje hidraulinih sistema, Mainski
fakultet Zenica,
Savic, V., (1987) Mainska hidraulika, Mainski fakultet, Zenica,
Savic, V., (1988) Uljna hidraulika, Dom tampe, Zenica,
Momirski, M., (1981) Elementi teorije skeletnih konstrukcija, Fakultet tehnikih
nauka, Novi Sad
PREDICTION OF DAM BEHAVIOUR USING MULTIPLE LINEAR

REGRESSION AND RADIAL BASIS FUNCTION NEURAL NETWORK
Vesna Rankovi1, Nenad Grujovi2, Goran Milovanovi3, Dejan Divac4, Nikola
Milivojevi5
Summary: The safety control of dam is supported by monitoring activities and is based
on mathematical models. The variations of hydrostatic pressure and temperature are
the main variables that should be taken into account when analyzing the results of the
concrete dam observations. Deterministic models based on mechanical principles are
often difficult to construct. Statistical procedures, such as multiple linear regression
(MLR), have been applied to dam safety to determine the influence of external loads
on the structure deformation. The relations between the loads and dam behavior are
nonlinear. Radial basis function (RBF) neural network can be successfully applied to
function approximation, forecast, and dynamic systems identification. Neural network
modeling from measured data is effective tool for the approximation of uncertain
nonlinear systems. This paper presents novel approach based on the use of RBF
network to estimate dam behaviour. Mathematical models based on experimental data
are developed. MLR and RBF neural network models for prediction of dam behaviour
have been compared with the measured data on the basis of correlation coefficient.
Key words: Dam Safety, Multiple Linear Regression, Radial Basis Function Neural
Networks, Dam Behaviour, Modeling
1. INTRODUCTION
The variations of hydrostatic pressure, temperature and other unexpected
unknown causes, such as the time effects, are the main variables that should be taken
into account when analyzing the behaviour of the dam. Different models have been
developed and used to analyze dam behaviour. These are grouped into two general
categories: deterministic and statistical models, [1]. The deterministic modeling
1
Ph.D., Vesna Rankovi, Kragujevac, Faculty of Mechanical Engineering, University of Kragujevac,

(vesnar@kg.ac.rs)
2
Ph.D., Nenad Grujovi, Kragujevac, Faculty of Mechanical Engineering, University of Kragujevac,
(gruja@kg.ac.rs)
3
MSc., Goran Milanovi, Mrkonji Grad, HES Vrbas, (g.milanovic@hesvrbas.com)
4
Ph.D., Dejan Divac, Belgrade, Jaroslav erni Institute for the Development of Water Resources,
(ddivac@eunet.rs)
5
Ph.D., Nikola Milivojevi, Belgrade, Jaroslav erni Institute for the Development of Water Resources,
(nikola.milivojevic@gmail.com)
179
Vesna Rankovi, Nenad Grujovi, Goran Milanovi, Dejan Divac, Nikola Milivojevi
requires solving differential equations for which closed form solutions may be difficult
or impossible to obtain, [2]. The advantages of the statistical method, such as multiple
linear regression, consist in simplicity of formulation and speed of execution.
The radial displacement of one or several points of the dam is is an important
behaviour indicator. The radial displacement in any point of the dam is nonlinear
function of hydrostatic pressure and temperature and other unexpected unknown
causes. One of successful neural network applications is to model complex nonlinear
behaviour.
Mata [3] compared multiple linear regression and multilayer perceptron (MLP)
neural network models for the prediction of the upstreamdownstream displacement of
arch dam recorded by a pendulum. The models are generated on the basis of
experimental data of time histories of reservoir level and external temperature and of
structural responses. An improved neuro-wavelet modeling methodology to model and
forecast the horizontal displacement of the dam was applied by Cao et al., [4].
The aim of this paper is to construct a MLR and RBF neural network models to
predict the radial displacement of arch dam and demonstrate its application to
identifying complex non-linear relationships between input and output variables.
The RBF neural network has the capability of universal approximation [5-7] as
multilayer perceptron. In most of the literature, RBF neural networks are considered as
a smooth transition between fuzzy logic and neural networks. RBF network is
especially interesting because of their simple but adaptive structure and effective
learning ability. Although RBF networks usually need more neurons in hidden layer
than MLP, they are easier to initialize and train than MLP networks [5].
2. CASE STUDY
The Bocac dam, on river Vrbas, is a medium size dam (Fig. 1).
Fig. 1 Upstream face of dam and cross-section through block 8

Dam is located in Republic Srpska about of 25 km from the city of Banja Luka.
The dam was constructed between 1976 and 1981. It is a double curvature arch dam
and height is 66 m, crest length 221.4 m. The crest thickness is 3m-13m while the
maximum thickness of the dam base is 14.4m. The minimum, normal and maximum
operating levels are 254, 282 and 283 m above sea level (asl), respectively. The total
capacity of reservoir is 52.7 106 u m3 . The dam is equipped with a monitoring system
180
Prediction of dam behaviour using multiple linear regression and radial basis function neural network
to measure a particular parameters, such as: concrete, water and air temperatures,
reservoir water level, horizontal and vertical displacements, rotations, foundation
displacements, movements of joints, strain, stress, uplift pressure, foundation
displacements and seepage.
The three pendulums were installed to measure radial and tangential
deformations. In this paper it is analyzed the radial displacement of point V1 at block 8
with the proposed methods. A data set include 392 data samples. The available set of
data was divided into two sections as training and test set. Data from January 2000 to
December 2008 are used to train, and data from January 2009 to December 2010 are
used to test, (Fig.2).
Fig. 2 The radial displacement of the point V1 at block 8

3. OVERVIEW OF THE MLR, RBF NEURAL NETWORK
3.1 Multiple linear regression
Consider a training data set
xi
^x1i x2i ...xNi `
^ x , y , x , y
1
,..., xp , y p `
, where
is a vector of input variables and y i is the corresponding output
value, p is the number of training data points.

The multiple linear regression model is given by:
y mMLR E0 E1x1 E 2 x2 ... EN xN
(1)
where E i , i 0,1,..., N represent unknown parameters.

The E i can be estimated by minimizing the sum of the squares of the errors:
y 1 y m1
y 2 y m 2 ... y p y mp
2
where y mi denote the MLR output value from the i-th input element:
y mi E0 E1x1i E 2 x2i ... E N xNi
(2)
(3)
181
The matrix form Eq. 2 is:
y X y X
T
H
1 x11
1 x12

1 x1p
x21 xN1
x22 xN 2
T
where: X
, ^E 0 E1 ... EN ` , y

x2 p xNp
The least squares estimator of is given by:
X X
T
1
(4)
^y
y 2 ... y p `
XT y
(5)
3.2 RBF network
RBF network consists of one hidden layer and takes Gaussian functions as its
basis functions. The typical structure of an RBF network with N inputs, s hidden
neurons and one output is shown in Fig. 3.
Fig. 3 Structure of RBF neural network

The inputs x1 , x2 , , xN are applied to all neurons in the hidden layer. The
output ai 1 of each hidden unit i is then computed by:
ai 1
- ni21
, i
(6)
1,2,...s
where:
ni
N
Zi , j 1 - x j
j 1
0.5
Zi , j 1 is center of the Gaussian functions for i-th hidden neuron and bi 1

where sp is the radius or spread of all radial basis functions.
182
(7)
bi 1
ln 0.5
sp
Prediction of dam behaviour using multiple linear regression and radial basis function neural network
The output of the RBF neural network is given by:

y mRBF
Z
i 1
1,i 2 i 1
b1 2
(8)
where Z1,i 2 is the weight from the i-th hidden layer neuron to the output neuron and
b1 2 is the bias at output neuron.
There are different learning strategies which can be used to design of an RBF
network, depending on how the centers of the radial basis functions of the network are
specified. The most popular existing learning strategies include: fixed centers selected
at random, self-organized selection of centers, and supervised selection of centers. In
this paper, the first strategy is used to determine the parameters of the RBF network.
The learning method requires training data set. In this method, the centers of the
Gaussian functions Zi , j 1 are randomly selected from the training data set. Spread of
radial basis functions sp is important parameter for constructing an RBF neural
network. If spread is small, the RBF is very steep so that the neuron with the weight
vector closest to the input will have a much larger output than other neurons. A larger
spread leads to a large area around the input vector where hidden neurons will
respond with significant outputs. In this paper the Matlab Neural Network Toolbox is
used for the implementation of the RBF network. The Matlab function newrb is applied.
After obtaining Zi , j 1 values for all hidden neurons, the weights Z1,i 2 and bias b1 2 are
determined using multiple linear regression techniques.
4. SIMULATION RESULTS
Statistical model gives the displacement as the sum of three terms: the first is
related to the hydrostatic pressure, the second term takes into account function of time
and the third term is due to air temperature change. The general statistical model used
for the prediction of the displacement of a point in a dam is:
y mMLR h, t , s
E 0 E1h E 2 h 2 E 3 h3 E 4 h 4 E 5 e t E 6 t
E 7 cos d E 8 sin d E 9 sin2 d E10 sin d cos d
(9)
2S j
is the
365
season varying between 0 and 2S , j represents the number of days since January 1st.
The input variables ( x1, x2 ,..., x10 ) of the MLR model are h, h 2 , h3 ,
where h is water level, t is the elapsed time expressed in years, d
h 4 , e t , t , cos d , sin d , sin2 d , sin d cos d .
The MLR model for the prediction of the radial displacement of point V1 is:
5 2
9 3
11 4
y mMLR
V 2 h, t , s 1.7591 10 h 5.8059 10 h 2.2973 10 h
6.6749cos d 4.6632sin d
(10)
It can be seen from Eq. (10) that the elapsed time did not have a significant
effect on the radial displacement of point V1. Correlation coefficients of 0.9595 for the
183
training and 0.9748 for the test set are obtained between the measured and MLR
modeled values of the radial displacement.
Based on the MLR analysis results, the inputs of the RBF model are the h 2 ,
3
4
h , h , cos d and sin d . The output of the model is the radial displacement of
point V1. Different models were constructed and tested in order to determine the
values of sp and goal. The spread and goal are varied from 1 to 2 and from 0.05 to 0.1,
respectively. It is found that RBF with spread equal to 1.7 and goal equal to 0.07
produced the correlation coefficients of 0.9637 and 0.9766 between the predicted and
the measured values of the radial displacement of point V1 for the training and test
sets, respectively.
5. CONCLUSION
In this paper, RBF neural network and MLR model were developed to predict
the radial displacement of the arch-dam. The performance of the RBF network and
MLR model were tested using correlation coefficients. Results of simulation, presented
in this paper, show that the application of the neural network to prediction of radial
displacement gives a slightly higher coefficient of correlation values for training and
test sets. Proposed approach based on RBF network can be a very efficient tool and
useful alternative for the computation of seepage, stress, or crack opening of dam.
ACKNOWLEDGMENT: The part of this research is supported by Ministry of Science in Serbia,
Grants III41007 and TR37013.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
184
ICOLD. (2003). Methods of analysis for the prediction and the verification of dam
behaviour. Tech. rep. Swiss Committee on Dams.
Szostak-Chrzanowski, A., Chrzanowski, A., Massira, M. (2005). Use of
deformation monitoring results in solving geomechanical problemscase studies.
Engineering Geology, vol. 79 , no.1-2, p. 3 12.
Mata, J. (2011). Interpretation of concrete dam behaviour with artificial neural
network and multiple linear regression models. Engineering Structures, vol. 33,
no. 3, p. 903910.
Cao, M., Qiao, P., Ren, Q. (2009). Improved hybrid wavelet neural network
methodology for time-varying behavior prediction of engineering structures.
Neural Computing & Applications, vol. 18, no. 7, p. 821832.
Park, J., Sandberg, I. W. (1991). Universal approximation using radial basis
function networks. Neural Computation, vol. 3, no. 2, p. 246257.
Leonard, J.A., Kramer, M.A., Ungar, L.H. (1992). Using Radial Basis Functions to
Approximate a Function and Its Error Bounds. IEEE Transactions on Neural
Networks, vol. 3, no. 4, p. 622630.
Schilling, R.J., Carroll, J. J., Al-Ajlouni, A. F. (2001). Approximation of nonlinear
systems using radial basis function neural networks. IEEE Transactions on Neural
Networks, vol. 12, no. 1, p. 114.
SOME ASPECTS IN FAILURE ANALYSIS OF CRANES

Nenad Zrni 1, Sran Bonjak 2, Vlada Gai 3, Miodrag Arsi 4
Summary: Failure of a structural or mechanical component of cranes usually can be
associated with materials-related problems and/or design-related, as well as the
fabrication-related problems or inadequate structural maintenance. Also, crane
components and structure experience a spectrum of stresses while operating.
Therefore, about ten percent of material handling high-performance machines failures
can be attributed to fatigue failure. In most cases these failures were unexpected and
lead to catastrophic consequences. This paper discusses some aspects in failure
analysis of cranes, particularly high-performance ones, gives the background for failure
analysis and presents some typical examples of failure. The aim of this article is to
encourage practitioners in the failure investigation process to look beyond the
metallurgical issues and to also examine the loads and stresses.
Key words: cranes, failure analysis, finite elements method
1. INTRODUCTION
Exploitation of heavy duty and high-capacity lifting/conveying and earthmoving
machines such as cranes (tower cranes, container cranes, ship unloaders, gantry
cranes, elevators, conveyors, etc.), bucket wheel excavators and stacker/reclaimers
under the action of highly pronounced dynamic loads, may lead to failures of their
structural parts, substructures and subassemblies - plastic deformations, cracks and
fractures. Failures of the cranes structural parts unavoidably lead to serious damages
or total collapses; these accidents are often followed by very high financial losses
(millions of ) and possibly serious injuries or crane-related fatalities. A lot of data
reveals on the serious consequences of cranes accidents. For instance, more than
500 US construction workers died in crane accidents between 1984 and 1994,
according to a study of Occupational Safety and Health Administration (OSHA), while
the International Union of Operating Engineers revealed that 502 workers died in 480
separate accidents - major causes included assembly/dismantling (12 percent), boom
buckling (8 percent), rigging failure (7 percent) and overturn (7 percent). [1]. The US
Department of Labor OSHA in May 2010 estimated there are as many as 82 fatalities
1
Associate Professor, University of Belgrade, Faculty of Mechanical Engineering, nzrnic@mas.bg.ac.rs

Professor, University of Belgrade, Faculty of Mechanical Engineering, sbosnjak@mas.bg.ac.rs
Assistant, University of Belgrade, Faculty of Mechanical Engineering, vgasic@mas.bg.ac.rs
4
Senior Research Associate, Institute for Testing Materials, Belgrade, miodrag.arsic@institutims.rs
2
3
185
annually associated with cranes in construction, and said a more up-to-date standard
would help prevent them. The increase of cranes numbers was followed by the
increase of accidents, so in 2009, according to the source [1], have been reported 303
accidents with 197 deaths (in the same period, according to Aviation Safety Network http://aviation-safety.net/, have been reported 30 aircraft accidents with 757 fatalities).
Statistics reveal that crane accidents are the most common cause for construction site
death. There are over 125,000 cranes used in operation in the construction injury and
another 100,000 in general and maritime industries. The causes of crane-related
deaths in US construction industry in the period 1992-1996 are given in Fig. 1 [2].
Fig. 1 Causes of crane-related fatalities in US construction (1992-2006)[2]

* Included 64 struck by falling booms/jibs, ** Included 21 falls from cranes, 9 falls from crane
baskets, 8 from crane loads, ***Other causes included 9 highway incidents.
Types of cranes usually involved in statistical data are mobile cranes, tower
cranes, floating or barge cranes and overhead cranes. At least 71% of all crane-related
accidents involved mobile cranes, Fig. 2a. Mobile cranes are involved e.g. in 63% of all
crane collapses, tower cranes are involved in 5% of all crane-related incidents, while
other and unspecified cranes were involved in 24% of all crane related incidents [2].
The reasons of the crane collapses are shown in Fig. 2b [2]. According to [3], there is
30 plus major accidents annually worldwide with around 50 deaths. Also, since the
year 2000 there have been over 1112 tower crane accidents which have resulted in
over 778 deaths and countless injuries [3]. Of course, many accidents are never
reported on, so the data can perhaps be doubled. Worldwide tower crane accident
statistics in the year 2009 gives 176 accidents resulting in 75 deaths, while in the year
2010 we have 112 accidents with 62 deaths [3]! The incidents analyses statistics show
that a non-negligible number of reasons leading to the accidents remain unknown.
Mechanical and civil engineering structures are always designed to carry their
own dead weight, superimposed loads and environmental loads such as e.g. wind or
waves. These loads are usually treated as maximum loads not varying with time and
hence as static loads. In some cases, the applied load involves not only static
components but also contains a component varying with time which is a dynamic load.
In the past, the effects of dynamic loading have often been evaluated by use of an
equivalent static load, or by an impact factor, or by a modification of the factor of safety
186
Some Aspects in Failure Analysis of Cranes
[4]. Many developments have been carried out in order to try to quantify the effects
produced by dynamic loading. Examples of structures where it is particularly important
to consider dynamic loading effects are the construction of tall buildings, long bridges
under wind-loading conditions, buildings in earthquake zones, high-performance
cranes etc. Exploitation of high-performance cranes in insensitive working conditions
provides fertile ground for the occurrence of fatigue cracks, while extreme
environmental conditions may also cause disastrous consequences, even when
machines are out of operation. Brittle fracture occurrence of vital parts of the structure
is also possible in the installation stage of the machine [5]. The above examples
confirm that a failure with disastrous consequences is possible at any stage of the
product life cycle [5]. In addition, they confirm the factual existence of four main
reasons for the collapse of high-capacity lifting/conveying and earthmoving machines
[6]: (1) design faults, the so-called designing-in defects, [6,7]; (2) manufacture faults
causing the so-called manufacturing-in defects, [6,7]; (3) exploitation faults
according to [7], these causes can be named operating-in defects; and (4) extreme
environmental impacts unusual occurrences (extreme storm, earthquake, fire)
according to [7], these causes can be named environment-in defects. Of course,
machine failures are often the result of a combination of several different causes [6].
Fig. 2 a) Overturning of the mobile crane, b) Statistics of crane collapses

2. SOME EXAMPLES IN CRANES FAILURE
Several different examples of crane failures and accidents, reported in the last
decade, can be found in the following references and explained shortly in this paper.
Failure analysis of mobile harbor crane wheel hub is presented in [8], while the failure
analysis demonstrated that the mechanism of failure was fatigue due to the serviceinduced growth of a fatigue crack and stress-concentrating effects of that crack on
motioninduced stresses. Failure analysis of a large ball bearing of a dockside crane is
shown in [9]. Analysis of the bearing showed that faulty hardening of the bearing
caused pitting of the raceways, which led to failure of the bearing. It should be
mentioned that in both papers [8,9] the authors didn't carry out a thorough structural
analysis and/or e.g. employ FEM to evaluate the failure. Practically they used a
''mettalurgical'' approach to analyze failure. On the other hand some failure case
studies encompassed a more detailed structural analysis including FEM [3, 10, 11, 12,
13]. For instance fatigue damage analysis and repair procedure at the large 250 kN
portal crane placed in the shipyard is shown in [10]. The fatigue cracks occurred at
several critical points, i.e., on the bottom of the tower and on both legs of the portal,
187
Fig. 3a). Previous attempts of repair by simple welding of cracks were not successful,
because new cracks were detected soon after the repair, Fig. 3b).When cracks
reached the critical length (300500 mm), the exploitation of the crane was stopped
and detailed analysis was carried out. First cracks were detected soon after the crane
was placed in the shipyard. The crane was still in usage with the reduced allowed
carrying capacity (50 kN instead of nominal 250 kN), but the cracks continued to grow.
In order to find the source of crack initiation and growth, the complete documentation
and calculations were checked. It was found that the calculations were performed by
using the finite element method and simple beam elements, without taking into account
the stress concentrations, the influence of inertial forces and the wind. Those facts led
to perform the complete static and fatigue analysis, what involved measurements of
real stresses during the typical maneuvers.
Fig. 3 a) Location of fatigue cracks, b) Cracks after repair

The catastrophic collapse of the crane on the Milwaukee Brewers baseball
stadium retractable roof project (Miller Park) that could be the most awesome lift
accident of all time is shown in [11], while the investigation of the cause and origin of
the collapse of an overhead tower crane at an office building construction site in
Bellevue, USA, is given in [12]. Failure analysis of container crane boom failure
occurred while the boom were being lowered is shown in [13]. The most recent failure
analysis of a tower crane that was used during the construction of a hydroelectric
power plant is given by the authors of this paper in [3]. Due to the breakaway of the
gusset plate of the counterjib truss structure, Fig. 4a) and 4b), plastification occurred
and the counterjib ballast fell from the crane. This caused the cranes loss of static
stability whereupon the jib leaned (at a distance of approximately 29 m from the axis of
rotation) onto the structure of the adjacent portal crane. Because of this there was no
total overturning of the crane.
Fig. 4 a) Fractured gusset plate, b) 3D model of the joint fracture

188
Some Aspects in Failure Analysis of Cranes
In order to diagnose the cause of the gusset plate fracture the authors in [3]
carried out FEA to calculate the working stress in the gusset plate accompanied by
chemical composition and mechanical properties testing and microstructure testing. It
was concluded that the fracture of the gusset plate originated from the cumulative
influence of the following factors, such as influence of the parent metal hardening in
the fracture zone.
3. CONCLUSION
Failure of a structural or mechanical component usually can be associated with
materials-related problems and/or design-related problems (which may include,
depending on the definition of design, unexpected service environment). Materials
failure analysis investigations are usually carried out first, and the objective is to
determine what the material can reveal about the cause of failure [14]. This type of
failure analysis consists of first evaluating the failed material for evidence of the failure
mechanism such as fatigue, overload, corrosion and environmentally assisted cracking.
But, at the same time an engineering review of the component design and
service application to determine the loads, displacements, temperatures, vibrations,
and other service-related factors is required. It may also include numerous analytical,
classical, and computer techniques that are available to assist in a structural design
failure analysis. However, traditional analytic techniques have their limitations in that
only relatively simple and idealized structures can be analyzed using simplified loading
and materials property assumptions. When complex designs, such as highperformance material handling and mining machines, transient loadings, and nonlinear
material behavior need to be evaluated, computer-based techniques are used. This is
where finite element analysis (FEA) is most applicable and provides considerable
assistance in design analysis as well as failure analysis [14]. Finite element analysis is
one of the most common tools used by design engineers in failure analysis
[3,10,11,12,13]. It has application in the structural/mechanical fields to determine the
stress, strain, and displacements of structures subjected to different types of loading.
Therefore, in the failure investigation process it is necessary to look beyond the
metallurgical issues and to also examine the loads and stresses.
The majority of large structures such as cranes have significant welding and
hence small defects (cracks) are already present and in [15] is assumed that small
cracks are present and that crack propagation is the only part of the fatigue process.
Because of non-uniform temperature gradient and local elastic/plastic deformations
during the welding and cooling process, residual stresses are found in all welded
structures. The residual stresses profiles and their magnitudes are difficult to quantify.
Therefore it is usual for welded structures to ignore the effects of mean stress and
whether the stress is tensile or compressive. An investigation given in [15] after
analyzing failures of over sixty materials handling machines found that about ten
percent of shiploader, stacker, reclaimer and stacker/reclaimer failures can be
attributed to fatigue failure. Similar can be concluded for other type of material handling
machines. In most cases these failures were unexpected and lead to catastrophic
consequences. The process of fatigue means that steel strength deteriorates under the
action of cyclic loads and this may ultimately lead to cracking and the unexpected
failure of structures.
189
ACKNOWLEDGMENT
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
190
CraneAccidents.com, official site, from http://craneaccidents.com/stats.htm,

McCann, M. (2010). Understanding crane accident failure: A report on causes of
deaths in crane-related accidents, Crane and Rigging Conference. Presented at
CRC 2010, May 27, Houston, Texas, US.
Zrni, N., Bonjak, S., Gai, V., Arsi, M. (2011). Failure analysis of the tower
crane counterjib. Procedia Engineering, Elsevier, accepted for publication.
Kuntiyawichai, K., Burdekin, FM. (2003) Engineering assessment of cracked
structures subjected to dynamic loads using fracture mechanics assessment.
Engineering Fracture Mechanics, Vol. 70, p. 1991-2014.
Hadianfard, MJ, Hadianfard, MA. (2007) Structural failure of a telescopic
shiploader during installation. Journal of Failure Analysis and Prevention, Vol. 7:
p. 28291.
Bonjak S., Arsi, M., Zrni, N., Rakin, M., Panteli M. (2011) Bucket wheel
excavator: Integrity assessment of the bucket wheel boom tie-rod welded joint.
Engineering Failure Analysis. Vol. 18. p. 212222.
Gagg, CR. (2005) Failure of components and products by engineered-in defects.
Case studies. Engineering Failure Analysis. Vol. 12.10001026.
Shirokoff, J. Mobile harbor crane wheel hub fatigue failure. (2003) Practical
Failure Aanalysis.; Vol. 3(1). p. 81-84.
Ost, W., De Baets, P., De Waele, W. (2004) Failure of a large ball bearing of a
dockside crane. Engineering Failure Analysis. Vol. 11. p. 33553.
Domazet, ., Krstulovi-Opara, L., Stupalo, S. (2005) Fatigue cracks and failures
in cement industry, shipbuilding and power plant facilities. Engineering Failure
Analysis. Vol. 12. p. 819833.
Ross B., McDonald, B., Vijay Saraf, SE. (2007) Big blue goes down. The Miller
Park crane accident. Engineering Failure Analysis. Vol. 14. p. 942961.
McDonald, B., Ross, B., Carnahan, RA. The Bellevue Crane Disaster. (2011)
Engineering Failure Analysis, doi: 10.1016/j.engfailanal.2010.09.003.
Soderberg, E. (2008) Container crane boom collapse: Cause and prevention. Port
Technology International. Vol. 39. p. 60-65.
Service, T. (2002) ASM Handbook Vol. 11, Failure Analysis and Prevention (ASM
International). Finite Element Modeling in Failure Analysis. p. 380-389.
Morgan, R. Gatto, F., Ford, P. (2005). Fatigue of Complex Structures for the
Mining Industry. Proceedings Australian Structural Engineering Conference.
Sydney, N.S.W.: Engineers Australia. p. 753-759.
SOME ASPECTS TO DESIGN OF AIRCRAFT STRUCTURES WITH

RESPECTS TO FATIGUE AND FRACTURE MECHANICS
Stevan Maksimovi1, Ivana Vasovi2, Mirko Maksimovi3
Summary: Attention in this work is focused on developing computation methods of
damaged aircraft structural components with respect fatigue and fracture mechanics.
Computation method is based on combining singular finite elements to determine
stress intensity factors for cracked structural components with corresponding crack
growth lows that include effect of load spectra on number of cycles or blocks up to
failure. To demonstrate efficient computation procedure in fatigue life estimation here
numerical examples are included. Attention in this work is focused on design of aircraft
wing-fuselage joints. Computation procedure to strength analyze with respects fatigue
and fracture mechanic is applied to cracked aircraft lug type structure. Computation
results are compared with correspond experiments. Good agreement between
computation and experimental results is obtained
Key words: Fatigue, Fracture mechanics, Aircraft structures, Damaged lugs, Finite
Elements,Crack growth, Strain energy density method
1. INTRODUCTION
Damage tolerance application to the light training aircraft structural components
is limited to critical parts. A part, that if it fails, alone may cause the loss of an aircraft is
classified as a critical part. This definition means, that aircraft wing-fuselage
attachments must be comply with the damage tolerance requirements [1,2]. The main
goal is a safe life design, i.e. a slow crack growth structure not requiring any insection
during its full life. The Damage Tolerance approach assumes the components have a
preexisting flaw from which a crack will grow under dynamic loads. This assumption
makes it possible to account for in-service or manufacturing defects in determining the
dynamic life. The Damage Tolerance Methodology uses fracture mechanics to predict
the fatigue crack growth in a structure. In the design analysis of a slow crack growth
structure it is most important to make correct estimates for the early portion of the
crack growth process, because it is there the life is. In most cases this implies that
maximum accuracy is needed for small corner cracks.
Ph.D, Stevan Maksimovi, Belgrade, VTI, Ratka Resanovia 1, (s.maksimovic@open.telekom.rs)

Ph.D student, Ivana Vasovi, GOA institute Belgrade, (ivana.vasovic@gmail.com)
3
Ph.D student, Mirko Maksimovi, Water supply and Sewerage, Belgrade, (mirkom@open.telekom.rs)
2
191
S. Maksimovi, I. Vasovi, M. Maksimovi
Attachment lugs are particularly critical components in crack initiation and

growth because of their inherently high stress concentration levels near the lug hole.
For these reasons, it is important to develop analytical/numerical as well as
experimental procedures for assessing and designing damage tolerant attachment lugs
to ensure the operational safety of aircraft. Over the years, several extensive studies
[3-5] have been made on lug fatigue performance, involving both experimental and
numerical means.
In the work of fatigue crack growth and fracture behavior of attachment lugs
[6,7], an accurate calculation of the stress intensity factor is essential. Over the years
several methods have evolved to compute the stress intensity factors for structural
components containing cracks. These methods include analytical as well as
experimental approach. The experimental backtracking approach was used to derive
empirically the stress intensity factors for structural components using the growth rate
data of through-the-thickness cracks for simple geometry subjected to constantamplitude loading.
The finite element method can be applied easily to the propagation of the crack
under constant amplitude loading condition, but it is difficult to use under random
loading condition due to the complexity of modeling the retardation behavior of the
crack (crack closure effect suggested by Elber), which can be introduced by
overloadings.
Accurate stress-intensity factor (SIF) solutions are required to conduct thorough
damage tolerance analyses of structures containing cracks. Exact closed form SIF
solutions for cracks in three-dimensional solids are often lacking for complex
configurations; therefore, approximate solutions must be used. Over the past two
decades, considerable effort has been placed on developing computationally efficient
methods which provide highly accurate SIF solutions for cracks in three-dimensional
bodies.
The purpose of this investigation was to test the accuracy of the crack growth
models. All necessary parameters, such as material property data, stress intensity
solutions, and the load spectrum, were defined. To determine residual life of damaged
structural components here are used two crack growth methods: conventional
Forman`s crack growth method and crack growth model based on the strain energy
density method. The last method uses the low cycle fatigue properties in the crack
growth model.
2. COMPUTATION PROCEDURES TO LIFE ESTIMATIONS
An analysis to show complience with a crack growth criterion, as required by a
damage tolerance specification, involves following steps:
Calculation of the stress intensity factors
Calculation of the critical crack size
Compilation of crack growth data
Adoption of a model for prediction of fatigue crack growth

In the design analysis of a slow crack growth structure it is most important to
make correct estimates for the early portion of the crack growth process, because it is
there the life is. In most cases this implies that maximum accuracy is needed for small
corner cracks. Stress intensity Factor (SIF) is an important fracture parameter to
192
Some aspects to design of aircraft structures with respects fatigue and fracture mechanics
predict the behaviour of the joint in the presence of cracks using Linear Elastic
Fracture Mechanics (LEFM). For a successful implementation of the damage tolerance
philosophy to the design and in-service operation of structures subjected to fatigue
loading it is crucial to have reliable crack growth prediction tools
3. STRESS INTENSITY FACTOR OF CRACKED LUGS
In general geometry of notched structural components and loading is too
complex for the stress intensity factor (SIF) to be solved analytically. The SIF
calculation is further complicated because it is a function of the position along the crack
front, crack size and shape, type loading and geometry of the structure. In this work
analytic and FEM were used to perform linear fracture mechanics analysis of the pinlug assembly. Analytic results are obtained using relations derived in this paper. Good
agreement between finite element and analytic results is obtained. It is very important
because we can to use analytic derived expresions in crack growth analyses. Lugs are
essential components of an aircraft for which proof of damage tolerance has to be
undertaken. Since the literature does not contain the stress intensity solution for lugs
which are required for proof of damage tolerance, the problem posed in the following
investigation are: selection of a suitable method of determining othe SIF, determination
of SIF as a function of crack length for various form of lug and setting up a complete
formula for calculation of the SIF for lug, allowing essential parameters. The stress
intensity factors are the key parameters to estimate the characteristic of the cracked
structure. Based on the stress intensity factors, fatigue crack growth and structural life
predictions have been investigated. The lug dimensions are defined in Fig. 1.
=t
W
Output Set: Case 12Step 0.500000

Contour: Plate Top Equivalent Stre
29.01
27.28
25.55
23.82
22.09
2R
20.36
18.63
16.9
15.17
13.45
11.72
9.987
8.258
6.529
Y
Z
4.8
3.071
1.342
r
F
Contact pin/load FE model
Fig 1 Geometry and loading of lugs

To obtain stress intensity factor for the lugs it is possible to start with general
expression for the SIF in the next form
(3.1)
K YSUM V S a
193
where: Y correction function, a- the crack length. This function is essential in

determining of the the stress intensity factor. Primary, this function depends on stress
concentration factor, kt and geometric ratio a/b. The correction function is defined using
experimental and numerical investigations. This correction function can be defined in
the next form [13]:
YSUM
k
b
1.12 kt A
k Q
a
A
b
(3.2)
er a b
w 2 R
2
(3.3)
(3.4)
2 R
2 R
3.22 10.39
7.67
w
w
a
U 103
b
a
103
b
2 R
2 R
0.72 0.52
0.23
H
H
(3.5)
(3.6)
a
1.8951
b
A 0.026 e
(3.7)
(3.8)
The stress concentration factor kt is very important in calculation of correction

function, eq. 3.2. In this investigation. A contact finite element stress analysis was used
to analyze the load transfer between the pin and lug.
4. CRACK GROWTH ANALYSIS USING CONVENTIONAL APPROACH
For crack growth analysis and fatigue life estimations have been used various
conventional crack growth models. Many of these models achieve correct solutions of
crack growth analyses for cracked structural elements under cyclic loads of constant
amplitude. Hovever, for construction under cyclic loads of variable amplitude in form of
load spectrum such as in aircraft cases it is necessary tu include the effects of shape
of load spectra and its effects of estimation life of structural elements [9].
Forman, Newman and others [8] developed the NASGRO equation, which is an
equation often used to describe crack growth. This equation describes the crack
growth curve in terms of the crack length a , the number of cycles N, the stress ratio R,
the stress intensity factor range DK, and material constants, C, n, p, q through best fits
of the da/dN - DK data.
194
'K th
n 1
1 f

'K
C
'K
q
1 R
K max
1
Kc
da
dN
(4.1)
where: a - crack length, N-number of cycles, C, n, p, q are experimentally derived

material parameters , K is the stress intensity factor (SIF), K th is the threshold stress
intensity factor, R is the stress ratio,
K c - is the critical stress intensity factor. The
Newman closure function isone of these terms and is define as f :

f
K op
K max
max
, A0 A1 R A2 R 2 A3 R 3 ; R t 0
A0 A1 R ;
(4.2)
2 d R 0
and the coefficients are given by:
A0
A1
A2
A3
S Smax
(0.825 34D 0.05D 2 ) cos(
2 V0
(0.415 0.071D ) Smsx / V 0

1 A0 A1 A3
2 A0 A1 1
where: D is the plane stress/strain constraint factor, (Vmax/V0) is the ratio of maximum
stress to the flow stress. The threshold stress intensity factor range is calculated by the
following empirical equation:
1
'K th
1 f
a 2
'K th 0
/
a 1 (1 Ao )(1 R)
(1 Cth R )
(4.3)
Relation (4.1) represents one general crack growth models based on

conventional approach. This relation can be transformed to conventional Forman`s
crack growth model [5]. In region III rapid and unstable crack growth occurs, so
Forman at al. Proposed equation for region III as well as for region II, [9].
da
dN
C 'K
1 R KC 'K
n
(4.4)
where KC is the fracture toughness. Forman`s equation has been developed to model
of unstable crack growth domain (III).
5. CRACK GROWTH MODEL BASED ON STRAIN ENERGY DENSITY METHOD
In this work fatigue crack growth method based on energy concept is
considered and then it is necessary to determine the energy absorbed till failure. This
energy can be calculated by using cyclic stress-strain curve. Function between stress
195
and strain, as recommended by Ramberg-Osgood provides good description of elasticplastic behavior of material, and may be expressed as
1
'H
'V
'V n /
2 / .
E
2k
(5.1)
where E is the modulus of elasticity, 'H/2 is strain amplitude and 'V/2 is stress
amplitude. Equation (5.1) enables the calculation of the stress-strain distribution by
knowing low cyclic fatigue properties. As a result the energy absorbed till failure
become [10,11]
Wc
4
V /f H /f
/
1 n
(5.2)
where Vf/ is cyclic yield strength and Hf/ - fatigue ductility coefficient. Given the fact that
strain energy density method is considered, the energy absorbed till failure must be
determined after the energy concept is based on the following fact: The energy
absorbed per unit growth of crack is equal to the plastic energy dissipated within the
process zone per cycle. This energy concept is expressed by
(5.3)
Wc Ga = Zp,
where Wc is energy absorbed till failure, Zp- the plastic energy and a - the crack length.
In equation (5.3) it is necessary just to determine the plastic energy dissipated in the
process zone Zp. By integration of equation for the cyclic plastic strain energy density
in the units of Joule per cycle per unit volume >10@ from zero to the length of the
process zone ahead of crack tip d* it is possible to determine the plastic energy
dissipated in the process zone Zp. After integration relation of the plastic energy
dissipated in the process zone becomes
Zp
1 n /
/
1 n
'K I2 \
E I n/
(5.4)
where 'KI is the range of stress intensity factor, \ - constant depending on the strain
hardening exponent n/, In/ - the non-dimensional parameter depending on n/.
Fatigue crack growth rate can be obtained by substituting Eq. (5.2) and Eq. (5.4) in Eq.
(5.3)
da
dN
1 n \ 'K
/
4E I n/ V H
/
f
/
f
'K th ,
2
(5.5)
where 'Kth is the range of threshold stress intensity factor and is function of stress ratio
i.e.
'Kth= 'Kth0(1-R)J,
(5.6)
'Kth0 is the range of threshold stress intensity factor for the stress ratio R = 0 and J is
coefficient (usually, J = 0.71). Finally number of cycles till failure can be determined by
integration of relation for fatigue crack growth rate
196
ac
a0
da
'K I 'K th 2
4 E I n / V /f H /f
1 n \
/
(5.7)
and
'K I
Y S S a,
(5.8)
Equation (5.7) enables us to determine crack growth life of different structural

component. Very important fact is that equation (5.7) is easy for application since low
cyclic material properties (n/, Vf/, Hf) available in literature are used as parameters. The
only important point is stress intensity factor which, depending on the geometry
complexity and the type of loading, could be determined by using analytical and/or
numerical approaches.
6.
NUMERICAL VALIDATION
To illustrate computation procedures in damage tolerance analysis and residual

life estimations of damged structural components here are numerical examples
included.
6.1 Life estimation of damaged structural elements
Subject of this analyses are cracked aircraft lugs under cyclic load of conctant
amplitude end spectra. For that purpose conventional Forman crack growth model and
crack growth model based on strain energy density method are used. Material of lugs
is Aluminum alloy 7075 T7351 with the next material properties:
Vm=432 N/mm2 Tensile strength of material
V02=334 N/mm2
KIC=2225 [N/mm3/2]
Dynamic material properties (Forman`s constants): CF=3* 10-7, nF=2.39.
Vf/=613 P,
Hf/=0.35, n/=0.121.
Cyclic material properties:
The stress intensity factors (SIF`s) of cracked lugs are determined for nominal
stress levels: Vg = Vmax=98.1 N/mm2 and Vmin=9.81 N/mm2. These stresses are
determined in net cross-section of lug. The corresponding forces of lugs are defined
as, Fmax= Vg (w-2R) t = 63716 N and Fmin= 6371.16 N, that are loaded of lugs. For
stress analyses contact pin/lug finite element model is used.
For cracked lugs defined in Table 6.1, with initial cracks a0, SIF`s are determined using
finite elements, Table 6.2. To obtain high-quality results of SIF`s cracked lugs are
modeled by singular finite elements around crack tip.
197
83.3
160
t = 15 mm
44.4
R20
R41.6
F = 6371.6 daN
Fig 6.1: Geometry of cracked

lug 2
Fig. 6.2. Finite Element Model of cracked lug

with stress distribution
Table 6.1 Geometric parameters of lugs [13]

Lug
No.
2
6
7
2R
40
40
40
W
83.3
83.3
83.3
Dimensions [mm]
H
44.4
57.1
33.3
L
160
160
160
t
15
15
15
The stress intensity factors of cracked lugs are calculated under stress level: Vg
= Vmax=98.1 N/mm2, or corresponding axial force, Fmax= Vg (w-2R) t = 63716 N. In
present finite element analysis of cracked lug is modeled with special singular quarterpoint six-node finite elements around crack tip, Fig. 6.2. The load the model, a
concentrated force, Fmax, was applied at the center of the pin and reacted at the other
and of the lug. Spring elements were used to connect the pin and lug at each pairs of
nodes having identical nodal coordinates all around the periphery. The area of contact
was determined iteratively by assigning a very high stiffness to spring elements which
were in compression and very low stiffness (essentially zero) to spring elements which
were in tension. The stress intensity factors of lugs, analytic and finite elements, for
through-the-thickness cracks are shown in Table 6.2. Analytic results are obtained
using relations from previous sections, eq. (3.1).
Table 6.2 Comparisons analytic with FEM results of SIF
Lug No.
a >mm@
K IFEM
max
.
K IANAL
max
2
6
7
5.00
5.33
4.16
68.784
68.124
94.72
65.621
70.246
93.64
198
From above Table 6.2 is evident good agreement between analytic and finite
element results for determination of stress intensity factors. Accuracy of SIF`s is very
important in precise crack growth analyses and life estimation of cracked lugs. That
means that proposed analytic model for determination of SIF`s is adequate in crack
growth analyses.In design process is very important to know how any geometric
parameters of lug have the effects on fracture mechanics parameters. In Fig. 6.4 are
shown dependence SIF, Kmax, and height of head of lug H. In this analysis geometric
properties of lugs are given in Table 6.1. From Fig. 6.4 is evident increasing of SIF`s
with increasing crack length and reducing with increasing height of lug`s head.
In Fig. 6.3 are shown computation and experimental results of cracked lug No.
2 as defined in Fig. 6.1 and Table 6.1. In this computation analysis Forman crack
growth model is used. Good agreement between computation and experimental results
is obtained. It is evident that computation Forman`s crack growth model is to a small
extent conservative for longer crack, Fig. 6.3.
Lug No.2
) 20
m
m
(
a 15
Eksperiment
Proracun
10
5
0
0
10 12 14
Nf x 10
Fig.6.3 Comparisons computational with experimental crack growth results for lug No.2
(H=44.4 mm); kt=2.8
Kmax [N mm0.5]
1 400
H=44. 4 mm
H=57. 1 mm
H=33. 3 mm
1 1 50
900
650
400
0
0.25
a/b
0.5
0.75
Fig. 6.4 The effects of head length (H) in function of crack length to SIF of Lugs
denoted as: 7 (H=33.3), 2 (H=44.4), 6 (H=57.1)
199
In Fig. 6.5 are shown results of crack growth results for cracked lug using two
methods: (1) conventional Forman`s method i (2) strain energy density method (GED) [7].
a [mm]
22.5
20
Forman
GED
17.5
15
12.5
10
7.5
N [ciklusa]
2.5
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
12000
Fig. 6.5 Comparisons crack growth resultats of cracked lug using two crack growth
models: (1) Forman`s and (2) Strain Energy Density (GED) methods
Results in Fig. 6.5 shows good correlation crack growth results of cracked lugs
using two crack growth methods; (1) conventional Forman`s method and (2) crack
growth model based on strain energy density method (GED).
7. CONCLUSIONS
Predictions and experimental investigations for fatigue life of an attachment lug
under load spectrum were performed. From this investigation followings are concluded:
A model for the fatigue crack growth is included which incorporates the low cycle
fatigue properties of the material.
Comparisons of the predicted crack growth rate using strain energy method method
with experimental data and conventional Forman`s model points out the fact that
this model could be effective used for residual life estimations
The stress intensity factor of cracked lug is well defined by analytical method since
there is really minor difference when compared results obtained by singular finite
elements.
Acknowledgment: This work was financially supported by the Ministry of Science and
Technological Development of Serbia under Project No. TR-34028.
200
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
Airplane Damage Tolerance Requirements, MIL-A-83444, Air Force

Aeronautical Systems Division, July 1974.
Brussat, T.R., Kathiresan, K and Rud, J.L., Damage tolerance assessment lugs,
Engineering Fracture Mechanics, 23, 1067-1084, 1986
Maksimovi, S., Fatigue Life Analysis of Aircraft Structural Components, Scientific
Technical Review, 2005.
Maksimovi K., Nikoli-Stanojevi V., Maksimovi S., MODELING OF THE
SURFACE CRACKS AND FATIGUE LIFE ESTIMATION, ECF 16, 16th
European Conference of Fracture, ECF 16, Alexandroupolis, Greece, 2006.
Maksimovi, S., Boljanovi, S, Maksimovi, K., Fatigue life prediction of Structural
Components under variable amplitude loads, FATIGUE 2002, 8th International
Fatigue Congress (IFC8), Stocholm, 2-6. June, 2002.
Antoni, N and Gaisne F., Analytical modeling for static stress analysis of pinloaded lugs with bush fitting, Applied Mathematical Modeling, 35, 2011, 1-21.
Liu, Y.Y and Lin F.S., A mathematical equation relating low cycle fatigue data to
fatigue crack propagation rates. Int. J. Fatigue, Vol. 6 (1984), pp.31-36
Forman, R. G., and Mettu, S. R., Behavior of Surface and Corner Cracks
Subjected to Tensile and Bending Loads in Ti-6Al-4V Alloy, Fracture Mechanics:
Twenty-second Symposium, Vol. 1, ASTM STP 1131, H. A. Ernst, A. Saxena, and
D. L., McDowell, eds., American Society for Testing and Materials, Philadelphia,
1992, pp. 519-546.
Forman, R.G., V.E. Kearney and R. M. Engle, Numerical analysis of crack
propagation in cyclic loaded structures, J. Bas. Engng. Trans. ASME 89, 459,
1967.
Liu S.B and Tan C.L., Boundary element contact mechanics analysis of pinloaded lugs with single cracks, Engineering Fracture Mechanics, Vol. 48,No.5,
717-725, 1994.
Maksimovi, S., Boljanovi, S., Fatigue Life Prediction of Structural Components
Based on Local Strain and an Energy Crack Growth Models, WSEAS
TRANSACTIONS on APPLIED and THEORETICAL MECHANICS, Issue 2,
Volume 1, 2006, pp 196-205.
Maksimovi, K., Maksimovi, S., Analytic approach for determination of fracture
mechanics parameters and crack growth analysis of 3-D surface cracked
structural elements, Technical Diagnostic, Vol. III, No. 1, 2004.
Maksimovi K., Damage tolerance analysis of aircraft constructions under
dynamic loading, Master Thesis, Faculty of Mechanical Engineering, University of
Kragujevac, 2003.
201
STATE-OF-THE-ART IN LIFE CYCLE ASSESSEMENT AS A CORE

OF LIFE CYCLE DESIGN
Andrija Vujii 1, Nenad Zrni2
Summary: This paper deals with the state-of-the-art of Life Cycle Assessment (LCA)
as a core of Life Cycle Design (LCD) of a product. Background of LCA is given, as well
as some modern approaches to LCA studies. The paper focuses on the practical
methods in LCA including the list of main widespread software and some other
computer tools with the basic descriptions of their abilities. Finally, a short survey of
critics of LCA methods is presented, as well as some proposals for improving the
method.
Key words: Life Cycle Design, Life Cycle Assessment, LCA Software
1
INTRODUCTION
Growing interest in environmental condition and concept of sustainable

development resulted in appearance of design disciplines with goal to develop
solutions for decreasing the impact of human activities and industrial products on
environment. Among disciplines such as Ecodesign and Design for Environment, one
is established a completely new design approach Life Cycle Design [1].
Based on fundamentals of LCD, all products and services must be designed to
scope all of their life cycles stages. Processes needed for manufacturing, distributing
and disposing of the product at the end, are considered to be one. Beside the design of
product itself, this approach designs product system, in a way that it determents all
possible events in product life cycle.
Having this in mind, LCD discipline makes a radical discontinuity with traditional
design process [2]. The purpose of this is to design product life cycles, in order to
indentify and efficiently combat environmental impacts. In other words, it is possible to
create product and minimize the input of raw materials and energy, and the impact of
all emissions and waste.
Before design of any new product life cycle, and before any improvement of
existing product life cycles, it is of vital importance to estimate the present effects on
environment. The only way to both quantitatively and qualitatively calculate the harm of
all effects of products and human activities on environment is to conduct Life Cycle
Assessment or LCA [3].
1
2
Mag.Tech.Sc., Andrija Vujii, Belgrade, Dunav Insurance Company, avujicic@dunav.com

Assoc.Prof.Dr.Sc., Nenad Zrni, Belgrade, Faculty of Mechanical Engineering, nzrnic@mas.bg.ac.rs
203
Vujicic A., Zrnic N.
2.
LIFE CYCLE ASSESSEMENT
As the core of LCD the LCA is quantitative tool for assessment of

environmental impacts of products and services such as climate change, global
warming, ozone depletion, (smog) creation, eutrophication, acidification, toxicological
impact on humans and ecosystems, the depletion of resources, and others. It is
systematic approach for analyzing the entire life cycle stages of products from material
extraction through manufacturing, use and eventually disposal or recycling preferably.
Therefore it is often called a cradle-to-grave analysis.
The first form of LCA appeared in 1970 in the U.S. as a method for evaluation
of resource and energy use. This methodology was developed to scope resource and
energy consumption with air and land emissions through life cycle of product and
packages. After 1990 and work conducted be SETAC, the LCA method was
inaugurated. Finally, after publishing the principles of conducting the LCA by ISO
(1997), the wide acceptance of this method started [4].
2.1 LCA Framework
The framework of LCA method consists of four phases (ISO14040-43):
1.
2.
3.
4.
Goal and scope definition

Life Cycle Inventory
Life Cycle Impact assessment
Interpretation
Fig. 1 LCA Framework [5]

2.1.1
Goal and scope definition
Goal and scope definition is the first stage of LCA in which all general decisions
are made. For definition of goal, the application, purpose and intended audience and
comparative analysis is determent. If the study is planned to be published the critical
review is a must. In order to set scope definition the methodology is defined. Several
factors are determent, from functional unit, through data quality requirements, and all
rounded with system boundaries. Functional unit can be presented as quantified
definition of the function of a product, which enables different products to be compared
and analyzed. Data quality requirement is a factor which describes data acquisition
204
State-of-the-art in Life Cycle Assessment as core of Life Cycle Design
(data measured or calculated) and are data sufficient to represent the system. System
boundaries define which processes will be assessed. Boundaries are defined with cutoff criteria.
There are four main options used to define the system boundaries (see Fig 2,
modified chart [5]):
Cradle to Grave: This assessment consists of all processes from the raw material
extraction through the production, distribution and use phase, up to the end of life.
Cradle to Gate: Includes all processes necessary to determine the environmental
impact of the production of a product, and thus term gate is used as factory gate.
Gate to Grave: Goal of this assessment is to determent the environmental impacts of
a product once it leaves the factory.
Gate to Gate: includes the processes from the production phase only and it is used to
determine the environmental impacts of a single production step or process.
Fig 2. System boundaries

2.1.2
Life Cycle Inventory
Inventory analysis is a methodology for estimating the consumption of

resources and the quantities of waste flows and emissions caused by products life
cycle. It is LCA phase where all inputs and outputs are collected for a given product
system or process in order to make Life Cycle Inventory LCI. It is iterative process with
data classification and calculation as most important steps. The classification step is
most intensive and complex of all LCA. Here all processes need to be classified as
raw material or other inputs, products or wastes, emissions to air, water and soil or
other environmental aspects.
2.1.3
Life Cycle Impact Assessment
Life Cycle Impact Assessment (LCIA) provides indicators and the basis for
analyzing the potential contributions of the resource extractions and emissions. The
result of the LCIA is an evaluation of a product life cycle, on a functional unit basis, in
terms of recognized environmental impact categories, such as global warming,
acidification, ozone depletion, etc. The emissions are converted to impacts potential
205
through processes of classification and characterization.

In order to conduct LCIA different methods can be used, from problem-oriented
to the damage-oriented approach. Most common methods are TRACI (Tool for the
Reduction and Assessment of Chemical and other environmental Impacts) and CML
(Dutch: Centrum voor Milieukunde Leiden ). The TRACI method is developed from US
EPA (Environmental Protection Agency) while CML is product of University of Leiden
Netherlands. Beside these methods, other common methods are EDIP, Ecoinvent and
Ecoindicator.
2.1.4
Interpretation
The goal of the life cycle interpretation phase is to draw conclusions, identify
limitations and make recommendations for the intended audience of the LCA. This is
iterative phase in which results are checked for consistency with goal and scope, and
significant issues are indentified. Significant issues can consist of inventory elements
such as energy consumption, major flows, emissions etc.
Fig. 3 Example of classification and characterization [5]

2.2 LCA Software
Conduction life cycle assessment can be very complex process which often
requires comprehensive modelling and mathematical base. Nevertheless the powerful
software and database of input/output flows is essential for quality of LCA. As the
interest for LCA method grew, the number of suppliers of LCA software tools on the
market raised. There is choice of software intended for any type of users and designed
for different types of LCA.
Most common LCA tools intended for practitioners are GaBi from PE
International (Germany) and SimaPro developed by PR Consultants (Netherlands),
with 60% and 30% share respectively. Among other most popular are Ecopilot, EcoLab
EcoScan, Ecobilan, EDIP PC-tool, EPS, JEMAI-LCA, GEMIS and others. There are
also LCA tools developed at product level like BEES and ATHENA, etc.
206
State-of-the-art in Life Cycle Assessment as core of Life Cycle Design
2.2.1
GaBi, PE International
GaBi (Ger: Ganzheitlichen Bilanzierung) is most common LCA tool on the

market with largest LCA database for product and process analysis. GaBis approach
to conducting LCA is based on concept of plans and flows.
With in GaBi system, boundaries are presented as a plan, while the flows that
enter the product system coming from the natural system or that leave the system are
called elementary flows. The LCI is the input/output list of all elementary flows
associated with the system. Flows are characterized by mass, energy and costs,
where last feature allows for additional option of conducting Life Cycle Cost (LCC)
assessment.
For the purpose of scientific research, free of charge GaBi Educational version
is available upon request.
2.2.2
SimaPro, PR Consultants
Second most common LCA tool on the market is SimaPro. This software is
suitable for its functional based LCA, screening and accounting LCAs, and other partial
LCA studies.
Within SimaPro input/output data are stored as processes in database or in
project. Modeling of the life cycle is done in steps, as part of a project: the products are
modeled as assembly. The functional unit is modeled in the life cycle, end of scenarios
are modeled separately as disposal, reuse, disassembly scenarios [6].
Demo version is available for download.
2.3 Critical review on LCA
LCA has become very important tool for environmental policy makers,
designers and industry in general. Therefore the proper conduction of the assessment
is vital, in order to avoid misinterpretations, inappropriate use and false conclusions.
Since LCA in most cases works with modeling approach, thus the process
description may not refer to actual practice. Especially in cases when study is not
intended for publishing data are unverifiable and may be inconsistent or incorrect. In
the absence of a formal materials balance accounting system, data inconsistency or
errors may not be detected. Therefore it must be insisted that all data could be
published or accessible even in cases of declared secrecy.
The process data may be synthesized, using models based on laws of
thermodynamics and chemistry. For example the connection between CO2 and fossil
fuel burning. Although this synthesized data may not be incorrect, they could distort the
actual situation [7].
Based on principle that every material product has certain emission and must
eventually become a waste it is important in order to have review or comparison of
products, to take into account its environmental impacts from cradle to grave. With
this not only the inputs and emissions of products in production phase are treated, but
also the future environmental impacts [8].
Also common problem of LCA are differences between US methods on one
side and European and Japanese on the other. It is essential that in process for
unification of methods common mass units and mass balances must be used.
207
3.
CONCLUSION
For relatively short period of existence LCA method has made strong
contribution to concept of sustainable development. Today it is integrative part of every
product designing, eco labeling, process optimization, environmental research and
business efficiency.
From engineerings point the LCA method is an asset in planning phase,
especially for ping pointing process with high environmental impact, and guidance for
ecological efficiency and technological alternatives. It can also be important tool for
estimating SPI and KPI and decision making.
Although there are certain limitations on LCA, discussion on this methodology
is leading amount interested parties is on its improvement and wider acceptance, and
definitely not on its dismissal. With increase of accessible data on chemical and
material production data the accuracy of method will further improve. Also with building
inventory bases on energy and material productions from less developed countries the
LCA could be globally applicable.
Accepting the principles of sustainable development, while mastering the skills
of Ecodesign and Life cycle assessment is best path for engineers, designers and
students of technical faculties. Designing the products with low environmental impact,
which will use renewable energy sources will be mandatory in near future. Therefore,
those engineers and , companies which decide to make transition based on concept of
sustainable development will have significant edge over those who remain
conservative.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
208
Vezzoli C., Manzini E., (2007), Design and Innovation for Sustainability,
Politecnico di Milano, INDACO Department, p. 51-72
Zrni N., Bonjak S., orevi M., (2010), Design for modernization A way for
implementation of eco improvements of ports cranes, Machine Design, Faculty of
technical scineces, Novi Sad, p. 07-13
Azapagic A., Clift R., (1999), The application of life cycle assessment to process
optimisation, Computers and Chemical Engineering 23, Elsevier, p.15091526
Franklin W.E., (1995),Life Cycle Assessment a remarkable tool in the era of
sustainable resource and environmental management, Resources, Conservation
and Recycling 14, Elsevier, p. 05-08
GaBi, (2009), Handbook for Life Cycle Assessment (LCA) - Using the GaBi
Education Software Package, PE International & Universitt Stuttgart
Jnbrink A.K., Wolf-Wats C., Erixon M., Olsson P., Walln E., (2000), LCA
Software Survey, Swedish Environmental Research Institute, Stockholm, p. 20-21
Vujii A.,( 2010), Life Cycle Assessment of Machinery and Devices of
Interruptive Transport, Magister thesis in Serbian, Faculty of Mechanical
Engineering, Belgrade
Ayres R., (1995), Life cycle analysis: A critique, Resources, Conservation and
Recycling 14, Elsevier, p. 199-223
STRESS INTENSITY FACTOR AND ITS EFFECT OF STRUCTURAL

INTEGRITY
Mersida Manjgo1, Ljubica Milovi2, Zijah Burzi3
Summary: Analysis of the behaviour of brittle body with a crack fracture mechanics
has opened up new possibilities in providing security metal structures. Standards BS
7448 Part-1 and ASTM E399 for determining the fracture toughness plane strain, KIc,
the process of developing a linear-elastic fracture mechanics (LEFM) is rounded to its
application to real structures. Test specimens with a crack shows local behaviour of the
material around the crack tip and the assumption that the material is sufficiently
homogeneous specimen, which means that the results of the behaviour can be treated
globally meaning that they can be directly transferred to an appropriate structure.
Fracture mechanics specimens are specimens, a crack in the welded joints as the
most critical errors occur.
Keywords: welded joint, crack, stress intensity factor.
1.
INTRODUCTION
Assessing of structural integrity starts with the basic principles of design that
the components in the exploitation under the load voltage is set lower than the stress
fracture components. And while considered unchanged starting material and its
properties are sufficiently homogeneous structure, a form of components intact, the
integrity assessment is a relatively easy task. Problems arise when one of the three
requirements are disrupted. Thus, in many constructions, steel characteristics are
changed with the change of temperature and there is reduced plasticity at low
temperature, i.e. steel is brittle: no plastic deformation until the appearance of fracture.
As an example of insufficient homogeneity of the material, we have weld, which,
because of the welding process has three different structures and mechanical
characteristics (base metal, heat affected zone and weld metal). The most important
form of damage to components is cracks. Fracture mechanics has demonstrated the
importance of crack by introducing a new parameter, stress intensity factor (K) and its
impact on the integrity of the components shown by the average stress () and crack
length (a), taking into account the geometrical shape of the components of Y factor: [1]
1
doc.dr. Mersida Manjgo, Mostar, Univerzitet Demal Bijedi Mainski fakultet Mostar BiH,
mersida.manjgo@unmo.ba
2
doc.dr. Ljubica Milovi, Beograd, Univerzitet u Beogradu, Tehnoloki fakultet, Srbija, acibulj@tmf.bg.ac.rs
3
prof.dr. Zijah Burzi,Beograd, Vojnotehniki institut Beograd, Srbija, zijah_burzic@vektor.net
209
2.
Y V a
>Pa m @
MATERIAL AND WELDING TECHNOLOGY
For experimental tests we have used steel .4730, JESENICE. The chemical
composition of materials are given in Table 1, and mechanical characteristics are given
in Table 2.
Table 1 Chemical composition of steel .4730
Melt
14-9245
C
0,25
Si
0,26
Mn
0,63
% mas.
P
S
0,015
0,004
Cr
0,97
Mo
0,16
Table 2. The mechanical composition of steel . .4730 able to supply

Melt
14-9245
Yield stress
Rp0,2, MPa
420
Ultimate strength,
Rm, MPa
565
Elongation A, %
25,6
Impact ergy,
KU3, J
60
For welding .4730 steel plates 16 mm we selected MAG (CO2)

semiautomatic welding process. As an additional material, we used steel wire with
diameter 1.20 mm, Sarajevo. Welded joint is abutted with regular x-symmetrical
groove, a groove preparation was done according to standard JUS C. T3.030
Table 3 The parameters of the welding process
Activity
Number of passes in the weld
Temp. predgrijvanja, Tp > qC@
Average current, I > A@
Medium voltage, U >V@
Welding speed, v>cm/min@
The degree of thermal efficiency, K
The amount of heat input, Q >J/cm@
3.
Test X
8
170
160
24
22,5
0,7
7680
DETERMINATION OF FRACTURE TOUGHNESS KIC
Test specimens with a crack shows local behaviour of the material around the
crack tip and the assumption that the material is sufficiently homogeneous specimen.
Bearing in mind the structure of the weld, the test procedure appears reliable enough,
because the crack tip in the development of fracture can pass through areas of
different structure and mechanical properties of welded joints. This requires detailed
analysis of welded joints in terms of application of fracture mechanics. Influence of
heterogeneity of structure and mechanical properties of welded joints is reflected
through the location of fatigue cracks and the characteristics of the area through which
210
Stress Intensity Factor And Its Effect Of Structural Integrity
fracture develops. In order to visualize the impact of heterogeneity of structure and

mechanical properties of welded joints in the process of welding, low alloy steel has
been studied to improve .4730. Fracture toughness in plane strain, were performed to
determine the critical stress intensity factor, KIc. To test tubes were used in three-point
bending (SEB), whose geometry is determined by the standard ASTM E399 [2].
Testing itself, was done at room temperature on a dynamometer
electromechanical SCHENCK TREBEL RM 400th Opening of the crack tip (COD) was
registered using a special extension meter "piston-GAGE" DD1 with r 0.001 mm of
measurement accuracy. We made three groups of specimens, depending on the point
of incision. They are as follows :
Group I - specimen notched in the base metal (BM)
Group II - specimens notched in the weld metal (WM). and
Group III - specimen notched in the heat affected zone (HAZ).
According to standard ASTM E399 first, we tried to create a praline fatigue that
is, we tried to determine nominal marginal force FL, i.e. the maximal force of the fatigue
start is defined by equation:
FL
B b 2 RT
2 L
where: B - specimen width, b - the lygament, L-span between the supports, and the RT
- the effective yield stress, which is determined by the equation.
RT
R p 0 , 2 Rm
2
As the demands for the fulfilment of conditions of plane strain condition:
K
B t 2,5 Ic
R
p 0, 2
are not met, instead of applying linear elastic fracture mechanics -defined by ASTM
E399 standard, we tried to use elastic-plastic fracture mechanics defined by ASTM
E813 [3]. The aim of using the elastic-plastic fracture mechanics is that the value of
critical stress intensity factor KIc, is determined indirectly through the critical J integral
JIc in other words we are to monitor crack growth in terms of plasticity.
Test procedure is reflected in obtaining R-curve, or J - 'a curve, which consists
of the value of J integral for the uniform crack growth 'a. The experiment was
conducted using a test specimen successive partial unloading, or the single specimen
compliance method, as defined by ASTM E1152 [4].
The aim of the unloading compliance method is to register the size of the crack,
'a, which occurs during the test. As a rule, the crack growth occurs when the force
exceeds the maximum in the diagram F-G. It should be noted that the fundamental
diagram derived in the coordinate system of the force F - G crack opening, and not in
the system of forces F - shifting points of attack power ' as it is required by standards
ASTM E813 and ASTM E1152. Figure 1 presents examples of diagrams F-G, for
specimens with notches in the base metal (BM), weld (WM) and heat affected zone
(HAZ).
211
b) Diagram F-G
specimen X-MZ
a) Diagram F-G
specimen OM1
c) Dijagram F-G
specimen X-ZUT
Fig. 1 Typical diagrams of the force F-move the crack opening [5]
Obtained from the force diagram F-G crack opening displacement,
measurements and calculations, we have obtained two sets of data:
Consumed work at different stages of the cycle (area under the curve), and
Change in compliance (change in slope of the line of elastic unloading).
Visible relief on the force curve F-G crack opening displacement, are used to
determine the specimen compliance for a three-point bending at the current crack
length a. For compliance, which is presented by G ratio increase and the increase in
the F line of relief, it is possible to determine the crack length over expression:
'ai
b c c
' ai 1i i 1 i 1 i
2 ci
ai-1 - previous crack length.

Ci=tgDi - inclination of the observed lines of relief
Ci-1=tgDi-1 - slope of the previous line of relief
Permissiveness was measured directly using the analogue-digital plotters HP
7090. In addition, the diagrams of the force F - crack opening displacement G directly
recorded during the experiment, and they were the basis for calculating all the
parameters needed for the construction of a J-curve '.
J integral is calculated from the relationship:
where:
Ji
2 Ai Ai 1 ai ai 1
1
J i 1 b
B
bi

where: A - the area under the curve., B - specimen thickness., b - length of the
ligament.
Based on the obtained data, we have constructed J - ' a curve on which is
constructed the regression line according to ASTM E813. Obtained from the regression
line, we get critical J integral, JIc. The typical example of curves J - ' obtained by
testing a specimen notched in: BM, HAZ and MZ is given in Figure 2.
212
Stress Intensity Factor And Its Effect Of Structural Integrity
1000
2
OM-1
JIc = 107,3 kJ/m
700
X1 - M[
1000
JIc = 119,8 kJ/m
JIc = 88,2 kJ/m
500
400
200
600
J - Integral, kJ/m
J - Integral, kJ/m
800
J - Integral, kJ/m
X1-ZUT
600
800
600
400
200
JIc
JIc
0
0
Rast prsline, 'a, mm
10
a) Diagram J - 'a
specimen OM
300
200
100
JIc
0
0
400
0
0
10
b) Diagram J-'a
specimen X-MZ
c) Diagram J-'a
specimen X-ZUT
Fig. 2 Diagrams J - ' a test tube in the BM, WM and HAZ [5]
Knowing the value of the critical JIc integrals, we can calculate from the value
of critical stress intensity factor or plane strain fracture toughness KIc, by using the
dependency:
K Ic
J Ic E
1 Q 2
i.e.
K Ic
V S ac
and by entering the values of conventional yield strength, Rp0,2 = V, assuming that the
shape factor is equal to one, we have calculated approximate values for the critical
crack length ac. Table 4 gives calculated values of plane strain fracture toughness KIc,
as well as the critical crack length.
Table 4 The critical stress intensity factor
Mark
specimen
BM-1
BM-2
X1-WM
X2-WM
X1-HAZ
X2-HAZ
4.
Crack
length, a,
mm
17,03
17,20
15,40
14,70
14,80
17,20
Critical J-integral,
2
JIc, kJ/m
107,3
115,3
119,8
116,1
88,2
89,7
The critical stress

intensity factor KIc,
MPam
148,3
153,7
156,7
154,3
134,5
135,6
The critical crack

length, ac, mm
38,7
41,6
40,4
39,2
29,8
30,2
CONCLUSION
Based on these results we can conclude:
Contribution to the study of fracture mechanics as a modern scientific
discipline is reflected in the fact that it describes global behaviour of
materials in the presence of crack like defects. Unlike conventional
mechanical tests.
Fracture mechanics defines two new characteristics of materials, such as
critical stress intensity factor, KIc, and the crack tip opening displacement
COD. From these data, we can quantitatively determine the critical crack
length, ac, which is very important data of material behaviour in the
presence of crack like defects.
213
In the specific case a decisive influence on the values of KIc, i.e. critical
crack length has the installation point meaning the structures of
heterogeneity. Best tenacity toward the spread of crack and fracture
toughness have the notched specimens in the weld metal what was to be
expected, given the overall best features of the weld metal.
REFERENCES
[1]
[2]
[3]
[4]
[5]
214
Stojan Sedmak, IFMASS 9,

ASTM E399. Standard Test Method for Plane Strain Fracture Toughness of
metallic Materials; 1983.
ASTM E813. Standard Test Method for JIC; 1989.
ASTM E1152. Standard Test Method for Determining J-R Curves; 1987.
Manjgo M., Uticaj koliine uneene toplote na mehanika i eksploatacijska
svojstva zavarenog spoja elika .473, Magistarski rad, Zenica, 2000.
VIBRATORY STRESS RELIEVING OF TANK FLANGS

Vukojevi Nedeljko1, Hadikaduni Fuad2, Pavi Mate3
Summary: The procedure of residual stress relieving on structures can be done in
several ways, the best results can be achieved with heat annealing treatment, but due
to limitations on the dimensions of structures that need to be treated alternative
methods have been developed. One of these alternative procedures is vibrational
relaxation process. The paper presents method and results of performing the
vibrational relaxation of residual stresses. Measurements of residual stress are made
i.e. the definition of "zero state" as well as measurement of changes in residual stress
caused by vibrational relaxation, which was performed in order to eliminate or
decrease the size of residual stresses in the zone of welds flanges and vertical tanks
with floating roof.
Key words: vibratory stress relief, residual stresses, strain gauge measurements,
dynamic analysis, tanks
1. INTRODUCTION
There are two kinds of residual stresses in the welding technology: forced,
resulting from assembling of parts with free ends, where the effect is present only on
the observed parts, and reactive which occurs in parts whose ends are trapped and
they are prevented from forming, and transferred internal stresses in the whole welded
structure. Residual stresses, formed in above mentioned ways, are multiaxial and
combined with other stresses or low temperatures could create conditions for the
occurrence of brittle fracture.
In the past 60 years, VSR has grown from a little known area to respective
primary process, which was established as an alternative to heat treatment of castings,
manufactures, pieces requiring additional machining, and non-metallic materials.
Effects of reduction of residual stresses in various previous studies have a wide
range from 20% to 95%. Effect of vibrational relaxation method depends on many
factors, such as type of construction, structural dimensions, weight and rigidity of the
structure, characteristics of VSR equipment, suspension of construction, etc.
Dr. Sc Nedeljko Vukojevi, Zenica, BiH, Faculty of Mechanical Engineering, (vukojevicn@mf.unze.ba)

Mr. Sc. Fuad Hadikaduni, Zenica, BiH, Faculty of Mechanical Engineering, (hfuad@mf.unze.ba)
3
Grad. Ing. El. Mate Pavi, Slavonski Brod, Croatia, Vibro metal d.o.o., (pavic@vam-ing.hr )
2
215
Vukojevi N., Hadikaduni F., Pavi M.
2. DESIGN OF VSR TECHNOLOGY

The process of performing VSR technology (after welding) at the connection
point of tank hole and tank flange, and performance of monitoring technology took
place at the following stages: preliminary dynamic analysis of tank structural
construction and flange, preparation and installation of measuring gauges, detailed
measurements of residual stresses using the ''hole drilling method, monitoring of
changes in stress during VSR procedure, positioning of force exciter at two orthogonal
locations at flange, monitoring the process of relaxation, postprocessing of datas,
measurement of residual stresses after applied procedure.
Residual stress relaxation, using the VSR is carried out at three flanges on the
tank, which have the same geometrical properties, Figure 1. Flanges are attached to
the sheet metal by welding which is performed by specially projected technology.
Flange is a light opening 750 mm and with a thickness of 23 mm, and a tank
thickness of 31 mm in the zone of the flange.
Fig. 1 Flange on the tank of 85.000 m3 storage capacity

Materials steel quality of container and tank sheets are S3555J2G3 according
to EN10025 whose mechanical properties are shown in Table 1.
Table 1 Mechanical properties of S355J2G3 material (EN10025)
Material
S355J2G3
Standard val.
216
Yield strength
Tensile
strength
min.ReH[N/mm2]
Rm[N/mm2]
315
490-630
Elongation
min. [%]
A5
Axial direction
20
Impact toughness
work
KV, -20,min [J],
Axial direction
27
Vibratory Stress Relieving of Tank Flangs
DYNAMIC ANALYSIS OF CONSTRUCTION
In order to analyze the possibilities for the application of VSR methodology for
reduction and redistribution of residual stresses, as a result of connecting technology
for flange inlet hole to sheet metal, a dynamic analysis of structures are made.
Numerical model of complete structure was performed with 13 517 points, or 13 100 of
plate finite elements. The dimensions of the entire structure, thickness of individual
segments of shheets and flanges are taken from the submitted design documentation.
Based on dynamic analysis of the tank construction and flanges, an analysis of
structure movements in the zone of force action is made. According to the specification
of available equipment range and impulse force, the bending analysis of the structure
and information on the movement of 0,041 [cm] in the vertical plane and the movement
of the maximum of 0,023 [cm] in the horizontal plane is made. Based on analysis of
their natural forms of oscillations in the zone of flanges and stiffeners values of 43.69
[Hz], 82.46 [Hz] and 120.6 [Hz] are determined. The structural response of the vertical
and horizontal planes at the places of force inducer are presented in Figure 3.
Fig. 2 Numerical model of the tank and a detail of the flang
a) responses in vertical plane
b) responses in horizontal plane
Fig. 3 Frequency response analysis of structure
217
DESCRIPTION OF VSR TREATMENT OF THE TANK
On the basis of specific scientific parameters, the geometrical characteristics of

flanges with sheets, and the geometry and the quantity of welds, it can be considered
that places of weld joints of the flange and tank sheets are with greatest concentrations
of stresses. VSR equipment of Meta-Lax type, with working frequency range 0-100 Hz
and 20-100% eccentricity is used. The equipment consists of the exciter force
transducer and acquisition devices. With the force exciter a ''scanning'' of the
construction of the first flange is conducted, which showed ''sensitivity'' of structure in
the range of 45-80 Hz. Therefore, the applied frequencies were in that range, lasting
30-40 minutes. Based on these indicators, parametric procedures are shown in Table 2.
Table 2 VSR tank treatment parameters
Flang
P1
P2
P3
Vibrator position
PV1
PV2
f1 (Hz)
f2 (Hz)
65
80
65
88
62
84
Vibrator
eccentricity
(%)
100
100
100
Strain gauges are installed on the following flange measuring places:

Z1 - place to measure the residual stress before VSR treatment,
Z2 - place to measure the residual stresses after VSR treatment,
MM1 and MM2 - places to measure the change of stresses during the
vibrational relaxation for the two orthogonal directions on the flange no. P1,
vibrational relaxation for the two orthogonal directions on the flange no. P2,,
vibrational relaxation for the two orthogonal directions on the flange no. P3.
5
MEASUREMENT OF STRESSES DURING VSR TREATMENT
All three flanges are treated in the same conditions, with the same position of
the vibrator, where the rate varied from ~ 60-90 Hz, Table 2. The following diagram
shows the change in stress at the measuring point M1 occurred as a result of vibration.
Vertical tanks with floating roof for liquid fuel storaging are welded steel
structures of large dimensions and as such are usually subjected to the treatment of
residual stresses relaxatiion. In Table 3 the comparative values of the measured
changes in residual stresses during vibrational relaxation are given. After the results of
measurements analysis, given in Table 3, can be concluded that the effect of applied
VSR in the change of residual stresses, at measurement locations on MM1 and MM6,
is fairly uniform and is in the range 23-33 [MPa].
Measured strains are negative, indicating that the initial stress was tensile in
character which is common in welded joints.
218
Vibratory Stress Relieving of Tank Flangs
Stress values at flange No. 1 during VSR treatment

Measuring point M1
30,0
20,0
Stress, MPa
10,0
0,0
-10,0
-20,0
-30,0
-40,0
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Time, sec
s1-max
s2-min
sef
Fig. 4 Stress values, flange No.1 Measuring point M1

Table 3 The results of stress measurement at the tank
Measuring point
Max. main stress
V1max, [MPa]
Min. main stress
V2min, [MPa]
Equivalent stress
Vekvi, [MPa]
Main stress direction
M, []
P1
P2
P3
MM1
MM2
MM3
MM4
MM5
MM6
-10
-20,7
-8
-7
5,5
-7,3
-28
-38,4
-30
-25,4
-35
-21
24
33
27
23
32
25
180
90
90
180
6. STRAIN GAUGE MEASUREMENT OF TANK RESIDUAL STRESSES

Measurements of residual stresses are derived in order to determine the
parameters at the beginning and end of the process. Measurements of residual stress
were conducted using the system for automatic determination of residual stresses
using strain gauges "MTS-3000" where used hole drilling method defined by ASTM
837-01.
Data acquisition was performed using a measurement system "Spider 8-30,
triaxial measuring gauges, and software "Catman 5.0 Professional. Measurement
points for determination of residual stresses are presented in Figure 4. Table 4 shows
that the highest value of principal stress before treatment vibrational relaxation is in the
weld zone. All the measured residual stresses are of tensile character.
Equivalent stress according to the hypothesis Von Misses.
219
Fig 5. Residual stress measuring points of flange P1

Table 4. Comparative results of measurement of residual stresses
7.
Z1
V1 max
[MPa]
132
V2 min
[MPa]
109
Vekvk
[MPa]
122
Vekvh
[MPa]
132
Z2
104
84
95
104
Sample
Measuring point
Before VSR
After VSR
CONCLUSIONS
Applying the above described methodology of vibrational relaxation of residual

stresses on weld flange and the tank, the following can be concluded: a) using hole
drilling measurement method before vibrational relaxation, the presence of residual
tensile stress intensity of 122 [MPa] in the zone of welded joint is indicated, b) after
vibrational relaxation the presence of residual tensile stress intensity of 95 [MPa] in the
zone of welded joint is indicated, c) measurement in changes of residual stresses
during the vibrational relaxation showed a value of 23-33 [MPa]. The origin of
measured residual stresses are not only because of chosen welding technology, but
may be a result of incarceration, grinding, etc. The comparison of measurement results
of residual stress relaxation before and after the relaxation ilustrates that the reduction
of residual stress, using the vibrational relaxation of the tank is in the range of 18 -28%.
REFERENCES
[1] Vukojevi, D. (1998.). Teorija elastinosti sa eksperimentalnim metodama,
Mainski fakultet u Zenici.
[2] Hoffmann, K. (1987.). Eine Einfrung in die Technik des Messens mit
DehnungsmeEstreifen, Hottinger Baldwin Messtechnik GmbH, Darmstadt.
[3] A. G. Youtsos. (2006). Residual Stress and Its Effects on Fatigue and Fracture.
Proceedings of a Special Symposium held within the 16th European Conference of
Fracture ECF16, Aleandroupolis, Greece.
k
h
Vekv - equivalent stress calculated according to the hypothesis of Von Misses.

Vekv - equivalent stress calculated according to the hypothesis of Tresca.
220
VIBRODIAGNOSTICS OF ROTATION MACHINES

Ranko Antunovic1
Summary: With the advent of increasingly complex and demanding technical systems
is becoming crucial need for the development and application of appropriate diagnostic
system to provide high availability and reliability of these systems.
Direct supervision and diagnostic analysis of the situation, with the specified diagnostic
methods, we get good insight on the current state of technical systems and the
maintenance actions carried out when needed.
Diagnose the operating mode represents a multiple measurement and analysis of the
response to output signals from which to infer the character of the dynamic behavior of
the system and determine the cause of malfunction. I have a diagnostic approach
based on the fact that any dynamic disorder caused by the movement or vibration is
exactly determined character.
1
INTRODUCTION
Occurrence of certain faults machine produces a fixed excitation power that

produces a specific oscillatory movement. The analysis of vibration response with
sensors system can determine the character of excitation force and determine the
cause of machine malfunction.
Ph.D. Ranko Antunovic, Assoc. prof. University of East Sarajevo, Faculty of Mechanical Engineering, Vuka
Karadzica 30, Lukavica, East Sarajevo, B i H
221
Ranko Antunovic
Fig. 1 Dependence of vibration excitation force

2.
MONITORING SYSTEMS
In order to have available the necessary information for diagnostic assessment and
analysis of technical systems there must be systems for the collection of all relevant
data. With the existence of a system for collecting and analyzing relevant data, there
must be an organizational management system machines that clearly defines who gets
to see what information and who comes to the appropriate decision.
Fig. 2 Eddy Current Proximity Probes and Choosing a Mounting Position

Supervisory instrumentation is likely to signal analysis and conditioning, and
communication with the computer and expert systems, other than the protection of
performing diagnostic tests which reveal the cause of the defect and provide measures
to eliminate it.
222
Vibrodiagnostics of Rotation Machines
Fig. 2 Ways of monitoring the implementation of diagnostic systems

3
ADVANCED METHODS VIBRODIAGNOSTICS
Permanent monitoring of vibration can be easily designed if the monitored level

of vibration (usually the level of vibration velocity) with an additional built-in relays that
can signal exceeding a preset level of vibration. More complex than monitoring
vibration monitoring vibration on the bearings and shaft, followed by phase angle of
vibration with respect to the shaft and has additional capabilities of monitoring process
parameters of the machine. However it is possible to extract methods vibrodiagnostic
indicating the early onset of damage to the technical system. This will be shown some
of vibrodiagnostics methods commonly used for early detection of defects and
monitoring a particular problem in the work machine.
3.1 Orbital (dual) analysis
The orbit is mainly used for monitoring and analysis of rotor vibration in
machines with plain bearings and it is a dynamic path of movement sleeve [1].
223
Ranko Antunovic
Fig. 3 Preview formation orbit

3.2 SED analysis
Selected envelope detection (SED) is a powerful analytical method that extracts
and analyzes viskofrekventne modulated signals. It is used for damage detection of
rolling bearings, gears, for monitoring cavitation, inductance motors, etc. [3].
The following figure shows the spectral envelop the view that in a simple way
nedvosnislen indicates damage of rolling bearings [2].
Fig. 4 Envelope Spectrum
224
3.3 Cepstral analysis

By rotary machines are frequently present, the basic dynamic modulation
phenomenon lead to the creation of harmonic frequency components around the main
phenomena. This analysis is used to detect damage gears, beds, etc. [2].
Fig. 5 Way of obtaining kepstra

3.4 SEE Analysis
SEE SKF's method for detection of bearing failures and gears in the early stages of
failure, as well as to detect cavitation, friction, cracks, electrical problems etc. This
method focuses on calculating the total energy spectrum for the high frequency range,
typically frequencies between 150-500 kHz. It actually gives enveloping acoustic
emission spectrum [4].
Fig. 6 Showing envelope acceleration and SEE spectrum for damage to the inner track
rolling of bearing
3.5. Modal analysis
In principle, modal analysis can be operationalized and classical modal
analysis. Operational modal analysis is carried out in the paper machine. By carrying
out this analysis we can determine the complex transfer function of the system caused
by the forces in actual service [7].
225
Ranko Antunovic
Fig. 7 Virtual machine shifts obtained using operational modal analysis

4. ANALYSIS OF RESULTS OF STUDY
This will be displayed diagnostic assessment turbogenerators TPP, using modern
systems of technical diagnostics.
1. Installed power unit: 300 MW
2. Plant / machine: turbine
Fig. 8 Dispositional scheme turbogenerators in TPP

3. Identified problems/shortcomings: Exploitation staff noticed an increased
overall level of vibration on the generator bearings. On the bed No. 7 overall
level of vibration in the vertical direction was 6.2 mm/s
4. Results
The measured overall vibration levels in bearings turbogenerators in all three
measurement directions are shown in the table below.
Table 1 Overall of vibrations in bearings
H
Bearings
1
2
3
4
5
6
7
8
226
RMS
[mm/s]
3,4
2,1
1,4
2,0
3,2
3,3
3,8
6,4
RMS
[mm/s]
1,8
1,8
3,6
1,9
2,7
4,6
5,8
5,2
RMS
[mm/s]
1,8
2,1
2,2
2,3
2,1
6,5
9,3
4,1
Note: Evaluation of the state machine according to the standard ISO 10816 are
declared as: are allowed.
-Spectral analysis
Spectral analysis was performed on the bearings 6.7 and 8 at the station where the
measurement of the dominant overall vibration levels.
Fig. 9 Spectral plot vertical vibrations in bearings

From the spectral plot dynamic damages can be identified: unbalance and
misalignment.
Lets show Polar plot 1X harmonic vibration in 6-th and 7-th of bearings,
various in parameters process.
Fig. 10 Polar trend plots of 6 and 7 bearings

Can be seen that B1-vector of thermal imbalance, has little bearing on the
dynamic balance (the overall level of vibration to slight changes) and C1-vector of
unbalanced magnetic field, which significantly affects the dynamic state machine,
therefore it is in phase with the residual mass asymmetry of the rotor in response to the
beds 7th.
227
Ranko Antunovic
CONCLUSION
The progress of computer and measuring techniques, and continuous price

reduction of development of computer hardware, allows today a significant step
forward in monitoring and diagnosis of technical systems with financial investment is
very acceptable. The application of modern systems of technical diagnostics, can be
coupled with analytical methods, much better to explore the dynamic behavior of
machines. Thus, one can timely detect errors in operation, following the change of
state machines, then set depending on the operating modes, failure to diagnose the
causes of the events and store all relevant information about the state of the system.
LITERATURE
[1]
Bently Nevada, Predictive maintenance: How the pieces fit together, Orbit, Vol.7,
No.2, June 1986
[2] Multi-channel & multi-task Analyzer-PULSE, Bruel & Kjaer advanced training
course, held in Budapest, Hungary, on 14-16 June 2005
[3] Vibracion Diagnostic Guide, SKF Reliability System, Application Note CM5003
[4] REVOLUTIONS, A Publication of SKF Condition Monitoring, Vol.7, Number 3
[5] Enveloping Technology For Fast And Accurate Bearing Analzsis,
REVOLUTIONS, A Publication of SKF Condition Monitoring, Vol.3, Number 2
[6] Cepstrum Analysis, Bruel & Kjaer, Licture Note BA 7076
[7] Ward Heylen, Stefan Lammens, Paul Sas, Modal Analysis Theory and Testing,
Kathoolieke Universiteit Leuven, 1997
[8] Agnes Muszynska, Ph.D., VIBRACIONAL DIAGNOSTICS OF ROTATING
MACHINERY MALFUNCTIONS, The Course in Rotor Dynamics and Vibration in
Turbomachinery, 21-25 Septembar 1992
[9] R.Antunovi, M.Skoko, OPTIMIZACIJA METODA TEHNIKE DIJAGNOSTIKE,
YUKOCigre 2007. god.
[10] mr
Ranko
Antunovi,
IMPLEMENTACIJA
CDS
(KOMPJUTERSKO
DIJAGNOSTIKOG SISTEMA) U TE GACKO, 8 Kongres JISA, Herceg Novi,
jun 2003
228
B. PRODUCTION TECHNOLOGIES
AND ENGINEERING

ADVANCED CUTTING TECHNOLOGY HIGH-PRESSURE JET

ASSISTED MACHINING
Kramar D.1, Sokovi M.2, Kopa J.3
Summary: High-pressure jet assisted machining (HPJAM) is starting to be established
as a method for substantial increase of removal rate and productivity in the metal
cutting industry. Cooling lubrication with high pressures in turning operations is an
effective method for providing higher productivity, reducing temperature in the cutting
zone and improving chip control depending on the pressure and flow rate of the fluid
jet. The aim of this paper is to compare the capabilities of dry, conventional flooding
and HPJAM of hard-to-machine materials. The performances of different cutting
conditions are compared on the basis of chip breakability, technological windows which
yield particular operational ranges, cooling lubrication efficiency by measuring local
temperature, tool wear, and cutting forces.
Key words: machinability, technological windows, tool wear, chip formation
1. INTRODUCTION
In order to reduce the consumption of cutting fluids the trend in machining is
moving towards dry or near dry cutting. But even with the most advanced tool materials
and coatings, it seems that lubricants in the case of hard-to-cut materials such as
hardened steels used for moulds, Cr-Co alloys used for prosthesis, Ti-based, and Nibased alloys used in gas turbines and in the aerospace industry cannot be completely
avoided. In order to keep increasing the machining performance, different assistance
methods have been developed to replace the conventional process. One of them is
the high pressure jet assisted machining (HPJAM), which aims at upgrading
conventional machining by using the thermal and mechanical properties of a high
pressure jet of water or emulsion directed into the cutting zone.
Pigott [1] was the first author to discuss the use of HPJAM in steel turning with
high speed steel tools. He injected the coolant at a pressure of 2.76 MPa directly at the
clearance of the tool. He found the temperature drop and that tool life increased by
eight times. In addition, it was found that the use of a HPJAM led to an improved
surface roughness compared to the conventional cooling. Moreover, the low speed of
the coolant does not allow the lubricant to reach the cutting edge; a situation that
1
Assist. Prof. Dr. Kramar D., Lj., Slovenia, Fac. of Mech. Eng., University of Lj., (davorin.kramar@fs.uni-lj.si)
Prof. dr. Sokovi M., Ljubljana, Slovenia, Fac. of Mech. Eng., University of Lj., (mirko.sokovic@fs.uni-lj.si)
3
Prof. dr. Kopa J., Ljubljana, Slovenia, Fac. of Mech. Eng., University of Lj., (janez.kopac@fs.uni-lj.si)
2
231
Kramar D., Sokovi M., Kopac J.
favours the formation of built up edge and unfavourable heat dissipation due to which
the chips are cooled to a higher extent than the tool and the workpiece. In Yankoff [2]
solution, the orifice is placed atop the exposed surface of the insert for ejecting high
speed coolant at 20 MPa across the rake face and beneath the chips. Experiments
reported in [1, 2] were conducted within a limited pressure range. Since that time
numerous researches have been performed on the field of HPJAM of different
advanced materials [3-5], using higher fluid pressures and flow rates.
The machining of hard-to-machine materials with coated carbide tools,
conventional turning parameters and conventional cooling, usually results in significant
problems concerning extremely long chips and severe adhesion wear mechanisms. By
applying HPJAM at reduced flow rates, the friction and the heat induced in tool-chip
interface can be reduced. Based on this technology turning of hard-to-machine
materials with conventional cutting speeds and low cost coated carbide tools can be
performed. The aim of this investigation is to compare the capabilities of dry,
conventional and HPJAM turning of hard-to-machine materials. The performances of
different cutting conditions are compared on the basis of chip breakability and cutting
forces, technological windows which yield particular operational ranges, cooling
efficiency, and tool wear.
2. HPJAM WORKING PRINCIPLE
To summarize statements indicated in introduction, it can be seen that HPJAM
is applied differently depending on the type of equipment available: either employing
pressures higher than 150 MPa and flow rates lower than 6 l/min while using small
nozzle or lower pressures up to 30 MPa and high flow rates which can reach 50 l/min
with larger nozzles. Due to their higher flexibility and non additional operations,
external nozzles are widely utilized compared to internal channels which can also be
found in the literature [3]. The simplest but efficient way to inject the jet into the cutting
zone is at the tool-chip interface. Compared to the conventional cooling, the idea of
HPJAM is to inject a high pressure jet of coolant in the cutting zone. The lathe should
be fitted with high pressure equipment. This involves high pressure pump, high
pressure tubing, and outlet nozzle fixed beside tool holder. A pump is supplied with
filtered water or emulsion. A complete machine tool set is presented on Fig. 1.
Fig. 1 HPJAM system components
232
Advanced cutting technology High-Pressure Jet Assisted Machining
The high-pressure jet of coolant can be applied in two ways:

With an external nozzle: the jet is injected directly in between the rake face and the
chip or can be directed to the gap between the flank face and the workpiece.
Through internal channels: the fluid is injected through the tool using small holes in
the insert.
The supply of high-pressure jet
between the rake face and the chip
decreases the length of their contact.
On the other hand, the cutting zone can
be reached by injecting the coolant
below the flank face of the tool. The
following method is the most efficient to
reduce the temperature in the cutting
zone, but has no influence on chip
breakability. In this investigation jet is
Fig. 2 Delivery of HP jet through an
applied on the rake face through an
external nozzle
external nozzle that can provide higher
machining performances and is easier
to set on the conventional lathe (Fig. 2).
3. EXPERIMENTAL WORK
All the experiments were conducted in longitudinal turning process on a
conventional lathe, fitted with a Hammelmann high pressure plunger pump of 150 MPa
pressure and 8 l/min flow capacity. The fluid utilized was the Vasco 5000 cooling
lubricant from Blaser, a 5.5% emulsion without chlorine on the basis of vegetable oil
mixed with water (pH 8.5-9.2). Standard sapphire orifices, commonly used in water jet
cutting applications, were set in a custom made clamping device that enabled an
accurate jet adjustments. The jet was directed normal to the cutting edge at a low
angle (about 5-6) with the tool rake face (Fig. 1 and 2). The nozzle was located 22
mm far from the tool tip in order to assure its use in the core zone of the jet and avoid
variations in the diameter of the jet and radial distribution of the pressure.
In the experiments hard chromium platted and surface induction hardened
steel C45E (AISI 1045) and Ni alloy - Inconel 718 were tested. Both materials are
known to induce chip control problems. That makes them suitable for HPJAM
machining. The depth of the hardened surface layer for C45E was between 1.5 and 1.8
mm with a hardness of 58 HRc. The cutting tool inserts used in the experiments with
C45E were coated carbide cutting tools SANDVIK SNMA 120408 with Al2O3
coating, while Inconel with hardness 36 38 HRc was machined with coated carbide
cutting tools SANDVIK SNMG 12 04 08-23 with TiAlN coating. The inserts for C45E
machining were flat-faced, while inserts for Inconel had positive geometry with rake
angle of 13.
3.1 Experimental methodology
Considering the studies described in Introduction, it is clear that injecting a
pressurized fluid in the cutting zone can drastically enhance the cutting process
performance. Therefore in the first step, screening experiments were conducted in
233
order to determine coolant pressures that yield adequate chip breakability and cooling
capability. Within this experimental step the influence of coolant pressure on the chip
formation and cutting forces was analysed. The next step was to determine the region
of operability for selected tool-material pair (TMP), which sets the boundaries of the
process cutting speed and feed rate in conventional flooding and HPJAM. The selected
TMP methodology is based on the monitoring and analysis of the specific cutting
pressure, chipbreaking mechanism and surface roughness while varying the cutting
speed first and then the feed rate. The latter has been applied to HPJAM as well as
flooding conditions. In the third step, experiments were performed with the cutting
speed and the feed rate belonging to the cross-section of overlapped technological
windows for particular cooling condition determined in the previous step. By measuring
tool wear and local cutting temperature the assessment of cooling capability was
conducted. Within this experimental step the depth of cut was also kept constant.
4. RESULTS AND DISCUSSION
In order to compare HPJAM to conventional flooding results are presented in
manner of benefits and drawbacks. Some observations and results are gained on both
materials, while some of them are only obtained on selected one. The following
sections describe main machinability criterions analysed in our researches.
4.1 Cutting forces
In the initial experiments the influence of the cooling conditions and coolant
pressure levels on the cutting forces was analysed. The feed and radial force
decreased as soon as the HPJAM was applied but no real trend could be noticed with
the increase of the pressure (Fig. 3). In the case of the main cutting force it was more
difficult to verify a significant trend, whereas the small variations observed can be
considered to be within the margin of measurement error.
Fig. 3 The influence of the cooling conditions and coolant pressure on the feed and
radial force in hardened C45E turning
4.2 Region of operability
The procedure for regions of operability technological windows definition for
both materials and selected tools is described in detail in [6]. In this paper only results
are presented.
234
Fig. 4 Technological window for TMP in HPJAM and conventional cooling of hardened
steel C45E
Firstly the TMP methodology was performed on hard chromium platted and
surface induction hardened steel C45E. According to the maximum rotation speed
limitation of the spindle, cutting speeds higher than 200 m/min were not tested and the
upper limit was fixed to 158 m/min for safety reasons. Fig. 4 shows operational areas
for TMP for the case of conventional cooling and HPJAM conditions. An extension of
the operational area for both cutting parameters is achieved with HPJAM. More
specifically, a 45% increase in the maximum achievable cutting speed and
approximately 25% increase in the maximum achievable feed rate are shown.
In HPJAM of Inconel 718 two different orifice diameters were tested, namely
0.25 mm and 0.4 mm. The pressures were also set on two levels, 50 MPa and 130
MPa. These settings gave 4 different flows of coolant. TMP methodology was
performed for all 4 conditions and the influences of cooling on the cutting capability
were analysed.
Fig. 5 Region of operability for four HPJAM conditions and for conventional cooling
showing also a consumption of coolant, and jet momentum (Inconel 718)
235
At pressure 50 MPa no evident enlargement of the technological window in

comparison to conventional flooding was achieved when smaller nozzle was applied
(Fig. 5). This is probably because of a low contribution of HPJAM in chips breakability
and insufficient cooling with such low jet momentum. The flow rate of coolant in this
pressure/nozzle (50 MPa/0.25 mm) combination was only 0.6 l/min. At pressure of 130
MPa with the same nozzle, better results were achieved. The technological window
was enlarged on the side of lower cutting speeds and lower feed rates. The results with
a considerably better productivity and flexibility were achieved with a larger nozzle (0.4
mm). Even with lower pressure level the cutting performance was increased, which is
shown with the technological window in Fig. 5 (red field). Both, higher and lower cutting
speeds and feed rates compared to conventional cooling can be used with such a
nozzle for both of the applied high pressures. Fig. 5 also shows the consumption of
coolant at different cooling conditions. The highest improvement in the cutting
performance in the connection with the lowest coolant consumption is achieved with a
larger nozzle and lower pressure. In this case almost a quarter of cutting fluid is used
compared to conventional cooling.
4.3 Tool wear and tool life
The machinability can be characterised by tool wear and tool life. In order to assess
and compare the capability of cooling conditions, experiments were carried out within
the common cross-section of overlapped technological windows for a particular TMP
and cooling condition (Fig. 4, 5). In this step of investigation the following cutting
conditions were employed for machining of hardened steel C45E:
Depth of cut ap = 2 mm
Feed rate f = 0,25 mm/rev
Cutting speed vc = 98.5 m/min
Workpiece diameter d = 30 mm
Cutting fluid concentration c = 5.5 %
Crater wear changes the effective rake face angle and leads to a changing
cutting behaviour. Excessive crater wear weakens the tool just behind the cutting edge.
As a result, a sudden break down of the cutting edge can occur. The flank wear occurs
as a flattened area on the cutting tool flank face. The width of flank wear, referred to as
the flank wear land width VB, is traditionally considered a good indicator of the state of
the cutting tool. An increase in VB leads to increased friction between the cutting tool
and the workpiece. Consequently, a higher specific cutting force and more heat are
generated. It was discovered that a flank wear land width of VB = 0.1 mm does not
lead to workpiece damage. In HPJAM condition the distribution of the wear along the
flank face was also quite uniform, however some marks of notch wear at the depth of
the cut line can also be noticed. Crater wear was also present on the rake face. The
changes in the specific cutting force and roughness were noticed as a consequence of
the increased tool wear.
Fig. 6 shows the tool flank wear trend in conventional cooling and HPJAM. For
the selected criteria VB = 0.1 mm, tool life in the case of HPJAM of hardened steel
C45E was about 10 minutes, which is approximately 5 times longer than in the case of
conventional flooding. It should be pointed out that the consumption of cutting fluid in
that case was more than 4 times lower as in the case of conventional cooling.
236
Fig. 6 Tool wear development in conventional and HPJAM of hardened C45E

Similar tests as in case of steel C45E were performed also on Inconel 718. In
order to assess and compare the capability of cooling conditions, experiments were
carried out with HPJAM pressure set on 50 MPa, and nozzle of 0.3 mm diameter was
used. The following cutting conditions were employed in this step of investigation:
Depth of cut ap = 2 mm
Feed rate f = 0,25 mm/rev
Cutting speed vc = 55 m/min
Workpiece diameter d = 140 mm
Cutting fluid concentration c = 5.5 %
Tool flank wear development for both cooling conditions is shown in Fig. 7. In
conventional cooling, for the chosen tool wear criteria of VB = 0.3 mm, the tool is
rejected within 4.5 min. In HPJAM tool life was more than 8 min. A sudden increase of
tool wear is typical for flooding condition, while in HPJAM a more uniform increase was
noticed. The changes in the specific cutting force and roughness were noticed as a
consequence of the increasing tool wear again.
Fig. 7 Tool wear development in conventional and HPJAM of Inconel 718
237
4.4 Cooling capability

In case of Inconel 718 cutting also temperature below the cutting zone was
measured. A thermocouple was directly embedded in the insert to reach this aim. The
main drawback of this measurement technique is that it does not reflect the maximum
temperature and its exact location and hence cannot be used to measure the cutting
temperature as the average overall temperature. The conducted measurement is thus
qualitative, employed to assess the heat transmitted into the tool through a fixed
reference point.
500
HP jet
Flood
HPJAM
dry
temperature [C]
400
> 400 C
300
~ 200 C
200
~ 130 C
100
~ 60 C
0
0
10
20
30
40
50
60
70
time [s]
Fig. 8 Comparison in temperature measurements between three cooling conditions

The efficiency of HPJAM in limiting the heat transmitted to the tool can be illustrated by
the online measurements shown in Fig. 8. It can be stated that a 30% decrease in tool
temperature compared to conventional cooling is achievable with a 450% reduction in
coolant flow. The thermal advance is substantial compared to dry cutting with an
approximate decrease of tool temperature by 70%. In consideration of the pressures
used, the high speeds of the jet are governing a high convective heat transfer through
the turbulent behaviour of the fluid. This attributes to the fluid increased heat
dissipation capabilities, which prevents the cutting tool from excessive heating.
5. CONCLUSIONS
The presented research is based on experimental observation of two hard-tomachine materials, in turning processes, where the main difference is in the utilized
cooling. The experimental work has proved that the turning of both materials,
chromium platted and surface induction hardened steel C45E and Inconel 718 with
coated carbide tools at conventional cutting speeds is not possible in dry cutting
conditions. The process capabilities of conventional and HPJAM methods are
compared with respect to chip breakability, cutting forces, and tool life. Both processes
are characterized with technological windows that yield the operational area according
to the employed tool-material pair. The major concluding remarks related to the
HPJAM precedence over conventional cooling in the turning of both hard-to-machine
materials with coated carbide tools are:
An extension of the region of operability for a given tool-material pair. More
238
specific, approximately 35% increase in both the maximum achievable

cutting speed and the maximum achievable feed rate were obtained for
hardened steel C45E. In the case of Inconel 718 more than 20% increase in
the maximum achievable cutting speed and almost 30% increase in the
maximum achievable feed rate at both pressures were shown.
A significant increase in chip breakability.
A five times increase in tool life (VB=0.1mm), from 2 minutes to 10 minutes
was achieved for C45E and almost double tool life (VB=0.3mm), from 4.5
minutes to more than 8 minutes was achieved in case of Inconel 718
cutting.
Temperature measurements exposed a major improvement compared to
dry cutting and conventional cooling.
All machining advantages mentioned above were achieved with a reduction
of coolant consumption by several times.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
Pigott, R.J. S. (1953). Method of applying cutting liquids, US Patent 2,653.

Yankoff, G.K. (1986). Method and apparatus for machining, US Patent 4.621.547.
Schoenig, F. C., Khan, K.A., Atherton, A., R. Lindeke. (1993). Machining of
Titanium using water jet assistance through the insert, 7th American Water-Jet
Conference, August 28-31, Seattle, Washington, USA, p. 801-813.
Kaminski, J., Alvelid, B. (2000). Temperature reduction in the cutting zone in
water-jet assisted turning, Journal of Materials Processing Technology, vol. 106,
1-3, p. 68-73.
Ezugwu, E.O., Bonney, J. (2004). Effect of high-pressure coolant supply when
machining nickel-base, Inconel 718, alloy with coated carbide tools, Journal of
Materials Processing Technology, vol. 153-154, (10), p. 1045-1050.
Kramar, D., Kopa, J. (2009). High performance manufacturing aspects of hardto-machine materials, Advances in Production Engineering & Management Vol. 4,
1-2, p. 3-14.
239
ANALYSIS OF THE EFFECTS OF APPLYING NEW MATERIALS

Milan Milovanovi1, Milentije Stefanovi2
Summary: The paper shows the results of researches on introduction of new materials
(new generation plastic masses, high strength steel sheet metal) with the intention to
satisfy the regulations, improve the construction or monitor the worldwide trends in car
industry. The application of new materials can influence the existing technologies;
therefore, when introducing the new materials it is necessary to determine the level of
technologies modification, i.e. the need for introduction of new technologies. The paper
presents some of considered influential parameters. In order to determine the effects of
new materials application, it was necessary to develop the investigation methods,
which will be shown in the paper.
Key words: car body, new materials.
1. INTRODUCTION
One of the reasons for introducing new materials, starting with product
development process, is the reduction of weight of each drive unit and vehicle as a
whole while satisfying ever stricter demands. The introduction of new materials
inevitably leads to rapid changes of applied technologies, both in the designing
process and in the processes of manufacture, finalization and, afterwards, exploitation
and recycling processes. Regarding passenger car bodies, there are two dominant
influences from the regulations aspect: safety and noise. Passenger car body, as a
carrying construction, is one of the major car elements. On modern passenger
vehicles, self-supporting car body constructions are most frequently applied. Car
bodies are usually made of thin-wall sheet metals, which are attached by appropriate
procedures (welding, soldering, gluing etc). The carrying construction is influenced by
many factors: adopted vehicle style, vehicle concept, vehicle size, vehicle version,
application of modern materials, safety regulations influence etc. This paper will
analyze only the influence of new materials application on the carrying construction,
need for introduction of new materials and influence on the existing technologies.
2. DEVELOPMENT AND APPLICATION OF NEW MATERIALS
Development and application of new materials for car bodies has been
1
2
dr Milan Milovanovi, ZASTAVA VEHICLES GROUP, Kragujevac, SERBIA, siljakg@sbb.rs

prof. dr Milentije Stefanovi, Faculty of Mechanical Engineering Kragujevac, SERBIA, stefan@kg.ac.rs
241
Milan Milovanovi, Milentije Stefanovi
adjusted to general society demands, economic resources saving, energy sources and
ecology preservation, improvement of passengers safety etc. For that purpose, new
materials are applied with the aim of reducing weight: high strength sheet metals,
aluminium sheet metals, titan and its alloys, sandwich materials etc. The reduction of
weight along with fulfilment of demands regarding car body stiffness decreases the fuel
consumption and makes space for additional assembling of elements for improvement
of active and passive safety. However, simple replacement and introduction of new
materials demand a series of technological adjustments in the existing production
process, i.e. formation of completely new forming systems. The main concept of
standard selection of materials for car body parts implies the fulfilment of the following
demands:
Low costs (materials and production costs)
Quality and quality reliability
Supply reliability
Functional properties (weight reduction effect, strength, resistance to
corrosion)
Productivity in mass production (formability, weldability, ability to be
attached and painted, change of conventional production process
Utilization of waste and recycling.
2.1 High strength sheet metals (HSS)
They make possible the manufacture of car body of increased stiffness and
reduced weight, with satisfactory performances regarding standard durability tests
(energy accumulation at impact, oscillatory comfort etc.). They were first applied in the
70-ties of the last century, as a response to energy crisis and have been increasingly
applied ever since, especially in countries with developed industry and for very
complex products, such as vehicles. It has been estimated that the application of high
strength sheet metals will rise to 43%. Thereat, car body weight is reduced for 10-20%.
In comparison with classic low-carbon sheet metals for deep drawing, these sheet
metals have unfavourable formability properties and higher cost for the same weight.
They are used for obtaining parts which do not satisfy high deformation degree. When
considering the application of HSS, technically speaking, there are two cases:
Possibility for application of HSS within the existing technologies and
Designing of new technology which will include the application of HSS for
manufacture of new model car body parts.
Generally limiting factors in application of HSS can be systematically shown as
a set of following properties: resistance to fracture, ability to maintain form, adaptability
to tool shape. Researches aimed at determining the need for developing locally
produced HSS, named PHZ 260, which would be used for car body parts production,
were realized in Zastava. With the purpose of determining the effects of HSS
application, three solution variants for side door strengthening were made, which are:
Variant 1 (strengthening in the door zone made of low-carbon steel
0146b, support thickness 1.6 mm)
Variant 2 (same material class, support thickness 2.0 mm)
Variant 3 (newly developed HSS)
Version 3 gave the most favourable results. HSS could also be applied for
supports of less complex form. However, based on obtained results it can be observed
242
Analysis of the Effects of Applying New Materials
that the replacement of materials is not sufficient itself to solve the numerous problems
in car body designing. New solutions involving introduction of new materials need to be
combined with improved constructive solutions. In the actual case - strengthening of
the side door zone it is necessary to find the solution which will include forming of
closed support section with introduction of HSS as a material for closed support
elements.
Fig. 1 Analysis of effects of HSS application

2.2 Application of Al-alloy sheet metals
Application of aluminium for manufacture of some passenger car body parts
(hood, doors, outer coatings etc.) started by applying alloy Al-Mg-Zn in the 80-ties.
According to the researches of the Japanese Association for Light Materials from 1990,
the share of aluminium alloys in the total cars mass was 8.3%. On the world market,
there are already cars whose entire car body is made of these alloys (Audi 8, Mazda
AZ550, PorscheEXP...). The share of special aluminium alloys in the total cars mass is
expected to be around 15%, and around 25% for high performance cars.
MAXX
Fig. 2 Application of Al-alloys (whole car body or carrying construction)
Small weight, resistance to corrosion and recyclability are the main properties
of Al-alloys, especially regarding application in car industry. Three groups of Al-alloys
are most often used for car body parts: Al-Cu(series 2000), alloys Al-Mg-Si (series
243
6000) and alloys Al-Mg (series 5000). Al-alloys are less formable compared with steel
sheet metals. Cars producers are interested in applying aluminium for passenger car
bodies manufacture. Audi manufactured the model A8 whose entire car body was
made of aluminium. By applying Al, car body weight can be reduced for 30-50%. The
application of aluminium requires the change of style, where outer form must be
adjusted to aluminium properties. Extremely rounded form is needed, with as few
sharp edges as possible. Aluminium application is becoming interesting for the
manufacture of small cars as well, in combination with plastic. Fig. 3 shows the exterior
of vehicle MAXX, and Fig. 2 the carrying construction of that vehicle made of Al. By the
application of aluminium profiles, the desired carrying construction was obtained which
allows unrestrained vehicle exterior.
MAXX
Yugo
Fig. 3 Application of Al-alloys
In Zastava cars, the concept of car body parts manufacture out of aluminium
sheet metals was applied. Figure 3 shows front outer mudguard of vehicle Yugo. The
existing pressing tools were used for manufacture of parts. Sheet metals thickness was
increased but not as much as requested by aluminium parts. The entire car body was
manufactured, with aluminium parts, and it underwent the existing car body protection
process. Aluminium parts satisfied the requested quality level completely. This was the
verification of the concept of manufacturing aluminium parts which are hung onto steel
car body.
FIAT Tipo
BMFT-carrying construction
Fig. 4 Application of plastic masses for car body parts manufacture

244
Analysis of the Effects of Applying New Materials
2.3 Application of plastic

Plastic masses are increasingly more applied for cars, for manufacturing both
the exterior parts and the interior parts of the vehicle. There is a constant tendency of
raw materials producers to apply plastic for manufacture of car body coatings, carrying
construction and, afterwards, the entire car body. There are many examples of
manufacturing outer coatings, see Fig. 3, of parts which are not under load and need to
be replaced quickly and frequently. With the aim of reducing vehicle mass by applying
plastic masses, comprehensive researches are being carried out, especially on the
side of raw materials producers. Initially, application was limited to outer coatings. On
vehicle FIAT Tipo, front inner mudguard coating and the fifth door were made out of
plastic masses, see Fig. 4. At that time, front inner coating (mudguard) made out of
plastic mass was an innovation, especially since the sheet metal pressed parts were
eliminated. On this point, the application really made sense because this part has a
very complex form, and plastic masses are also resistant to corrosion. The fifth door
was also in focus, due to its weight and complex form. In the course of application of
plastic masses on the fifth door, connecting of door coatings and hanging of other
elements onto it was a huge problem. There were some attempts to make the carrying
construction out of plastic masses, see Fig. 4, where the result of BMFT project,
realized by VW, is shown.
application points
recyclable materials
Fig. 5 Points of plastic masses application

Fig. 5 shows some points where it is possible to apply plastic masses. New
possibilities are found on a daily basis. However, car industry is facing the problem of
recycling its products, especially after their expiry. This was an additional limitation for
application of plastic masses.
In recent times, new solutions for plastic masses application in passenger car
body have appeared. Valid regulations, which are constantly becoming stricter and are
changing all the time, especially regarding safety, have created the possibility to apply
plastic masses for preservation of car body joints. These demands are even stricter for
cars which are manufactured or are about to be manufactured serially. This paper will
look into the materials of company Dow automotive, i.e. the application of their
materials for strengthening car body junctions. The advantages of applying these
materials are the following:
Possibility to satisfy regulations
245
Minimal changes on construction, i.e. existing car body parts.

Fig. 6 shows the points where it is possible to apply these materials, as
recommended by company Dow. The selected points are generally sensitive on most
car body types. The idea is to increase the stiffness of car body as a whole by
increasing the stiffness of car body joints. With the purpose of determining the effects
of application of new generation plastic masses, appropriate static and quasi-static
investigation of car body joints were carried out. In addition to determining the
influence on stiffness, the effects of secondary influences related to noise and
vibrations of observed car body joints were measured. By the application of plastic
masses of new generation, increase of stiffness of car body joints, and especially of
car body as a whole, was obtained. The application of these materials does not require
the change of existing technologies.
Fig. 6 Application of plastic masses of new generation

3. CONCLUSION
Development and application of new materials in car industry is a necessity.
Both material producers and the entire cooperation must be involved into such trends.
The responses of car producers are various, but they are all following the trend of
introducing new materials.
LITERATURE
[1]
[2]
[3]
246
Hayashi, H., Nakagawa, T. (1994). Recent trends in sheet metals and their
formability in manufacturing panels, Journal of Material Processing technology, 46
(1994), str. 455- 487.
Advertising materials of company Dow
Milovanovi, M. (2000). Passenger car body, Monograph, Cars Institute,
Kragujevac, 2000.
APPLICATION OF RAPID PROTOTYPING RESINS FOR

PHOTOELASTIC TESTING
Tomasz Kudasik1, Tadeusz Markowski2, Olimpia Markowska3, Sawomir
Miechowicz4
Summary: The possibility of application of chosen Rapid Prototyping (RP) resins for
photoelastic testing in this paper is presented. The aim of the work was to compare
usefulness of available resin test models to be directly applied for photoelastic
experiment. The RP resin test models were compared with vacuum casted resin model
(as a reference model) commonly used for photoelastic testing of machine parts. The
models for tests were fabricated using chosen Rapid Prototyping (RP) techniques:
Stereolitography (3D Systems Sl5170 epoxy resin), Polyjet (Objet Fullcure 720 epoxy
resin) and vacuum casting method (Epidian 5 epoxy resin). In the paper the evaluation
process of fabricated RP models was described. The experiment assumptions, the
fields of possible application of chosen RP materials, medical and technological
consideration of complex models testing in this article were pointed.
Key words: photoelastic test, rapid prototyping (RP)
1.
INTRODUCTION
Photoelasticity tests together with the finite-element method are known

methods of measuring stress distribution in the elements of complex structures [1,2,3].
In the photoelastic method, transparent materials are used to observe stress
distribution in models. Those transparent materials are characterized with special
abilities that allow for illustrating the stress distribution in a form of an isochromatic
image. Apart from traditionally used epoxy resins (e.g. Epidian), nowadays rapid
prototyping materials are used (Stereolitography, PolyJet). With the rapid prototyping
method, it is possible to create models which reproduce the inner structure of objects;
such a construction of a model allows for achieving the analysis results exceeding
anything that was possible so far [4]. It is essential to examine the properties of those
1
Phd, Tomasz Kudasik, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959 Rzeszow,
Poland, tkudasik@prz.edu.pl
2
Prof., Tadeusz Markowski, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959
Rzeszow, Poland, tmarkow@prz.edu.pl
3
Msc, Olimpia Markowska, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959 Rzeszow,
Poland, omarkowska@prz.edu.pl
4
Phd, Slawomir Miechowicz, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959
Rzeszow, Poland, smiechow@prz.edu.pl
247
Tomasz Kudasik, Tadeusz Markowski, Olimpia Markowska, Sawomir Miechowicz
new materials to see if they are useful in photoelasticity tests.

2. PHOTOELASTICITY
The phenomenon of photoelasticity is based on the temporary double refraction
effect. This effect occurs in some transparent materials under load. The incident ray
passing through the photoelastic material is split into two differently refracted rays. On
leaving the material, the rays experience a phase retardation originating in the various
speeds of the rays in the area of birefringence.
Isochromatics are the geometric locus of the fixed points of the principal stress
1 and 2 difference. The isochromatics images allow for determining the distribution
of the principal stress difference in the tested area of the model, on the grounds of the
Wertheim law:
V1 V 2
m KV m
2W max
const
Where: m the order of isochromatics sets the value of the relative retardation
of the incident ray components in proportion to the phase perigon (2S,)
KV m
is the model elastooptic constant, it sets the value of the isochromatics
order (MPa).
The optical effect is not in fact stress as such: isochromatics are the geometric
locus of the fixed points of the principal stress 1 and 2 difference, determines directly
value of Wekstr, therefore illustrates the effort level in accordance with the Wmax
hypothesis. The transmission polariscope used to experiment is shown in figure 1.
Fig. 1 The transmission polariscope used for experiment

248
Application of rapid prototyping resins for photoelastic testing
3. PHOTOELASTIC RESIN ANALYSIS

Samples made of the following materials were tested:
Epidian 5 epoxy resin the most common epoxy resin used in photoelastics
research, due to the low price and tested and proved photoleastics
properties,
SL-5170 epoxy resin used in Stereolitography (rapid prototyping method),
FC-720 epoxy resin - used in PolyJet (rapid prototyping method).
The materials were prepared in a form of equal epoxy resin sample plates,
each 10x70x5mm in size. The samples made of SL-5170 and FC-720 epoxy resins
were produced in the process of rapid prototyping. They were used as models to
produce the Epidian 5 epoxy resin sample via the vacuum casting method with silicon
mould (fig. 2).
Fig. 2 Epidian 5 epoxy resin sample - produced in the vacuum casting method with
silicon mould, SLA SL-5170 epoxy resin sample, PJ FC-720 epoxy resin sample
A load was applied to all the samples in a test stand created especially for the
experiment (Fig. 3). The linear displacement L, which value was measured out with the
micrometer calliper, was used as the input function.
249
Fig. 3 The test stand used to apply the load

The transmission polariscope (Fig.1) was used to monitor the stress in the form
of isochromatic lines. The tested sample was placed between the polariscope and the
analyzer. The built-in quarter-wave plates allowed for acquiring the isochromatic lines
distribution image free from interference.
The most important parameter of the isochromatic lines distribution is the so
called isochromatics order m. It is determined by observing the gradually appearing
sequences of colours:
m = 1 black (zero stress), yellow, red, blue
m = 2 yellow, red, green
The following isochromatics orders are described as repetitive sequences of
colours of m=2 order, while the higher the order of the isochromatics, the more pale the
colours become.
The load applied and the orders of isochromatics for individual samples are
shown in the table 1 and the figure 4 below.
250
Application of rapid prototyping resins for photoelastic testing
Table 1 The value of isochromatics order in relations to the load applied

L=0,25 mm
L=0,5 mm
L=0,75 mm
L=1 mm
Epidian 5 m
10
FC-720 m
1.5
2.5
3.0
SL-5170 m
1.5
3.5
Fig. 4 The isochromatics lines distribution in the tested samples

The presented results show, that the Epidian 5 samples are characterized with
the best photoelastic properties, with higher isochromatics orders produced. The SL5170 and FC-720 epoxy resins have weaker photoelastic properties, nevertheless the
intensity of the temporary double refraction effect was satisfying and those materials
are suitable for photoelastic experiments.
4.
CONCLUSION
It cannot be ultimately determined which of the tested epoxy resins is the most
suitable for photoelastic research. Epidian 5 has the best photoelastic properties, yet
the models can only be produced via vacuum casting method. It means that the
Epidian 5 epoxy resin can be used to create only fulfilled structure models. The other
tested resins can be used for rapid prototyping technology. In these methods models
can reproduce the inner structure of tested elements [5]. Therefore, they can be used
to analyze the stress distribution in e.g. medical models in which reconstruction of the
inner structure of tissues, together with empty spaces inside is crucial [6]. It was not
possible to produce models with the bone inner structure with traditional materials.
Consequently, older models were very simplified [7]. The advances in Rapid
Prototyping methods of producing medical models from computed tomography (CT)
data, give a new quality in experimental research.
251
ACKNOWLEDGEMENTS
This work was financed from the Polish science budget resources as the
research project NN501268134.
LITERATURE
[1] Doroszkiewicz, R. (1975). Elastooptyka. Stan i rozwj polaryzacyjno-optycznych
metod dowiadczalnej analizy napre. Polska Akademia Nauk, Instytut
Podstawowych Problemw Techniki, PWN, Warszawa Pozna.
[2] Kopkowicz, M. (2003). Metody dowiadczalne bada konstrukcji. Oficyna Wyd.
Politechniki Rzeszowskiej, Rzeszw.
[3] Pindera, J. T. (1953). Zarys elastooptyki. PWT, Warszawa.
[4] Miechowicz, S., Marciniec, A. (2004). Stereolitography the choice for medical
modeling. Acta of Bioengineering and Biomechanics, nr 6/1.
[5] Petzold, R., Zeilhofer, H.F., Kalender, W.A. (1999). Rapid prototyping technology
in medicine basic and applications, Comput. Med. Imaging Graphics, nr 23,
(277284),
[6] Raic, K., Jansen, T., Rymon-Lipinski, B., et al. (2002). Fast generation of
stereolithographic models. Biomedical Technology, nr 47, p. 83-85.
[7] Bdziski, R. (1997). Biomechanika inynierska. Wydawnictwo Politechniki
Wrocawskiej, Wrocaw.
252
AUDITING PROCESS DESIGN COMPANY LOGISTICS SYSTEM

Slavica Cvetkovi1
Summary: A particular problem relates to the process of auditing enterprises logistics
system process. The importance of auditing, optimization and process improvements
for supply-chain management is particularly important because based on the results of
this process in organization, strategic decisions and plan for significant resources are
made.
Key words: logistics audit, handling the company, methods of analysis, using the
results of the logistic audit
1.
INTRODUCTION
Logistics Company is a function of business management which aims to add,

to the traditional values of assets such as: product type, quality, price and so on, the
new values: availability of required goods at the place and moment in which the
consumers manifest the needs for these goods.
In any case, this definition provides a fundamental understanding of logistics,
emphasizing the spatial and temporal dimension of the area it covers.
The goal of logistics can be generally defined as: meeting customer needs by
delivering the required product, the required quality at the right place at the right time,
at a minimum price.
For more efficiantly handling logistics system, companies need a good
knowledge of the existing system. For this, it is possible to use logistics audit and after
the implementation of his proposed changes, we can achieve increased performance
of the logistics system. [1]
2.
COMPANY LOGISTIC AUDIT
A firm logistic audit is standardized electoral and the projective process

project, which is related to the logistical functions of the system for handling company.
The main goal of logistics audit, in addition to review and detailed list of current
logistics summary of the main bottlenecks, is to find logistic system problematic places
and propose solutions for removal. Logistic audit provides an overview of the required
solutions of some logistical problems and potential possible revenue (time and financial
1
Slavica Cvetkovic, Fakultet tehnickih nauka, Kosovska Mitrovica, Srbija

The work was done under the project no. 042006, funded by the Ministry of Science of Serbia.
253
Slavica Cvetkovi
savings). The selected areas are specifically named and limited.[2]

Logistic audit the methodological approach to the evaluation of the condition
and operation of logistics system performance. It is a complex logistical process
diagnostics and functions of the company. The aim of the implementation of logistic
audit is to get enough information as is necessary for innovation management system
company with logistics activities.
If we decide to implement the logistics of the audit, it is necessary to select
areas where its implementation has been successful. In the case of complex logistics
audit its size is ten core areas of logistics systems company. The solution for selecting
specific fields is the basic model of micro logistics companies
information flow,
material flow
the flow of finance).
-Figure 1.(
Aspects of defining specific areas are the basic logistics processes, that result
from the basic micrologistic model of some Company:
Strategic Logistics
Acquisition Logistics
Production logistics
Logistics service process
Maintenance
Logistics Sales
Distribution
Technical Resources Logistics
Information Logistics
Reverse logistics
In addition to the complex logistic audit of the micrologistic system, all ten
areas are evaluated. Within each evaluated area, there are ten processes that have to
be evaluated. in any evaluation process, it is necessary to answer ten questions. This
means that, in total, the thousand questions need to be answered at, which is an
adequate amount of questions in order to achieve a more detailed review of the
implementation of complex logistics audit. [3]
The Principle of issues construction in the logistics audit is based on the
branched hierarchical structure with the coefficient 10. Each question has a question
code in itself, made from combinations of three numbers. Specific numbers are used
for classifying questions on specific process and specific areas. Each question is
divided by areas: the average one, the information one, the decisive one, the
managerial one or the quality one. This mechanism is used for evaluation section.
The formulation of questions in the logistic audit has its explicitly side,
represented by the question and its expected answer, and together the hidden implicit
side. Questions regarding system handling the company can be defined explicitly and
implicitly. There are the testing issues here, through which the propriety of the answers
to the question that is already set, is checked.
In case the company don't decides for complex logistics audit, but has interests
in the logistics audit only for selected areas, the logistic audit is adjusted to the agreed
areas. The selected areas are specifically designated and limited.
254
Auditing Process Design Company Logistics System
Fig. 1 The basic model of micro logistics firm [2]

2.1 Release logistics audit
The first step is to highlight the needs of the logistics audit firm. The following
step is contact with the company, which offers specific audit services and on the basis
of common conversation is selected subject to audit. It can be arranged complex
logistics audit, or logistic audit can be translated only in selected areas. According to
the agreed scope of the logistics audit you select one, or all those auditors who need to
be informed about necessary data and implement the logistics audit.
After the completion of the audit follows the logistic evaluation of its results,
then some specific shortcomings are named and then suggest the correct procedures
255
Slavica Cvetkovi
in order to remove the shortcomings. Using the logistics audit is not just getting
detailed views on the current state of logistics in the firm, but also finding bottlenecks
problematic logistics system places and proposal for removing in order to demonstrate
the potential logistical changes for the purpose of accomplishment of more efficient
performance of the company. [5]
Proposal of appropriate treatment are the correct procedures that are
compared with the importance and charging time. Approval of the proposal follows the
correct procedures by firm management. Enter the appropriate treatment is
implemented through the pre-defined harmonogram. It is necessary to monitor the
changes of entering the appropriate treatment, as these can significantly impact the
company strategy.[6]
2.2 Logistic audit as a method of analysis
The design of logistics systems can use the system analysis, SWOT analysis,
or for example, the simulated models. For these purposes it is possible to use the
logistics of the audit.
Logistic audit as a method of analysis is directly focused on the logistics
activities of all the above methods of analysis has the highest comparative ability. It
can be compared to systematic analysis in the field of logistics, where the system is a
complex logistics system has been designed to cover all the basic spheres of the
logistics processes. It is more siutable for the conceptual proposals than for concrete
proposals algorithms logistics activities.
The logistics audit is "used" as an effective tool for designing logistics systems.
Results from implementation of the logistics audit could be used in choosing
the design paradigm, so called in the case, if the system requires radical change,
which can be realized with the new system, or reengineering of the basic system.
Logistic audit shows, at which level is the current system, if it is functioning as a logistic
system and whether to do only partial change. In this case, the basic paradigm has
focused on several of case studies process areas, which stipulated in the audit if the
results of the logistic are problematic.
From another aspect, the logistics audit can be defined as the implementation
of the SWOT analysis because from the evaluating the response to current issues is
clear that processes are evaluated highly, that is, they are approaching the maximum
assessment and the processes that has been low evaluated.
256
Auditing Process Design Company Logistics System
Fig. 2 The choice of paradigm, analysis and synthesis of the proposals [7]
But, for the proposal of the logistic processes of their innovations or engineering, the
logistics audit is not enough and upon it other methods, by which we achieve further
information on technology, elementary processes, service processes, planning
257
Slavica Cvetkovi
process, of the deployment etc., are built. Logistic audit could be used as a proposal of
the appropriate procedure, but not as detailed proposal of the logistics processes and
activities.
2.
CONCLUSION
Besides detailed review about the current state of logistics activities in a

company and finding bottlenecks, the proposal of their removing should be offered. It is
also possible to use it in designing logistics systems as a quick preparatory analysis of
the logistics system, or directly as a method of analysis, oriented to key logistics
activities of firms.
Significant opportunities to increase cash and success of the management, are
hidden in the use of new methods and procedures in managing the company's
controlling [8]. Logistic audit is also plotted as an effective tool in handling company.
Challenges in the logistics audit can be used as a suggestion of correct procedure in
handling the company, or as the basis for implementation of proposed changes in the
process of the firm.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
258
Vinko Jevti, Tehnika logistka, skripta 2003 Mainski fakultet, Ni

Boi V., Uvod u logistiku marketinga, Beograd, 2004., str. 13.
www.efqw.com
www.krajinaklas.com/
www.krajinaklas.com/
Mr. Zdenko Adelsberger, Auditiranje procesa za upravljanje rizicima prema ISO
31000:2009
Curtis C. Verschoor, AUDIT COMMITTEE ESSENTIALS, John Wiley & Sons,
Inc., ISBN 978-0-471-69959-0
Vidovi, M., "Mogunost poboljanja performasi logistikih sistema optimizacijom
operativnog planiranja nekih klasa pretovarnih procesa", Doktorska disertacija,
Saobraajni fakultet, 1997. god.
MAJSTOROVI, V.: Upravljanje proizvodnjom i projektima, Sveuilite u Mostaru,
Mostar, 2001.
CAD/CAM TECHNOLOGY IN DENTAL MEDICINE

Andonovic Vladan1, Vrtanoski Gligorce2
Summary: CAD/CAM is an acronym that means computer-aided design/computeraided manufacturing. CAD/CAM technology has been used in the manufacturing
industry for many years. Today, CAD/CAM technology is used in dentistry to help
dentists and dental lab technicians fabricate precise shapes and sizes for dental
restorations, including inlays, onlays, crowns, and bridges. Dentists use CAD/CAM
technology to provide their patients with durable, well-fitted single and multiple tooth
restorations in a more efficient manner than traditional lab-fabricated restorations.
In this paper it will be explained how dentists and dental lab technicians use CAD/CAM
technology to design the anatomical features, size and shape of a tooth restoration on
a computer. The CAD/CAM computer screen displays a 3-D custom image of the teeth
and gums, allowing the dentist to use a cursor to draw the precise design of the tooth
restoration. The CAD/CAM machine fabricates the restoration through a milling
chamber that crafts the tooth-like ceramic material into a precise replica of the drawing.
Key words: CAD CAM technology, dentists, crowns, bridges
1. INTRODUCTION
Dental CAD/CAM technology is available for dental practices and dental
laboratories, enabling dentists and their staff (or a laboratory technician) to design
restorations on a computer screen [1]. The CAD/CAM computer displays a 3-D custom
image of prepared tooth or teeth obtained by digitally capturing the preparations with
an optical scanner. Alternatively, the 3-D images can be obtained by scanning a
traditional model obtained from conventional impressions of the preparations. The
dentist or laboratory technician then uses those 3-D images and CAD software to draw
and design the final restoration [2,3]. The amount of time it takes for a dentist, in-office
restoration designer or laboratory technician to design a restoration varies based on
skill, experience, and complexity of case and treatment. Some cases could take
minutes, while others could require a half-hour or more of design time to ensure
quality. Once the final restoration is designed, the crown, inlay, onlay, veneer or bridge
is milled from a single block of ceramic material in a milling chamber. The restoration
1
PhD student, Andonovic Vladan, Stip R. Macedonia, University of Stip - Faculty of Mechanical
Engineering, (vladan.andonovik@ugd.edu.mk)
2
PhD, Vrtanoski Gligorce, Skopje R. Macedonia, University of Skopje - Faculty of Mechanical Engineering,
(gliso@mf.edu.mk)
259
Andonovic Vladan, Vrtanoski Gligorce
then can be customized with stains and glazes to create a more natural look, before
being fired in an oven (similar to ceramics and pottery), and then finished and polished.
2. CAD/CAM SYSTEM CONFIGURATIONS
2.1 CAD/CAM in the Dental Office
In-office CAD/CAM systems allow the dentist to take a digital impression and
design the restoration, all in the doctor's office. Once the restoration has been milled, it
is often polished or glazed before placement. Another approach is to send the scan or
digital impression to the laboratory [1,2,3]. A laboratory technician designs the
restoration on the computer and communicates that to the milling machine. The milling
machine can be in the lab or outsourced to a milling center. Once the milling machine
has completed the restoration, it may require layering porcelain and staining and
glazing.
2.2 In office complete CAD/CAM systems Chairside Intraoral Scan and Mill
These systems allow the dentist in one patient visit to:
Do an intraoral scan.
Design the restoration.
Mill the restoration.
Stain and glaze the restoration.
Seat the restoration.
Manufacturer and vendor are: CEREC (Sirona), E4D Dentist (D4D Technologies).
2.3 In office Intraoral Scanners - Chairside Intraoral Scan and Outsource Milling
These systems allow the dentist to:
Do an intraoral scan.
View an interactive digital model.
Design the restoration, in some cases.
Final steps to complete restoration:
Virtual model trimming done by design center, lab, or SLA
(stereolithography apparatus) facility.
Stereolithographic model and dies are created by SLA facility.
If zirconia restoration, coping can be made by milling center.
If PFM restoration, wax coping can be made by milling center or SLA facility.
Model and/or coping sent to laboratory to finalize restoration.
Manufacturer and vendor are: Lava Chairside Oral Scanner C.O.S. (3M ESPE),
iTero (Cadent).
2.4 CAD/CAM in the Dental Laboratory
Laboratory CAD/CAM systems can be standalone systems that allow dental
laboratories to scan, design, and mill restorations; alternatively, restorations can be
outsourced to a milling center where a coping is made. Laboratory CAD/CAM systems
may also accept intraoral scans from dental practices and fabricate the final
restoration.
260
CAD/CAM technology in dental medicine
2.5 Laboratory CAD/CAM systems

Laboratory CAD and laboratory CAD/CAM systems include following:
Use traditional impression techniques, have the model scanned, and make
the restoration using CAD/CAM technology.
Use any intraoral scanner and send the file to a participating lab or milling
center for design and finishing.
Manufacturer and vendor are: Cercon (DENTSPLY Ceramco), Everest (KaVo),
Lava Precision Solutions (3M ESPE), Sirona InLab (Sirona), Zeno (Wieland)
2.6 Laboratory CAD systems
Some laboratories have made the investment in digital scanning and design
technology. However, to complete the restoration, the lab will outsource the design to
another company to mill. Send the restoration back to the design CAD lab for finishing,
if applicable. Manufacturer and vendor are: Authorized Lava Milling Center (3M ESPE),
Cercon (DENTSPLY Ceramco), GC Advanced Technologies Milling Center (GC
America),Katana (Noritake), Procera (Nobel Biocare), Vericore (Whip Mix), Zeno
(Wieland).
2.7 Milling machines
Every dental CAD system saves the model in a format that is suitable for
manufacturing using CNC Machining Centres, laser sintering machines and 3D printing
systems i.e. has export option for 3D printing and rapid prototyping machines. Today,
there are 3, 4 and 5-axis CAM milling machines to help us manufacture simple and
anatomical copings and bridge frameworks, inlays, onlays, dental bars, implant
bridges, and customised abutments. Every completely CAM system uses industry
standard formats to enable easy model sharing and connectivity between other
CAD/CAM systems and CNC machines. 3-axis machining module offers fast, effcient
cutting strategies for most dental applications [5,6,7]. The optional 4- and 5-axis
module gives the added flexibility needed to cut undercut regions and other more
complex shapes such as dental bars, implant bridges and customised abutments. On
the figure 1 it is given a part of one CAD software where we can see how is easy
manipulating with our dental model.
Fig. 1 Part of one dental CAD software [5]

On the figure 2, we can see how milling tool is in contact with dental materials
i.e. tool in work.
261
Fig. 2 Contact between tool and dental material [5]

Toolpath at dental milling systems is fully automated and as a result of this it is
produced high finished parts and reduced cutting times, as shown on figure 3.
Fig. 3 Toolpath at dental milling systems [5]

Almost every dental CAM milling system use materials such as: Titanium,
zirconium or zirconia dental PMMA, nano composites, Al-Oxide (Aluminum Oxide),
dental wax and alloys, chrome-cobalt (CoCr), titan and composites, as can be seen on
the figure 4.
Fig. 4 Different dental materials used in CAM milling systems [5]

Dental CAM system recognizes the margin line as the most critical part of the
model and applies unique cutting strategies in this area to ensure high accuracy,
therefore minimizing hand finishing, as shown on the figure 5.
262
CAD/CAM technology in dental medicine
Fig. 5 Dental CAM system recognizes the margin line [5]

On the figure 6 it is shown dental milling tool in action producing fast and
accurate finished dental products.
Fig. 6 Dental milling tool in action [5]

As end or finished product of dental milling machine, 3-, 4- and 5-axis CAM
milling program, can be simple and anatomical copings and bridge frameworks, inlays,
onlays, dental bars, implant bridges, and customised abutments, figure 7 and figure 8.
Fig. 7 End product of dental milling machine [5]
263
Fig. 8 Finished product of dental milling machine [5]

3. CONCLUSION
All-ceramic restorations, including those fabricated at a dental laboratory using
CAD/CAM technology, tend to be a more expensive restorative solution. However,
even though the materials for CAD/CAM restorations might cost more, the expense
incurred by the dental laboratory and/or the dentist may not be passed onto the patient.
Also, there is no additional fee or cost to have a restoration placed in one visit as
opposed to two. Insurance reimbursement is similar for in-office, same day dentistry or
laboratory-fabricated restorations.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
264
Daskalaki, A. (2009). Dental computing and applications: Advanced techniques

for Clinical Dentistry. Max Plank Institute for Molecular Genetics. Germany.
Jef, Z. (2007). Zirconia ceramic in dental CAD/CAM: How CAD/CAM technology
enables zirconia to replace metal in restorative dentistry. Journal of Dental
Technology. p. 17-24
M. Angeles F.T. et al. (2009). CAD/CAM dental systems in implant dentistry:
Update. Med Oral Patol Oral Bucal, 14(3):E141-5.
E. Dianne R. (2006). Dental CAD/CAM systems:A 20-year success story. The
Journal of the American Dental Association, vol. 137, 5S-6S.
Delcam, from http://www.delcam.com, accessed on 2010-08-07.
Dentalcompare, from http://www.dentalcompare.com, accessed on 2011-02-03.
Astra tech, from http://www.astratech.us, accessed on 2011-03-04.
CHANGES VOLTAGE COMPACTION POLYPROPYLENE FILLED

WITH GLASS POWDER
Zoran Janju1, Aleksandar Petrovi2, Aleksandar Jovovi3, Radica ProkiCvetkovi4, Predrag Ili5
Summary: The paper addresses changes in mechanical properties of composite
materials made from waste polypropylene and waste glass, without going into detail
regarding chemical reactions and physical characteristics of the materials combined.
The main goal of the research conducted was to demonstrate that polymer waste
materials might be used as raw or usable materials obtained through municipal waste
management. Repeated use of polypropylene, achieved by material recycling and
addition of glass additive, was examined with respect to the reduction in tensile
strength of the material analysed. Research investigation was conducted using the
standard test specimens made from the base material (polypropylene) with 0%, 5%,
10%, 15%, 20%, 25% or 30% by weight of glass powder additive, with glass particle
size of less than 0.5 mm. Test specimens were examined after each of the six
repeated recycling processes. Research investigation included measurement of all
compaction test parameters of the materials examined. Measured data enabled a
voltage compaction change pattern to be determined for polypropylene-based
composite containing different amount of glass additive, analysed after each of the six
repeated recycling processes. Results obtained are deemed to be useful in pointing
out future research directions.
Key words: Recycling, polypropylene, voltage compaction.
1. INTRODUCTION
Growing concerns about environmental degradation, primarily caused by rapid
depletion of natural resources as well as accumulation of waste, have initiated many
actions oriented towards the use of environmentally friendly materials and their
1
Master of science, Zoran Janju, Banja Luka, City administration Banja Luka, (mrjzoran@teol.net)
Doctor of science, Aleksandar Petrovi, Belgrade, Faculty of Mechanical Engineering, University of
Belgrade, (apetrovic@mas.bg.ac.rs)
3
Doctor of science, Aleksandar Jovovi, Belgrade, Faculty of Mechanical Engineering, University of
Belgrade, (ajovovic@mas.bg.ac.rs)
4
Doctor of science, Radica Proki-Cvetkovi, Belgrade, Faculty of Mechanical Engineering, University of
Belgrade, (rprokic@mas.bg.ac.rs)
5
Doctor of science, Predrag Ili, Banja Luka, Institution of protection, ecology and informatics,
(pedja_1979@yahoo.co.uk)
2
265
Zoran Janju, Aleksandar Petrovi, Aleksandar Jovovi, Radica Proki Cvetkovi, Predrag Ili
recycling at the end of the products' useful life. In general, waste polymers represent
significant portion of the overall quantity of municipal waste [6]. Due to their favourable
recycling properties, thermoplastic polymers are the most commonly used polymers,
while duroplast is used less often [4]. Polyethylene (PE) and polypropylene (PP)
represent 60-70% of total waste polymers, while polystyrene (PS) contributes with 1015%, polyvinylchloride (PVC) with 15% and polyethylene terephthalate (PET) with 5%
to total polymer waste [2].
In recent years, increased use of composite materials i.e. materials
representing a combination of two or more materials differing in form and composition,
combined in order to achieve desired properties of the newly formed material, has
been recorded. The most commonly used are fibrous composites which contain a
fibrous phase that enhances mechanical properties of the composite formed. Besides
artificial fibres, natural fibres such are hemp, jute and flax fibres, as well as poultry
feather, waste paper, cardboard, textile and other similar fibers are more frequently
used nowadays [3], [5]. However, the use of two different materials for the formation of
composite material inevitably increases the complexity of composite recycling, since all
constituent materials i.e. fibres and matrices need to be separated before the recycling
is performed [1]. The said difficulty may be overcome through the use of particulate
(granular) composites which may be reused in their integral form. The paper presents
a change in the voltage compaction of one such composite occurring after multiple
composite recycling (hereinafter processing cycles), as well as the effects of different
content of particulate additive on the voltage compaction variation of the particulate
composite analysed.
In addition, rather then considering possible utilization of the material in
question, numerous research investigations have mainly considered ways to increase
tensile strength of polypropylene. This was mainly achieved through copolymerization
i.e. addition of additives or modifying agents or the formation of fibrous or laminar
composites.
2. TEST SAMPLES
Experimental investigation was conducted to examine a change in the voltage
compaction of the glass reinforced waste polypropylene occurring after multiple
repeated processing cycles of the polypropylene-based composite containing different
content of glass additive. Glass powder additive used in the experiments conducted
was obtained by waste glass grinding.
Mechanical and other impurities in the waste polypropylene and waste glass
were removed by washing and drying, in that manner enabling the materials to be
properly prepared for recycling. Waste polypropylene and waste glass used in the
experiments conducted were obtained by waste sampling at Banja Luka municipal
solid waste landfill. Waste sampling was carried out at several locations at the
municipal landfill, in that manner enabling the collected waste polypropylene and glass
to be of diverse age and quality.
Experimental investigation was focused on examining changes in mechanical
properties of the composite materials generated from waste polypropylene and waste
glass, without going into detail regarding chemical reactions and physical
characteristics of the materials combined.
266
Changes voltage compaction polypropylene filled with glass powder
The main objective of the research performed was to reach conclusions on the
possibility, quality and reliability of repeated composite use, achieved through partial
component separation and repeated use of the components obtained. In this manner,
a change in mechanical properties (voltage compaction) of the composite, occurring as
a result of multiple composite processing cycles, was estimated.
Experiments were performed with polypropylene powder in order to achieve
homogeneous mixing with glass powder during composite formation. Glass powder
used in the experiments was characterized by particle size of less than 0.5 mm.
Polypropylene powder (matrix phase) and glass powder (dispersed phase) mixing was
performed mechanically. Prior to further use, polypropylene and glass powders were
dried out in order to remove the excess moisture.
Test specimens were made from the base material and various fractions of
glass powder additive (5%, 10%, 15%, 20%, 25% and 30% by weight). A certain
number of composite specimens containing 10%, 20% or 30% by weight of glass
additive was not recycled after the initial set of experiments had been performed, while
the rest was grinded and prepared for repeated casting and testing. In such manner,
specimens made of base material as well as those made from the composites
containing different amount of glass additive were re-manufactured after each of the
six material processing cycles. In addition, different colour additives were added to the
specimens with different content of glass additive in order to make them more
distinguishable.
3. COMPACTION TEST RESULTS
In general, comapaction testing is conducted by subjecting a standard
dimension test specimen to a gradually increasing tensile load until reaching a
breakpoint (fracture). During the compaction testing, all changes occurring in the
specimen are recorded and used to determine mechanical properties of the material
examined. Compaction testing was performed by applying a gradually increasing
compaction load, as specified in the standard for determination of compaction
properties of plastic materials.
Dimensions of the compaction specimen used in the experiments were the
following: Length of the specimen l = 22 mm; Height of the specimen h = 20 mm; Initial
gauge thickness s = 4 mm.
Experiments were performed in such manner as to enable compaction
properties of the specimens made from the base material, as well as those made from
the composite material produced after each material processing cycle, to be
determined. Material obtained after each processing cycle was used to manufacture
three composite specimens which were then subjected to compaction testing.
Similarly, three specimens were made from the base material and used in the
experiments mentioned.
Experimental results obtained may be summarized as follows: maximum
compaction load F[N], as indicated by a measuring device . Compaction voltage Vp
[N/mm2], calculated from the expression: V p F max , where: A l so initial cross
Ao
sectional area in the gauge section.

Mean compaction voltage values are obtained as arithmetic means of the
compaction voltage values determined for each of the three test specimens
267
manufactured after each processing cycle of the material containing specific content of
the glass powder. Mean compaction voltage values are indicated in Table 1.
Table 1 Mean values of compaction voltage obtained by calculation
Glass
powder
content
(% by
weight)
0%
5%
10%
15%
20%
25%
30%
Mean compaction voltage VP [N/mm2]

Base
mat.
1st proc.
cyc.
2nd proc.
cyc.
3rd proc.
cyc.
4th proc.
cyc.
5th proc.
cyc.
6th proc.
cyc.
54.92
52.65
51.52
53.41
55.30
59.85
60.98
55.87
57.95
57.95
57.95
56.82
56.82
55.49
57.58
57.95
57.58
58.33
57.58
56.44
58.33
57.95
58.71
59.85
59.85
60.98
62.12
62.12
63.26
62.5
62.88
4. COMPACTION VOLTAGE DATA PROCESSING

Based on the experimental results presented, functional dependence of
compaction voltage on the glass powder content and number of repeated material
processing cycles was analysed. Analysis was performed using the least-squares
method and was carried out in two stages. The first stage of the analysis performed
was focused on determining a functional dependence which would most accurately
describe a correlation between the compaction voltage and the glass powder content
of the composite analysed. The correlation was to be determined for each processing
cycle examined. The following dependence between the parameters specified was
obtained:
(1)

where: y(s,x) compaction voltage; x glass powder content, by weight; a, b, c
coefficients describing the variable (x); s number of repeated material processing
cycles. As indicated in Table 3, coefficients a, b and c describing the variable (x) have
specific values for each processing cycle analyzed. The functional dependency
obtained (1) showed good agreement with mean values of compaction voltage, as
confirmed by standard deviation (2) and coefficient of determination (R2) shown in
Table 2.
Table 2 Tensile strength change model and related coefficients
Processing
cycle
a(s)
Base material 53.27
1st proc. cyc. 55.88
2nd proc. cyc. 57.96
268
b(s)
-0.015416488
-0.017465904
-0.015425126
c(s)
0.004586536
0.0042231955
0.0036609222
Coefficient of Standard
determination deviation
R2
0.86526705
0.99996371
0.99999813
2
9.608082
3.517736
2.380759
Changes voltage compaction polypropylene filled with glass powder
3rd proc. cyc.

4th proc. cyc.
5th proc. cyc.
6th proc. cyc.
58.16
57.94
57.02
56.78
-0.019295891
-0.01599839
0.014475026
0.01108406
0.0043160625
0.0040007363
-0.0015413783
-7.847311610-4
0.98332189
0.99993625
0.98808786
0.99967724
2.257738
3.754872
3.626022
4.51548
The second stage of the analysis performed was focused on determining a functional
correlation between the coefficients describing the variable (x) and the number of
material processing cycles i.e. and Investigation performed indicated
the following type of dependence between the parameters mentioned:
(2)

where: z(s) value of the coefficient or ; A, B, C, D, E and F
values of the coefficients from the equation (2) which depend on the coefficient which
is being determined i.e. or .
Values of A, B, C, D, E and F are shown in Table 3 (r2max= 0.996; 2max=2.54).
Table 3 Coefficients a(s), b(s) and c(s)
Coefficients a(s), b(s) and c(s)
a
b
c
A
53.256959
-0.015339
0.0045689
B
2.5057376
-0.0315013
0.0048400
C
D
E
F
0.6001802 -0.5119487 0.0878057 -0.0046734
0.051153
-0.027792 0.0059806 -0.000434
-0.0091865 0.0051384 -0.0011124 0.0000808
Results obtained are presented in Figure 2. The composite materials

containing 5%, 15% and 25% by weight of glass powder were only addressed in the
numerical analysis conducted and were not the subject of experiments performed.
Maximum error was calculated to be 3.22%, which was considered acceptable.
64
Basic material
1st recycling
2nd recycling
3rd recycling
4th recycling
5th recycling
6th recycling
Compaction voltage (N/mm2)
62
60
58
56
54
52
0
10
15
20
25
30
Glass powder content (%)
Fig. 2 A change in the compaction voltage of the composite material after different
number of repeated material processing cycles, depending on the glass powder
content of the composite
269
Graphic and table view show that compression increases in first and second recycling
(5% per cycle), and afterwards through other cycles of processing it slowly decreases
or stagnates in materials with lower share of glass powder, while in materials with more
than 15% of glass powder it continuously increases. It can also be noticed that
compression is relatively constant up to 10% of glass powder, and it further increases
in percentage lower than 3%. Relatively simple mathematical forms describe changes.
5. CONCLUSION
Based on the experimental investigation conducted and analyses performed,
the following is concluded:
1. Polypropylene-based composite containing up to 30% by weight of glass
additive (glass powder particulate composite) exhibits a stable change in
compaction voltage. The said enables relatively simple expressions to be used
to calculate compaction voltage of the composites made with different glass
powder content.
2. Composite materials analysed in the research investigation performed may be
repeatedly processed (recycled) without the component separation. Research
results presented in the paper have demonstrated that repeated material
recycling (six repeated processing cycles) does not significantly affect
compaction voltage of the material examined.
3. Based on the analyses performed, analytical expressions have been derived
that enable tensile strength of the composites made with different content of
glass additive to be calculated, as well as the compaction voltage of the
materials obtained after different number of repeated recycling processes.
4. Analytical expressions derived allow the future users of recycled municipal
waste materials (polypropylene and glass) to estimate possible utilization of
those materials in the construction industry or their use for other purposes.
LITERATURE
[1] Alcock, B., Cabrera, N.O., Barkoula, N.-M., Wang, Z., Peijs, T. (2008). "The effect
of temperature and strain rate on the impact performance of recyclable allpolypropylene composites", Composites: Part B 39, pp. 537547.
[2] Cavalieri, F., Padella F. (2002). "Development of composite materials by
mechanochemical recycling of post-consumer plastic waste", Waste Management
22, pp. 913916.
[3] Huda, S., Yang Y. (2008). "Composites from ground chicken quill and
polypropylene", Composites Science and Technology 68, pp. 790798.
[4] Pickering, S.J. (2006). "Recycling technologies for thermoset composite materials
- current status", Composites: Part A 37, pp. 12061215.
[5] Bullions, T.A., Hoffman, D., Gillespie, R.A., Price-O Brien, J., Loos, A.C. (2006).
"Contributions of feather fibers and various cellulose fibers to the mechanical
properties of polypropylene matrix composites", Composites Science and
Technology 66, pp. 102114.
[6] Janju, Z., ljivi, M. (2009). The reliability of use of recycled thermoplast, Journal
for Tehnology of plasticity, Novi Sad, 27-35.
270
CHANGING THE WALL THICKNESS PROFILE IN THE PROCESS OF

PROFILING
Miletic Ostoja1, Todic Mladen2
Summary: Analysis of test results obtained by the process of profiling showed the
changes in the thickness of the wall profile. The character of the flow deformation is
determined by the deformation zone where was a free material strengthening and wos
stretching of elastic profile. Deformation of materials on the transfer of the profile has a
complex character depending on where the usual degree of deformation does not
include all influential factors in particular the change of wall thickness profile. For
accurate analysis can be applied depending on changes in the thickness of the wall
profile obtained by experimental analysis. In the zone of bending appear high
expression of stress which causes large uneven deformation, their non-monotonicity.
Studying local deformations, their inconsistency, it is necessary to take into account
the structural properties of workpiece materials. So, it is necessary to experimentally
determine the character of the deformation of the section. The relations for the
calculation of the degree of deformation of the profile section should provide
mechanical properties profiles and their standardization in engineering practice.
Key words: profiling, rollers, complex profiles, relative wall thickness
1
INTRODUCTION
In the process of profiling the correlation surface of mechanical strain [1] of the
following factors are: the relative width of the shelf profile B/so and secondary bending
angle Dsr. It tends to avoid changes in the conditions of the process of profiling, through
all the experiments is avoided by the change in the relative gap z/so among the rolls,
thus was excluded error results. Used were rollers with varying radius curves in scale
and from 1 mm to 9 mm, without changing their clearance.
2.
INFLUENTIAL FAKTORINA CHANGES IN THE THICKNESS
General deformation bands in flat roller profiling consists of strain due to

bending of its own bar and cross extension and thinning of the wall thickness. With the
participation of thinning of wall thickness profiles, or cross extension, is linked not only
on determination of the width of the strip but also on determination of surface of

1
2
Dr Ostoja Miletic, Banjaluka, Faculty of Mechanical Engineering, (miletic@urc.rs.ba)

Dr Mladen Todic, Banjaluka, Faculty of Mechanical Engineering, (mladentodic@unibl.rs)
271
Miletic Ostoja, Todic Mladen
mechanical deformation of the section. Based on the experimental plan and results
(Table 1).
Table 1 Plan experiments and results
Line
On
1
2
3
4
5
6
7
8
9
10
11
12
Plan the experiment

r/to
b/to
D
4.3
10
30
0.3
20
30
2.3
15
60
0.3
10
90
4.3
20
90
2.3
15
60
2.3
15
60
0.3
20
90
0.3
10
30
4.3
20
30
2.3
15
60
4.3
10
90
xo
1
1
1
1
1
1
1
1
1
1
1
1
Part No
x1
x2
1
-1
-1
1
0
0
-1
-1
1
1
0
0
0
0
-1
1
-1
-1
1
1
0
0
1
-1
x3
-1
-1
0
1
1
0
0
1
-1
-1
0
1
Results
ln's/so
's/so
0.01
-4.6052
0.05
-2.3358
0.03
-3.5066
0.07
-2.6593
0.02
-3.9121
0.03
-3.5066
0.03
-3.5066
0.09
-2.4080
0.05
-2.9958
0.01
-4.6052
0.035
-3.3494
0.02
-3.9121
Obtained the dependence of relative changes of thickness profile in the

profiling of the parameters in the form of profiling,
's
so
0,032D
0 , 5256
r

so
b

so
0 , 5635
0 , 0906
(1)
where:
D - angle of bending of the regime profiling
b - width of the strip, piece
r bending radius
so band thickness, piece
Based on this correlation has been established by which the nomogram the
regime profiling can determine the maximum value of relative wall thinning profile, it is
necessary in determining width pripremka and the design and construction of the
rollers, Figure 1.
272
Changing the Wall Thickness Profile in The Process of Profiling
Fig. 1 Nomogram for determining the maximum changes in the thickness of the wall
profile
Clearance between the rollers z/so=1 is equal to the thickness of the wall piece.
However, this clearance is not the same in transverse sections of the complex
configuration of axial rolls, where is the contact surface of the tool and workpiece
material complex and different directions of that surface what causes further stressdeformation state.
273
Miletic Ostoja, Todic Mladen
3.
CONCLUSION
If during the construction of calibrated rolls, without taking into account the
complexity of the configuration profile, there is increasing lateral deformations of that
profile leads to different strains of certain elements of the profile. The consequences of
extreme strain differences in the profile elements affect and on correct filing geometry
of profile dimensions profile on its quality. Transversally extension, or change of wall
thickness profile, reduces the efficient production of closed-caliber and more difficult to
define width piece.
Fig. 2 Directions of the forces and form of contact surface of workpiece materials and
tools
LITERATURE
[1]
[2]
[3]
[4]
274
Miletic O. (2002). Konstitutivne relacije procesa neprekidnog profilisanja

kompozitne ljuske i ploce, Univerzitet u Banjoj Luci Masinski fakultet.
Miletic O, Todic M. Uticaj triboloskih efekata kontaktnih povrsina na fenomen
ojacavanja, Procedings 2008, p.179-183.
Todic M. (2006), Position of neutral surface stress and strain in composite
laminates, PhD thesis, Faculty of Mechanical Engineering Banja Luka.
Todic M. Miletic O, Maric D. Tvrdoca u funkciji naponsko-deformacionog stanja
dvoslojnog kompozita, DEMI 2009, p. 189-195.
COMPARATIVE INVESTIGATION OF DIFFERENT LUBRICANTS

FOR BULK METAL FORMING OPERATIONS
Plavka Skakun1, Miroslav Planak2, Dragia Viloti3, Mladomir Milutinovi4, Dejan
Movrin5, Ognjan Luanin6
Summary: Friction between die and workpiece material has influence in almost every
metal forming process. Reduction of friction leads to reduction of required forming load
of certain process. It also influences in positive way die wear (it reduces it) and quality
of the final product. That is the reason why research of different kind of lubricants is
very important and has great impact on metal forming processes. Conventionally, in
bulk cold metal forming processes, coating of zinc phosphate is applied. But zinc
phosphate coating is not environmental friendly, due to sludge accumulation in baths
and its associated content of heavy metals. That is the reason why new,
environmentally harmless lubricants have been developed.
In this paper, determination of friction coefficient for different lubricants is
presented. Both environmentally harmless and environmentally hazardous lubricants
were used. The experimental method for friction coefficient determination was ring
compression test. Results of the test were used to compare values of friction
coefficient for different type of lubricants. Four different types of lubrication were
applied.
Key words: friction, metal formign processes, environmetal friendly lubricants
1. INTRODUCTION
The goal of lubrication is to reduce friction, lower the deformation forces and
prolong tool life. Lubrication system in bulk cold forming processes must have high
performances because of the large surface expansion of workpiece and high normal
pressure between workpiece and tool. It is also important to avoid breakdown of
lubricant film and metal to metal contact between tool and workpiece. For that reason,
efficient lubrication system is required to ensure quality production.
Mr Plavka Skakun, Novi Sad, Faculty of Technical Sciences, (plavkas@uns.ac.rs)

Dr Miroslav Planak, Novi Sad, Faculty of Technical Sciences, (plancak@uns.ac.rs)
3
Dr Dragia Viloti, Novi Sad, Faculty of Technical Sciences, (vilotic@uns.ac.rs)
4
Mr Mladomir Milutinovi, Novi Sad, Faculty of Technical Sciences, (mladomil@uns.ac.rs)
5
Dipl. In. Dejan Movrin, Novi Sad, Faculty of Technical Sciences, (movrin@uns.ac.rs)
6
Dr Ognjan Luanin, Novi Sad, Faculty of Technical Sciences, (luzanin@uns.ac.rs)
2
275
P. Skakun, M. Planak, D. Viloti, M. Milutinovi, D. Movrin, O. Luanin
A common and very widespread way of lubrication in cold metal forming is use
of phospate coatings. Phosphate coatings are usually used in combination with some
additional lubricant, and this way of lubrication for a long period of time had been the
best solution for extrusion processes. However, from environmental point of view,
phospate coatings need to be replaced because of content of heavy metals in baths
and large amounts of waste waters. That is the reason why new ecological friendly
lubricants have been developed.
According to [1] environmental problems in metal forming tribology can be
divided in following areas: a) health and safety of people, b) influence on equipment
and buildings and c) destruction and/or disposal of waste and remaining products.
Improvement efforts should be concentrated on elimination of hazardous chemicals
and reduction of waste, including prolonging tool and lubricant life, recovery and reuse
of lubricants and minimal quantity lubrication.
In this paper a part of results of comparative investigation of different lubricants
for bulk metal forming operations is presented. Both conventional and new lubricants
were involved in investigation. Main goal was to show that new, ecology friendly
lubricant can reduce friction in bulk forming process in a same way as conventional
one.
2. CONVENTIONAL TRIBOLOGICAL SYSTEMS IN COLD FORMING
As it is mentioned before, conventional way of lubrication in cold extrusion
processes is phosphate coating in combination with some aditional lubricant. It is well
known that wide application of cold extrusion started when use of phosphate coatings
was developed.
Phosphate coatings contains havy metals, usually zinc. There are many
reasons why zinc phosphate coatings were widely applied, but two main reasons are:
between zinc phospate coating and subsequent lubricant reaction takes
place. For that reason zinc phosphate coating have much better lubrication
characteristics under high pressure
the crystal structure of coating posseses plastic properties under the
conditions of temperature, pressure and shear stress which exists in cold
forming operations.
Phases of zinc phosphate forming are: alkali degrease, rinse, acid pickle
(sulphuric or hydrochloric acid), rinse (cold water), rinse (hot water), application of zinc
phosphate solution, rinse (cold water), rinse (hot water), lubrication, drying.
This procedure, from environmental point of view, has several disadvantages
[1]: a) sludge of heavy metal phosphates b) large water requirement in the rinse baths
c) periodic replacement of baths for degreasing, neutralizing, pickling and lubrication is
required and d) large amounts of waste water, containing grease, acid and soap.
Disadvantages of conventional lubrication in cold bulk metal forming had
initiated developement and implementation of new tribological systems.
3. NEW TRYBOLOGICAL SYSTEMS IN COLD METAL FORMING
A number of lubrication systems for cold forming have been developed as an
alternative to phosphate coatings and soap. According to [1] they may be organized
276
Comparative investigation of different lubricants for bulk metal forming operations
into the following groups:

New conversion coatings (electrolytic phosphate coating, microporous
coating)
Lubrication without conversion coating (dual bath systems, single bath
systems)
3.1 New conversion coatings
Many disadvantages related to the process of forming phosphate coatings are
reduced by electrolytic phosphating. The main advantages of this process are: use of
acid for pickling may be avoided by electrochemical pickling, phosphating bath is
sludge free, the treatment time is shorter, the working environment is improved and
electrochemical procedure makes possible to phosphate high alloyed steel and
stainless steels. Electrolytic phosphating ensures much more uniform and finer
crystalline coating with smaller film thickness.
There are different techniques for forming microporous coating. One of them is
described in paper [2]. Surfaces with pores of micrometer size for entrapping lubricant
are generated by electrochemical deposition of an alloy, consisting of two immiscible
metals, of which one metal is subsequently etched away leaving 5 m layers with a
sponge-like structure. When the lubricant is subsequently applied to the porous
coating, it is trapped in the pores acting as numerous small lubricant reservoirs.
3.2 Lubrication without conversion coating dual and single bath systems
The dual bath systems form two layer coating: a base coating adhering to the
metal surface and an overcoating to further reduce friction. Nakamura at al. [8] have
tested a number of alternatives, among them a dual bath system too.
Several single bath systems have also been developed. One of them has been
developed in Nihon Parkerizing as a water based lubricant [5]. This product consist of
an inorganic salt as the base component and a wax as the lubricant. The application
method is called dry in place and consists of a simple dip and dry process forming a
double coating consisting of a lubricant carrier as the base with a lubricant film on top.
Similar lubricant, Bonderlube FL 741, was used in experiment which is
described in this paper. Schematic view of a single bath system application is
presented in Fig. 1.
4. EXPERIMENTAL INVESTIGATION
For the experimental prediction of friction conditions ring compression was
employed.
4.1 Ring compression test
The coefficient of friction is generally determined experimentaly. One test that
has gained wide acceptance, particularly for bulk deformation processes, is the ring
comperssion test [7]. The idea for this test was given by Kunogi [3], later improved by
Mail and CockCroft [4], Avitzur [6] and many other researches. Basic technique of this
test relates dimensional changes of a ring specimen to the coefficient of friction. When
a ring specimen is compressed between two flat dies, high friction coefficint results in
277
inward flow of the material, while low friction coefficient results in an outward flow of
the material. To determine friction coefficient it is necessary to measure change in
height and in inner diameter, to calculate percentage value of nominal strains h i d
and to compare h -d curve to calibration curves. In this research calibration curves,
calculated acording Avitzurs equations [6], were used for friction coeficient
determination.
Wax and Metallic soap
Water evaporation
Inorganic compound
2 layers
Metal
Metal
Metal
Coating
Drying
Film
Fig. 1 Schematic view of single bath system application [9]

4.2 Determination of friction coefficient
Friction coefficient was determined for four different types of lubricants:
oil for cold forging
phosphated surface with MoS2
phospated surface with MoS2 dissolved in oil and
Bonderlube FL 741 lubricant.
Material of rings was 4320.
The most common geometry of a ring has proportion of outer diameter to inner
diameter to height 6:3:2, so this proportion was used in current experimental
investigation too. Dimensions of rings were: outer diameter D = 18 mm, inner diameter
d = 9mm and hight h = 6 mm. Experiment was conducted on Sack&Kieselbach
hydraulic press, nominal force 6.3 MN, using flat dies.
Rings before compression are presented in Fig.2, and after compression in
Fig.3. Frist ring was lubricated with oil, secon ring had phosphated coating (for this ring
two types of lubricants were used) and last ring had Bonderlube coating.
For each lubricating conditions the experiment was repeated three times. In
Fig. 4 solid lines present friction calibration curves for three different coefficient of
friction (0.12, 0.1, 0.08). Four other curves are experimentaly obtained functions d.=
f(h) for four different tribolgical conditions. As it can be seen from the diagram, low
friction coefficient is achieved with application of phosphated coating and MoS2
disolved in oil and with application of Bonderlube coating. Preferences are on the side
of the Bonderlube because of positive environmental and economical effects (less
process steps, reduction of waste water, no phosphate sludge).
278
Comparative investigation of different lubricants for bulk metal forming operations
a)
b)
c)
Fig. 2 Rings before compression,(a) without coating,(b) with phosphate coating,(c) with
Bonderlube coating
a)
b)
c)
Fig 3. Rings after compression , (a) without coating, (b)with phosphate coating, (c)with
Bonderlube coating
0.35
0.3
0.25
=0.12
0.2
=0.1
=0.08
0.15
phosphat. +MoS2
Bonderlube
0.1
oil
0.05
phosphat.+MoS2+oil
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
-0.05
-0.1
Fig. 4 Calibration and experimental curves
279
5. CONCLUSION
Friction coefficients for four different types of lubricants were determined. Ring
compression test was used for determination of friction coefficient. Three convencional
lubricants were used (oil for cold forging, phosphate coating with MoS2, phosphate
coating with MoS2 disolved in oil) and one environmental friendly lubricant , Bonderlube
FL 741. Comparing with conventional ways of lubrication there are several advantages
in both economical and environmental way, when Bonderlube FL 741 lubricant is used.
Experimental results showed that value of friction coefficient for Bonderlube lubricant is
low (0.08), compared to conventional lubricants (0.08 0.12).
ACKNOWLEDGEMENT
This paper is a part of the investigation within the project EUREKA E!5005
financed by Serbian Ministry of Science and Technological Development. Authors are
very grateful for the financial support.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
280
Bay, N., Azushima, A., Groche, P., Ishibashi, I.,Merklein, M., Morishita, M. (2010).
Environmentally Benign Tribo-systems for Metal Forming. CIRP Annals
Manufacturing Technology, vol. 59, no. 2, p. 760-780.
Arentoft, M., Bay, N., Tang, P.T., Jensen, J.D. (2009). A New Lubricant Carrier for
Metal Forming. CIRP Annals Manufacturing Technology, vol. 58, no.1, p.243246
Kunogi, M. (1956). A new method of cold extrusion. Journal of Science Research
Institute, vol. 50, p.215
Male, A.T., Cockcroft, M.G. (1964/65). A Method for the Determination of the
Coefficient of Friction of Metals under Conditions of Bulk Plastic Deformation.
Journal of the Institute of Metals, vol.93, p.38-46
Yoshida, M., Imai., Y., Yamaguchi, H., Nagata, S. (2003). Nihon Parkerizing,
Avitzur, B. (1968). Metal Forming:Processes and Analysis. Mc Graw Hill Book
Company, USA.
Kalpakjian, S., Schmid, S.R. (2008). Manufacturing Processes for Engineering
Materials.Pearson Prentice Hall, Singapore
Nakamura, T.,Sumioka, Y., Sagisaka, Y., Ishibashi I., Sekizava, M. (2008).
st
Lubrication Performance of Environmentally Friendly Lubricants for Forging. 1
th
Report Proceedings of 59 Japanese Joint Conference for the Technology of
Plasticity, p.579-580
Bay, N., Nakamura, T., Schmid, S. (2010). Green Lubricants for Metal Forming.
Proceedings of 4th International Conference on Tribology in Manufacturing
Process 2010, p.5-28.
COMPLEXITY ANALYSIS OF CAD/CAM SYSTEMS INTEGRATION IN

ERP BUSINESS ENVIRONMENT
Borislav Kovljeni1
Summary: Traditionally separated development of CAD/CAM and ERP systems
resulted that, in current attempt to full digital integration in the company, we have a
complex problem of interconnection. Significantly increased the amount of data in
defining the modern products, and the scope and speed of information exchange in the
dynamic business environment, further complicate process of efficient enterprise
integration. This paper presents a brief analysis of the basic integration problems,
possible approaches to solving them, and author's research and development activities
in domain of the observed problems.
Key words: CAD/CAM, ERP, Enterprise Integration
1.
INTRODUCTION
The application of modern computer tools in all business segments was

significantly improved and accelerated activities of design and production, and
generally, all activities in the product lifecycle management (PLM). On the other hand,
trough the current trends, like customisation, intelligent, ecological, ergonomic and
aesthetically attractive products, these technologies enable the management of large
amounts of information and data that are generated in the production and business
processes. Therefore, problem of information flow at the level of the entire business
system is actual more than ever, [2, 3]. Traditionally concentrated on solving partial
problems, manufacturers of various application software solutions today try, each in
their field, offer a wider integrated information platform, usually through the integration
and/or development of various additional software modules, Fig. 1. Among these
applications, CAD/CAM systems are particularly important, as generators of the
majority of processed data in the product lifecycle. Hence, their integration into wider
business environment is the basis for the integration of other tools. In the absence of
standardization and wider analytical approach to information flows within the
production and business processes, software solutions available today, both in the field
of CAD/CAM systems and in the ERP domain, become more complex. It's normally
resulting that their effective integration becomes more complicated. Separated
software automation of various business activities resulted in current situation where
1
Mr. Borislav Kovljeni, dipl. ma. ing., PhD student, Mainski fakultet u Beogradu, Kraljice Marije 16, 11120
Beograd 35, e-mail: bkovljenic@gmail.com
281
Borislav Kovljeni
companies need to use different applications which duplicate large percentage of data
and functions. In larger systems, duplication of data normally leads to the inefficiency
and unsustainable of their management. Analysis of complexity is an important step in
building process of successful IT (Information Technology) integration.
EDM - Engineering Data

Management
PDM - Product Data
Management
ERP - Enterprise Resource
Planning
PPS - Project, Planning and
Control System
PLM Product Lifecycle
Management...
Fig. 1 Trend of contemporary increasing complexity of informations content and

flow
2.
BASIC PROBLEMS OF INTEGRATION
Because of the mentioned separate development, and in principle different

types of data processed in CAD/CAM and ERP systems, nowadays we haven't
satisfactory solutions for their integration. From a wider principal and technological
problems, author extracts following basic integration problems:
Lack of standardization in the field of digital product description, both in the
constructive-technological and commercially-profit business sectors.
Many current heterogeneous applications as a result of the traditional partial
approach to programming automation of business activities, all in the
absence of mentioned standardization.
Commercialisation of the software products, concerned with commercial
market game between software developers.
The complexity of modern products and different models and formats for
data storing.
Increasing of complexity requires significant material and human resources.
Concentration of research and development activities in the commercial
sector ignores the wider interests.
Reduced competition due to globalisation in the field of software
developers...
An important characteristic of modern software solutions for digital integration
at the enterprise level is their difficult implementation in small and medium enterprises.
Complexity and high prices are main reasons. Among others, this is consequence of
their first appearance in the large production systems such as automotive and
aerospace industries.
282
Complexity analysis of CAD/CAM systems integration in ERP bussines environment
3.
APPROACHES TO PROBLEM SOLVING
Generally, in order to reduce the complexity of the observed integration, author

as possible solution sees a redefinition of existing concepts of CAD/CAM and ERP
systems in terms of standardization of digital product description, standardization of
business process activities, and implementation of models and technologies oriented
to open and widely applicable platforms like Web-based systems, XML and SQL
languages, etc.. In this sense, to be a perspective, future CAD/CAM platforms must
provide solutions for an open systems interconnection in order to enable easy
integration in wider business environment. Examples of such solutions and activities
are implementation of the STEP standardization [4], or providing programming
interface with CAD/CAM such as Web.Link, J-Link or Protoolkit from PTC [5]. Maybe,
some international projects like that in the field of Digital Factory [1] will contribute to
wider mutual cooperation between software developers.
Each company, depending on it's own needs and costs, individually solves
problem of digital data integration in order to achieve effective realization of business
processes. As general guidelines for proper selection of CAD/CAM and ERP solutions
are observed the following:
Choice of platform and software adequate to the specific production system.
Strategic approach to information organization of the company.
Development of own IT resources.
Maximum use of modern and standardized solutions wherever is possible.
Minimize the information content according principle - one data in one place.
Decoupled partial software solutions in terms of their functional
independence.
On the other hand, manufacturers of CAD/CAM and ERP systems continue to
market competition by developing separate commercial solutions, not contributing to a
significant increase of their integration. There are examples that many firms based
their business on complexity of observed integration. Before implementation, potential
users must have this in mind and analyse the impact of this tools on other business
processes.
4.
CURRENT RESEARCH
Through working on doctoral dissertation and practical projects author

investigates new platform of integration that would be independent of individual
CAD/CAM and ERP software solutions. The data model is based on a single database
for all data used by a variety of business activities. In that sense, redefining the existing
model for data storing in CAD/CAM systems becomes inevitable, primarily because of
the complexity and scope of information flows. Specifically, the implementation of the
above mentioned principles for minimizing of information content and decoupling
partial solutions, should enable minimum possible system complexity and thus the
efficient CAD/CAM integration in ERP environment.
Database and database management systems are technologically perfected
and proven concept that successfully handles large volumes of data. All modern digital
integrated systems for managing large data amounts use this concept. Database
concept is seems as only solution for handling an increasing amount of data in modern
CAD/CAM systems. It is easy to implement mentioned principles for minimizing
283
Borislav Kovljeni
information content, as a necessary prerequisite for reducing the complexity of the

observed system and increasing the efficiency of integration. It is most popular concept
for easy data access and probably new paradigm for future CAD/CAM systems, Fig. 2.
(Digital Factory)
CAD/CAM
ERP
EDM
Standardized
description of
products
Standardized
descriptions of
busines processes
Unified database
PLM
PPS
...
Fig. 2 Integration model to minimize complexity (amount of data)

5.
CONCLUSIONS
Modern business conditions require wide application of computer support and

efficient digital contents integration and sharing. In such circumstances, companies
that require computer support tools for production and business activities are limited on
partial commercial programs available on the market and their subsequent internal
integration. Thereby, very few of them are able to independently develop solutions for
meet specific requirements and increase performance of digital business integration. In
production-business systems one of the most important and complicated tasks is
integration of CAD/CAM tools in the wider business environment. Viewed from the
aspect of the future CAD/CAM platforms, seems that standardization of digital
description for product model and business processes and implementation of wider
used IT concepts are necessary prerequisites for better integration. The current
requirement for fast processing of large amounts of data, consequently requires
incorporation some principles of integration in the initial concept development of
CAD/CAM systems.
REFERENCES
[1]
[2]
[3]
[4]
[5]
284
E. Westkmper, Factories of the Future beyond 2013, A view from Research: The
role of ICT, Fraunhofer-Institut fr Produktionstechnik Und Automatisierung (IPA),
Stuttgart, Germany
PLM (Product Lifecycle Management) Market: Key Research Findings 2010,
Yano Research Institute Ltd., 2-46-2 Honcho, Nakano-Ku, Tokyo 164-8620,
Japan
SeokRyul Kim, Daniel Weissmann, Middleware-based Integration of Multiple CAD
and PDM Systems into Virtual Reality Environment, Computer-Aided Design &
Applications, Vol. 3, No. 5, 2006, pp.: 547-556.
http://ng.tc184-sc4.org
http://www.ptc.com/
COMPUTER CONTROLLED EXPERIMENTAL DEVICE FOR

INVESTIGATIONS OF TRIBOLOGICAL INFLUENCES IN SHEET
METAL FORMING
S. Aleksandrovi1, T. Vujinovi2, M. Stefanovi3, V. Lazi4, D. Adamovi5
Summary: Sheet metal forming, especially deep drawing process, is influenced by
many factors. Blank holding force and drawbead displacement are two of them that
can be controlled during the forming process.
For this purpose, electro-hydraulic computerized sheet-metal strip sliding device has
been constructed. Basic characteristic of this device is realization of variable contact
pressure and drawbead height as functions of time or stripe displacement. There are
both, pressure and drawbead, ten linear and nonlinear functions. Additional features
consist of the ability to measure drawing force, contact pressure, drawbead
displacement etc.
Presented in the paper are the device overview and the first results of steel sheet
stripe sliding tribological physical model.
Key words: deep drawing, drawbead, variable contact conditions, tribology
1. INTRODUCTION
Technology of deep drawing of thin sheet metals is extremely important in
modern industry. Due to the development of new materials of more complex formability
and raising of the technological requirements to the higher level, the need for
realisation of complete control of forming process increases. In order to succeed in
that, it is necessary to identify, out of a large number of influential factors, the ones
which can be influenced throughout the forming process. It was determined that there
are only two such factors: contact pressure on flange and drawbead height [1].
Process control through active complex (closed-loop) systems requires
constant dynamic feedback between the given function of the objective, controlled and
controlling variables [2]. The functions of the objective and controlled variable can be
different: wrinkle height, thinning in critical zone, flange motion, flange thickness
change, friction force, forming force, stress in work piece wall etc. The given objective
1
Dr S. Aleksandrovi, red. prof., Faculty of Mechanical Engineering, Kragujevac, Serbia, srba@kg.ac.rs

Mr T. Vujinovi, dipl. ing., FAM Jelingrad, Banja Luka, RS, BiH
3
Dr M. Stefanovi, red. prof., Faculty of Mechanical Engineering, Kragujevac, Serbia, stefan@kg.ac.rs
4
Dr V. Lazi, red. prof., Faculty of Mechanical Engineering, Kragujevac, Serbia, vlazic@kg.ac.rs
5
Dr D. Adamovi, vanr. prof., Faculty of Mechanical Engineering, Kragujevac, Serbia, adam@kg.ac.rs
2
285
S. Aleksandrovi, T. Vujinovi, M. Stefanovi, V. Lazi, D. Adamovi
functions are defined either by computer simulations or by previous experiments.

Pressure on flange and drawbead height present the controlling effects. High velocity
of reacting to controlled values change and robust controlling hardware and software
apparatus are required, which all implies significant investments [2, 3, 4].
There is also the alternative a much simpler approach used in this paper.
However, first it is necessary to define optimal functions of pressure and drawbead
height according to proper criterion (drawing depth, piece quality etc.). This often
requires comprehensive experiments [3, 4] in order to identify the character of
specified factors influence. With such information, it is possible to form the controlling
apparatus for practical application whose main objective is to realise previously defined
optimal functions of pressure and drawbead height. Such equipment requires
considerably smaller investments regarding hardware and software and is far more
accessible to a wide range of users.
Application of constant height drawbeads is still most often applied and well
known [5, 6]. The same goes for application of constant blank holding force on flange.
The main reasons for this are smaller forming process costs. However, due to the
development of new materials of more complex formability properties, in most cases it
is not possible to accomplish the satisfactory results by classical methods.
There are also some new ideas, such as application of drawbeads in which the
o
angle between drawbead axis and sheet metal plane is different from 90 [7]. There is
also the increased interest in many numeric simulations and virtual application of
drawbeads in processes of complex work pieces forming [8].
The application of blank holding force without draw beads is the subject of
separate researches based on the same aforementioned principles [9].
In this paper, the emphasis is on the display of properties of apparatus for
investigation of the character of the connection between drawing force and various
influences combination, installed at the Faculty of Mechanical Engineering in
Kragujevac. The properties include friction conditions (dry, application of lubricant),
drawbead geometry (two rounding radii), one variable function of pressure, variable
functions of drawbead height and corresponding constant values of both pressure and
drawbead height.
2. EXPERIMENTAL APPARATUS
2.1 THE DEVICE
General block scheme of the apparatus is shown in fig. 1, central part
assembley in fig. 2, and physical appearance in fig. 3.
Sheet metal stripe is positioned vertically between contact pairs, drawbead
and die, which are variable, fig. 3 and fig. 5. In fig. 3, the positions are as follows: 1right cylinder 70x40 mm, 2-bolt M8x80, 3-tube 1, 4-die nut, 5-bolt M8x30, 6-plate 2, 7die guide, 8-stud bolt 12x55 mm, 9-lower plate, 10-upper plate, 11-die, 12-drawbead,
13-bolt M6x45, 14-nut M6, 15-washer A6,4, 16-lateral guide, 17-plate 1, 18-guide, 19stud bolt 12x45 mm, 20-tube 2, 21-drawbead nut, 22-bolt M8x100 mm.
The needed stripe width is 30 mm, and recommended length is 250 mm.
Drawbead and die are variable, which enables monitoring of the influence of various
rounding radii.
286
Computer Controlled Experimental Device for Investigations of Tribological Influences in Sheet

Metal Forming
Fig. 1 Block scheme of experimental apparatus
Fig. 2 Assembley of main device part

287
Fig. 3 Physical appearance of experimental apparatus

Drawing force is obtained from laboratory press ERICHSEN 142/12 in range 020 kN, as well as stress signal for measuring the force of proper sensor. Hydrocylinders for drawbead motion and pressure realization are fed by aggregate
ERICHSEN of nominal pressure 100 bars and flow 1,5 l/s. The oil from the aggregate
runs through the series of controllable proportional hydro valves to both cylinders.
Measuring and pressure controlling branch (fig.1) consists of pressure sensor
which gives the current true value signal and control unit (micro-controller) which
receives the given desired value from the software and sends signal D/A to the
convertor. The received analogous signal is transmitted to the control card of the
proper hydro-valve connected to the pressure cylinder.
In controlling branch, due to drawbead motion, the current true drawbead
position is read by rotation encoder. After processing, the signals are sent to the
control unit (micro-controller), and then to the card for control of hydro-valve for
drawbead cylinder. One signal is related to the direction change, and the other one to
the value of drawbead motion function. For measuring and reading the true drawbead
position, supporting branch with inductive sensor and proper amplifier is made.
All true values signals are brought into PC computer with integrated A/D card
and proper original software, which enables monitoring of all values, their memorizing,
presentation as well as generating of pressure and drawbead motion functions
necessary for micro-controller performance.
2.2 PREVIOUSLY GIVEN AND REALLY REALISED FUNCTION OF PRESSURE
AND DRAWBEAD HEIGHT
For the needs of planned comprehensive experiment, 6 variable dependencies
of both pressure and drawbead motions on time, as given functions, were defined. In
figures 4 to 6, those functions are marked with numbers 1 to 6. Dependencies 5 and 6
are linear, and 1, 2, 3 and 4 non-linear. Functions were defined based on empiric
values of minimal and maximal pressure (0-20 MPa) and drawbead height (0-8 mm).
288
Computer Controlled Experimental Device for Investigations of Tribological Influences in Sheet

Metal Forming
R1: h = t2/8100 - t/30 + 8
R2: h = -t2/8100 + 2t/30
10
theoretical
experimental
R1
R2
6
4
2
0
20
40
60
80
12
Drawbead displacement h, mm
12
10
theoretical
experimental
R4
8
6
4
2
0
100 120 140 160 180 200
R3: h = 2t2/2025 - 8t/45 + 8

R4: h = -2t2/2025 + 8t/45
R3
0
20
40
60
80
Time t, s
100 120 140 160 180 200
Time t, s
Fig. 4 Previously defined and really realised dependences of drawbead height (R) on
time
25
R5: h = 8 - 2t/45
R6: h = 2t/45
10
theoretical
experimental
R5
R6
6
4
P2
15
10
P1
2
0
P1: p = 2t2/8100 - 14t/90 + 20

P2: p = -t2/4050 + 63t/405
20
Pressure p, MPa
12
20
40
60
80
100 120 140 160 180 200
20
40
60
80
Time t, s
100 120 140 160 180 200
Time t, s
Fig. 5 Previously defined and really realised dependences of drawbead height (R)and
contact pressure (P) on time
25
P3
15
10
5
0
P5: p = 20 - t/9
P6: p = t/9
20
Pressure p, MPa
Pressure p, MPa
20
25
P3: p = -t2/405 + 4t/9

P4: p = t2/450 - 2t/5 + 20
P5
P6
15
10
5
P4
0
20
40
60
80
100 120 140 160 180 200
Time t, s
20
40
60
80
100 120 140 160 180 200
Time t, s
Fig. 6 Previously defined and really realised dependences of contact pressure (P) on
time
289
3. CONCLUSION
Based on the concise presentation of the developed apparatus and the
preliminary results, it can be concluded that specified computerised device for testing
various influences on process of stripe sliding over drawbead enables accurate
registering of the influence of variable pressure, variable drawbead height, drawbead
geometry and friction conditions on drawing force.
Based on preliminary results obtained by investigating steel sheet metal
DC04 on the basis of character of drawing force response, it is possible to make the
following conclusions:
a) due to favourable combination of simultaneous performance of contact
pressure change, change of drawbead height, drawbead geometry and friction
conditions, it is possible to influence precisely the course of sheet metal forming
process according to the desired forming force criterion,
b) by such investigations, with rather simple apparatus, it is possible to define
significant data for numerical simulations and immediate application in practice at deep
drawing of complex geometry parts.
LITERATURE
[1] S. Wagner (1998). Tribology in drawing car body parts. 11th International
colloquium: Industrial and automotive lubrication, Technische Akademie Esslingen,
Proceedings Vol. III, p. 2365-2372.
[2] M. Liewald (2008). Current Trends in Research on Sheet Metal Forming at the
Institute for Metal Forming Technology (IFU) at the University Stuttgart. Papers of the
International Conference on New Developments in Sheet Metal Forming, IFU
Stuttgart, p. 263-288.
[3] C. Blaich, M. Liewald (2008). New Approach for Closed-Loop Control of Deep
Drawing Processes. Papers of the International Conference on New Developments in
Sheet Metal Forming, IFU Stuttgart, p. 363-384.
[4] J. R. Michler, K. J. Weinmann, A. R. Kashani, S. A. Majlessi (1994). A strip-drawing
simulator with computer-controlled drawbead penetration and blankholder pressure,
Journal of Materials Processing Technology, no. 43, p. 177-194.
[5] J. A. Waller (1978). Press Tools and Presswork, Portcullis Press Ltd, Great Britain.
[6] M. Stefanovic (1994). Tribology of deep drawing, monograph, Yugoslav Society for
Tribology and Faculty of Mechanical Engineering, Kragujevac, (In Serbian).
[7] L. M. Smith, Y. J. Zhou, D. J. Zhou, C. Du, C. Wanintrudal (2009). A new
experimental test aparatus for angle binder draw bead simulations, Journal of Materials
Processing Technology, no. 209, p. 4942-4948.
[8] G. H. Bae, J. H. Song, H. Huh, S. H. Kim, S. H. Park (2007). Simulation-based
prediction model of the draw-bead restraining force and its application to sheet metal
forming process, Journal of Materials Processing Technology, no. 187, p. 123-127.
[9] S. Aleksandrovic (2005). Blank holding force and deep drawing process control,
monograph, Faculty of Mechanical Engineering, Kragujevac, (In Serbian).
290
DEPOSITION OF TITANIUM NITRIDE LAYERS BY REACTIVE

PLASMA SPRAYING
Radu Alexandru Rou1, Viorel Aurel erban2, Mihaela Popescu3, Uu Drago4
Cosmin Locovei5
Summary: Cutting tools are subjected to the intense wear solicitations. Usually, they
either are subject to superficial heat treatments or are covered with various materials
with high mechanical properties. In recent years, thermal spraying is used increasingly
in engineering area because of the large range of materials that can be used for the
coatings. Titanium nitride is a ceramic material with high hardness which is used to
cover the cutting tools (drills, mills), increasing their lifetime. The paper presents the
results obtained after deposition of titanium nitride layers by reactive plasma spraying
(RPS). As deposition material was used titanium powder and as substratum was used
titanium alloy (Ti6Al4V). Following the demonstration program have been achieved
layers deposited with a thickness between 68,5 and 81,4 m. Also, are presented the
macroscopic and microscopic (scanning electron microscopy) images of the deposited
layers.
Key words: thermal spraying, titanium nitride, surface engineering, deposited layers
1.
INTRODUCTION
Surface engineering and coating technology include those processes which

modify the surface material to improve their behavior, performance, life time and the
economic production. This means the modification of the component surface to confer
different properties from those of the base materials (substrates) [1]. Life time and
performance of basic mechanical components (shafts, gears and other components)
are conditioned largely by their geometry and metallurgy, but there are outstanding
opportunities to optimize the performance of these components by modifying their
surface. At the parts subject to wear is possible to reduce friction and prolong the life
time of the equipment by covering with special materials [2].
Dr. Eng. Radu Alexandru Rosu, Politehnica University of Timisoara, e-mail: rrosu@isim.ro
Prof. Dr. Eng. Viorel Aurel erban, Politehnica University of Timisoara, serban@mec.upt.ro
3
Prof.assoc. Mihaela Popescu, Politehnica University of Timisoara, e-mail: hela.popescu@yahoo.com
4
Lecturer. Dr. Eng. Uu Drago Politehnica University of Timisoara, e-mail: dutu@eng.upt.ro
5
Assist. Dr. Eng. Cosmin Locovei Politehnica University of Timisoara, e-mail: clocovei@mec.upt.ro
2
291
Radu Alexandru Rou, Viorel Aurel erban, Mihaela Popescu, Uu Drago, Cosmin Locovei
Thermal spraying knows a increased applicability due to the numerous

advantages. Currently, there are industries where using thermal coating technologies
is essential to obtain characteristics required by the users. Among the areas most
advanced components for use in coated material is noted in particular the aerospace
and automotive industry [3]. Generally, the ceramic materials can not be used as wires
or bars, their use in powder form does not constitute a disadvantage in thermal
spraying because the materials used often in this process are powders. On the other
hand, using powder ceramic materials their chemical composition can be well
diversified, which led to new materials with special properties.
Titanium alloys are increasingly used in many industries (aerospace,
aeronautical, medical) due to good corrosion resistance and low specific density than
other alloys. The disadvantage of titanium alloys is that they have a low wear
resistance. Therefore, their surface is covered with various materials to increase the
wear resistance [4]. Titanium nitride (TiN) is a ceramic material which has high
hardness and is often used to cover titanium alloys, steel, aluminum substrates to
increase the surface characteristics of the components.
The most common applications are titanium nitride coatings is to increase
wear and corrosion resistance of the cutting tools (drills, mills) so increasing their
lifetime more than three times [5, 6].
The paper presents the experimental results after the deposition of titanium
nitride layers using titanium powder by reactive thermal spraying method.
2.
MATERIALS USED
For experiments was used pure titanium powder with the particles size ranging
between 5 and 15 m. As substrate it was used titanium disks 30x 5mm diameter. The
samples were blasted with corundum with the average particle size between 0.8 and 2
mm at a pressure of 5 bar and blasting distance of 50 mm. After the blasting process
the samples were cleaned with alcohol.
3.
EXPERIMENTAL PROCEDURE
For experiments was used the plasma thermal spraying equipment from
National Institute for Research and Development in Welding and Material Testing, ISIM
Timisoara supplied by Sulzer Metco. The principle of deposition titanium nitride layers
by reactive plasma thermal spraying process is based on chemical reactions which
occur during the thermal spraying process. As plasmagen gas was used nitrogen.
Also, the transport gas of the powder and cooling was used nitrogen. As a result of
chemical reactions between the particles of titanium melted in the plasma gas stream
(15000C) and nitrogen are obtained TiN compounds with higher hardness compared
with the substrate [7].
292
Deposition of titanium nitride layers by Reactive Plasma Spraying
Fig. 1 The principle of reactive plasma spraying process [7]

In table 1 are presented the parameters used to deposit titanium nitride layers
by reactive plasma spraying.
Table 1. Parameters used to deposit titanium nitride layers
Ip
[A]
320-330
4.
Ua
[V]
90-100
QP
[l/min]
40-45
Qtr
[l/min]
6-7
mp
[g/min]
15-17
dP
[mm]
1505
No of passes
3
4.1 Macroscopic analysis

Figure 2 shows the macroscopic aspect of the deposited layers by reactive
plasma thermal spraying.
a)
b)
Fig. 2 Macroscopic image of the deposited layers: a) general view, b) detail

The macroscopic analysis shows that deposited layer has spherical voids. This
is due to agglomeration of titanium melted powder which reach the substrate surface at
high speed.
293
4.2 Microscopic analysis

The biphasic structure of + of titanium alloy (Ti6Al4V) used as substrate is
presented in Figure 3.
Fig. 3 Microscopic structure of titanium alloy (Ti6Al4V) used as substrate, 200x

Figure 4 presents the SEM images of the deposited layers by reactive plasma
thermal spraying method.
a)
b)
Fig. 4 SEM micrograph of titanium nitride layers deposited by reactive plasma thermal
spraying method (a) 500x, (b) 1000x
The microscopic analysis shows that the layer is composed of lamellar
particles. Also is observed that the layer is non-uniform presenting pores on the
surface.
294
Deposition of titanium nitride layers by Reactive Plasma Spraying
4.3. Layers thickness determination

The microlayers thickness were performed using Leptoscop Pocket device,
the obtained values are listed in Table 2.
Table 2 Microlayers thickness values
Microlayers thickness [m]
Individual values
78,3; 68,5; 81,4; 77,8; 79,1
Mean value
77,02
4.4. Roughness determination of the deposited layers

The surface roughness determination was made by Surftest 201 (SJ-201)
device from Mitutoyo, the results are showed in Table 3.
Table 3 Roughness values of the deposited layers
Average roughness Ra [m]
Individual values
5,43; 5,68; 5,44; 5,98; 5,70
Mean value
5,64
4.5 Hardness tests

HV1 microhardness tests were performed on the microlayers surfaces, the
microhardness values are summarized in Table 4
Table 4 Hardness values
No.
Ti6Al4V alloy
1
2
3
5.
348
HV1 Hardness values

Deposited layer
633
686
641
CONCLUSIONS
1. Thermal spraying knows a increased applicability due to the numerous
advantages, currently, there are industries where using thermal coating
technologies is essential to obtain characteristics required by the users
2. The macroscopic and microscopic analysis shows that the deposited layers by
reactive plasma spraying presents thermal defects such as pores. This is due
to agglomeration of titanium melted powder which reaches the substrate
surface at high speeds.
3. The thickness of the deposited layers by reactive thermal spraying method is
have values between 68.5 and 79.1 m, thickness which posses hardness
between 633 and 686 HV1, hardness which ensures a high wear resistant of
the components.
295
Acknowledgement
This work was partially supported by the strategic grant POSDRU/89/1.5/S/57649,
Project ID 57649 (PERFORM-ERA), co-financed by the European Social Fund
Investing in People, within the Sectoral Operational Programme Human Resources
Development 2007-2013.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
296
ASM Thermal Spray Society (2006). International Thermal Spray &Surface

Engineering, Vol 1 No 2
Clarke, A., Partridge, A. (2006). A Strategic Review of the Surface Engineering
Industry in the UK
Ashok Kumar, Yip-Wah Chung, John J. Moore, Gary L. Doll, Kyoshi Yatsui,
D.S. Misra, (2002) Surface Engineering:Science and Technology, TMS 2002
Annual Meeting in Seattle, Washington, USA
Hirokazu, T., Yasutaka, A. (2009). Study of titanium nitride deposition by
supersonic plasma spraying, Vacuum 83, 98101
Peng Z., Miao, H., Longhao, Q., Gong, J., Yang,
S., Liu, C. (2003).
Microstructure and mechanical properties of titanium nitride coatings for
cemented carbide cutting tools by pulsed high energy density plasma, Chinese
Science Bulletin Vol. 48 No.13 p. 1316-1320
Alessandro, V., Walid, K., Hassen, J., (2008). Titanium Nitride Coating of RF
Ceramic Windows by Reactive DC Magnetron Sputtering, Proceedings of
EPAC08, Genova, Italy
Zou, D., Yan, D., Xiao, L., Dong Y (2008). Characterization of nanostructured TiN
coatings fabricated by reactive plasma spraying, Surface & Coatings Technology,
vol. 202, no. 10, p. 1928-1934
DESIGN AND FABRICATION OF MEDICAL MODELS WITH RAPID

PROTOTYPING TECHNIQUES AND VACUUM CASTING
Sawomir Miechowicz1, Tadeusz Markowski2, Tomasz Kudasik3, Olimpia
Markowska4
Summary: The possibility of application of modern Rapid Prototyping techniques for
medical model fabrication in this paper is presented. The aim of the work was the
cranial bone model fabrication for photoelastic tests. The bone tissue modelling is a
problem due to a specific, complex geometry and difficulties with material for
experimental aims. The Computer Tomography (CT) and Cone Beam Computer
Tomography (CBCT) imaging were applied for the 3D model preparation. The model of
the tissues was fabricated using chosen Rapid Prototyping (RP) techniques and
Vacuum Casting (VC) method. The data processing of the medical data (DICOM),
model design and fabrication process was described. The experiment assumptions,
the fields of possible application, medical and technological consideration of complex
model fabrication with RP methods in this article were pointed.
Key words: CT, CBCT , Rapid Prototyping (RP), Vacuum Casting
1.
INTRODUCTION
Rapid prototyping methods can be used to create spatial models from precise
3D-CAD virtual objects. For medical objects designing, the data needed, may be
obtained from CT, MR or rotational angiography (angio-CT) [1,2,3]. With the use of
Rapid Prototyping method, it is possible to create models that reproduce the inner
structures of object. In case of the experimental research, based on medical models,
the precision is a crucial factor. The result of analysis depends mainly on the accuracy
of the model. The aim of the study is:
x to present the application of reverse engineering and rapid prototyping
techniques to medical model manufacturing,
1
Phd, Slawomir Miechowicz, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959

Rzeszow, Poland, smiechow@prz.edu.pl
2
Prof., Tadeusz Markowski, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959
Rzeszow, Poland, tmarkow@prz.edu.pl
3
Phd, Tomasz Kudasik, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959 Rzeszow,
Poland, tkudasik@prz.edu.pl
4
Msc Olimpia Markowska, Rzeszow University of Technology, Powstancow Warszawy 8, 35-959 Rzeszow,
Poland, omarkowska@prz.edu.pl
297
Sawomir Miechowicz, Tadeusz Markowski, Tomasz Kudasik, Olimpia Markowska
x
x
2.
to manufacture high resolution cranium model using rapid prototyping

method based on CT data,
to manufacture high resolution vacuum casted cranium model for
experimental test.
DATA ACQUISITION AND CRANIUM MODEL DESIGN
The CT examination of cranium was performed

Sensation 16 scanner (fig. 1) and dedicated CT protocol
thickness 0.6 mm, reconstruction increment 0.5 mm, sharp
512 pixels, reconstruction FOV (Field of View) adjusted to
get the highest spatial resolution possible.
using Siemens Somatom

(collimation 0.6 mm, slice
kernel U90U, matrix 512 x
dimensions of cranium) to
Fig. 1 Siemens Somatom Sensation 16 CT scanner and the volume rendering of bone
and soft tissues of head
CT images in DICOM format were then imported into a modelling program (3DDoctor by Able Software) and after segmentation, a 3D polygonal mesh model of
cranium was generated (fig. 2). The final virtual model in STL format was imported into
SLS system software.
Fig. 2 A STL surface model of cranium generated based on the CT data acquired with
Siemens Somatom Sensation 16 CT scanner
298
Design & Fabrication of Medical Models With Rapid Prototyping Techniques & Vacuum Casting
3.
SLS METHOD
The method of Selective Laser Sintering (SLS) of powder consists in

hardening the surface of the powder layers by means of laser beam. The sintering
process is performed in a chamber filled with powder, where a platform together with
the fabricated model move down by a value, which equals the layers thickness.
Hardening of subsequent layers causes partial melting of the previous layer which
makes it possible to obtain coherent model structure. During the sintering process the
model is dipped in the powder and that is why it is not necessary to apply additional
supportive structures. The models of skull on a scale of 2:1 were made with SLS EOS
Formiga P100 device (Bibus Menos Poland). The applied material was PA 2200
polyamide powder. The cranial bones model is presented in figure 3. Using vacuum
casting method, based on SLS model, the VC medical model of cranium can be finally
manufactured.
Fig. 3 Model of skull fabricated with SLS method

4. VACUUM CASTING METHOD
Vacuum casting is one of the rapid tooling techniques [4,5]. It is used in
creating the prototypes of machine elements and medical models. In this method
silicone moulds are prepared based on master models, that usually have been built
with rapid prototyping methods [6]. In this case, the cranial bones models were used as
base models.
The steps of the silicone mould production process:
- preparing the moulding box of the right volume,
- preparing the gating and venting systems models,
- constructing the model sets by connecting the gating and venting systems
models with the rapid prototyping base model,
- placing the model set into the moulding box,
- preparing the liquid silicone rubber,
- pouring the liquid silicone rubber over the model set in the moulding box,
- de-aerating in a vacuum chamber,
- putting the silicone moulds into a heating chamber,
- separating the moulding box from the silicone mould block.
299
The first step is to analyze the shape and the size of the base model, so the
moulding box is of a right size. The model set should be stable inside the moulding
box. The box should not be too big to minimize the amount of the silicone needed.
The gating and venting systems are very important. It is possible to design
them during the stage of the STL modelling, and they can be created with the rapid
prototyping method along with the model. They can also be constructed with any other
forming method.
In case the complex surfaces modelling, i.e. a model of cranium, there are
obviously some open spaces and the silicone can enter them easily. Nevertheless,
there are also some places the silicone cannot access.
The next step is to prepare the silicone rubber. To produce the moulds, the
liquid silicone and the MM240 TV A+B hardener was used (10:1 ratio). The
components need to be well mixed, then de-aerated in a vacuum casting machine
(UHG 400) in sub atmospheric pressure of 250 hPa (fig. 4).
Fig. 4 A vacuum casting UHG 400 machine (Department of Mechanical Engineering,

The Rzeszow University of Technology)
The silicone solution needs to be poured into the moulding box and be deaerated again. After the 24 hour curing period, the silicone cast material is ready for
the next steps of the process. An example of the cured silicone mould of cranium
upper part is shown in fig. 5. The silicone mould reflects the outer and inner surface of
the base model (together with hollow spaces). To prove that silicone filled the inside of
the model, two moulds were created.
300
Design & Fabrication of Medical Models With Rapid Prototyping Techniques & Vacuum Casting
Fig. 5 The inside and outside silicone mould of skulls fragment after separation
The moulds can be separated again only after full polymerization of the resin.
The process of separating mould of skull top part model is shown in figure 6 (left).
After taking the model out of the mould it is necessary to clean it precisely and remove
the venting and pouring gates and channels. The ready model, made of epoxy resin
Epidian 5 is presented in the figure 6 (right).
Fig. 6 The process of separation of mould of skull top part from model (left). Model
made of Epidian 5 resin ready for photoelastic experiment (right)
The model prepared in this way can be used for experimental tests. The
additional advantage of silicone moulds is their high durability, which allows for using
them many times to fabricate a greater number of models [6].
In case of preparing models for photoelastic experiment, the moulds were filled
with the Epidian 5 epoxy resin. After the 24 hour polymerization period, the moulds
were cut and the epoxy resin cranial bones model was removed. Some silicone
remains inside the model, so in this particular case, it is not possible to re-use the
silicone mould. The silicone rubber inside the model can simulate soft tissue [7].
Thanks to this simulation, the results of the experimental analysis can be more precise.
The Vacuum Casting models can be manufactured with variety of biocompatible
materials [8,9].
301
5.
CONCLUSION
Reverse engineering and rapid prototyping may utilize data obtained by

commonly used spiral CT scanners.
Manufacturing of medical models using reverse engineering techniques should
be more commonly used, reducing time needed for surgery treatment and improving
the efficiency of healing process.
Insufficient spatial resolution of CT is the main factor limiting the accuracy of
medical models created using rapid prototyping based on CT data. The use of CBCT
data and CT scanners can improve radically the resolution of manufactured models
based on these data.
ACKNOWLEDGEMENTS
This work was financed from the Polish science budget resources as the
research project NN501268134.
LITERATURE
[1] Miechowicz, S., Marciniec, A. (2004). Stereolitography the choice for medical
modelling. Acta of Bioengineering and Biomechanics, nr 6/1.
[2] Petzold, R., Zeilhofer, H.F., Kalender, W.A. (1999). Rapid prototyping technology
in medicine basic and applications, Computer Medicine Imaging Graphics, nr 23,
(277284),
[3] Raic, K., Jansen, T., Rymon-Lipinski, B., et al. (2002). Fast generation of
stereolithographic models. Biomedical Technology, nr 47, p. 83-85.
[4] Bdziski, R. (1997). Biomechanika inynierska, Wydawnictwo Politechniki
Wrocawskiej, Wrocaw.
[5] Budzik, G. (2009). Odwzorowanie powierzchni krzywoliniowych opatek czci
gorcej silnikw lotniczych w procesie szybkiego prototypowania. Oficyna
Wydawnicza Politechniki Rzeszowskiej, Rzeszw.
[6] Budzik, G. (2007). Possibilities of utilizing 3DP technology for foundry mould
making, Archives of Foundry Engineering, Vol. 7, Issue 2/2007, p. 65-68.
[7] Kopkowicz, M. (2003). Metody dowiadczalne bada konstrukcji. Oficyna
Wydawnicza Politechniki Rzeszowskiej, Rzeszw.
[8] Muller, A., Krishnan, KG., Uhl E et al. (2003). The application of rapid prototyping
techniques in cranial reconstruction and preoperative planning in neurosurgery,
Journal of Craniofacial Surgery, 14(6), p. 899-914.
[9] Markowski, T., Markowska, O., Kudasik, T., Miechowicz, S. (2008). Wykorzystanie
tomografii komputerowej i metody rapid prototyping dla potrzeb otolaryngologii,
Archives Of Foundary Enginieering, Vol. 8, Special Issue 1/2008, p. 199-202.
302
DEVELOPMENT AN EXPERT SYSTEM FOR MACHINING FIXTURE

DESIGN
Djordje Vukelic1, Branko Tadic2, Janko Hodolic3, Igor Budak4, Milovan Lazarevic5
Summary: This paper presents an expert system for machining fixture design. System
provides new fixture construction design for specified input parameters on basis of
adequate production guidelines. Paper provides applied methodology basic structure,
specific systems segments review, and example of systems implementation in
industrial production. In closing, there are conclusions, developed systems advantages
and disadvantages, and directions for future research.
Key words: fixture, expert system, automation.
1. INTRODUCTION
Fixture design process is complex, intuitive, and long-term and mostly depends
on designers knowledge and experience. This inflicts the need for new technologies
implementation in fixture design process. New technologies have the main aim to
reduce time and costs in new fixture construction design and find optimal solutions for
specific manufacturing conditions. This can be successfully solved by adequate fixture
design system development [1, 7]. There are numerous examples of system appliance
in world in field of fixture design where knowledge is presented in shape of regulations.
Regulations can be understood as knowledge elements or as elementary amount of
knowledge from specific fixture design domain. Regulations presents logic relation and
it can be presented as:If X Then Y. This means: If assumption X exists then Y can be
concluded. For example:
If locating surface is a hole Then locating element is long round pin
In other words, rule is logical expression with type If-Then with meaning if there
is some kind of premise (or group of premises), then there is a conclusion (or group of
1
Dr. Sc. Djordje Vukelic, University of Novi Sad, Faculty of Technical Sciences, Department for Production
Engineering, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia, vukelic@uns.ac.rs
2
Dr. Sc. Branko Tadic, University of Kragujevac, Faculty of Mechanical Engineering, Department for
Production Engineering, Sestre Janjic 6, 34000 Kragujevac, Serbia, btadic@kg.ac.rs
3
Dr. Sc. Janko Hodolic, University of Novi Sad, Faculty of Technical Sciences, Department for Production
Engineering, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia, hodolic@uns.ac.rs
4
Dr. Sc. Igor Budak, University of Novi Sad, Faculty of Technical Sciences, Department for Production
Engineering, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia, budaki@uns.ac.rs
5
Dr. Sc. Milovan Lazarevic, University of Novi Sad, Faculty of Technical Sciences, Department for Industrial
Engineering and Management, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia, laza@uns.ac.rs
303
Djordje Vukelic, Branko Tadic, Janko Hodolic, Igor Budak, Milovan Lazarevic
conclusions), and then action (or group of actions) can be taken. This kind of
knowledge expression is very natural and it suits the tendency for knowledge to
become modular [8]. In addition, it satisfies requirement for easy knowledge base
modification because new rules can be independently added from others and system is
transparent, i.e. ways for making conclusions are easy to explain. Because of its
shape, production rules are suitable way for presenting logic based knowledge.
Conclusions are carried out by comparison of regulations group with group of facts or
knowledge about the current situation.
Boerma and Kals [2] proposed a system for automatic selection of setups and
datum based on feature tolerances for prismatic parts. The system automatically
selects the positioning, clamping, and supporting faces for each setup. Dai et al. [3]
described rule-based reasoning a modular element database creation method, which
can be used effectively for integrating with a CAD system and for modelling fixture
subassemblies. Darvishi and Gill [4] developed a fixture design expert system which is
based on examining the design goals to be achieved and then creating rules to satisfy
these imposed specifications. Lin and Yang [5] developed an expert system for a
modular fixture. Sistem provide a fixturing procedure analysis for workpieces with
different shapes such as L, I, T, and U types for face milling machining operation. Nee
et al. [6] presented a feature-based classification scheme for fixtures using a 3D solid
modeller, a feature extractor and an object-oriented expert system shell. Senthil Kumar
et al. [8] developed expert fixture design system, which integrated setup planning and
fixture design for an automated manufacturing environment. Senthil Kumar et al. [9]
presented classification model and classification rule for conceptual design of fixture in
milling and drilling operation, for prismatic part. Stampfer [10] discussed an automated
setup and fixture planning method for machining box-shaped parts at horizontal
machining centres.
The above mentioned expert and knowledge systems for fixtures design have
some deficiencies in common. Production rules are developed mostly for locating and
clamping elements. Elements can be chosen by repeating of regulations until all
suitable elements are picked out from database. If large number of elements satisfies
required demands, designer decides which solution will apply. Elements selection is
done based on their functions. Whenever regulation is used, first is to check elements
function and after that its geometrical characteristics. Main disadvantage of all the
researches is that they only give conception or partial solution. Fixture concept solution
is a solution that defines clamping or locating scheme, i.e. it defines locating and
clamping surfaces, and locating and clamping elements position, not elements
themselves. Fixture construction partial solution is a solution that defines actual
locating and clamping elements.
Main aim of this paper is development of an expert system for fixture design
that will secure fixture elements selection from all the elements functional groups. This
includes selection of clamping and locating elements, fixture body elements, tool
guiding elements, tool adjustment elements, connection elements, and additional
elements. With this, overall fixture design process would function with expert system
support.
304
Development an expert system for machining fixture design
2. GLOBAL SYSTEM STRUCTURE

Fig. 1 shows basic structure of an expert system for fixture design. Basic
structure consists of knowledge base, inference engine, working memory, user
interface, knowledge acquisition subsystem, special interfaces and explanation
subsystem.
Fig. 1 Structure of an expert system for fixture design

Knowledge base has expert's knowledge from fixture design field. Knowledge
input goes through knowledge acquisition subsystem. During the system functioning
knowledge is not changing, its only changing before and after system exploitation.
Working memory consists present data of characteristics that expert system solves.
This data is variable and its values reflect present state in problem solving process.
Inference engine is a program that solves the problem based on variable data and
knowledge that is built in knowledge base, so it carries out systems assignment.
Communication and result presentation (fixture elements) between system and user
(designer) goes through user interface. During the processing there are facts,
propositions, conclusions and other relevant data in working memory connected to
description of workpiece and manufacturing characteristics performed on workpiece.
These data can change, generate, and they can lose its importance during the working
time. In systems based on rules, data in working memory are organized in form of
statements, very similar to rule clauses. At the basis of initial data in working memory
and rules in knowledge base, expert system can bring out conclusions in relation with
selection of adequate fixture elements. In conclusion process, inference engine
attempts to find a solution based on initial data in working memory and knowledge in
knowledge base. User inputs initial data and they are stored in expert systems working
memory. Inference engine applies knowledge from knowledge base to those data, and
generates new data in working memory. With this, group of data in working memory is
expanded. New condition in working memory can be enough to solve a problem and in
that case, conclusion process is finished. Opposite to that, expanded group of data is
processed again with usage of knowledge from knowledge base, and this leads to new
change in working memory condition. Process is continuing with iterations until working
memory condition is adequate for solution or until solution cannot be found. Inference
engine can require input of additional data from user, if this is necessary, during the
search for solution.
305
3. SYSTEM FUNCTIONING
Expert systems structure for automated fixture design is explained in previous
chapter and to fulfil this explanation it is necessary to show how system works in real
conditions. In this chapter example of system testing is shown with real manufacturing
workpiece example. Workpiece for which was necessary to design fixture is shown on
Fig. 2. Drilling operation is performed on workpiece for two holes 30mm at vertical
drilling machine.
1000.1
57
30
94
10
45
1360.5
58
1900.2
1400.1
2100.08
Fig. 2 Workpiece
First step in system apply is input information coding and with that selection of
individual fixture elements can be done. Factors that have influence on fixture
construction can be joined with correct defining of input information. Input information's
can be divided into two general groups:
manufacturing characteristics (type of machining, main machine tool group,
machine tool sub-group, type of machine tool, specific machine tool which is
selected for machining, number of workpieces being machined at the same
time, number of tools, number of machining surfaces, method of connecting
fixture with machine tool, batch size, forces and moments acting during
machining process),
workpiece characteristics (workpiece shape, overall workpiece dimensions,
number of degrees of freedom arrested with locating elements, workpiece
locating method, shape of locating surfaces, integrality of locating surfaces,
quality of locating surfaces, type of locating surfaces, characteristic
dimensions of locating surfaces, shape of clamping surfaces, clamping
scheme in particular directions, clamping drive in particular directions,
intensity of clamping force in particular directions, types of clamping
surfaces by particular directions, characteristic dimensions of clamping
surfaces).
Segment of forms (computer reviews) how coding can be done is shown on
Fig. 3. At basis of input information's labels are generated from needed fixture for
observed workpiece manufacturing operation. Label consists of certain number of
306
Development an expert system for machining fixture design
codes and it presents the key for searching of fixture elements database. Locking
mechanism selects adequate fixture elements that will be used for final fixture structure
and it represents them with a specific form (Fig. 4).
Fig. 3 Coding of input information
Fig. 4 Fixture elements form
Element list is form based on choice decision developed criteria (production

rules) for every element from individual functional groups. Label, and its individual
attributes carry appropriate information - choice criteria. At basis of these criteria
selection of fixture elements from all functional groups is done. After data base search
potential fixture elements preview is generated if they exist. When several specific fixture
elements solutions are obtained for the observed workpiece manufacturing operation,
technological and economical analysis is being conducted for offered solutions. For goal
functions, it is possible to set different parameters. It is suitable to chosen production,
accuracy and price for parameters. It is expected that fixture has highest productions and
accuracy, and lowest costs (price). Outputs are certain fixture elements for fixture
assembling. After fixture assembling fixture structure is obtained, as it is shown on Fig. 5.
After fixture analysis and fixture that can do the function is being identified, procedure for
generation of certain technical documentation is carried out.
Fig. 5 Output results fixture for drilling operation

4. CONCLUSION
Suggested system verification is carried out in several manufacturing systems.
System provides selection of fixture elements and new fixture design. System provides
307
quick fixture solution when there is need for new products, and with that, it improves
technical and economical parameters of the completely manufacturing process.
System provided satisfactory results for prismatic and rotating workpieces for milling
and drilling operations. In following research stages system will be developed for other
manufacturing operations (turning, grinding etc.). Author's idea was to implement
system in it first stage for manufacturing operations of drilling and milling. This was not
without of reason. Particularly for these manufacturing operations, it is often necessary
to design fixture. For other manufacturing operations, (turning, grinding, etc.) most of
the operations can be done with usage of universal fixtures that are delivered together
with machine tool.
LITERATURE
[1]
Ameri, F., Summers, J. D. (2008). An Ontology for Representation of Fixture

Design Knowledge. Computer-Aided Design and Applications, Vol. 5, No. 5, pp.
601-611.
[2] Boerma, J. R., Kals, H. J. J. (1989). Fixture Design with FIXES: the Automatic
Selection of Positioning, Clamping and Support Features for Prismatic Parts.
CIRP Annals - Manufacturing Technology, Vol. 38, No. 1, pp. 399-402.
[3] Dai, J. R., Nee, A. Y. C., Fuh, J. Y. H., Senthil Kumar, A. (1997). An approach to
automating modular fixture design and assembly, Proceedings of the Institution of
Mechanical Engineers, Part B: Journal of Engineering Manufacture, Vol. 211, No.
7, pp. 509-521.
[4] Darvishi, A. R., Gill, K. F. (1988). Knowledge representation database for the
development of a fixture design expert system, Proceedings of the Institution of
Mechanical Engineers, Part B: Management and Engineering, Vol. 202, No. B1,
pp. 37-49.
[5] Lin, Z. C., Yang, C. B. (1995). An expert system for fixturing design for face
milling using modular fixture, The International Journal of Advanced
Manufacturing Technology, Vol. 10, No. 6, pp. 379-388.
[6] Nee A. Y. C., Senthil Kumar A., Prombanpong S., Puah, K. Y. (1992). A FeatureBased Classification Scheme for Fixtures, CIRP Annals - Manufacturing
Technology, Vol. 41, No 1, pp. 189-192.
[7] Pehlivan, S., Summers, J. D. (2008). A review of computer-aided fixture design
with respect to information support requirements. International Journal of
Production Research, Vol. 46, No. 4, pp. 929-947.
[8] Senthil Kumar, A., Nee, A. Y. C., Prombanpong, S. (1992). Expert fixture-design
system for an automated manufacturing environment. Computer-Aided Design,
Vol. 24, No. 6, pp. 316-326.
[9] Senthil Kumar, A., Subramaniam, V., Teck, T. B. (2000). Conceptual Design of
Fixtures Using Machine Learning Techniques. The International Journal of
Advanced Manufacturing Technology, Vol. 16, No. 3, pp. 176-181.
[10] Stampfer, M. (2009). Automated setup and fixture planning system for boxshaped parts, The International Journal of Advanced Manufacturing Technology,
Vol. 45, No. 5-6, pp. 540-552.
308
DEVELOPMENT MODEL FOR CONTROL METAL CUTTING

PROCESS
Bogdan Nedi1), Gordana Globoki Laki2)
Summary: All processes and phenomena that are happening in the localized cutting
zone have, not static, bat expressively dynamic character, caused by variable cutting
depths, dynamic character of the shaving formation process, creation and destruction
of layers on the cutting wedge, instability of the system: tool - machined piece equipment - machine, machined piece geometry, etc.
In this paper is presented analysis of the model of cutting process management based
on oscillatory motion of machining system elements. Cutting process is determined by
dynamic character of chip forming process and oscillations of tool - work piece accessories - machine system elements.
Key words: metal cutting, control, dynamic, chip
1
INTRODUCTION
Development of adaptively managed machining systems is tightly connected

with the development of sensors and measuring devices. Discontinue - direct (off-line,
quality of working surface, wear, accuracy etc.) and continual - indirect (on-line, indirect
measuring of wear through wear resistance, vibration, noise, accuracy, temperature
etc.) systems for measuring output parameters of cutting process are simultaneously
developed. Proper functioning of monitoring system is necessary for development of
automatic machining systems, intelligent systems for supervision, control and
management. It requires adequate analysis of failures and well chosen sensors, not
only wear sensors, but also sensors for estimating functionality of working process:
quality, accuracy, wear, reliability etc.
Reason for development of monitoring and on-line systems for control cutting
process is because the automatic systems, numeric devices, have to react quickly and
effectively, and adjust to any new situation caused by unexpected effects:
1)
2)
Dr Bogdan Nedi, red. prof., Mainski fakultet, Kragujevac, Serbia, nedic@kg.ac.rs

Dr Gordana Globoki Laki, vanr. prof., Mainski fakultet, Banja Luka, RS, BiH, gnm@urc.rs.ba
Acknowledgments. The authors would like to thank to the Ministry of Science and Technological
Development of the Republic of Serbia. Paper is result of technological project TR35034; The research of
modern non-conventional technologies application in manufacturing companies with the aim of
increase efficiency of use, product quality, reduce of costs and save energy and materials; which is
supported by Ministry of Science and Technological Development of the Republic of Serbia..
309
Bogdan Nedi, Gordana Laki Globoki
sedimentation, unexpected flow of chip, tool damages, change of rigidity of tool - work
piece - accessories - machine system etc. Influence of wear on behavior of machining
system elements is usually known, but intensity of tool wear is not.
Off-line system for control and optimization has the purpose to determine all
initial machining parameters and laws of their change. On this level, before and during
machining, analyses, modeling and predicting the behavior of machining system
elements are conducted. Proper dimensions and characteristics of tool, way of holding
workpiece, and other parameters are chosen.
A
MACHINING
PROCESS
Environment:
- machine,
- fixture, ...
INPUT
OUTPUT
A= f(s, v, , , Rm, ..)
A = f( bxh, l, ...)
CUTTING
ZONE
A = f( Rm, a/l, d, ...)
CONTROL
PANEL
ACC ili ACO

on-line Control
( s, v )
a/l
IDENTIFICATION
OPTIMIZATION
off-line Control
(Bxh, l, , , c, Rm)
SENSORS MEASURING SYSTEM
DATA
BASE
CALCULATION,
MODELING,
SIMULATION
Fig. 1 Model of control cutting process based on dynamic characteristics of machining

process elements [3]
To form monitoring system and system for control cutting process it's
necessary to:
describe mathematically and verify experimentally:
x dynamic character of chip forming process,
x oscillatory behavior of tool,
310
Development Model for Control Metal Cutting Process
x oscillatory behavior of work piece,

verify experimentally sensors of cutting process
develop adequate monitoring and control systems.
2
DYNAMIC CHARACTER OF CHIP FORMING PROCESS
Cutting process is highly dynamic. Oscillations formed during the process are
unplanned, unwanted, and they can cause inaccuracy, shortening of tools duration and
damages of machine elements. Basic problem with studying cutting process is the
problem of describing dynamic character of chip forming process. Cutting process can
be described as series of processes conducted one after another: elastic deformation,
plastic deformation, destruction of material and additional deformation - twisting and
chip removing.
Starting from elementarily of chip phenomenon, photographs and existing
models of chip forming, cutting process can be observed as compression of material
under cutting wedge, shear of material in cutting plane and shear in sliding plane.
These processes are determined by variable pressure and shear tensions depending
on way x of cutting wedge (figure 2).
Fig. 2 Elementarily of chip segments forming [3, 5]

Based on this picture, with the proper analyses of forces we can find the term
for the way x=xmax when shear begins:
G cosJ
xmax
R
cosJ cosI
tgI 1 m tgI
W
cos(I J )
s
If the speed of cutting wedge is known we can find the period T0 and frequency
fst of chip segments forming:
311
60 x max
1000 v
T0
fst
R
1 1000 v
1000 v
cosJ cosI
tgI 1 m tgI
T0
60 xmax
60 G cosJ
cos(I J )
W
s
Experimental research of influence of working regime and tool geometry on

value of chip compression factor, shear angle and frequency of chip forming show
great match between estimated model and experimental data..
3
DYNAMIC BEHAVIOUR OF TOOL AND WORK PIECE
Knowing the parameters of oscillatory motion of tool and workpiece enables us

to predict it's behavior during cutting process. With purpose to describe their dynamic
behavior tool and work piece can be approximated with systems with one degree of
freedom. This way parameters of free damped oscillation can be mathematically
modeled and experimentally determined.
Placing acceleration sensor directly under tool peak and in the proper places of
workpiece and leading them out of position of equilibrium, picture 3., during
experimental research signals of damped oscillations were received [2, 3].
AD convertor
bxh
B&K
8303
Amplifier
l
Acceleration,
m/s
j
1000
Acceleration, m/s 2
4
g
800
600
400
200
0
3
2.5
2
1.5
1
-200
0.5
-400
-600
3.5
-800
2
0
0
4
10
Time,
2000
4000
6000
12
8000
10000
Frequency,
12000
Hz
s x 10
Fig. 3 Measuring of parameters of free damped oscillations

Analyzing signals and using proper computer programs, knowing circle
frequency p, nondimensional damping coefficient G and damping factor b were
determined at cutters with most frequently used dimensions of cross sections of
holders.
Starting from these measurement it is possible to model oscillations of the work
piece and the tool. It is particularly important for objects of greater length and smaller
diameter with intensive axial rigidity change, and the tools with great bracket (la). It
means that frequencies of free damped oscillations can be written as
fp = f(lp, la, d, bxh, a/lp, Rmp, ... ).
312
Development Model for Control Metal Cutting Process
4.
DETERMINATION OF SENSORS AND PARAMETERS OF DYNAMIC

CHARACTER OF CUTTING PROCESS
Cutting resistances and acceleration, as it was determined analyzing great

number of papers, are reliable and mostly used parameter for process identification
during machining. That's the reason why great deal of conducted investigations refers
to analysis of machining process through dynamic characteristics of cutting resistance
signal and acceleration signal.
Based on extensive experimental research, "static" and dynamic part of signal
were analyzed depending on machining regime and tool geometry. During these
investigations equipment contains of piezo dynamometer for cutting resistance
measuring, acceleration sensor, proper computer equipment and software. (picture 4).
Using amplitude-frequency analysis of cutting resistance signal and tool peak
acceleration existence of characteristically frequency of oscillations of tool and work
piece and the frequency of chip segments forming. It enabled analysis of cutting process
based on dynamic characteristics of cutting parameters after knowing the origin of
oscillations, the reasons of forming the oscillations of machining process elements.
During the analysis influence of oscillation of other machining process
elements (machine parts etc.), oscillations of work piece due to eccentricity, and air
gaps in holding elements cannot be ignored. That's the reason why research of
dynamic character of cutting process included research in lower frequencies regions of
oscillations, the frequency of signal acquisition was great deal lower. It was also shown
that analyzing the cutting resistance signal in this case, changes during cutting process
which resulted due to tool wear could be identified.
Accelerometer
Dinamometer
KISTLER
Measuring 300 Hz
Fi
..
x
..
z
Computation
F23
F2
FR
. ..
F3
Acquisition 42 KHz
Fi
Amplifier
t (s )
F12
F1
..
z
..
x
PR
Computation
fpi
Api
AD converter
Fig. 4 System for identification of parameters of dynamic character of cutting process [1, 3]
Management based on amplitude-frequency characteristics and analyses
doesn't enable on-line system for monitoring on this level of computer techniques, but
in recent future terms for that will certainly be created.
313
CONCLUSION
Cutting zone is characterized by dynamic behavior of tool that can be

described with the term for characteristic frequency depending on tool dimensions
(bxh, l), dynamic behavior of work piece that can be described with changeable rigidity
and own frequency depending on work piece dimensions, machining point a/l and
characteristics of the material. It is also characterized by dynamic character of chip
segments forming process depending on characteristics of work piece material, tool
geometry and machining regime.
Cutting zone is effected by the surrounding also characterized with dynamic
character, machining system elements, machine elements, accessories etc. have
specific rigidity and can be described by characteristic oscillatory frequencies.
Managing this system, machining process, from aspect of dynamic
characteristics, suppose estimation, modeling and experimental research. Magnitudes
have to be determined and mathematical models estimated for these research.
Next level considers development of modules for identification of changes and
cutting process management and determination of origin of these changes manifested
at measured magnitudes. If cutting process is continual, without side effects like
unwanted vibrations, noise, tool damages etc., we could say that cutting process is
running in anticipated course, and that process can be managed continually, on- line
by determined criteria ACC, or optimizing ACO
However, if it is not possible to achieve wanted effects, identification of existed
effects must be done and origin determined. It's necessary to stop cutting process and
compact input parameters: change the tools, increase rigidity of work piece, change
machining technology etc. In that case it is needed to conduct some calculations,
modeling and simulation of cutting process from dynamic aspect using data base.
Calculation and modeling can be done starting from derived terms. In that case
development of cutting process simulation modulus in conditions of dynamic forced
oscillations and changeable rigidity of the workpiece. It considers development of new
CAD/CAM software. That's the way of getting references for defining technological
methods that would ensure wanted effects.
6. LITERATURE
[1]
[2]
[3]
[4]
[5]
314
Nedi, B., Dinamika procesa rezanja, monografija, Mainski fakultet, Kragujevac,

2006.
Nedi, B., Vesi, N., etkovi, M., Dinamics characteristics of metal cutting process,
Balkantrib 05, 5-th International conference on tribology, Kragujevac, 2005.
Nedi, B., Development of methods for metal cutting process management through
dynamic characteristics of cutting parameters, PhD dissertation, Faculty of
Mechanical Engineering, Kragujevac, 1998.
Nedi, B., Free damped oscillations of tool during turning, YUTRIB '97, Kopaonik Ni, 1997
Jaspers, S., Metal Cutting Mechanics and Material Behaviour, Technische
Universiteit Eindhoven, Eindhoven, 1999.
DEVELOPMENT OF EPC PROCESS FOR MANUFACTURING PARTS

IN AUTOMOTIVE INDUSTRY
Aurel Prsti1, Zagorka Acimovi-Pavlovi2, Zvonko Guliija3, Mirjana Stojanovi4
Summary: In this paper, some of the results of research are presented, by applying
casting technology with evaporative pattern for obtaining castings (EPC process). This
technology of castings enables obtaining complex parts, intended to, for example,
automotive industry. It means that it is characterized by large series, low production
expenses and high quality of castings. The investment into equipment and space are
not high, the process is flexible and high productivity can be achieved.
By research, a satisfactory quality is achieved and necessary, relevant information for
industrial production of silumine casting are obtained.
Key words: evaporative pattern, castings quality, refractory coating, layer tchickness
1.
INTRODUCTION
Evaporative patterns casting (EPC process) is patented by H.F. Shroyer in

1958. Since then, up to today, development and practical application goes on with
changeable success. Mainly, the problems of the development of this process are lack
of appropriate materials for making evaporative patterns and refractory pattern
coatings. Unlike casting in the sand moulds, the process uses patterns and pouring
systems, which remain in the mould after its making until pouring of metal. This justifies
the title full mould casting. In the contact with liquid metal, the pattern is split in a
relatively short time. At the same time, the castings crystallization takes place. As the
consequence of the pattern splitting, a great quantity of gaseous and liquid products is
produced. If the conditions of their elimination from the mould are not fulfilled, many
defects will appear on the castings, which are considered characteristic for this
process. >1-2@
Important factors on patterns decomposition and evaporation process,
besides temperature and patterns density, are also the type and refractory coat layers
thickness which the evaporable pattern is been covered with, type and size of sand
1
M.Sc., Aurel Prsti, Belgrade, AMI Belgrade, Kneza Miloa 86, 11000 Belgrade, Serbia lucijap@ amiline.rs
Ph.D., Zagorka Aimovi, Belgrade, Universtiy of Belgrade, Faculty of Technology and Metallurgy, 11120
Belgrade Karnegijeva 4, Serbia zagorka@tmf.bg.ac.rs
3
Ph.D., Zvonko Guliija, Institute for Technology of Nuclear and other Mineral Raw materials - ITNMS,
z.gulisija@itnms.ac.rs
4
Ph.D.,Mirjana Stojanovi, Institute for Technology of Nuclear and other Mineral Raw materials - ITNMS,
m.stojanovic@itnms.ac.rs
2
315
Aurel Prsti, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi
grain for modeling, respectively permeability of sand for modeling, castings and gating
of moulds construction. The patterns density and permeability of refractory coat and
sandy cast determine polymers evaporation velocity. The velocity of liquid metal
coming into the cast and its contact with the pattern is regulated by proper defining the
gating of moulds.
In order to obtain castings of a priori desired quality, critical process
parameters should be determined for each particular polymer pattern, as well as the
type of alloy for casting. That requires long-lasting researches with a goal to achieve
optimization of EPC process and obtain the castings of a priori specified properties. In
order to understand correctly the EPC process optimization it is necessary to know that
various types of castings structure determine their different properties. Besides this
dependency for obtaining the castings of a priori specified properties there also should
be determined the fundamental structure dependency on technology, which implies
critical process parameters control and control of useful castings properties, and a
special consideration in this paper was given to that matter. >3,4@
2.
EXPERIMENT
A four series of experiments were done with the goal of analyzing the possibilities
for appliance of refractory coats based on talc (series mark: T), cordierite (series mark:
C), zircon (series mark: Z) and mica base (series mark: M) in EPC process. The coats
compositions were defined (table 1) and the coats components preparation methods
were determined. Grinding and fine grinding of refractory loaders of talc, cordierite,
zircon and mica was done in mill with balls of Cr Ni steel, capacity 20 kg/h, with mill
load of 70% and grinding time 45-60 minutes.
Compositions of used refractory coats on talc, (T), cordierite (C), zircon (Z) and
mica (M) base were:
Refractory powder: T; C; Z; M
Granulation of refractory powder/m /: 35-40
Powder quantity /%/: 85-90
Binding material: bentonite 3,5-5%; Bindal H 3-8%; sodium silicate (water
glass) 2-4%
Coat maintenance material: dextrin 0,3-0,5%; lucel 0,3-0,5%
3
Solvent for achieving the needed coat density: 2 g/cm : water
At applying refractory coat on polymer pattern by techniques of immersion into
the tank with coat, overflowing and coating with brush, a special attention was given to
coats quality control. The basic criteria for quality evaluation of this type of refractory
products were-pertinence for applying, drying behavior, resistance to attrition,
sedimentation and penetration.
The experimental parameters of EPC process regarding which were done the
selection of composition and preparation of refractory coats series: T, C, Z, M were:
Tested alloy: AlSi6Mg3
Preparation methods of liquid die: refinement by compounds based NaCl
and KCl in quantity of 0,1% on die mass; degasification by briquette C2Cl6 in
quantity of 0,3% on die mass; modification by sodium in quantity of 0,05%.
Casting temperature: 755-780C
316
Development of epc process for manufacturing parts in automotive industry
Evaporable polystyrene pattern: density 20-25 kg/m3; pattern construction:

plate (200x50x20) mm and staged probe with different wall thickness: 10
mm; 20 mm; 30 mm; 40 mm; 50 mm; polystyrene grain size 1-1,5 mm
Mounting pattern for casting: "cluster" with four patterns-plates set on
central runner gate and "cluster" with two staged probes set on central
runner gate
Gating of moulds: central runner gate (40x40x400) mm; ingates (20x20x10)
mm, 2 pieces.
Dry quartz sand for cast production: grain size: 0,17 mm; 0,26 mm; 0,35 mm.
The process parameters for production of refractory coats series T, C, Z, M,
methods of coat applying on patterns and drying, when optimal results of testing the
structure and properties of obtained aluminum alloy castings were:
3
Coat densities: 2g/cm
Coat temperature: 25C
The way of coats excess remove from the pattern after pulling out from
the tank for lining: patterns are been seeped, in vertical position, 5-10 s, and
then set 5 s under 45 angle in order to coat layers on patterns surface get
equally even
Slowly coat mixing in tank during the coat applying on pattern: velocity 1
revolutions/min
Methods of coat applying on pattern: "cluster" immersion into tank with

coat; overflowing; coating with brush
Drying: 10h
Coat layers thickness on the pattern after drying: 0,6-1,5 mm.

EPC process phases are shown on Fig. 1.
a) coats excess remove from the pattern
b) staged probes prepared for

casting
317
c) casting
e)easy coat remove from the

casting
Fig. 1 EPC process phases

3.
RESULTS AND DISCUSSION
By controlling the critical process parameters for refractory coats production

and controlling the coats properties it was determined that the coats of all series
comply with conditions for appliance in EPC process. It was determined that coats
were easy to apply on polymer patterns, being equally lining at overflowing and
immersion, were easy to been coat with brush, without any mark of brush, leakage,
drops and clots formation. After drying, the coat surface was smooth; coats layers
were of equal thickness everywhere on patterns surface, without any bubbles, crazing,
peelings or attrition. The coat quality and refractory loaders homogeneity in the coat
depends on coat preparation. In order to achieve even coat layers thickness on
patterns surface it is necessary to slowly constantly mix the coat during its applying on
patterns, to maintain defined density (2g/cm3) and temperature (25C) of coat. On the
contrary, the coat compositions inhomogeneity appears, Fig.2.
318
Development of epc process for manufacturing parts in automotive industry
a) homogenous composition coats
b) non-homogenous composition coats
Fig. 2 Refractory coats

In order to observe the effects of casting process, evaluation of certain operation
phases and analysis of applied refractory coats influence, a visual control of the
obtained castings was done, testing their structural and mechanical properties. After
pulling out the founded "clusters" from the cast, their surface is covered with coat layer
which is easy to be broken and removed from it, so the cleaning is not necessary,
which significantly reduces the production costs. The refractory coats of all series have
demonstrated positive effects on the surface quality shiny and smooth castings
surfaces were obtained. The castings are true copy of the patterns (dimensionally are
precise) which indicates that the decomposition and evaporation of polystyrene pattern
was in totality, and that the gating of moulds solution was satisfactory. It was noted
that the lower castings parts of all series have flat and sharp edges, clean and shiny
surface. At some castings from the series with coat layers of higher thickness (above
1,5 mm) the upper castings surfaces are a bit uneven and folded, and also on certain
castings parts a surface roughness appears, and more often at the castings from the
3
series with patterns density above 20 kg/m .
The study results of castings structural and mechanical characteristics were
within the limits predicted by the standards for this type of alloys. That would be the
castings from series with used polystyrene patterns up to 20 kg/m3, refractory coats of
less thickness layers, below 1 mm, applied quartz sand for modeling with its grain size
above 0,26 mm, casting temperature within the limits of 760-780C and casting velocity
which enabled even decomposition and evaporation of polystyrene, with complete
elimination of gassy products from patterns evaporation, without any cast falling in and
liquid metal penetration into sand.
On the other hand, castings from the series with applied patterns of densities
above 20 kg/m3 and coat thickness above 1,5 mm have expressed subsurface and
volumetric porosity too. This indicates that the reasons for these type of errors are
primarily the polystyrene pattern, and next the refractory coat and high casting velocity.
319
4.
CONCLUSION
In order to attain a quality and cost-effective castings production by the EPC

casting process, it is necessary to attain the balance in the following system:
evaporable polymeric pattern- liquid metal-refractory coating sand mold during metal
inflow, polymeric pattern decomposition and evapouration, castings formation and
solidification. All this points to complexity of the castings solidification conditions by the
EPC casting process, as well as to the necessity to determine the corelation between
the casting parameters, structure and properties of castings.
REFERENCES
[1]
[2]
[3]
[4]
R. Monroe: Expandable Pattern Casting, AFS Inc. (1994), SAD

M.M.Risti: Principles of materials science, Monographs. Vol. DCXVII, No 36, Serbian
Academy of Sciences and Arts, Belgrade (1995).
. Josipovi: The influence of the technological parameters on the quality of refractory
coatings based on the ceramic materials for the sand moulds and cores, Master paper,
Faculty of Technology and Metallurgy Belgrade (2003), (in Serbian).
Z.Aimovi, Lj.Andri, V.Miloevi, S. Milievi: Refractory coat based on cordierite for
application in new evaporate pattern casting process, Ceramic International, 2011 (in
press)
ACKNOWLEDGE
These investigations were conducted under the Project 34002 funded by the Ministry
of Science and Technological Development of the Republic Serbia
320
EFFECT OF SURFACE ROUGHNESS ON WOOD ADHESION

Obuina Muro1, kalji Nedim2, Smaji Selver3
Summary: This paper presents research impact planed surface roughness of beech,
fir, oak and thermally modified beech-wood (220C) samples, the strength of bonding.
The samples were planed in the radial direction. Tools for planing had a double-edged,
planed and the displacement at a speed of 12 m/min. The depth of the milling
operation was 1.00 mm, rake angle blade tool was = 15. at 6000/min, 5000/min,
and 8000/min. Diameter of the cutting edge was = 125 mm. Surface roughness is
measured by the length of treatment using an electrical-mechanical profilmetra
Mitutoyo SJ-201. Samples were glued with PVAc glue. The lowest roughness had
samples of oak, then fir and beech the highest, and thermally treated beech. The
highest shear bond strength, they had samples of beech, followed by oak, fir, and had
the lowest strength of the samples we did thermally modified beech.
Key words: roughness, bonding strength, planing, cutting conditions..
1.
INTRODUCTION
Besides the physical, mechanical and anatomical wood properties, surface

quality, detail and finished products is influenced by numerous factors such as the
direction of cutting edge, geometry of the blade, and its sharpness, thickness of the
cut part, lack of precision of the sharpening tool, technological parameters (speed of
cutting, speed of movement, etc.) (Richter et al, 1995). It is assumed that the surface
roughness of cut surface Softwood sawmills and increases with increasing radius of
the wedge (Keturakis, 2007). Quality of treatment involves the precision machining and
quality of machined surface. These two mutually dependent processing quality
indicators, which are dependent on many factors, represent the most important
condition for achieving the required product quality. The technological processing of
wood glue holds very important place. It significantly affects the practical value and
quality of wood products. During bonding simultaneously and in parallel to place more
complex physical and chemical processes, which are a consequence of the properties
of wood and glue, the interaction between wood and glue, and paste elements of the
technological process. As we know, wood is a biological material is hygroscopic,
anisotropic, inhomogeneous and porous, with it, and biologically degradable. In
1
Doc.dr. Obuina Muro, Sarajevo, Mainski fakultet, obucina@mef.unsa.ba

Doc.dr. kalji Nedim, Sarajevo, Mainski fakultet, skaljic@mef.unsa.ba
3
Dipl.ing.ma. Smaji Selver, Sarajevo, Dallas Company, selver_smajic@hotmail.com
2
321
Obuina Muro, kalji Nedim, Smaji Selver
addition to wood, and adhesives are relatively complex. It is as a result of physical and
chemical processes, during the bonding substantially change passing from liquid to
harden the state. Otherwise, the properties of wood and glue are fundamentally
different, which complicates the temporal stability and bonding strength of adhesive
joints. The main goal when gluing wood to achieve a combination of high strength and
long-lasting strength, which in principle should not be less than the strength of the
wood. The strength of the glued joint a lot depends on the interaction between wood
and adhesives, and adhesion of the adhesive. The quality of machined surfaces of
wood depends on the anatomical structure (tree species), and cutting parameters
(depth of operation, velocity, displacement, rotation number of tools, interventions by
the blade, blade angle,...). Tests show that smooth surfaces require a relatively small
amount of paint for surface protection (Marian et al, 1958). The paper is accompanied
by the quality of surface finish (roughness) for milling and its influence on the strength
of the glued joint using a PVAc adhesive. Used waterproof PVAc adhesive is widely
used in the bonding range. The results are presented in tables and graphs on the basis
of these comments and give the derived conclusions.
2.
MATERIAL AND METHODS
Experimental measurements were performed on samples of oak, fir, beech and

thermally modified beech wood. The dimensions of the original samples before planing
and gluing were: 70x21x600 mm steamed beech, oak and fir, a 70 x 21x 500 mm
thermally modified beech-wood at 220C. Moisture content of samples was: Steamed
beech: 9.18%, Fir: 18.27%, Oak: 9.31%, Thermally modified beech-wood: 4.47%.
Samples were radial texture, before planing were conditioned at 20 C and relative
humidity of 65 5%. Experimental samples were planed in the radial direction. Given
that the paper dealt with four types, a little more attention was paid to thermally
modified beech wood. Thermally modified beech-wood is a product that is obtained
through intensive treatment of wood to the unusually high temperatures. This warm
wood permanently changes a number of its chemical and physical properties. Changes
in properties are generally caused by thermal degradation of hemicellulose. These
changes are starting to emerge already at temperatures around 150C and they are
continuing with the continuous increase of temperature. Beech used in the experiment,
heat-treated at 220C. As a result, the natural tendency of wood to warping, swelling
and shrinkage is reduced by about 50%. Thermo timber thus becomes much more
stable than normal wood in varying climatic and environmental conditions. For bonding
it was used PVAc adhesive manufacturers Kleiberit Germany, type Klebit 314.
According to technical specification adhesive properties are: Onecomponent D4
waterproof adhesive for bonding according to DIN / EN 204. The quality of machined
surface, followed by the group of samples were planed down the processing of these
parameters. Machining process was conducted with a cabinet planer (Weinig
Powermat 400). Only the top spindle of the machine with two knives was used at 125
mm tool diameter. The knives were made of industry standard high-speed steel.Planed
o
at the speed of displacement of 12m/min. The used knife rake angle was 15 and the
depth of cut was 1.0 mm. The number of revolutions was 5000/min, 6000/min and
8000/min.
322
Effect Of Surface Roughness On Wood Adhesion
Fig. 1 Prepared samples for planing
Fig. 2 Star press
After planing, the samples were glued. Application with the adhesive was 150
2
g/m , Manual, according to manufacturer's recipe. Bonding was performed on a cold
star presses, molding pressure was 0.7 MPa. After sticking samples were conditioned
for seven days. The temperature at the facility where the samples were glued with
PVAc adhesive was 19C and relative humidity 49%. After the samples conditioned,
accessed the crop specimens for testing shear strength. As a measuring device for
testing the surface roughness was used electromechanical profilmeter Mitotuyo SJ201. Measurements were made along the length of wire and profile obtained is
Fig. 3 Profile roughness (Beech planed at 6000 /min)

3.
Comparing the results obtained for all four types of wood, can be determined,
as far as roughness, that with the increase in processing speed or by changing the
number of revolutions of the shaft, the surface roughness increases. When the
samples were planed at 6000/min, the smallest roughness had samples of oak, then
fir, beech, and maximum, as well as heat-treated beech.Planed samples at 5000/min,
the smallest roughness had samples of steamed beech and oak, while the maximum
roughness of the samples were heat-treated fir and beech.
323
When samples planed at 8000/min, the smallest roughness had oak and
thermally modified beech wood, while the highest roughness had steamed beech and
fir.
Table 1 shows the minimum and maximum values of roughness, which are
obtained by experimental measurements.
Table 1. Minimum and maximum values of roughness at different rpm, obtained by
experiment TMBW- thermally modified beech wood, BW-beech wood, OW-oak wood,
FW- fir wood
RPM
/min
5000
/min
6000
/min
8000
/min
Wood species
1
2
3
4
1
2
3
4
1
2
3
4
TMBW
BW
OW
FW
TMBW
BW
OW
FW
TMBW
BW
OW
FW
Roughness Ra [m]
minimum
maximum
Num.of
Samples
10
10
10
10
10
10
10
10
10
10
10
10
1,81
1,86
1,23
1,14
2,99
2,91
2,2
3,58
1,02
2,47
1,02
3,11
5,45
3,77
3,11
4,07
7,34
7,95
4,53
9,03
2,91
4,98
2,18
4,92
mean
value
3,469
2,581
2,15
3,468
5,605
5,756
3,154
6,641
1,929
3,579
1,498
3,971
st
dev
3,53
3,71
3,99
3,77
0,18
1,82
1,79
1,46
1,40
1,38
1,28
0,99
Fig. 4 Effect of wood species and number of revolutions on the quality of machined
surface
324
Effect Of Surface Roughness On Wood Adhesion
Table 2 Display medium strength in relation to the type of glue, speed and type of wood
TMBW- thermally modified beech wood, BW-beech wood, OW-oak wood, FW- fir wood
Serial
number
II
III
Wood
species
RPM [min-1]
v [m/min]
glue
Vsr [MPa]
ST
DEV
Note
OW
5000
12
PVAc
8.64
4.23
shearing
FW
5000
12
PVAc
6.73
1.5
shearing
BW
5000
12
PVAc
14.17
3.5
shearing
TMBW
5000
12
PVAc
3.42
2.71
shearing
OW
6000
12
PVAc
6.14
1.99
shearing
FW
6000
12
PVAc
7.25
1.52
shearing
BW
6000
12
PVAc
11.92
2.93
shearing
TMBW
6000
12
PVAc
2.19
2.81
shearing
OW
8000
12
PVAc
9.12
1.82
shearing
FW
8000
12
PVAc
6.04
1.91
shearing
BW
8000
12
PVAc
10.15
2.01
shearing
TMBW
8000
12
PVAc
4.83
2.35
shearing
Figure 5 presents the results of shear strength of specimens glued PVAc

adhesive. The Figure shows that the minimum strength independent of the number of
revolutions that has been used for milling a tube made of thermally modified beech, as
well as tubes made of fir.
Fig 5. The mean value of shear strength of PVAc adhesive glue patterns depending on
the number of revolutions
325
By far the greatest strengththat is was obtained ,which can be seen in the
figure, showed a specimen made of beech, and the dependence of the mode of
processing (speed and number of tools) in proportion to show results. Proved to be the
best combination of glued beech, which is planed to Powermat 500 planer with the
head above the workpiece and the four blades of the tool, as well as the number of
revolutions of 8000 /min.
4.
CONCLUSIONS
On the basis of the experimental research of influence the quality of surface

finish on the strength of the glued joint, the respective samples, can be defined by the
following conclusions: The first factor that influenced the strength of the glued joint was
the treatment regimen. Another factor that significantly affected the results obtained by
the surface roughness was investigated after different treatment regimes. Planed
samples at 6000 /min, the smallest roughness had samples of oak, then fir and beech
the highest, and thermally treated beech. According to the procedure or treatment
regimen, which was set up, and the type of wood, we come to the conclusion that the
greatest strength, seeing the number of rotations and glue, show patterns that are
made from steamed beech planed at 5000/min, then samples of oak planed in
8000/min, fir planed at 6000/min, a minimum strength of the samples that had been
made of thermally treated beech wood over the entire rotation.
REFERENCES
[1] kalji N., Obuina M., Beljo Lui R., avlovi A. Utjecaj posmine brzine i vrste
drva na hrapavost obraene povrine, Drvna Industrija, Zagreb, 2009.
[2] Obuina M., Smaji S., kalji N., Beljo Lui R. Effect of Rotation Speed and
Wood Species on Roughness of Machined Surface, Zadar, Croatia, 2010.
[3] Aydin I. Activation of wood surfaces for glue bonds by mechanicak pretreatment
and its effects on some veneer surfaces and plywod panels, Trabzon, Turkey,
2004.
[4] Backovi M. Lijepljenje u tehnologijama prerade drveta, Sarajevo, 1996.g.
[5] Gindl M., Tschegg S. Significance of the acidity of wood to the surface free
energy components of different wood species Wiena, 2002.g.
[6] Kamke F.A., Resnik J., ernek M., Penetration od Liquid UF adhesive into Beech
wood, Virginia USA, 1998.
[7] Marra A. Technology of wood bonding, New York, 1992.g.
[8] Richter, K.; Feist, W.C.; Knabe, M.T. 1995: The effect of surface roughness on the
performance of finishes. Part 1 Roughness characterization and strain
performance. Forest Prod. J. 45(7/8):91-97.
[9] Keturakis, G.; Juodeikiene, I. 2007: Investigation of Milled Wood Surface
Roughness. Materials Science (Mediagotyra), 13 (1).
[10] Marian, J.E.; Stumbo, D.A.; Maxey, C.W. 1958: Surface texture of wood as related
to glue-joint strenght. Forest Prod. J. 12: 345-351
326
FORMING LIMIT INDICATORS IN METAL FORMING

M. Stefanovi1, D. Viloti2, M. Planak3, S. Aleksandrovi4, Z.Gulisija5, D. Adamovi6
Summary: Precise information on forming limit indicators is very important in
designing the technological processes of metal forming. The significance of stress
state and strain history for limit strains realization is emphasized above all. For the
case of upsetting, indicators in the system of principle surface strains are specified, as
well as classic example of FLD, as a dependence of limit strain on stress ratio
coefficient. For the case of sheet metal forming deep drawing, the example of
determining FLD at classic and two-phase - proportional forming is shown. The
specified experimental results are related to the area of stretching and the area of pure
deep drawing of axis-symmetrical pieces.
Key words: Metal forming, Forming limit, Strain path, Upsetting, Deep drawing
1.
INTRODUCTION
When designing the technological processes of metal forming, it is extremely

important to understand the concept of limit formability, which can be defined as the
ability of materials to achieve permanent shape changes, i.e. ability of materials to
accomplish maximal strains in the given forming conditions. The criterion for defining
limit formability can be either fracture or forming instability (appearance of
localizations). The influence of certain factors on the value of limit strain, as a
numerical indicator of materials formability, can be implicitly expressed, by formability
function [1]:
M el
DM
f ( H M , S M , To , M , TV )
(1)
where:
DM material formability,
M el
limit strain,
H M type of material, defined by a particular chemical content,

1
prof. dr Milentije Stefanovi, Faculty of Mechanical Engineering, Kragujevac, Serbia, stefan@ kg.ac.rs
prof. dr Dragia Viloti, professor, Faculty of Technical Sciences Novi Sad, Serbia, vilotic@uns.ac.rs
3
prof. dr Miroslav Plancak, Faculty of Technical Sciences Novi Sad, Serbia, plancak@uns.ac.rs
4
prof. dr Srbislav Aleksandrovi, Faculty of Mechanical Engineering Kragujevac, Serbia, srba@kg.ac.rs
5
prof. dr Zvonko Gulisija, ITNMS Institute, Beograd, Serbia, zgulisija@itnms.ac.rs
6
ass. prof. dr Dragan Adamovi, Faculty of Mechanical Engineering Kragujevac, Serbia, adam@kg.ac.rs
2
327
M. Stefanovic, D. Vilotic, M. Plancak, S. Aleksandrovic, Z. Gulisija, D. Adamovic
S M structural state of the material,

To forming temperature,
x
strain speed,
TV stress state determined by stress tensor.

For established and unchangeable forming conditions (material, speed,
temperature), the possibility for changing limit strain values is, obviously, most
efficiently influenced by the change of stress state, i.e. el f (T ) . For that purpose,
executive elements of the forming system should be created in such a way that the
available formability potential could be exploited up to its maximum. Control of forming
system tools construction, geometry of initial work piece shape and tribological
conditions should have, as a final result, generation of appropriate stress
components in the pressure zone and avoidance of stress in tension zone.
At experimental determining of formability properties, it is necessary to comply
with the conditions of so called strain history. This means that limit strain values must
be established in conditions which satisfy proportional forming rules. Such stress is
realized when outer forces, which load the observed element, increase proportionally
to one general, constant parameter, from the beginning of forming. Due to this, the
ratio of diagonal components of stress tensor deviator (and principle stresses ratio, as
well) is a constant value. Also, the concurrence of main stress and strain axes must
exist throughout the process. In more complex cases, these conditions are not fulfilled,
so the entire process should be divided into several phases stages, within which the
specified conditions are satisfied.
2.
BULK DEFORMATION PROCESSES
In the area of bulk metal forming, there are two methodologies for defining
forming limit diagram. The first methodology is related to determining the limit forming
f (H1 ) .
curve, as principle strains function in the moment of material fracture H 2
Experimental determining of dependence
H2
f (H1 ) is performed by measuring the
elements of measuring grid per forming phases (stages) in fracture zone. At cylinder
compression by flat dies, measuring zone is marked in meridional plane of the piece.
Figure 1 shows the forming limit diagram (FLD) for the case of compression of
cylindrical specimen by conical dies [3]. According to the obtained results, the
dependence among main strains can be presented in the form of expression 2 = 0,150.51. Dependence 2 = - 0.5 1 is valid for completely homogenous uniform forming.
328
Forming limits indicators in metal forming
Fig. 1. Limit strains curve at compression by conical dies

The second procedure is based on establishing the connection between effective
limit strain and stress state indicator in the critical point of the investigated specimen in
which the material structure is damaged (strain localization or fracture). Basically, the
ratio of the first invariant of stress tensor and second invariant of stress tensor deviator
is taken as the indicator of stress state. In the papers of associates from the Laboratory
for metal forming of the Faculty for Technical Sciences in Novi Sad, in the limit
formability area at cold forming, the following ratio is taken as the indicator of stress
state [1], [2]:
3V m
I1
(2)
Ve
3 ID2
where:
I1 - the first invariant of stress tensor,

I D 2 - the second invariant of stress tensor deviator,
Vm
Ve
- mean hydrostatic pressure,

- equivalent (effective) stress.
In disproportional (non-monotonous) processes, stress state indicator

constantly changes throughout the forming process; therefore it is necessary to
introduce mean value of indicator av [1],[2] [5]:
E av
M el
M el
E (M )dM
e
(3)
In case that material damage occurs on free surface of the formed piece
(plane stress state), indicator of stress state E in critical point of the specimen is
determined by the application of forming theory [1], [6] :
329
1 2D
2 D

2
1 2D 1 2D

1
2 D 2 D
1
In the given expression, coefficient D

components in two perpendicular directions:
(4)
is defined by change of strains

d
, i.e. by strain history

dz
f (z ) .
In the case of non-monotonous forming, it is necessary to establish the
dependence of stress state indicator on effective strain f (e ) , and then apply the
expression (3) to calculate mean value of this factor.
The paper [4] presents a simpler methodology for determining the mean value
of stress state indicator, which gives the results equivalent to the ones obtained by
previous methodology and is based on flow theory. The mean value of stress state
indicator in that case is determined based on limit values of main strains components:
E av
M el
(M1l M 2l )
(5)
where:
M1l and M 2l - main natural strains in fracture zone

M el - effective strain in the moment of specimen fracture
Fig. 2 shows the limit forming diagram for steel EN: 100 Cr6, determined by
application of basic methods of deformation: uniaxial tension (=1), torsion (=0) and
compression (=-1). After that, the data obtained by upsetting the cylinders of different
heights (A, B,C) by recessed dies was entered into diagram. Lines of strain history
are drawn in the diagram for particular specimen series: series A specimens of
dimensions 20x35 mm, series B specimens of dimensions 20x40 mm and series C
specimens of dimensions 20x50 mm. According to Fig. 2, the influence of stress
state indicator on limit formability of the material is obvious. Based on that diagram, it
can be concluded that lower values of stress state indicator (processes in which
compressive stress components dominate) provide higher values of limit strains and
vice versa processes in which tensile stress components dominate result in lower
values of limit strain. Because of intensive obstruction of radial material flowing at
cylinder upsetting by recessed dies, lower values of limit strain are realized compared
with data obtained for cylinder upsetting by flat dies. A slight increase of specimen
height leads to increase of limit strain (Fig. 2).
330
Fig. 2. FLD for steel 100 Cr6 and strain history at cylinder upsetting with
recessed dies [5]
Strain history and change of factor during the forming process (Fig. 2) show
that the change of stress state and considerable decrease of forming potential occur in
compression process. The advanced objective is, definitely, to retain the coefficient
in the area of negative values.
Fig. 3 FLD based stress formability index and effective strain [6]
Figure 3 shows the FLD in which the positions of some forming methods are
defined based on estimation of stress state indicators. Well-known theoretical solutions
for stress components were used in the analysis, based on which stress state indicator
E was determined. Vertical broken lines show conditionally that there is proportional
forming at investigation, with linear strain history and constant stress state indicator
331
( const. ), while horizontal arrows show possible moving of particular processes in

FLD conditioned by change of relevant process parameters (contact friction, geometry
of tools and specimen etc.).
3.
SHEET METAL FORMING
In the area of sheet metal forming by deep drawing, it is best to observe and
express the limit formability within the forming space of natural strains. The first
forming limit diagrams were formed in the 60-ties, for the needs of the car industry , in
order to establish the universal approach for estimation of limit values of strain,
independent on stress-strain scheme and geometry of work piece being formed.
Forming limit diagram can be formed in various ways, whereat the conditions of
proportional forming must be complied with. Figure 4. shows the FLD for low-carbon
steel sheet metal, intended for manufacture of passenger car-body parts. FLD was
determined by the Nakazima procedure [7]. The left part of the diagram, with the
negative second principle line strain, is typical for so called pure deep drawing, and the
right part is typical for stretching (both natural main strains are positive). The following
factors have the most significant influence on the limit curves position and shape:
material (type and thickness) experimental methodology, measuring grid size, strain
history.
Q
0,8
0,7
0,6
M
FRACTURE
0,5
localization zone
A
a
E
M
SAFE FORMING
0,1
-0,3
-0,2
-0,1
II

0,2
-0,4
C 0,3
wrinkling zone
M 2M
t
m=0
A /0/1/ M /
I
B /1/0/ M /
I
M
C /1/-1/ /
I
D /-0,5/1/ M /
I
E /-0,5/0/ M /
0,1
0,2
0,3
M M
t
m
Fig. 4 Forming Limit Diagram (FLD)
M M
t=1
m=1
C B
M =0
t=0
m=0,5
0,4
Q
M
I
M
Fig. 5 Different strain paths at twophase forming
Analytical description and understanding of the forming process are rather

simple regarding proportional forming. However, the actual processes most often take
place in conditions different from proportional forming conditions. This is especially
prominent in multi-operational processes of sheet metal forming. The strain path is
completely described by strain trajectory- the trajectory of points which are for the
corresponding measuring fields into the system of main linear strain M1 M 2 for each
332
moment of drawing. Figure 5 shows, respectively, the most often used linear strain
paths for two-phase process [8].
For plane stress state and proportional forming, or each phase of such
forming, the following is valid [8]:
M 2 / M1
(6)
const
or the coefficient of stress ratio for isotropic material can be applied:
V2
V1
where
2M2 M1
2M1 M2
V 1 ,V 2
2t 1
const
2t
(7)
are main stresses on sheet metal plane.
Figure 6 shows FLD forms for various forming histories. The most favourable
strain path is, obviously, uniaxial tension in the first phase (t=-0,5) and biaxial tension
in the second phase (t=1).
0,8
M
0,6
0,5
tm
on
ot.
0,4
t=
-0,5 -0,4 -0,3 -0,2 -0,1
0,1
t=
-1
5
-0,
t=
0,2
0,1
0,2
0,3
0,4
0,5
M
Fig. 6 Complex strain paths and corresponding FLD [7]

In Metal forming laboratory of the Faculty of Mechanical Engineering in
Kragujevac, the influence of strain history on limit formability of thin sheet metals has
been investigated for many years. In the course of that, standard two-phase
investigation schemes were used: the first phase involved the realization of pure
uniaxial or complete biaxial tension, and in the second phase very different forming
was realized [9].
At forming of sheet metal with dominant negative second strain, such as pure
deep drawing, the influence of strain history is more complex. Such procedures
integrate very different stress-strain schemes; therefore, the influence of previous
forming is not easy to illustrate. The results which are presented further are related to
classic, so called pure deep drawing of axis-symmetrical work piece of 50 mm
333
diameter and blank of 100 mm diameter. The results are shown for determining the
successful forming fields in the system blank holding force-drawing depth at
application of various contact conditions (dry surfaces D, lubrication by oil O,
application of polyethylene foil and oil as lubricants F+O), Fig. 7 [7], [10].
80
DC04
Do=100 mm, d=50 mm
70
D
O
F+O
wrinkling curve
fracture curve
no defects
60
h, mm
50
40
30
20
10
0
10
15
20
25
30
FD, kN
Fig.7 Dependence of drawing depth on blank holding force for one drawing phase
Successful forming field is bordered with wrinkle line (full line) on one side and
with fracture line (broken line) on the other side. It is obvious that only in case D there
are both borders. In other cases, for given drawing ratio, there is no fracture at
drawing.
Two-phase strain history was realised as follows: in the first phase, biaxial
uniform tension was performed up to the amount M1 M 2 0,189 , mark t=1
trajectories in Figure 6. Then, for the same conditions as previously, drawing was
performed and successful forming field was determined, Fig.8.
334
80
DC04
Do=100 mm, d=50 mm
70
D
O
F+O
wrinkling curve
fracture curve
no defects
60
h, mm
50
40
30
20
10
0
10
15
20
25
30
FD, kN
Fig.8 Dependence of drawing depth on blank holding force for

two phase forming
In case of ideal lubrication, similar ratios were obtained. However, the increase
of friction in contact leads to drastic reduction of successful forming field, e.g. the
largest depth decreases from 42 mm to 18 mm with the absence of any type of
lubrication. Full drawing depth at oil lubrication, at two-phase path, is realized when
blank holding force increases considerably, from 18 to 22 kN. Due to reduced sheet
metal thickness, total drawing force is smaller.
4.
CONCLUSION
Determining of forming limit diagram has practical significance, because it

enables the designing of bulk forming process with minimal number of forming
operations or phases, which reflects on final work piece quality and increase of
economic effects in production. Optimization of the actual technological process with
formability criterion requires previous experimental determining of forming limit diagram
and detailed analysis of stress-strain state per forming phases. In that sense, it is
possible to apply software packages for numerical simulation of bulk deformation
process, which should enable defining of the position of actual forming phase in
forming limit diagram.
For established and unchangeable forming conditions (material, speed,
temperature), the possibility for changing limit strain values is most efficiently
influenced by change of stress state.
At experimental determining of formability properties, it is necessary to comply
with the condition of so called strain history. This means that limit strain values must be
established in conditions which satisfy proportional forming rules.
In sheet metal forming by deep drawing, when determining FLD, proportional
forming rules must be complied with. It is reasonable to use FLD in conditions
equivalent to the experimental ones. In two-phase strain trajectories, previous uniaxial
335
tension raises the limit strains field in FLD. Thereat, larger local strains are realized, as
well as smaller drawing depths.
If FLD is obtained from sheet metal which was previously biaxial tensed, FLD
curve moves into the lower strains field and the position of FLD depends considerably
on previous strain ratio.
5.
LITERATURE
[1] Vujovi V., Shabaik A., Workability Criteria for Ductile Fracture Trans. ASME J.
Engng Mater. Technol. 1986, 108, pp. 245-249.
[2] Kolomogorov V. L.: Ductility and fracture (in Russian), Metalurgija, Moskva,
1977.
[3] Hartley O., Vilotic D., Planak M., Formability maps in cold forming with conical
dies Conference - ESAFORM 2005, Cluj-Napoca, 2005, Proceed. pp.527-530.
[4] Viloti D., Planak M., upkovi Dj., Alexandrov S., Aleksandrova N., Free
Surface Fracture in Three Upsetting Tests, Experimental Mechanics, (2006) 46:
pp.115-120.
[5] Krainik M., Investigation of material formability in cold bulk forming processes,
MSc work, FTN Novi Sad, 2008.
[6] Viloti D.: Behaviour of steel materials in different forming systems of cold bulk
forming, Doctoral thesis, Faculty of Technical Sciences, Novi Sad, 1987.
[7] Aleksandrovi S., Blank Holding Force and Deep Drawing Process Control,
Monograph, University of Kragujevac, Faculty of Mechanical Engineering,
Kragujevac, 2005.
[8] Stefanovi M., Tribology in Deep Drawing,Monograph, Yugoslav Tribology
Association, Kragujevac, 1994.
[9] Stefanovi M., Aleksandrovi M., Importance of Strain Analysis in TriboModelling in Deep Drawing, World tribology Congress 1997, London, Abstract of
papers, pp. 626.
[10] Aleksandrovic S., Stefanovic M., Significance of Strain Path in Conditions of
Variable Blank Holding Force in Deep Drawing, Journal for Technology of
o
Plasticity, Novi Sad, Vol. 30 (2005), N 1-2, pp.25-36.
336
HEATED TOOL BUTT WELDING OF POLYETHYLEN PIPES

Runev Dobre1, Gligore Vrtanoski2, Ljupo Trpkovski3
Summary: The paper shows a part of the investigation results of the heated tool butt
welding of PE100 pipes, with a diameter 160 mm and with thickness of 8.00 mm.
Wildings were performed on several tests, during which a part of the parameters of the
welding regime were changed. Changes were made over the time of heating of the
tool, time of affection and intensity of pressure of the pipes over the tool - plate.
Serial of examinations for establishing the quality of the welded joint were performed
over the tests. A visual control and test of resistance to internal pressure was
performed. Test specimens were made from the welded pipes and a tensile and
bending test was performed.
Key words: polyethylene pipes, PE 100, butt welding, heated tool - plat
1. INTRODUCTION
The use of pipe installations of plastic, polymer materials: pipes, knees,
pockets etc have been more and more implemented. For production of pipes and pipe
elements the most used polymer materials are: the polyethylene (PE) and
polyvinylchloride (PVC), also polyamide (PA) polypropylene (PP). [1.2]
The joining in production and installation of polymer elements is usually
performed by welding. According to DIN 1910 part 3, depending on the heat source
welding of polymers can be performed by: heated tool, heated gas, light beam,
ultrasonic, friction and high frequency heating. [3]
Welding of polymers is performed by combinations of heat and pressure, with
or without use of a filler material, depending on the shape of the joint and the thickness
of the basic material.
The development of the polymer welding begins in the year of 1935 with the
o
discover of polyvinylchloride (PVC). On 180 C PVC is in thermoplastic condition and
by use of pressure homogeneous joint is achieved. In 1938 welding by heated gas was
patented. In the second half of the fifties begin the intensive development and
implementation of the automatic polymer welding. [4.5]
1
Prof. PhD, Dobre Runev, Ss. Cyril and Methodius University Skopje, Faculty of Mechanical Engineering,
(dobre.runchev@mf.edu.mk)
2
Prof. PhD, Gligore Vrtanoski, Ss. Cyril and Methodius University Skopje, Faculty of Mechanical
Engineering, (gligorche.vrtanoski@mf.edu.mk)
3
MSc. Ljupo Trpkovski, Ss. Cyril and Methodius University Skopje, Faculty of Mechanical Engineering,
(lj.trpkovski@student.mf.edu.mk)
337
Dobre Runev, Gligore Vrtanoski, Ljupo Trpkovski
The last ten years the research of polymer welding has become more
intensive, because of the wider use of the polymer materials in the production of
important elements, as well as development of new kinds of polymer materials and
new technologies for their welding, like laser welding or hybrid polymer welding by
simultaneous perform of the laser beam and other heat source. [6.7]
Confirmation of the increased interest for research in this welding field is the
fact that from the beginning of 2007 DVS has begun with editing of a special scientific professional journal under title Joining plastics (Fgen von Kunststoffen), where the
latest scientific achievements are being presented. Few years before that, from 2004 to
2010 were issued more expert books about polymers and their joining. [6.9]
2. HEATED TOOL BUTT WELDING
This procedure is used for welding of polymers in shape of plates or pipes. In
the case of plate the joining is performed by heated tool butt welding or by overlapped
welding, and in the case of pipe joining it is performed by heated tool butt welding or
by overlapped welding so called telescopic weld.
The surfaces that are joining are being heated until soften, with the help of the
heated tool, which can have different shape: plate, ring etc. The tool is heated most
frequently by electrical resistance.
Depending on the shape and the dimensions of the welded elements, this
procedure can be performed manually, mechanically or automatically with a special
welding device.
The welding processes it self is performed in several phases, shown on Fig. 1.
a.
b.
c.
d.
f.
Fig. 1 Phases of heated tool plate shaped butt welding

At the beginning the pipes are being put in two part bends clamps for accepting
338
Heated tool butt welding of polyethylen pipes
(a), after that it moves and makes alignment of jointing surfaces (b). In the next phase
the pipes are being separated and between them we position fitting with a milling cutter
which simultaneously makes alignment of both of the pipes, which are at the same
time pressured with a constant strength of the tool. After the planned heating time is
passed, the plate is released and the heated and softened jointing surfaces again are
pressured one to another, with certain pressure in certain time (e), after what the pipes
are being released from the bends clamps and that means that the welding is finished
(f). The process of welding is performed in time measured from several seconds to
several minutes; it depends on the type of the polymer, the shape and the dimensions
of the welded pipes.
3. SAMPLE BUTT WELDING
Heated plate tool butt welding was performed on polyethylene pipes of PE 100,
with external diameter 160 mm and thickness of the wall of 8.00 mm.
The welding was performed according to the regime defined in DVS 2207-1,
[9], having in mind the recommendations of the welding equipment producer. The
welding regime is chosen depending on the type of the polymer, the shape and the
dimensions of the welded samples, the area of the cross section that is being welded.
Fig.2. shows a graphic representation of the welding regime, with all the
welding phases and the changes of the temperature and the pressure during time.
o
T [ C]
p1
p [N/mm
p5
p2
t1
t2
t3
t4
t5
t [s]
t6
t7
Fig. 2 Regime of heated tool plate shaped butt welding

The welding regime is consisted of the following parameters:
o
T [ C] heating temperature of the tool - plate,
p1 [N/mm2] jointing surfaces pressure on the heated tool during the
alignment,
p2 [N/mm2] jointing pressure on the heating element,
p5 [N/mm2] pressure between the jointing surfaces during cooling time,
t1 [s] alignment of jointing surfaces,
339
t2 [s] time of preheating of the jointing surfaces,

t3 [s] time of the removal of the heating plate,
t4 [s] time to achieve jointing pressure,
t5 [s] cooling time of the pieces to be welded at jointing pressure,
t6= t4 + t5 [s] welding time.
t7= t1 + t2 + t3 + t6 total welding time.
Larger number of samples was welded, some of the parameters were with
constant value, and others were changed in different levels.
Constant parameters were:
T = 210oC heating temperature of the tool plate,
p1 = 0.15 [N/mm2] jointing surfaces pressure on the heated tool during the
alignment,
p2 = 0.38 [N/mm2] jointing pressure on the heating element,
t1max = 8 [s] alignment of jointing surfaces,
t3max = 8 [s] time of the removal of the heating plate,
t5 = 18 [min] = 1080 [s] cooling time of the pieces to be welded at jointing
pressure variable parameters:
p5 = 1.9 2.3 [N/mm2] pressure between the jointing surfaces during cooling
time,
t2 = 112 - 124 [s] time of preheating of the jointing surfaces,
t4 = 7 - 9 [s] time to achieve jointing pressure.
Welding of the samples was performed by combination of the variable welding
parameters given in table 1.
Table 1 Samples welding regime
t2 [s]
t4 [s]
p [N/mm2]
S1
112
7
1.9
S2
124
7
1.9
S3
112
9
1.9
S4
124
9
1.9
S5
112
7
2.3
S6
124
7
2.3
S7
112
9
2.3
S8
124
9
2.3
S9
118
8
2.1
4. INVESTIGATION OF THE WELDED SAMPLES

More examinations were made upon the welded samples for establishing the
quality of the butt weld. A visual control of the weld was performed upon all of the
samples, according to DVS 2202-1.
The mechanical examinations of share strength and bending were performed
on a standard specimen made from each welded sample. The examination of share
strength was performed according to DVS 2207 part 5 and DIN 53455, the examination
by bending was performed according to DIN 53452.
An examination of impermeability by inner pressure according to ISO 1167 was
performed upon special welded samples, combining the parameters like: S1, S8 and
S9.
5. RESULTS AND ANALYSIS
By visual control of the welded samples and the control of the longitudinal cut
samples, from in and outside, flash shapes were noticed at the jointing, fig. 1e and 1f.
340
Heated tool butt welding of polyethylen pipes
No mistakes were established at the weld and its surrounding, according to

DVS 2202-1.
During the examinations by share strength all the samples broke at the basic
material far from the welded jointing, Fig 3.
Fig. 3 Share strength examined samples

The maximal hardness of the tested samples is in the interval Rm = 21.25 22.24 N/mm2, which are the standard values for this type of material. Capacity of the
welded joint is bigger than the capacity of the basic material and the welding does not
change the capacity of the basic material.
During the examination of bending 22 of total 27 welded samples, were bended
without breaking to angle of 160o, which is in the interval of standard values for this
type of material. Three samples showed a crack in the welded joint at angle of 120o,
and 2 samples showed a crack in the welded joint at the beginning at angle smaller the
10o, Fig.4.
Fig. 4 Samples examined by bending

Three separately examined samples, according to ISO 1167, were examined
on water impermeability by internal pressure on 20 oC temperature. All of the three
samples were prepared by the necessary supports and cups and than were sunk in
341
water pool. The test was finished with a resistance on internal pressure of 2.177 to
2.194 MPa during 60 minutes. All the samples resisted the test in determined time
without showing of permanent deformations, which confirmed the impermeability of the
weld and the basic material.
After the standard test, all the samples were tested on the maximal pressure of
the testing installation of 5.00 MPa. Permanent deformations were noticed at the basic
material, but a break explosion did not occur, which confirmed the impermeability of
the welded samples, Fig. 5.
Fig. 5 The samples before and after the examinations on impermeability

6. CONCLUSION
Butt welded joint of polyethylene pipes performed by heated plate perform a
high quality. By use of optimal parameters in the welding regime we get a welded joint
without mistakes, with hardness higher than the hardness of the basic material, and
flexibility like the basic material, impermeability like the basic material.
Considering the high quality of the welded joint and the high mechanical
characteristics, the polymer pipe installations heated tool butt welded can be used as
elements in installations under high working pressure.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
342
Kienle, U. (1998). Schweien von Kunstoffrohrmaterial, Jahrbuch Schweitechnik

99, DVS Verlag, p. 125-134.
Renneebrg, H., Schneider,W. (1998). Kunststoffe im Anlagebau (Werkstoffe,
Konstruktion, Schweiprozesse, Qualittssicherung), DVS Verlag.
DIN 1910 Teil 3:1977, Schweien von Kunststoffen Verfahren.
Hessel. J., Land, W., Schwarz, O. (1996). Schweien von Kunstoffen, Jahrbuch
Schweitechnik 97, DVS Verlag, p. 235-246.
Haberstroh, E., Wolters,L., Stagmaier, M. (2000) Kunststoffeschweien Teil 1,
Handwerkliche Verfahren, Schweissen & Schneiden, Heft 4, p. 214-224.
Potente, H. (2004). Fgen von Kunststoffe (Grundlagen, Verfahren, Anwendung),
Karl Hanser Verlag Mnchen Wien.
Wilhelm, G. (2004). Handbuch Kunststoff - Verbindungstechnik, Karl Hanser
Verlag Mnchen Wien.
Franck, A. (2006). Kunststoff - Kompendium, Karl Hanser Verlag Mnchen Wien.
N, N. (2006). Taschenbuch DVS-Merkbltter und Richtlinien Fgen von
Kunststoffen, DVS Verlag.
INCREASING PRODUCTION EFFICIENCY THROUGH CASTING

QUALITY IMPROVING BY ELECTROMAGNETIC FIELD APPLYING
Marija Mihailovi1, Aleksandra Patari2, Zvonko Guliija3, Miroslav Soki4
Summary: The improved production efficiency of heat treatable Al alloys, as the result
of applied electromagnetic field during continuous casting process was investigated.
The mechanical characterization of continuous casted Al-alloys EN AW 2007 and En
AW 2024 was done on testing machine Zwick/Roell Z 100. The microstructure of ascast samples was examined, as well. The results have shown that low frequency
electromagnetic field significantly influenced the microstructure and mechanical
properties of as cast ingots, and thus through increased castings quality, operation
time and energy savings, the production efficiency was improved.
Key words: Al alloy, electromagnetic field, casting, production efficiency
1. INTRODUCTION
The heat treatable Al alloys EN AW 2007 (AlCu4Pb Mg). and EN AW 2024
(AlCu4Mg1Mn) are designed for plastic processing and belong to high strength group
of alloys with wide application in almost all fields of industry. Their processing involves
a series of technological operations (modification, casting, homogenization, pressing
forming and heat treatment) >1,2@. Although long lasting and expensive process,
because of high consumption of electrical energy, the homogenization is unavoidable
in terms of obtaining an alloy with appropriate properties during the conventional
processing.
The main idea is to avoid or shorten this process in order to increase
production efficiency. The conventional horizontal or vertical direct chill casting due to
unbalanced strengthening conditions, leads to appearance of many defects. These
are: inhomogeneous microstructure, porosity, hot cracks, non-uniform grain size and
crystal segregation in the ingot. All this leads to deterioration of mechanical properties
of strength and toughness, first of all. In attempt to eliminate the specified
1
M.Sc., Marija Mihailovi, Belgrade, Institute for Technology of Nuclear and other Mineral Raw materials
ITNMS, m.mihailovic@itnms.ac.rs
2
M.Sc., Aleksandra Patari, Institute for Technology of Nuclear and other Mineral Raw materials ITNMS,
a.pataric@itnms.ac.rs
3
Ph.D., Zvonko Guliija, Institute for Technology of Nuclear and other Mineral Raw materials ITNMS,
4
Ph.D., Miroslav Soki, Institute for Technology of Nuclear and other Mineral Raw materials ITNMS,
m.sokic@itnms.ac.rs
343
Increasing production efficiency through casting quality improving by electromagnetic field applying
disadvantages, various methods have been applied worldwide: powder metallurgy,

ultra-sound and mechanical vibrations. Unfortunately, those procedures are either too
complicated and expensive or unsatisfactory efficient. It is shown in literature that the
procedure of casting under the influence of electromagnetic field is much easier and
more efficient >3-6 @.
During casting in the electromagnetic field presence, under the effect of
alternating current, an alternating magnetic field is generated by the inductor and the
melt can be inductively stirred. Generally speaking, improvement of microstructure,
reduction of porosity and stimulation of heterogeneous nucleation is caused by forced
convection generated by electromagnetic forces, resulting in colder melted metal flow
from edges towards the centre and vice versa, establishing more balanced
temperature field. The more uniform temperature through a cross section of ingot also
reduces tensions, which arise during strengthening. Reduction of metallurgical defects
in ingots can also be expected. This paper presents the investigation of applying
electromagnetic field in the procedure of Al alloys casting, with the purpose to influence
the crystallization process in order to obtain modified microstructure and improved
mechanical properties. The possibility to avoid some technological operations in
course of obtaining (modification) and processing (homogenization) of these alloys,
which would lead to significant economic effect, is also presented here.
2.
EXPERIMENTAL
The chemical composition of the used alloys EN AW 2007 and EN AW 2024 is

shown in Table 1.
Table 1 Chemical composition of the used alloys
Content/
Alloy
EN 2007
Si
Fe
Cu
Mn
Mg
Pb
Ti
0.30
0.30
3.92
0.80
0.95
1.10
0.01
EN 2024
0.09
0.22
4.10
0.60
1.28
0.01
The medium frequency induction furnace, with capacity of 100 kg was used for
melting alloys. The both ingots, of alloy 2007, with diameter 50 mm and of alloy
2024, with diameter 60 mm were obtained by horizontal continuous casting with
pulse draw-out. There was a drainpipe with graphite crystallizer that is intensively
cooled with water at the bottom of the furnace The low frequency magnetic field is
placed around the crystallizer itself, >5,6@.
The casting temperature was in the range between 710 and 720C and
average casting speed was 1,5 mm/s. The operating parameters during the casting of
ingots were strictly controlled and defined by various values of current (A), frequency
(Hz) and strength of electromagnetic field (At). The number of coil turns was N=40.
Frequencies were selected in the range from 1830 to 30 Hz.
The samples for microstructure analysis were taken on a cross section of an
ingot. The microstructure was examined after the usual metallographic preparation and
etching in Kellers reagent (revealing morphology of Al segregation-solid solution and
inter metallic phase) and anode oxidation with Barkers reagent (revealing size and
344
Marija Mihailovi, Aleksandra Patari, Zvonko Guliija, Miroslav Soki
shape of the grain in presence of dendrite segregation). For the quantitative

microstructure analysis the image analysis device Leica Q500MC was used.
The tensile mechanical testing of the alloys was performed on testing machine
Zwick/Roell.
3
The microstructure has shown that Al segregation from the solid solution
resulted in celluar/dendritic morphology. The structures of sample 1, casted without the
electromagnetic field presence, and samples 4 and 5, casted under the influence of
high frequency electromagnetic field, are more dendritic compared to the samples 2
and 3, which were casted with the presence of low-frequency electromagnetic field.
Obviously, the application of high frequency has no effect on the improvement of
microstructure. Quite oppositely, at 1830 Hz the structure was rougher compared to
one obtained without the influence of electromagnetic field. On the other hand, the
regions of extracted intermetallic phase, in the form of eutecticum or individually,
become finer with the electromagnetic field introduction and with frequency decreasing
from 50 Hz in sample 2 to 30 Hz in sample 3.
It is determined that the porosity of interdendritic type is reduced from sample 2
(casted at 50 Hz) towards the sample 3 (casted at 30 Hz), i.e. it is decreased with the
decrease of electromagnetic frequency. It is also found that the grain size is
significantly reduced from the sample 2 to the sample 3. This is important because the
less amount of porosity and the smaller grain size contributes the better quality of
ingots and thus to mechanical properties. Since it was concluded that there are no
positive effects of high frequency to the obtained microstructure, the further
investigation included only the mechanical properties of samples obtained by
application of lower frequencies, i.e. of 50 Hz and 30 Hz, and also of the one obtained
without the influence of electromagnetic field made of the same alloy EN AW 2024.
For mechanical testing, the Zwick/Roell Z 100 device was used. The samples
for this tensile testing were prepared according to JUS C.A4002 and for the hardness
testing according to JUS C. A4.103. The values of mechanical properties are given in
Table 2.
Table 2 Mechanical properties of alloy 2024 ingots
Sample
mark
2
3
Rp0.2
(Mpa)
198.1
246.6
Rm
(Mpa)
243.2
274.2
A
(%)
1.2
0.7
HB5/25//30
93.5
107.0
The stress-strain curves can be seen in Figure 1.
345
Increasing production efficiency through casting quality improving by electromagnetic field applying
(a) sample 2
(b) sample 3
Fig. 1 Stress-strain curves
Based on the previous microstructure analysis, the increasing trend of alloy

resistance properties could have been expected [7]. However, the decrease of
plasticity for specimen 3 can be interpreted by the appearance of rough continually
extracted intermetallic phase in interdendritic regions, Fig 2, in relation to specimen 3.
This continually grid of intermetallic phase particles has detrimental effect on
mechanical properties, especially on plasticity [8]. Furthermore, one should have in
mind the fact that the values of mechanical properties for specimens of alloy 2024
casted without the influence of electromagnetic field were: Rp0.2=162.5MPa, Rm=
179.9MPa and A=0.49%. i.e.they were the lowest.
Fig. 4 Intermetallic phase extracted in interdendritic regions, sample 3

CONCLUSION
The investigations results of alloy EN AW 2024 and EN AW 2007 ingots
obtained under different operating conditions, with or without the presence of
electromagnetic field, clearly show its positive effect on the microstructure and
mechanical characteristics. It has been proved that only the application of low
frequency electromagnetic field (d 50Hz) has the positive effects, which obviously
change strengthening conditions. Namely, when the frequency decreases (from 50 Hz
346
Marija Mihailovi, Aleksandra Patari, Zvonko Guliija, Miroslav Soki
in sample 2 to 30 Hz in sample 3), the grain size decreases as well, what is noticeable
through the finer microstructure and its uniformity throughout the cross-section. It is
also determined that the porosity of interdendritic type is reduced from sample 2 (50
Hz) towards the sample 3 (30 Hz).
All of this contributes the better mechanical properties and thus the quality of
ingots. The better mechanical properties in the initial stage of production process
provide less consumption of electrical energy. Obtained results indicate that some
steps in conventional technological process can be avoided, namely better surface
quality contributes the surface machine processing elimination. Besides, the finer
microstructure gives the possibility to shorten or eliminate the homogenization process,
one of the longest and the most expensive process during the production of aluminum
alloys. This is the relatively non expensive way in terms of investments needed, to
achieve both, the improved castings quality and the increased production efficiency.
ACKNOWLEDGEMENT
The authors wish to acknowledge the financial support from the Ministry of Science
and Technological Development of the Republic of Serbia through the project 34002.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Ptari, A., Guliija, Z., Markovi, S., Microstructure Examination of

Electromagnetic Casting 2024 Aluminum Alloy Ingots. Practical Metallography, 44
(2007) 6, p.290-298
Stefanovi, M., Guliija, Z., Jordovi, B., Patari, A., Hot Forging of High Strenght
Al Alloys. Journal for Technology of Plasticity, vol 32 (2007), No 1-2, p.29-38
Patari, A., Guliija, Z., Jordovi, B., Effect of Electromagnetic Field on the
rd
Microstructure of Continual Casting Al 2024 Alloy Ingots, 3 International
Conference Deformation Processing and Structure of Materials, September 2022, 2007, Belgrade, 141-149.
Patari, A., Guliija, Z., Mihailovi, M. (2009), The possiblity for application the new
production process for casting aluminum alloys, Machine Design 2009, Ed. Sinia
Kuzmanovi, University of Novi Sad, Faculty of Technical Sciences, p.431-434
Vives C., Ricou, R. Experimental Study of Continuous Electromagnetic Casting of
Aluminium Alloys; Metallurgical Transactions B, vol 16, 1995, p 373.
Meyer J.L., Ricou, R, A Comprehensive Study of the Induced Current, The
Electromagnetic Force Field, and the Velocity Field in a Complex
Electromagnetically Driven Flow Sistem, Metallurgical Transaction B, vol 18,
1997, p. 529.
Zhiqiang, C. Fei, J., Xingguo Z., Hai H., Junze J., Microstructures and mechanical
characteristics of electromagnetic casting and direct-chill 2024 aluminum alloys,
Materials Science and Engineering A237 (2002) p.133-137
Dong, J., Cui, J.Z., Le, Q.C., Liquidus semi-continuous casting, reheating and
thixoforming of a wrought aluminum alloy 7075, Materials Science and
Engineering A345 (2003) p.234-242
347
INFLUENCE OF ORGANIZATIONAL CULTURE ON BUSINESS

PERFORMANCE
Zorana Tanasi1, Goran Janji2, Bobrek Miroslav3, ivko Pejainovi4
Summary: Organizational culture is one of the key factors influencing the process of
organizational design and implementation organizational changes. As such, it is the
subject of research and analysis in this paper. In this context, the analysis of the core
values of employees which could affect its formation and reverse effect that it has on
performance of organizational processes has been done. Organizational cultures
associated with a group of special and in a particular way measurable parameters in a
company which represent the input requirements in the organization design.
Key words: organizational culture, design, human resources
1. INTRODUCTION
We are living in time of increasingly rapid, dramatic, complex and unpredictable
changes. Technological, market-economic, political, social, and global factors influence
the rapid changes as well as organizational theory and organizational design. There is
simply no way to escape the reality that individuals, groups, nations, states, and
international organizations are affected, albeit unequally, by increasingly complex and
rapid changes occurring at all levels of society. Implementation of methods of radical
change in business is becoming a necessity in the process of sustainability and
creating new competitive advantages. In order for organizations to be successful in
such conditions, they must incorporate in their structure flexibility and willingness to
change rapidly. Many theorists and practitioners have attempted to correct deficiencies
of traditional approach (functional) in the design of organizations by introducing new
organizational forms [1].
New technologies, information and communication systems, require new
people to manage them in order to avoid disasters and reduce risks. These are people
with new skills, new abilities and a desire for innovative and creative work.
Expectations from employees are increasing, not only to be experts in their fields, but
also to be accountable to the associates, project costs, etc. In other words, they are
responsible for managerial jobs. Individuals will need to have technical, analytical,
social and conceptual skills that should be harmonized with the environment and time.
1
Zorana Tanasi Ph. D., Banja Luka, Faculty of Mechanical Engineering, (tami9997@teol.net)
Goran Janji MSc., Banja Luka, Faculty of Mechanical Engineering, (goran.janjic@yahoo.com)
Miroslav Bobrek Ph. D. Prof., Banja Luka, Faculty of Mechanical Engineering, (m.bobrek@mp.vladars.net)
4
ivko Pejainovi Ph. D. Assistant Prof., Banja Luka, Faculty of Mechanical Engineering, (zivkop@teol.net)
2
3
349
Zorana Tanasi, Goran Janji, Miroslav Bobrek, ivko Pejainovi
The Japanese achieve this with the help of mentors, while the Americans rely more on
free thinking individuals. The seed of intelligence may be at an early stage of
organization development, but the seed will need to have a culture in which it will
develop and cultivate to become a successful organization.
Organizational culture consists of shared values, opinions and norms of
behavior that have emerged over time and which employees should adhere to. It is the
result of cumulative design decisions taken in the past and management behavior that
result from those decisions [2]. Thereby, organizational culture is a set of key
contingent conditions for the successful implementation of a defined strategy, but is
also affected by the behavior of employees. In order to manage culture, you need to
understand what culture is, what content culture covers, and how to assess it. It is
dangerous to oversimplify this concept because of the illusion that one is managing
culture when one is, in fact, managing only a manifestation of culture and, therefore,
not achieving ones goals [3]. Thinking about organisational culture at three distinct
levels as shown in Figure 1 makes it easier to gain awareness of the operative cultural
forces that affect new organisation design implementation [3], [4].
Artifacts and
behaviours
Visible organizational
structures and processes
(hard to decipher)
Espoused Values
Strategies, goals, philosophies

(espoused justifications)
Basic Underlying
Assumptions
Unconscious, taken for granted beliefs,

perceptions, thoughts and feelings
(ultimatte source of values and action)
Fig. 1 The Three Levels of Culture

2. NEW MODELS OF ORGANISATIONAL DESIGN
Without a model, it is difficult for a manager, owner or professional team to
describe or think about organization in a holistic way. They tend to think only about the
structure (i.e. organization chart). However, with such a narrow focus, they can't reveal
the necessary alignment of all the elements that make up completely functional
organization. Present models in organizational practices provide different perspectives
of organizational systems, but no one perspective is right, the one that makes sense
depends on circumstances, culture and context, among other things.
The modelling process allows a more complete (though not fully complete)
assessment of the appropriate product requirements. Systems model of organisational
design in common use are those conceived by consulting firms such as McKinsey, or
individuals such as Jay Galbraith, Marvin Weisbord, David Nadler, Warner Burke and
George Litwin (the Burke-Litwin model). Organisation design models do not result in a
350
Influence of organizational culture on business performance
product like a physical building or a garment that meets customer requirements. But
they do provide a good conceptual basis for developing an effective organisation
design. Different models present various perspectives of the organisational system, in
the same way that an astronomer standing on each of the planets would present a
different perspective of the universe [5].
One model that needs a special analysis is the Star Model. It was developed by
Jay R. Galbraith, the famous guru of organizational design, who emphasizes a holistic
view of the organization in a model of organizational architecture. The choice of model
also depends on how complex users want it to be how fancy, how usable, how costly
in terms of implementation, how adaptable to changing circumstances, and so on.
Galbraith's model emerged in 1973 and was revised a few times by Galbraith and his
followers based on new scientific discoveries and practical knowledge (Galbraith,
1773, 1977, 1994, 2002, 2005, 2008; Galbraith, Downey and Kates, 2002; Kates and
Galbraith, 2007 and others). The model has been adapted to new business conditions
and more dynamic environment. It is considered a leading theoretical model in the field
of organizational design.
STRATEGY
STRUKTURE
PEOPLE
REWARDS
PROCESSES
BEHAVIOUR
Performance
Culture
Fig. 2 Galbraiths Star Model

In the Star Model, as shown in Figure 2, design policies fall into five categories.
The first is strategy, which determines direction. The second is structure, which
determines the location of decision - making power. The third is processes, which has
to do with the flow of information (they are the means of responding to information
technologies). The fourth is rewards and reward systems, which influence the
motivation of people to perform and to address organizational goals. The fifth category
of the model is made up of policies relating to people (human resource policies), which
351
influence and frequently define the employees mind - sets and skills. It shows that
managers can influence performance and culture, but only by acting through the
design policies that affect behavior [5], [6].
New models are starting to emerge because organizations need to respond to
changes in society, technology, economics, environment, politics, laws, and all that
surrounds them. These models appear in the field of complexity theory, quantum
theory and non-western cultural traditions (rejection of traditional organizations) and
views. One way of developing an adaptive design is to look less at models and more at
design principles or questions that are applicable to specific business strategies. This
argues for a kind of a pattern book of organisation design models for an enterprise but
not a single model being used across the enterprise (Figure 3, Figure 4) [4].
Internal
External
Personality
Individual values
and beliefs
Individual
Personal integrity
"Walk the talk"
Character
Individual actions
and behaviours
Personal alignment
Values
alignment
Colective
Mission
alignment
Structural alignment
Culture
Group values and
beliefs
Social structures
Group actions and
behaviours
Group cohesion
enhanced capacity for
collective action
Group resilience
enhanced mission
assurance
Source: Wilber, K.: Summary of My Psychological Model or, Outline of An Integral

Psychology, 2000, in Barrett, R.: Building a Values-Driven Organization: A Whole-System
Approach to Cultural Transformation, Butterworth Heinemann, 2006.
Fig. 3 Ken Wilber's AQAL Model

Realistic appraisal of aspects of culture and group process followed by
effective management of them goes some way towards achieving either the desired
outcome of an organisation design or being successful in its realization.
352
Influence of organizational culture on business performance
Retention
(incremental
innovation)
Critical tasks
Single general manager
and executive team
Single, overarching vision
Multiple competing
strategies
People
Formal
organisation
Selection
Revolutionary
change
Critical tasks
People
Variation
(discontinuous or
architectural
innovation)
Culture
Formal
organisation
Critical tasks
People
Culture
Revolutionary
change
Culture
Formal
organisation
Multiple congruent
relations between:
Strategy
Critical tasks
Formal organisations
People
Culture
Source: Nadler, D., Tushman M.: The Organization of the Future: Strategic Imperatives and
Core Competencies for the 21st Century, Organization Dinamics, Vol.28, Issue 1, 1999.
Fig. 4 Nadler's Updated Congruence Model

Focus is on aspects of organisation design that are explicit, whereas much
about culture is implicit and difficult to describe because it is socially construed and
manifested in norms, behaviours, expectations and the way we do things round here.
Nevertheless, as Edgar Schein says in The Corporate Culture Survival Guide: Culture
matters. It matters because decisions made without awareness of the operative
cultural forces may have unanticipated and undesirable consequences [4].
In recent years, organizational culture is the component of the organization and
management studied most. In the Republic of Srpska, organizational culture as an
important management concept has not emerged yet. Given the great difference that
exists between our economy and the economies of developed countries, any delay
lasting less than a decade in the processing of a phenomenon in the management
area could be considered acceptable, which reminds us that it is high time to pay more
attention to this issue.
353
3. CONCLUSION
Organizational culture is one of the key factors influencing the process of
organization design and as such is a subject of research and analysis in this paper. It
determines how the company analyzes its internal and external environment and
assesses its own resources and abilities.
Further, coordination and communication as well as the process of creating an
innovative organization depend on the organizational culture. As a result of turbulent
environment of the organization, it may be noted that organizational culture is a factor
that is most closely associated with all dimensions of organizational design.
In an analysis of organizational culture, employees key values are obtained
that exert influence on its formation thus creating the possibility of taking measures to
change their beliefs in regards with the specific values of the organization.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
354
Tanasi, Z. (2010). Istraivanje modela za dizajn/redizajn socio-tehnikih sistema,

Mainski fakultet, Banja Luka, [doktorska disertacija]
Kates, A., Galbraith J. R. (2007).
Designing Your Organization, John
Wiley&Sons, Inc., San Francisco.
Schein, E. (2007). The Corporate Culture Survival Guide, Jossey Bass, San
Francisco.
Stanford N. (2007). Guide to Organisation Design Creating high-performing
and adaptable enterprises, FSC, London.
Eberhardt, R. (2000). Systems Architecting of Organizations, New York.
Galbraith J. R. (2002). Designing Organizations: An Executive Guide to Strategy,
Structure and Process, Jossey-Bass Publishers, San Francisco.
INNOVATION-THE KEY FACTOR IN ENTREPRENEURIAL CYCLES

Robert Molnar1, Drago Soldat2
Summary: This paper focuses on the innovation as a key factor in entrepreneurial
cycles particularly in those who deal with production. Taking into the consideration the
whole entrepreneurial cycle, the innovation is the most sensitive factor that determines
the products giving to them the soul and in most cases separating successful from
unsuccessful ones on the market. In the era of global market, competitiveness became
one of the most important factors for each company. Within an enterprise, the
engineers are those who are qualified and responsible to take care about the
competitiveness of the products that company produce. They have to be introduced
not just about the technology in sense of how to made smt. but even rather how to
made the better product than others. Better could be involved in product various
characteristics as quality, functionality, durability and any other ones which will
convince the custumer to buy it. The engineers should be more focused on the value
involved in the product rather than on its materialisation, simply because the custumer
buy the value for him/her. By neglecting innovation, companies become just a simple
economy subjects, without entrepreneurial elements, losing competitiveness and
endanger their future. In long-term perspective, the permanent innovations
implemented in the products and services as well as in the production process enable
companies to became or remain competitive on the market.
Key words: Innovation, Competitiveness, Product, Production process, Entrepreneurship.
1. INTRODUCTION
Structural changes that inevitably occur with the development of human society,
both personally and globally, affect the value systems and small, flexible organization put
in a more favorable position than large ones. Most of large companies, which are also
compete, competitors force to both innovation and continue to increase productivity.
Speed of changes taking place in human society, and on markets as well, increasing
competitive advantage of innovative and flexible companies, so they put at a
disadvantage those companies which are burdened by administration [1].
Last decades of the twentieth century have led to changes in manufacturing
1
Prof. dr Robert Molnar, Zrenjanin, Serbia, Technical College of Appplied Sciences in Zrenjanin,
(robert.molnar@vts-zr.edu.rs)
2
Prof. dr Drago Soldat, Zrenjanin, Serbia, Technical College of Appplied Sciences in Zrenjanin,
(biznisinvest@open.telekom.rs)
355
Robert Molnar, Drago Soldat
philosophy in many countries. Era of giant corporations, in terms of number of

employees engaged and technology, has been replaced by the era of small and
medium enterprises (hereinafter SMEs), which promote a spirit of creativity and
individualism, as opposed to the standardization and globalization, which are promoted
by large enterprises and multinational corporations.
Any dynamic economy relies on inventors and entrepreneurs with new ideas
aimed to change our everyday life. Large companies tend to remain large, and it is
natural that they resist to change themselves. Inventors, innovators and entrepreneurs,
with their new technologies threaten the state of "status quo" and they cause changes.
While running their businesses and seeking to grow, entrepreneurs create the driving
force for change.
2. THE ENTREPRENEURIAL CYCLE
An "entrepreneur" has been defined still around 1800. by the French economist
J.B. Say who described it as a man who allocate available resources in the area of lower
productivity, to those with higher one, ie. where higher profit is expected. Since then the
number of definitions have been developed around the concept of entrepreneur.
According to the Austrian economists J. Schumpeter, an entrepreneur is the key to
economic changes and development of a society. It makes new combinations of
production factors and apply them directly. In addition to these determinations it is
valuable to mention another one, created by F. Stankovi who defined that
"entrepreneurship is the name for activities that consist in the performance of innovation,
and individuals who carry these actions are named as entrepreneurs" >2@.
For the purposes of this study, prior entries would be extended to that, that
entrepreneur in his business must provide something new, different or better than the
others do, whether it is embodied in the product, service, management or whatever it is
about business in generall. The entrepreneur is the one who is able to identify market
needs, both present and future ones, and he is the bearer of risk that goes with any
entrepreneurial venture. Entrepreneurs are, therefore, the real individuals who do not
accept routine in the business, but always slightly decompose and reorganize
something. In fact, their assignement is to perform "creative destruction" which will be
the source to increase their own wealth. Entrepreneurs do not want to adjust
themselves to the changes, but strive to adjust the changes to their own wealth.
Moreover, they strive to be active agents of changes.
Because the changes could be used as an opportunity that is provided by
environment, there is a necessity of the presence of several factors in an entrepreneurial
cycle. As it is shown by Figure 1, the initiator of an entrepreneurial venture, in fact, is the
identified need in environment, that should be met. An entrepreneur must be motivated
to meet this need adequately, and in the initial period he runs through the complex
psychological processes within itself. Only when the entrepreneur decided to meet the
need of the environment he can access the next stage, ie. way of satisfying the identified
need. In order to justify its name, an entrepreneur must offer something new or better to
its environment, and thereby take over some risk. If there is not a component of a new or
better in meeting the needs of the environment, then we can not talk about
entrepreneurship, but simply about the ordinary economic activity.
356
Innovation-The Key Factor in Entrepreneurial Cycles
Fig. 1 The Entrepreneurial Cycle >3@

By an innovation, the entrepreneur provides a component of a new or better,
which later has to be incorporated in a product or service, in a framework of his
enterprise. When the product or service is sold on the market, it is considered that the
need has been met and the first entrepreneurial cycle is closed then. The cycles will be
repeated until unsatisfied customers are present in the environment, who send signals
that their needs being met adequately, or if not, the entrepreneur should correct his
innovation, and his product or service as well.
3. THE INNOVATION PROCESS
The real success in modern business no longer relies on the application of the
traditional methods and practices of industrial society. To produce more and cheaper
than others do, is no longer a guarantee for modern business success. Todays'
consumers want to pay for something new or better than they had before. Component of
new or better goes with an innovation and it has to be embodied in product or service.
According to P. Drucker "Every organisation, not just business, needs one core
competence: the innovation". At the beginning of the 20th century innovations become
the fruit of the planned researches. Until then, they have been created in the laboratories
of "lonely hunters" captured in their "forest of knowledge" >4@. But, modern societies are
developing so quickly that many of them can not wait for the ingenious mind of the few
individuals to light up, who will give us something new and/or better. Taking into the
consideration these facts, there are two approaches on defining innovation process:
357
1. The "Western" model

2. The "Eastern" model
According to the "Western" model there are 4 phases in innovation process.
They are: (i) Preparation 'sense' for the problem, (ii) Incubation, (iii) Illumination and
(iv) Verification. The model is based on the fact that human brain has two hemispheres
which function differently. Left brain is dedicated for rationality in human thoughts and
right one is for intuition. The innovation has become to reality by movement of thoughts
among these two hemispheres respecting some rules for each phase of innovation
process. The result is that the innovators, in Western civilization, are mainly those who
think logically, notice the problematic situation. They are capable to define the problem
out of it, and only then to find solutions passing through 4 stages described previously.
Using this model, entrepreneurs systematically work on innovation and it is the result of
hard work >5@.
According to "Eastern" model there are 9 phases in innovation process. Unlike
the "Western" model, innovation process is mostly based on intuition and innovation as
itself is unpredictible. In Eastern civilization innovators are usually people who are able
to detect the problem in the nature, ie. have the talent to "see" the problem, and to
interact with nature by trying to imitate it or learn from the nature how to solve the
problem. Later, through continuous listening of nature, where the inspiration has been
found, the own solutions are created, where one of them was imposed as the
dominant. The discovery of that amazing world is the 'light in the darkness' where
should to go. By this model, innovation is unpredictable >6@.
3.1 The Institutions for Supporting Innovations
The importance of R&D institutions for entrepreneurs and SME sector in all, is
indisputable. Considering the fact that SMEs have limited, ie. scarce resources to
perform even their basic functions, the vast majority of them do not have
capacities/resources, particularly in the area of R&D, which are necessary for creating
the development function of a company. So, R&D institutions have the role in the
sphere of production of new knowledge and its subsequent transfer to SME sector
compensating the previously mentioned drawbacks.
Science Park, Incubator Centre, Innovation Centre, Technology Center,
Research and Development Center and so on, are the names of the institutions
concerned with development and technology transfer, with the aim of increasing the
competitiveness of enterprises, and using the results of scientific research in the
market. Today, it is very often that some of the names of these institutions are used as
synonyms, although among them some differences exist, mostly in the domain of their
organization, as well as in technological level and management support they have. The
common feature of these institutions which operate under different names is gathering
of all participants in the innovation process and enable easier transfer of technology
from R&D institutions in the practice or commercialization of innovations in science,
technology and engineering.
Many developed countries and even less developed ones is being established
different types of R&D instituions in order to improve performances of their
entrepreneurs and SME sector at all. On behalf of that today we speak about new
branch of industry called "knowledge industry". The U.S. and Japan are leaders in this
area, where are special geographic areas are dedicated to produce new knowledge
358
Innovation-The Key Factor in Entrepreneurial Cycles
applicable in the industry as Silicon Valley and Route 128 in U.S as well as Kansai
Science City and Yokosuka Research Park in Japan.
Some innovation supporting institutions are presented by Figure 2, relating
their technology level and management support they have [7].
Fig. 2 Institutions for supporting innovations

3.2 Case Study Serbian SME and R&D sector linkage
Unlike to developed countries, R&D institutions in Serbia, proved as inefficient
institutions relating to SME sector, mostly because they usually produced knowledge
that hasn't been commercialized in an adequate manner. The modern role of R&D
institutions is to transfer the results of their researches into commercialization as soon
as possible, as well as they need to work more closely with SME sector who is "thirsty"
for new knowledge and who is a suitable partner for "experimentation".
By adoption of relevant legislation to R&D institutions in Serbia is given the
opportunity to focus more to the SME sector. Their role in providing relevant
knowledge, especially in the form of innovation is certainly the key to ensuring the
competitiveness of domestic SMEs.
The results of research conducted in framework of Non-finacial instituions
development in Serbia [2] showed some directions related to instituional support in
R&D area. The research embraced 33 institutions of non-financial support to SMEs in
Serbia, out of what 9 could be clasified as R&D ones (Note: There are numerous
institutions in Serbia that deal with R&D activities, as primary or secondary ones. But
only 15 was present in the Register of innovative activities, in part Record of innovation
projects, and NOSIC from Novi Sad, because of its strategic importance for the
development of R&D sector in Vojvodina).
R&D institutions can provide their services in the following areas: information
about new/available technical solutions, through partner institutions, information on
available markets, and markets for products and services. Beside that the sector can
offer services for performance and connection of the SME sector with international
markets and the diversification of some activities in companies. Further, in the field of
intellectual property protection in the form of training and consulting, then education
about nature and the redefinition of the business, as well as recognizing the capacities
of the own company. Development of intellectual tools for innovation by entrepreneurs
359
may not be related to the revolutionary organizational changes, product or technology.

It is enough to raise awareness of the continuous innovation in all phases of production
and operating on the principle of step by step, and ways to get out of systematic
innovation make a profit through increased competitive advantage over the
competition.
The research conducted clearly showed that one of the biggest weaknesses of
R&D institutions is their weak linkage with SME sector. Beside that, as an opportunity,
the sector sees in the larger development projects, primarily international ones. As a
general conclusion on state of R&D sector in Serbia is that Government has not made
serious efforts, except in the domain of legislation, to improve the position of nonfinancial institutions for SMEs support, particularly for those who are based on the
production of knowledge.
4. CONCLUSION
The importance of entrepreneurship and its holders entrepreneurs, is that they
carry out a silent revolution in which winning creativity and entrepreneurial spirit based
on innovation. Such people, entrepreneurs, are significantly different from the
capitalists of the 19th and 20th centuries, because they are with their teams create
new technologies, new products and services, new industries, they create new values
and, of course, new jobs.
The essence of entrepreneurship is just in applied innovations, which are the
basis for competitiveness and the possibility of gaining extra profit. Here is the
justification in the attitude of majority of the governments to help in R&D sector
institutions establishment, which will be available to SMEs as a part of non-financial
support.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
360
erniek, I., Nadasdi, F., Shaezy, S., Molnar, R. (2003). Integralno upravljanje
vrednostima. Domla-Publishing, Novi Sad.
Molnar, R. (2010). Razvoj sistemske nefinansijske podrke malim i srednjim
preduzeima i preduzetnitvu u Srbiji (doktorska disertacija). Fakultet tehnikih
nauka, Novi Sad.
Molnar, R. (2004). Preduzetniko osposobljavanje u velikim preduzeima.
Regionalni centar za razvoj malih i srednjih preduzea i preduzetnitva BANAT,
Zrenjanin.
Draker, P. (1991). Inovacije i preduzetnitvo praksa i principi. Privredni pregled,
Beograd.
Birman, E. (1987). Innovcio. Mszaki Knyvkiad, Budapest/Hungary.
Graff, L., Hills, D.C. (1993). Creativity and Innovation: East and West. World
Future Studies Federation Conference - Creativity Section. Turku (Finland),
August, 23-27. 1993.
Centre for Strategy and Evaluation Services (2002). Benchmarking of Business
Incubators. Final Report. European Commission - Enterprise Directorate General,
Brussels/Belgium.
LABORATORIES UNDER REQUIREMENTS OF DIRECTIVES AND

STANDARDS OF EUROPEAN UNION
Vid Jovievi1, Stevo Borojevi2, Gordana Globoki Laki3, Branislav
Sredanovi4,
Summary: This paper presents a model of the market of the European Union (EU) and
elements of regulatory techniques, related to technical industrial products. To apply
these models it is necessary to implement the requirements of EU Directives and
Standards relating to the companies, products and laboratories. The content of this
paper is predominantly related to laboratory and requirements to be met in order to be
a function of the CE marking of products.
Key words: EU Directives, CE mark, Accreditation
1. INTRODUCTION
The organizational model of EU market represents a model to be built. It is a
system with strong institutions that protect the rights of citizens and the regulated
market, which ensures free competition under strictly defined rules. These rules are
preventing monopoly, protecting the health and lives of citizens, protect the
environment, protect consumer interests and break down the barriers to free
movement of people, goods and services. The model also includes market surveillance
mechanisms that respond to any deviation from the adopted principles.
In order to ensure the free movement of technical industrial products the EU
has developed a new regulatory technique that includes:
a New Approach to regulation of products and
Global Approach to conformity assessment of products.
A New Approach for ensuring the free movement of technical industrial
products in the EU is based on the new approach directives. This regulatory technique
and strategy of the New Approach to technical harmonization and standardization set
by resolution of the Council of the EU, which was passed in May 1985th, and
innovated in July 1993. Year, based on the following settings:
Harmonization of legislation is limited to acceptance by directives of the
New Approach, relevant health and safety requirements, or requirements of
1
PhD. Vid Jovievi, Banjaluka,Faculty of Mechanical Engineering, vid.jovisevic@blic.net

MsC. Stevo Borojevi, Banjaluka, Faculty of Mechanical Engineering, stevoborojevic@hotmail.com
PhD. Gordana Globoki Laki, Banjaluka, Faculty of Mechanical Engineering, gnm@blic.net
4
Branislav Sredanovi, Banjaluka,Faculty of Mechanical Engineering, sredanovic@gmail.com
2
3
361
Vid Jovievi, Stevo Borojevi, Gordana Globoki Laki, Branislav Sredanovi
general interest, for products placed on the EU market must meet if they
want to take advantage of the free movement of goods;
Essential requirements relating to the safety of users, employees,
customers and products as well as environmental protection are inevitable;
Technical specifications of products safety, which meet the essential
requirements of the directives are given in the harmonized standards;
Determination of product compliance and certificate issue in collaboration with
a third party was performed by Notified Bodies, which are EU member states chosen
from bodies in its territory and which fulfill the directives prescribed conditions.
Notified Bodies depending on the requirements of the directive of the New
Approach could be:
Inspection Bodies,
Laboratories
Certification Bodies and
Approval Bodies.
Series of standards EN 45000 and the new ISO / IEC 17000 includes
procedures of accreditation assessment for different types of bodies to determine the
conformity of products (Fig. 1).
Fig. 1 Interaction between accreditation, certification and laboratories

For certification of certain products, for example metal parts, it is necessary to
perform pre-tests, carried out in accredited laboratories. This testing is performed
according to standards and directives of the European Union.
Laboratories used for testing must be accredited as required by international
standards BAS EN ISO/IEC17025:2005, in order to be able to follow the requirements
of new standards for the certification of products, and as a country which aspires to full
membership in the European Union, we are obliged to coordinate our rules, equipment
and behaviour with the European Union requirements.
Therefore, the establishment and continuous development of laboratories for
362
Laboratories under requirements of directives and standards of European Union
testing and certification of metal products is one of the necessary prerequisites of

comprehensive development, market protection and development of overall industrial
production.
2. LABORATORY PREPARATION
The first step for the project of accreditation is the laboratory management
decisions for the accreditation process. It is no longer a choice, but a necessity if the
laboratory wants to survive in the market of laboratory services and statutory
requirements. Laboratory should establish their quality of work, to achieve
differentiation from others, and to adapt to the needs of its users.
The management of laboratories according to their financial capabilities, the
number and expertise of staff determine the methods and scope of accreditation.
When considering the areas and methods that laboratory wants to accredit,
management and technical staff decide on the feasibility of accreditation methods, and
demands that are placed on the laboratory staff. Carefully considered and the financial
side of accreditation, the ratio of investment to economic viability in the future.
Currently that ratio is on the investment side, because there are no legal obligations for
mandatory testing in accredited laboratories.
Leadership of laboratories estimate duration of the project and appoint a
person to manage the project. If management deems it necessary, may be engaged
outside counsel or consultant. It should be noted that the standard BAS EN ISO / IEC
17025:2006, in contrast to other ISO standards, is specific to laboratories, and such a
consultant should possess the professional expertise and practical laboratory
experience.
3.
REQUIRED DOCUMENTATION
For accreditation process a large amount of documentation is required,

however the application of documents in the work lead to quickly discovering the
usefulness of a particular document. If some documents are not used, they should be
changed or put out of use. In preparing of the documents it is necessary to design
documents as required of employees who will use this document.
Quality system documentation makes representative person for quality
management, but technical documentation makes technical director of the laboratory.
Documentation describes the fulfilment of the requirements of standards BAS EN ISO /
IEC 17025:2006, which is related to: organization, quality system, document and
records management, expertise and training of personnel, technical competence and
testing methods.
Documentation of quality system is a novelty which standard BAS EN ISO /
IEC 17025:2006 made in laboratories. Procedures, work instructions and forms of
quality system need to be close to laboratory employees. These documents define the
responsibilities for certain tasks, communication with the customer, supply the
necessary materials, and everything else that affects the quality of services. The
technical documentation includes documentation related to testing methods, which
consists of:
Standard operating procedure (description of the purpose, principles and
363
4.
process methods, necessary equipment, quality assurance of results,

validation parameters and their acceptable values and a description of
validation experiments);
Report of validation (results obtained from method validation, and
confirmation that a particular method of the purpose);
Assessment of uncertainties (identification and quantification of sources of
uncertainty);
SUBMISSION OF APPLICATION FOR ACCREDITATION
In addition to the preparation of documentation for the method, in the laboratory

there are more jobs that are not directly related to accreditation, such as the calibration
of laboratory equipment, calibration and validation of instruments, which requires the
use of authorized service provider.
It is necessary to organize the logging of equipment, calibration and validation,
service instruments, and take care of certified materials and patterns, traceability, test
procedure and documentation that accompany the samples.
The most important indicator of professionalism and competence of
laboratories are participating in interlaboratory comparative testing. If the results of
such tests are not in acceptance limits, all previously described requirements are not
enough for obtain of accreditation.
An application for approval of accreditation shall be submitted to BATA, directly
or through a branch in the entities (Republic of Srpska and the Federation of B&H), on
the appropriate forms which BATA submit to the Body (Laboratories) at his request.
With the application for approval of the Accreditation Body must submit:
1. Evidence of legal status, ownership and all registered economic activities;
2. The decision of the competent authorities on eligibility for the performance
of activities (see the appropriate entity Law on Business Companies /
Enterprises);
3. Precisely defined area of the required accreditation;
4. List of sites with information that activities are performed on each of them;
5. Information, performing as one of the methods from pts. 3 on the ground;
6. Quality Manual;
7. Main list of management system documents;
8. Quality Policy Statement;
9. Details of venues in carrying out activities that are the subject of an
application for accreditation;
10. Evidence of independence and impartiality;
11. Organizational Chart of the Body and the scheme from which you can see
its place within the parent organization, if applicable;
12. List of staff with qualifications and work experience in the field for which
accreditation is sought;
13. Descriptions of responsibilities for each function of the Body;
14. Report on the executed internal audit;
15. Report on review of the system from the management;
16. Report on participation in the PT (proficiency testing), if applicable;
364
Laboratories under requirements of directives and standards of European Union
17. For Certified Bodies, structure, members and normative act about
Committee work;
18. Information on subcontracting, if used;
19. For Laboratories and Inspection bodies, technical data on measurement
and / or test equipment;
20. Information on internal calibration, if performed;
21. Copies of test reports / certificates about calibration / certificates of
conformity / report on the evaluation / reports or certificates of inspection;
22. For Certification Bodies, user manual for signs of compliance;
23. Evidence of payment of administrative fees and costs of registering and
processing applications;
24. Other documents in accordance with instructions shipped with the
appropriate forms for filing an application for accreditation;
5.
CONCLUSION
The main advantage of accreditation is the reliability of test results and

international recognition of test results. Internationally recognized results reduce the
need for additional testing, and facilitate buying and selling services in the market.
Accreditation is increased confidence in staff and in the laboratory.
In the legally regulated area, accreditation is a tool used by state institutions for
the allocation of authority for certain activities. As the importance of quality is still not
sufficiently recognized in our country, there is a possibility that the deficiency of
accreditation outweigh the advantages. The biggest drawback is the huge financial
costs for the process of accreditation, and other costs required to:
education of employees in the laboratory,
payment of external advisors (consultants),
purchasing a new equipment if the existing do not meet the prescribed
requirements,
take samples for interlaboratory testing and
calibration and validation of instruments.
To obtain the status of accredited laboratories is necessary to engage all
employees in the laboratory, and substantial financial investment.
The advantage of accredited laboratories in relation to non-accredited, it should
be easier to get jobs, but it seems that in our country accreditation has not yet been
sufficiently recognized, as a guarantee of reliable test results.
LITERATURE
[1]
[2]
Guide to the implementation of directives based on the New Approach and the
Global Approach (2000), Official Publicatons of the European Communities,
Brisel.
Jovievi, V., Mileti, O., Daki, P.(2003):Usaglaavanje proizvoda sa EU
direktivama u konkretnim uslovima, VIII Meunarodna konferencija MMA 2003,
Univerzitet u Novom Sadu, Fakultet tehnikih nauka, Institut za proizvodno
mainstvo, Novi Sad 26-27. juna 2003.
365
[3]
[4]
[5]
366
BAS EN ISO/IEC 17025:2006 Opti zahtjevi za osposobljenost ispitnih i

kalibracionih laboratorija.
Jovievi V. Borojevi S. (2007): Fabrika proizvodna kontrola u procesu CE
oznaavanja industrijskih proizvoda, 33. JUPITER Konferencija, Mainski fakultet,
Beograd, 2007.
Zakon o akreditaciji Bosne i Hercegovine, Slubeni glasnik BiH br.19/01.
LASER SHOCK PEENING OF N-155 SUPERALLOY EXPOSED TO

AGGRESSIVE MEDIUM
Sanja Petronic1, Andjelka Milosavljevic2, Biljana Grujic3, Radovan Radovanovic4
Radmila Pljakic5
Summary: Investigation was carried out on the iron base superalloy, that are widely
used at elevated temperatures and pressures. After their exposition to aggressive
mediums HCl and HNO3 acids, the laser surface treatment was applied with the aim
of materials recovery. The Nd3+ :YAG laser process characteristics were: wavelength
532 nm, pulse energy 37 mJ, while the pulse frequency varied 1, 2 and 4Hz. The
samples were observed by scanning electron microscopy (SEM) and analyzed by
energo-dispersive spectrometry (EDS). Vickers microhardness tests were performed.
In this paper, microstructures arising after various pulse frequencies were analyzed
and compared.
Key words: superalloy, laser, surface treatment, microhardness.
1. INTRODUCTION
N-155 alloy is an alloy with good ductility, strength, excellent oxidation and
corrosion resistance, and can be readily fabricated and machined. It is used in
numerous aircraft applications. Many superalloys are generally subjected to service in
aggressive environment. Surface morphology plays an important role in the
performance of machine parts that experience unfavorable service condition [1].
Lasers have been used for precise materials processing in micro and nano
manufacturing operations due to its non-contact nature and the high intensity [2]. Laser
shock peening (LSP) is relatively new technology to classical treatments for improving
fatigue, corrosion and wearing resistance of metals.
The effect on target material is mainly achieved through mechanical effect
produced by shock waves[3]. The principle of LSP is to use a high intensity laser and
suitable overlays to generate high pressure shock waves on the surface of the workpiece.
1
dr Sanja Petronic, Innovation center, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade,
Serbia, (e-mail: sanjapetronic@yahoo.com)
2
Prof. dr Andjelka Milosavljevic, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade, Serbia, (email: amilosavljevic@mas.bg.ac.rs)
3
dr Biljana Grujic, Innovation center, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade,
Serbia, (e-mail: bgrujic@mas.bg.ac.rs)
4
Prof. dr Radovan Radovanovic,Academy of Criminalistic and Police Study, 11080 Zemun, Cara Duana
196, (e-mail: radovan.radovanovic@kpa.edu.rs)
5
Mr Radmila Pljakic, High School of Mechanical Enginering, Trstenik
367
Sanja Petronic, Andjelka Milosavljevic, Biljana Grujic, Radovan Radovanovic
For LSP the surface of the specimen is pre-coated with a thin layer of material
highly absorptive at operating wavelength of the laser beam, such as black paint [4,5]
and transparent material, such as water or glass.
2. EXPERIMENT
The experiment was carried out on iron base superalloy samples with
dimensions of samples: 20x10x1.2 mm. Chemical composition of samples is listed in
Table 1.
Table 1. Chemical composition of superalloy N 155.
Ni
20.0
Cr
21.0
Mo
3.0
Mn
1.5
Si
0.5
Fe
30.0
C
0.15
Al
0.15
Co
20.0
W
1.2
Nb
2.0
The samples were exposed to aggressive mediums: the first group was
submerged in HCl acid for 72h, while the second group was submerged in NHO3 for
the same period.
Fig. 1 The experimental setup.

The experiment setup is presented in Fig. 1. The sample was coated with an
absorptive and protective layer (black paint), placed in the container filled with distilled
water (a transparent layer) and exposed to pulsed laser beam (Nd3+:YAG, wavelength
532 nm , pulse energy 37 mJ, and duration 10 ns). Protective overlay is used for two
reasons: (1) to absorb the incident thermal energy, expand and transfer the shock
wave to the metal target and (2) to protect the metal target from the heat influence of
the incident. The implementation of the transparent layer increases the plasma
pressure by a trapping-like effect on the plasma expansion.
The material was treated with nearly square laser spot, as there is no focus
point of release waves. The residual stress drop will not be as remarkable as that
treated by a round laser spot. Using square laser spots should attenuate the
simultaneous focalization of release waves and decrease the residual stress drop at
the centre of the laser shocked zone[3].
368
Laser shock peening of N-155 superalloy exposed to aggressive medium
The samples were exposed to various numbers of laser pulses with different
repetition frequencies. The microstructure was observed by scanning electron
microscope model JEOL JSM-5800, and analyzed by EDS. Microhardness
measurement was performed by Vickers using the apparatus model ZWIK Edman
Weltzar and under load of 0.5 N.
3. RESULT AND DISCUSSION
In this paper, the effect of laser surface treatment applied to superalloy N - 155
after being exposed to aggressive acid HCl and HNO3, was investigated. The aim was
to recover the surface structure after damages caused by these acids.
Figure 2 presents a base microstructure of superalloy N-155, and Figure 3a)
microstructure of superalloys N - 155 after the effects of aggressive acid HCl
intensively for 72 hours. The Figure 3a shows that, after the action of acid, the
microstructure is unfavourable, and the grain size is much bigger. Grain boundaries
decohesion is visible and a rather large gap between the grains are noticeable. The
structure is quite weak and susceptible to cracking and intergranular corrosion.
Fig. 2 The basic structure of the N-155
Fig. 3 Microstructure of N-155 superalloy exposed to HCl acid for 72h a) x500 b) x
3000 (SEM)
Figure 3b) presents the structure taken from Figure 3a) magnified 6 times. The
Table 2 lists the results of EDS analysis of points 1 and 2 in Figure 3b, and shows the
369
increased content of Si, especially at the grain boundary (point 2), where cohesion
occurred. It can be assumed that complex compound based on silicon are formed.
Also, the increased content of S and Mn suggests forming of MnS compound.
Table 2 The results of EDS analysis of points 1 and 2 in Figure 3
Si
1.67
2.03
Eds 1
Eds 2
S
1.06
1.28
Cr
21.57
20.82
Mn
1.94
1.61
Fe
29.95
30.05
Co
20.87
20.54
Ni
22.94
23.67
Fig. 4 Microstructure of N-155 superalloy exposed to HCl acid for 72h and laser treated
(1Hz, 12 pulses), x500 (SEM)
b)
a)
c)
Fig. 5 Microstructure of N-155 superalloy exposed to HCl acid for 72h and laser treated
by a)1Hz, 12 p b)1Hz,16p,c)4Hz, 12p. Magnification 3000. (SEM)
370
Laser shock peening of N-155 superalloy exposed to aggressive medium
Figure 4 shows the microstructure after laser mechanical treatment of material.

The laser beam was acted on recuperating the microstructure and creating something
finer grains. The laser process parameters were: pulse frequency - 1 Hz and the
number of pulses -12. Figures 5a) and 5b) show the microstructure magnified 3000x
after 12 and 16 pulses, respectively, and pulse frequency of 1 Hz. Comparing the
structure leads to the conclusion that the number of pulses recovering structure and
improve it. In Figure 5.b) is visible appearance of fine grain carbide. Mechanical
treatment by laser has a positive effect on the microstructure of reversing the
degradation of the area to be played by the action of acid HCl.
Table 2 The results of EDS analysis of points 3 and 4 in Figure 5c
eds 3
eds 4
Eds 5
Eds 6
Si
0.84
0.70
1.97
0.36
S
0.64
0.67
0.86
0.56
Cr
32.58
39.88
9.40
51.02
Mn
0.94
0.95
0.51
1.13
Fe
41.22
35.19
34.51
29.59
Co
12.66
12.33
23.38
11.67
Ni
11.12
10.28
29.36
5.68
Figure 5c) shows the laser surface treated structure by laser frequency of 4
Hz. It can be noticed that there has been melting of the structure, due to the higher
pulse frequency and material between the two pulses does not have time to cool down.
This surface treatment is called as thermomechanical, because it leads to the thermal
and mechanical changes in the material. Table 3 lists the results of EDS analysis at
points 3 and 4 in Figure 5b),and points 5 and 6 in Figure 5.c). The mechanical laser
treatment results in creating of carbide (Cr, Fe)xCy, as evidenced by increased content
of Fe and Cr. Due thermomechanical laser treatment (frequency 4 Hz) in the melted
parts, the contents of Co and Ni increase, while the content of Cr decreases. However,
in areas where there has been a purely mechanical laser treatment of material the
chromium carbides are identified. The appearance of sulphur is not related to Mn
which is in smaller quantities, but it is assumed that the complex particles MXCy + MS
were formed.
Fig. 6 Microstructure of N-155 superalloy exposed to HNO3 acid for 72h and laser
treated (1Hz, 12 pulses), x3000 (SEM)
In this paper, the Vickers microhardness of the base material (312 HV0.5), of the
material subjected to HCl acid for 72 hours (240 HV0.5), material exposed to HNO3 acid
371
(255 HV0.5), and microhardness of laser surface treated materials after 12 pulses (275
HV0.5) and after 16 pulses (285 HV0.5), were measured. It may be noticed that laser
surface treatment increases the microhardness after damages caused by acids, and
the value of the measured microhardness increases with increasing number of pulses.
4.
CONCLUSION
The samples were exposed to the aggressive acids, which caused a weakened
structure and susceptibility to cracking and intergranular corrosion. The laser beam
was operated on recuperating the microstructure and creating something smaller
grains. Comparing the structures after various numbers of applied pulses leads to the
conclusion that the higher number of pulse recovers and improves the structure, and
the appearance of fine grain carbides is notable. Mechanical laser treatment has a
positive effect on the microstructure of reversing the degradation of the area caused by
the action of acid HCl. Using the higher pulse frequency leads to melting of materials
and thermomechanical processing. Surface laser treatment increases the value of
microhardness which is reduced by the action of corrosive acids. Also, the
microhardness increases with the increasing of number of pulses.
Acknowledgements
This work was supported by the Ministry of Science of the Republic of Serbia under
contract number TR-35040.
LITERATURE
Sahaya Grinspan, A., Gnanamoorthy, R. (2007). Surface modification and fatigue
behavior of high-pressure oil jet-peened medium carbon steel, AISI 1040. J.
Manuf. Sci. Eng. 129, 3, 601-606.
[2] Hu, Y., Gong, C., Yao, Z., Hu, J. (2009).Investigation on the non-homogeneity of
residual stress field induced by laser shock peening. Surface & Coatings
Technology 203, 3503-3509.
[3] Warren, A.W., Guo, Y.B., Chen, S.C. (2008). Massive parallel laser shock
peening: Simulation, analysis, and validation. International Journal of Fatigue, 30,
188-197.
[4] Kovacevic, A., Petronic, S., Sedmak, A., Milosavljevic, A., Popovic, M. (2010).
Modifikacija mehanikih osobina austenitnih materijala superlegure nikla, eleza
i nerajui elik nanosekundnim laserskim impulsima, Fotonika 2010, Beograd,
book of abstract, 11.
[5] Petronic S. Uticaj termike i laserske obrade na promene u mikrostrukturi
superlegura nikla. PhD thesis 2010.
[1]
372
LEAN CONCEPT TOOLS IN PROCESS OF TECHNICAL SYSTEMS

OVERHAUL
Bogdan Mari1, Ranko Boikovi2
Abstract: The paper shows researches on influence of lean tools on duration of the
overhaul cycle in the process of VIPER turbo jet engines overhaul. The surveyed tools
are the Nagara system, Just-In-Time and Layout. A simulation of a working process
was done using Microsoft Project software tool in the real overhaul-production system.
The research results showed certain regularities in the overhaul cycle proceeding.
These regularities can certainly be used for evaluation of the analyzed lean concept
tools influence on the effectiveness and efficiency of a production process of every
overhaul-production system.
Key words: management, overhaul-production system, production process, lean tools.
1.
INTRODUCTION
Nowadays, under the conditions of modern market, the aims in a working

process of overhaul-production systems have to be set up for introduction of functional
organizational arrangement of a certain working process which will enable a high level
of flexibility in an organization and management oriented to a customer, with defined
quality system, competitive prices on the market and as short as possible delivery time.
The design of functional organizational arrangement, measured through the
effectiveness and efficiency of a working process, includes in itself organizational
arrangement of variety groups of influential factors that are interdependent in many
ways. Their effect intensity affects the successful result of every working process and
the development of overhaul-production systems. The final aim is to achieve as huge
as possible effects using minimal invested resources, in other words, costs per a unit of
technical system in overhaul process and/or services need to be as less as possible. In
that way, the managers who lead a working process are forced to constantly eliminate
those activities which cause loss for the enterprise using certain efficient methods,
procedures and tools. Additionally, the minimizations are also transmitted to individual
elements of organizational structure as spatial dimensions (object) retaining for that the
power and position on the market. Therefore, the industrial system is reduced to
meager, thin, relayed on something, dependant, and its structures become lean (lean
management, lean engineering, lean development, lean production, etc.). The
1
2
PhD Bogdan Mari, Bijeljina, Orao a.d., (maricbogdan@gmail.com)

PhD Ranko Boikovi, associate professor., Doboj, Faculty of Transport, (rankob@teol.net)
373
Bogdan Mari, Ranko Boikovi
approach to building this kind of state is referred to as a lean concept, lean approach or
lean doctrine.
The Lean concept has for its goal to eliminate loss sources or free spaces
and these are basically useless time and losses made due to early investments in
production of goods. Stock elimination is the largest problem because it requires the
synchronization of very close processes, those which connect different functions within
the industrial system.
This concept includes constant organization and management modernization.
Those are activities for reduction of production costs through reduction of all
manifested loss types and they are expressed in factory flows of materials, losses in
jams on equipments, in information, energy, inventory management, design etc. In that
way we create a system which will completely eliminate production losses, using what
is fundamental for the production and what has influence on production price increase:
equipment, material, labor force, time etc.
Systematic and constant researches on useless activities on all levels of
working process and their elimination or reduction indirectly affect the quality and costs
in industrial system. New methods for labor trainings tools and developed information
technology allows to accelerate the researches and ensures a continuous loss
reduction and product quality improvement due to customer requests in all elements of
a working cycle.
The Lean concept is referred to as a set of activities for the application of
principles, methods and tools for systematic detection of useless activities (losses and
mistakes) in a working process [2]. By their detection and elimination in industrial
systems we influence general work efficiency: we reduce operational cycles,
production cycles, material flows, we increase the utilization rate of space and energy,
we build productivity relations in working teams on all levels of organizational
structures, particularly for the purpose of production.
2.
TOOLS AND METHODS IN PROCEDURES OF LEAN CONCEPT DESIGN
Tools and methods which are used in the procedures of lean concept design
can be presented as significant elements for development of industrial system (Fig. 1).
As the Figure shows, which symbolically presents a house, the basis of everything is
consisted of working teams, high-motivated workers, and a 5 x why 1 x how
principle. The essence is not to be satisfied with the first answer but to search further
until we find cause of some phenomena. That is the best way to come to a solution.
The tools are given in random order, not in order of importance because they are all
important. In addition to these tools it is possible to create and design other tools which
give better results for certain conditions.
Diagrams of results and work process as well as the realization of VSM (Value
Stream Mapping) analysis are bases for encouraging managers to make decisions to
reduce losses and design lean concepts with given tools and methods. For this
analysis software tools as eVSM (electronic Value Stream Mapping) and MINITAB
statistical software can be very useful.
374
Lean Concept Tools in Process of Technical Systems Overhaul
TOOLS AND METHODS IN PROCEDURES OF LEAN CONCEPT DESIGN
JIT
SMED
TPM
TQM
Equipment
arrangement
Continuous
flows
Nagara
system
KAIZEN
5S
Work
standardization
Storage
location
Visual
systems
Diagrams and work process analyses (VSM)

Working teams / high-motivated workers / 5 whys / customer orientation
Fig. 1 Tools and methods in lean concept design

3.
THE INFLUENCE OF LEAN TOOLS ON OVERHAUL CYCLE DURATION
Reaching shorter duration of overhaul cycle with more effective and efficient
work processes in the reparation process of technical systems is in close relation with
application level of lean concept, its principles, tools and methods. Experimental
researches on its application are carried out in a real overhaul-production system in the
overhaul process of VIPER turbo jet engine with the simulation of a work process by
Microsoft Project software tools and they are oriented to three lean tools [3]:
1 Nagara system it is observed the allocation of workers serving more
workplaces or machines before and after introducing this tools by application of
the interdisciplinary of an individual. (A factor);
2 Just-In-time (JIT) in the research JIT was treated as a factor which observes
time reduction in the work processes applying certain methods, tools and
techniques, in other words, as a set of methods and procedures which has for
its aim to design a production process in the way to enable much more
precisely reply to delivery demands (B factor);
3 Layout reconfiguration Layout is observed before and after the introduction of
this tool (C factor).
The tools were treated as factors and their analysis of effects on overhaul cycle
3
time - Trc was processed by factor plan 2 . In Table 1, according to [3], mean times of
overhaul cycle - Trc (in hours) for all factors are shown. The analysis of their influence
and their correlation have been solving with MINITAB software which is used for a
complete regression analysis and variance analysis.
375
Table 1 Overhaul cycle time - Trc in hours: initial experiment matrix

JUST-IN-TIME
without
with
NAGARA SYSTEM
NAGARA SYSTEM
without
with
without
with
334.92
338.75
323.75
333.67
333.67
337.75
321.08
332.67
LAYOUT
LAYOUT 1
LAYOUT 2
with
The Diagram of the influence of observed factors on an overhaul cycle - Trc is

given in Figure 2.
Pareto Chart of the Effects
(response is Trc, Alpha = .05)
2.710
Factor
A
B
C
B
A
Name
NAGARA SY STEM
JUST-IN-TIME
LAY OUT
Term
AB
C
AC
ABC
BC
0
Effect
Lenth's PSE = 0.72
Fig. 2 Pareto diagram of the factor effects on overhaul cycle - Trc

The Diagram shows that the greatest impact on overhaul cycle duration has the
B factor, and the A factor comes after it and then the AB factor combination. It is
observed that the C factor and AC, ABC and BC factor combinations does not
significantly affect the overhaul cycle duration. The regression analysis discovers and
sorts the importance of individual factors according to the order:
1) Just-In-Time (JIT),
2) Nagara system,
3) Layout.
The behavior of main factor effects (Nagara system, Just-In-Time and Layout)
of overhaul cycle is shown in Figure 3.
376
Lean Concept Tools in Process of Technical Systems Overhaul
Main Effects Plot (data means) for Trc

NAGARA SYSTEM
JUST-IN-TIME
336
334
332
Mean of Trc
330
328
1
LAYOUT
336
334
332
330
328
1
Fig. 3 Effects of main factors during observation of overhaul cycle time

A layout of flat surface consisted of overhaul cycle time - Trc with two factors
are given in Figure 4.
Surface Plot of Trc vs JUST-IN-TIME, NAGARA SYSTEM
Surface Plot of Trc vs NAGARA SYSTEM, LAYOUT
335
Trc
335
Trc
330
325
330
325
2.0
320
1.0
1.5
1.5
NAGARA SYSTEM
2.0
2.0
320
JUST-IN-TIME
1.0
1.0
Surface Plot of Trc vs JUST-IN-TIME, LAYOUT
1.5
1.5
LAYOUT
NAGARA SYSTEM
1.0
2.0
Contour Plot of Trc vs JUST-IN-TIME, NAGARA SYSTEM

2.0
325.0
327.5
330.0
332.5
335.0
JUST-IN-TIME
1.8
335
Trc
330
1.6
T rc
< 325.0
- 327.5
- 330.0
- 332.5
- 335.0
- 337.5
> 337.5
Hold Values
LAYOUT 1
1.4
325
2.0
320
1.0
1.5
1.5
LAYOUT
2.0
JUST-IN-TIME
1.2
1.0
1.0
1.0
1.2
1.4
1.6
1.8
2.0
NAGARA SYSTEM
Fig. 4 Layout of regression flat surfaces

377
The regression equation of the behavior of overhaul cycle time - Trc is acquired
by regression analysis and is:
Trc = f(A,B,C) = 336 + 7.36 A 8.48 B 1.48 C
(1)
The effects of the application of these tools are:

x
x
x
x
4.
The overhaul cycle is decreased from 28.3 to 47.8% in relation to the traditional
approach to production process management in overhaul-production systems.
The number of doers (workers) in a overhaul-production system is decreased
for 51.5%. This particularly refers to the possibility to choose work potential of
qualified and trained workers in performance of key work processes.
It is avoided the system entropy which is created due to the presence of
incompetent workers and who reject and delay performing assignments
because of constant demands for help with state diagnoses.
The layout significantly regulates flows of work objects arranging them in
progressive and irretrievable flows.
CONCLUSION
The Lean concept presents a set of efficient and rational procedures in a

systemic usage of principles, methods and tools in industrial systems for detection and
elimination of useless activities (losses and mistakes) in work processes. This enables
necessary conditions for harmonious performance of enterprise functions in a certain
time and under certain environmental conditions. The Lean concept introduction affirms
the tools which lead to effectiveness and efficiency of a working process what helps to
create a productive atmosphere in overhaul-production systems to the satisfaction of
workers, users of products/services and society in general. However, further
improvements have to be directed towards Just-In-Time and the Nagara system. They
are effective for the future work of an overhaul-production system, especially the
Nagara system.
It is important to emphasize that the Lean concept is a continuous process and
that the whole philosophy of overhaul-production system has to be adjusted to the
constant process improvement and loss elimination in the system. Only in this way the
overhaul-production systems will have long-term benefits from the Lean concept, and
all that will contribute to the modernization of overall production and the rationality of
mentioned systems in efforts to achieve as better results as possible under real work
conditions.
REFERENCES
[1] Boikovi, R.: Lean Concept in Effective Production Systems, doctoral thesis,
Faculty of Technical Sciences, Novi Sad, 2005.
[2] James, J. & Frederick, M.; The Lean Company Making The Right Choices, Society
of Manufacturing Engineers, International Standard Book Number 0-87263-523-6,
Dearborn, Michigan, 2001.
[3] Mari, B.: Model of Production Process Management in Overhaul-production
Systems Based on Lean Concept, doctoral thesis, Faculty of Mechanical
Engineering, East Sarajevo, 2010.
378
NUMERICAL SIMULATION OF HOLE FLANGING OF CIRCULAR

SHEETS
Ranko Radonji1, Milan ljivi2, ivko Babi3, Milentije Stefanovi4
Summary: Hole flanging, widely used in the automotive industry, is one of the most
important processes in the sheet metal forming. Sometimes in this process, cracks
occur at the initial hole edge. In this study, numerical simulations have been used to
analyze the hole flanging of circular plates with a pre-determined initial hole at the
center of sheet metal. Necessary data on mechanical properties of the workpiece
material (St 14 according to DIN standard) for process modeling were determinated by
an experiment. According to obtained results, a model of the hole flanging process for
single sheet metal thickness was defined. Numerical simulation results include
dependency between the maximum punch load and punch stroke for different hole
diameters.
Key words: hole flanging, numerical simulation, sheet metal.
1. INTRODUCTION
Hole flanging is one of the sheet metal forming processes where a workpiece
prepared from sheet metal with previously sheared hole in its centre is rigidly clamped
around its periphery, and then a punch is used to force blank into a die to form collar
(hollow-flanged component). Forming capacity in this process is dependent on ductility
and surface quality of the hole edge.
Factors such as the edge condition before forming, punch shape, clearance,
shearing operation, and microstructure (volume fractions of the different phases) affect
the ability of the hole flange to stretch. There are two types of the hole flanging
process: hole flanging with ironing and hole flanging without sheet metal thickness
reduction.
Parts produced by the former are usually used for thread cutting at collar, while
parts obtained without sheet metal thickness reduction have adaptation as a support
for pipe joining by welding with other assembly parts. Products which in forming steps
also undergo a hole flanging operation are very common in car bodies, especially at
exaust gas systems, etc. In case of the hole flanging process without thickness
1
Mr Ranko Radonji, University of Banja Luka, Faculty of Mechanical Engineering, (ranko.radonjic@unibl.rs)

Dr Milan ljivi, University of Banja Luka, Faculty of Mechanical Engineering, (milansljivic@unibl.rs)
Dr ivko Babi, University of Banja Luka, Faculty of Mechanical Engineering, (zivkobabic@unibl.rs)
4
Dr Milentije Stefanovi, University of Kragujevac, Faculty of Mechanical Engineering, (stefan@kg.ac.rs)
2
3
379
Ranko Radonji, Milan ljivi, ivko Babi, Milentije Stefanovi
reduction, deformation mode at the edge of the hole is a combination of bending and
stretching. Hole flanging is a practical industrial process where cracks can arise on the
collar edge during the process, hence that part would not be useful for assembling.
A shematic drawing of the hole flanging process is given on fig. 1. Workpiece
shape with some important dimensions is drawn and given on fig. 2.
Fig. 1 Shematic drawing of the hole flanging

process (with stress-strain scheme)
Fig. 2 Workpiece with formed collar
Initial hole diameter which is needed to obtain collar with required dimensions
can be calculated by using the clause of equality of collar lenght by the central line
after forming process, and the width of the workpiece part which is necessary to obtain
collar.
t S
h rd
2 2
din # D1 2h 3.14rd 1.57t

D1 din
2
(1)
However, by using the formula (1) it is not possible to predict geometrical

dependencies at the workpiece which would ensure the forming process without the
cracks emergence. Model of the hole flanging process that will ensure the process
without cracks can be defined according to the experimental results or by the
application of modern numerical simulation.
In order to prepare adequate numerical simulation of the hole flanging process,
it is necessary to define mechanical properties of the sheet metal and formabilty
parameters, and according to that data calculate flow curve aproximation. The above
mentioned data are given in table 1.
380
Numerical simulation of hole flanging of circular sheets
Table 1 Mechanical properties of the sheet metal
Properties
Rp0.2 [MPa]
Rm [MPa]
Rp0.2 / Rm
A80
n - factor
flow curve
Material: DC04 (t=0.8 mm)

Angle
0
45
90
160.51
167.50
167.30
303.40
313.63
304.29
0.63
0.53
0.55
23.48
22.14
22.68
0.242
0.240
0.237
0.242
0.240
k=544.79
k=561.56
k=542.500.237
Average value
175.10
307.10
0.57
22.77
0.240
k=549.870.240
Also, to define the critical value for strain at the hole flanging process, a few
specimens were prepared with changeable initial hole dimaters and formed on the
adequate tool. For this purpose, a forming tool with changeable punch and die has
been designed according to experiment needs. Workpieces without cracks at the body
have been considered successfully formed, and one of them is shown on Fig. 3b.
According to the experimantal results of the hole flanging (for collar diameter 30,40 and
50 mm) critical value for strain was calculated by the formula:
Mcritical
ln
dp
d in
(2)
The flow curve equation and critical value for strain are inputs at the
preprocessing phase for numerical simulation.
(a)
(b)
Fig. 3 Hole flanged workpieces with 50 mm collar diameter; (a) a workpiece with initial
hole diamater of 22 mm (cracks occur), (b) a workpiece with initial hole diamater of
23 mm (successful forming)
3. NUMERICAL SIMULATION
LS Dyna software was used for numerical simulations of the hole flanging
process, and simulations were prepared at the Institute for Metal Forming Technology
in Stuttgart. Tool parts were designed by Catia software, and the punch had a
cylindrical shape with 3 mm radius. Shell elements were used for discretization of the
workpiece, while tool parts in the preprocessing phase were defined as rigid bodies. As
381
a material model in the simulations, 103*mat_anisotropic_viscoplastic was defined

because of possibility to circumscribe the critical strain value and represent the flow
role. Due to the axysimmetrical specimen shape, just one forth of the part was taken
into account in the simulations in order to decrease processing time.
punch
die
workpiece
Fig. 4 Tool parts and a workpiece at the preprocessing phase (defined finite elemente
grid)
After simulation on fig.5a, it can been seen that cracks arise at the workpiece
area which is close to collar edge in meridial direction. Also, the maximal value for
effective stress at the workpiece with cracks is higher (512.2 MPa) than the one at the
part without cracks (509.1 MPa). Comparing the results which are obtained by the
experiment it can be concluded that there are similarities. During the simulation, the
workpiece is bent twice, firstly around the punch profile and after that over the die
radius. Taking into account all simulation results, dependency between initial hole
dimeter din and collar diameter which is coincident to the punch diameter dp was
defined.
(a)
(b)
Fig. 5 Simulation of the hole flanging process for 50 mm collar diameter; (a) initial hole
diameter din= 22 mm, (b) din = 23 mm (sheet metal thickness is 0.8 mm)
In order to confirm accuracy, the simulation results were compared with the
experimental ones. Figure 6 shows curves which were defined according to the
simulation and experimental results. Both curves represent boundary dependence
between intial hole diameter and collar diameter. It means that drawn curves define
boundary between the hole flanging process without cracks and the area where cracks
382
Numerical simulation of hole flanging of circular sheets
will arise for the analyzed diameter spectrum (the forming process without cracks is
under the curves). The boundary curve which is drawn on the basis of simulation data
is linear while the second curve is with a higher order (for the experimental results).
Also, it can been conluded that there are small deviations between the accomplished
results in the above mentioned processses. From the engineering point of view, it is
possible to aproximate curve which is defined on the basis of the experimental results
by the first rank curve leaving the certain reserve of the ductility.
fracture
hole flanging without cracks
Fig. 6 Comparison of numerical simulation results with experimental results for

dependency between collar diameter dp and initial hole diameter din
A diagram of dependency between punch force and initial hole diameter for
hole flanging process with 50 mm collar diameter is shown on fig.7. Dependency is
approximately linear.
Fig. 7 Dependency between punch force and initial hole diameter
383
4. CONCLUSION
Numerical simulation possiblities for prediction of the fractures at the hole
flanging process were analyzed in this paper. Simulation results have been compared
with experimental results for exact domain of diameters, and there is good agreement.
During simulation, cracks arise on the collar edge in the meridial direction at the
damaged parts. According to the simulation results a boundary curve which divides
successful forming process from the hole flanging where cracks arise is approximately
linear. It is possible to predict by numerical simulation when damages at the workpiece
will apear. The analyzed diameter domain can be expanded according to needs, taking
into account the same curve gradient as it is given at the curve on fig.6 .
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
384
Altan, T., Vasquez, V. (1997). Status of process simulation using 2D and 3D finite
element method, Journal of Materials Processing Technology, Vol. 71(1997), p.
49-63.
Banabic, D., Bunge, H.-J., Pohlandt, K., Tekkaya A.E. (2000). Formability of
Metallic Materials, Springer-Verlag, Berlin.
Liewald, M. (2008). New Developments in Sheet Metal Forming, Institute for
Metal Forming Technology, University of Stuttgart, Stuttgart.
Radonjic, R. (2010). Determination of the formability limits in the hole flanging
process, University of Banja Luka, Faculty of Mechanical Engineering, Banja Luka
(master thesis).
Sljivic, M., Radonjic, R. (2009). Sheet metal forming technology, University of
Banja Luka, Faculty of Mechanical Engineering, Banja Luka.
Sljivic, M., Radonjic, R. (2010) Modeling of the hole flanging process,
International Conference on Innovative Technologies, IN-TECH 2010 Conference
Proceedings, p. 278-282.
ONE APPROACH TO THE DEVELOPMENT AND IMPLEMENTATION

OF FLEXIBLE MANUFACTURING SYSTEMS
Dejan Luki1, Velimir Todi 2, Mijodrag Miloevi3, Goran Jovii4
Summary: Flexible manufacturing systems (FMS) represent at the same time
production technology and production philosophy. FMS represent one of the ways that
the manufacturing system while achieving the high level of flexibility and automation,
that is considered to be agile. Design, selection and application of modern flexible
manufacturing systems, or its individual elements, requires a detailed analysis and
definition of a wide range of technical and economic basis for the selection of optimum
solution in terms of accuracy, reliability,mobility, flexibility, productivity and economy.
In this paper exposed an approach, or methodology for the development and
implementation of flexible manufacturing systems. The methodology is based on the
technological basis that are formed on the principles of group technology, application
of simulation techniques and multicriteria optimization.
Key words: flexible manufacturing system, development, application
1. INTRODUCTION
The basic characteristic of he development of modern production is reflected in
the continuous expansion of market demand in terms of product range and product
quality, reduce cost and delivery times. These requirements impose the solution of
manufacturing systems, which are based on the effects of automated large-batch and
mass-production manufacturing in terms of productivity and efficiency, and effects of
small batch and mass production in terms of flexibility, reliability and mobility, therefore
the development and application of FMS [1].
Basically FMS are intended to produce one or more families of products and
parts, with the development and implementation based on the application of the
concept of group technology [2].
The basic units of a modern FMS are numerically controlled machine tool,
especially at today's level when they are multi-functional machines that integrate
despite the different treatment processes (turning, milling, drilling, grinding, etc.), and
various technologies (heat treatment, metal forming, etc.). Integration of such
2
Prof. dr Velimir Todi, Novi Sad, Faculty of Technical Science, DPM, (e-mail: todvel@uns.ac.rs)
Mr Dejan Luki, Novi Sad, Faculty of Technical Science, DPM, (e-mail: lukicd@uns.ac.rs)
3
Mr Mijodrag Miloevi, Novi Sad, Faculty of Technical Science, DPM, (e-mail: mido@uns.ac.rs)
4
MSc Goran Jovii, Novi Sad, Faculty of Technical Science, DPM, (e-mail: goran.jovicic@uns.ac.rs)
1
385
Dejan Luki, Velimir Todi, Mijodrag Miloevi, Goran Jovii
machines by means of manipulation systems, measurement/control systems,

transport systems, as well as their control by computer, leads to creation of FMS of
various levels of complexity [3, 4].
In practice, there are many complex levels of FMS [5]:
Flexible manufacturing module (FMM)

Flexible manufacturing (assembly) cell (F(M/A)C)
Flexible manufacturing group (FMG)
Flexible manufacturing line (FML)
Flexible production systems (FPS)
Modern flexible manufacturing systems because of its high production and

technological capabilities, high prices and costs in terms of investment required to
ensure their rationale exploitations. Taking into account all the complex conditions of
production and marketing of products, you need a systems approach to the design and
implementation of FMS, or their elements, with different levels of automation and
flexibility.
2. REVIEW METHODOLOGY
The proposed approach or methodology consists of three stages or phases,
which significantly can be apply integral and simultaneously or separately (Fig. 1):
Technological basis for the development and implementation of flexible

manufacturing systems
Evaluation and selection of elements flexible manufacturing systems
Modeling and simulation of flexible manufacturing systems
Fig. 1 Phases of development of flexible manufacturing system

386
One approach to the development and implementation of flexible manufacturing systems
3. PHASES OF DEVELOPMENT FMS

3.1 Technological basis for the development and application of FMS
The first step in setting up the technological basis includes analysis of
production program and manufacturing processes appropriate products. Starting from
production program, or parts, appropriate technical-technological documentation
forming the appropriate part or product families.
By defining the content of group technology (GT) process for a particular part
families determine the types and levels of machining systems, while defining the type
and content of a particular group of operations will provide a basis for selecting the
type and level and other elements of FMS, such as tools, fixtures, measuring and
control systems, etc. GT process planning is performed based on complex part
representatives from the part families, that may be real or imaginary [6].
Group manufacturing processes and operations allow precise definition of the
process plan for any part, including the representative part of a part family. In order to
make rational and rapid determination cycle time of machining process is necessary to
precise machining process for corresponding representatives part, in depending on the
method used to determine this time. These methods are used for the determination of
the machining time (tg), auxiliary time (tp) and setup time (Tpz) for families of parts. They
are in practical implementation are used as analytical or graphical methods, or:
Semigraphical method (or just the graphical and analytical)

Method based on the representatives part of part families
In both cases, the methods provide approximately determination cycle and

setup time, which is in manufacturing practice, given satisfactory results in determining
the norms of time for certain machining process, and thus the basis for determining the
number of machining systems and establishing a degree of their utilization. In the
second paper, shown example of defining the technological basis for the development
FMS for manufacturing part families of shaft, through the application of the both
methods for defining cycle time of machining process.
Based on the geometric characteristics of parts from group, requirements
provided from process planing and characteristics of available resources choose the
level of complexity of elements FMSs.
3.2 Evaluation and selection of elements FMS
For the selection of most appropriate solutions FMS elements in the process of
their design and development using a number of multicriteria optimization or decision
making methods.
Multi-Criteria Decision Making (MCDM) refers to the decision problems where
inherent a very large number of different alternatives and criteria. In order to bring a
good decision, it is necessary to describe the alternative definition of appropriate
criteria. It is also necessary to define weights for each criterion, and the importance of
each criterion in relation to others. After that, in every aspect especially evaluate
alternatives based on strictly defined criteria or subjective evaluation. The way that
manifest that score depends on the chosen method used to solve the problem.
387
Today there are many methods of MCDM and depending on the method used
as a solution can be obtained: ranking of alternatives, the best alternative, or set of
alternatives that meets certain conditions.
The decision process consists of a certain stage, between which there are
similarities and differences, but almost all methods can be reduced to the next stage in
the process of decision making [7]:
Identifying and defining problems,

Defining a set of alternative solutions,
Defining a set of criteria for evaluating alternatives and their weights,
Evaluation of alternatives,
Selection of alternatives.
If the stages of decision making adds and two phases, the application of the
selected alternatives and evaluation of results that would determine whether the
problem is satisfactory, obtain the process making decision in general, or the process
of solving problems.
In the present approach, the problem of the choice of alternatives and criteria is
greatly facilitated by the fact that in the first stage determines the required number and
level of complexity of certain elements of FMS.
The existence more of MCDM, some of which should maximize and some
minimize, it means that to decisions are bring in conflict conditions and for solving
MCDM tasks must implement instruments that are more flexible than strict clean
mathematical optimization techniques. For this purpos have been developed a number
of methods MCDM which is as important can be separated: AHP, PROMETHEE,
ELECTRE, TOPSIS, Delphi, etc. [8].
The paper [9] is an example choice of machining system for turning process
part families of shaft, using multicriteria AHP methods.
Other than listed method for solving complex problems of making decision in
engineering practice often used multicriteria methods optimization that are based on:
methods of value assessment (method of total value, the methods of group selection
based on indicators tehnical-economic levels ,etc. ) and LSP method Logic Scoring of
Preference [10].
3.3 Modeling and Simulation of FMS
For designing and creating spatial structure of manufacturing systems, in
addition to the design of process planings, it is necessary to determine and other data,
or standards relating to time, technological capacity, material, surface manufacturing
facilities, energy, inland transport, and the best schedule of these manufacturing
resources, as elements of FMS.
These complex tasks that require the development of optimal solutions FMS
and related manufacturing processes, today successfully solved by modeling and
simulating of manufacturing systems and processes using appropriate software
systems [11].
Simulation of manufacturing processes using an appropriate software system
to simulate, as carefully and directed experiments on the computer, is done in the
general case when:
388
One approach to the development and implementation of flexible manufacturing systems
Design a new manufacturing system,

Advance the manufacturing process in the existing manufacturing system,
Introduces new program production in the existing manufacturing system.
In all the above cases, the simulation of manufacturing processes can be

separate four main stages of its realization [12]:
Assessment and collection of data from real manufacturing conditions

necessary for the development of simulation models,
Set the target simulation and development simulation model,
Simulation experiments on the computer,
The interpretation and application of simulation results.
In the modeling of manufacturing processes is necessary to define the contours

of manufacturing space in which to place elements of FMS and other parts of the
manufacturing system. In defining the tasks and goals of the simulation of
manufacturing processes at the core most important questions [13]:
What the technological mobility expected,

What is the best number and distribution of manufacturing resources,
What is the optimal level of manufacturing,
What are the best strategies of organization and realization of
manufacturing,
Where are the best location and the optimal size among baffers, etc.
Development of a simulation model, which refers to the model of the

production process, is solved in several iterations of the conceptual model to the final
model, which provides optimal results of the simulation. Thus developed model
represents the best solution manufacturing process, which is to adopt with additional
analysis, or if necessary, later corrected [11].
To simulate the process of manufacturing family of product using the program
system for simulation, the model of the manufacturing process must include the
following important information [14]:
The spatial distribution of elements FMSs,

The distance between the machining systems and other resources,
Defined making technology flows,
Defined time process and operation,
Defined way of performing the sequence of process and operation,
Defined objects for monitoring and recording the results of simulation,
Analysis of simulation results and others.
4. CONCLUSION
Exposed to approach the development and implementation of flexible
manufacturing system aims that for a specific product program and production
requirements make selection decisions of FMS with a high technically and
economically effects.
Three observed phase of FMS can be used integrally or individually to solve
certain tasks related to creating a technological basis, the simulation of manufacturing
389
processes and multi-criteria decision making in the selection of elements FMS.

Presented methodology was applied to set the basis for the development of
FMS for manufacturing of part families shafts.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Shivanand, H. K., Benal, M. M., Koti, V. (2006). Flexible Manufacturing System.

New age International (P) Limited, New Delphi.
Radharamanan, R. (1994), Group technology concepts as applied to flexible
manufacturing systems. International Journal of Production Economics. vol. 33,
no. 1-3, pp. 133-142.
Rekecki, J., Gatalo, R., Zeljkovi, M., i dr. (1989). Fleksibilni tehnoloki sistemi za
obradu rotacionih izradaka, Knjiga I, Stanje, tendencije i podloge za razvoj.
Fakultet tehnikih nauka, Novi Sad.
Satya, S. C., Douglas, N. H. (2008). The evolution of manufacturing cells. An
action research study. European Journal of Operational Research. vol. 188, no. 1,
pp. 153-168.
Stefanovi, M. (2006). CIM sistemi. Mainski fakultet, Kragujevac.
Mitrofanov, S. P. (1996). The Scientific Principles of Group Technology. National
Landing Library Translation, Yorks, UK, Boston Spa.
Triantaphyllou, E. (2000). Multi-Criteria Decision Making Methods, A Comparative
Study. Kluwer Academic Publishers, Dordrecht.
upi, M., Suknovi, M., Radojevi, G., Jovanovi, V. (2004). Specijalna poglavlja
iz teorije odluivanja: Kvantitativna analiza. Fakultet tehnikih nauka, Novi Sad.
Luki, D., Todi, V., Miloevi, M., Jovii, G., Vukman, J. (2011). Primena
viekriterijumskog odluivanja u izboru obradnog sistema. 37. Jupiter
konferencija, Mainski fakultet, Beograd.
Todi, V., Stani, J. (2002). Osnove optimizacije tehnolokih procesa izrade I
konstrukcije proizvoda. Fakultet tehnikih nauka, Novi Sad.
Law, A.M., Kelton, W.D. (2000). Simulation Modeling and Analysis, third edition.
McGraw-Hill, New York.
Todi, V., Luki, D., Miloevi, M., ivkovi, A. (2010). Podloge za razvoj
simulacionog modela procesa proizvodnje kotrljajnih leaja. asopis IMK-14
Istraivanje i razvoj, Vol.16, No, 35, pp.29-34.
Borojevi, S., Jovievi, V., Jokanovi, S. (2009): Modeling, simulation and
implementation of process planning. Journal of Production Engineering, Vol.12,
No.1, pp. 87-91, Faculty of Technical Science, Novi Sad.
Bangsow, S. (2010). Manufacturing Simulation with Plant Simulation and Simtalk:
Usage and Programming with Examples and Solutions. Springer-Verlag, Berlin.
Note: This paper present a part of researching at the Project number TR 35025, financed by
Ministry of Education and Science of Serbia.
390
PAPER ABOUT PAPERS IN THE AREA OF METAL FORMING

PRESENTED AT DEMI CONFERENCES HELD SO FAR
Milentije Stefanovi1, Srbislav Aleksandrovi2, Dragan Adamovi3
Summary:At the previous nine DEMI conferences, held in the period 1989-2009, the
specific area of production technologies metal forming was represented with 62
papers in total. Based on performed analysis, this paper presents the connections to
institutions to which the authors of papers belong to as well as scientific and technical
areas to which the published papers are related. In addition to that, the key results of
particular papers which, according to the authors, were most impressive during the
previous period of DEMI conferences, are specified. In addition to recommending that
more scientists from different institutions should participate in the conference, the
conclusions also indicate metal forming areas which are being developed intensively
worldwide.
Key words: DEMI conferences, metal forming,papers
1.
INTRODUCTION
The basis for writing this paper is the fact that the ongoing conference DEMI is
being held for the 10th time (the first one was held in 1998); therefore, there is a good
reason to talk about the Conference accomplishments up to now. Here, we shall
review only the papers published in the previous period related to metal forming area
as an area of specific production technologies, without including other numerous and
significant scientific areas. When selecting the papers to be analysed, the attention
was primarily paid to the connection of contents, methodology and theoretical
foundations which are typical for different metal forming procedures. There are
different approaches possible for each classification and determining of connection of
particular scientific areas; therefore, the list of papers here presented and analysed
can be changed.
Such reviews are extremely significant from many aspects, the most important,
probably, being the one that shows the accomplished level of development of a
particular scientific area in comparison with the results achieved in the surroundings
and worldwide. In addition to that, the contents of published papers indicate the
1
2
3
Prof. dr Milentije Stefanovi, Faculty of Mechanical Engineering Kragujevac, Serbia, stefan@kg.ac.rs

Prof. dr Srbislav Aleksandrovi, Faculty of Mechanical Engineering Kragujevac, Serbia, srba@kg.ac.rs
Ass.prof. dr Dragan Adamovi, Faculty of Mechanical Engineering Kragujevac,Serbia, adam@kg.ac.rs
391
M. Stefanovic ,S. Aleksandrovic, D. Adamovic
development of a particular economic area, association with industry, potential of

scientific and research institutions outside university etc.
2.
THE SELECTION AND ANALYSIS OF PUBLISHED PAPERS
At the previous nine DEMI Conferences, 62 papers in the area of metal forming
were published in total. Table 1. shows the list of authors per institutions which employ
them. Most papers, as expected, are from the Faculty of Mechanical Engineering in
Banja Luka (24), followed by authors from the Faculty of Mechanical Engineering in
Kragujevac (19) and the Faculty of Technical Sciences in Novi Sad (5). Experts from
the area of industry published only three papers. The authors from the Faculty of
Mechanical Engineering from east Sarajevo, Cars Institute Zastava from Kragujevac,
Institute for Technology of Nuclear and Other Mineral Raw Materials from Belgrade
etc. also took part in the Conferences.
The transformation of social capital into private capital has had a disastrous
influence on further existence of factory-related institutes which have been a
cooperative link between economy and universities for years. That is the reason for
decreasing number of scientific papers in this area and reduced possibilities for
transfer of new findings from universities research centres into industry. There are
many such examples in the Republic of Serbia, e.g. closed Institute for Ferrous
Metallurgy SARTID in Smederevo and the recent example of the Cars Institute Zastava
from Kragujevac.
Table 1 Review of published papers per institutions
Year when
conference
was held
1998.
1999.
2000.
2001.
2002.
2003.
2005.
2007.
2009.
Total
INSTITUTION
to which authors of papers in the area of metal forming belong
FME
FME
TF
HTS
Other*
Total of
Banja
Kraguje- Novi Industry Doboj
papers
Luka
vac
Sad
3
1
1
5
2
2
1
1
6
3
1
1
5
5
1
1
7
3
3
2
1
9
2
2
1
1
1
7
1
2
2
5
3
5
1
1
1
11
2
3
2
7
24
19
5
3
4
7
62
*FME East Sarajevo (2), IAZ Kragujevac (1), ITNMS Belgrade (1), FME Nis (1), TS Banja Luka
(1), Eftime Murgu Univ. Romania (1)
Table 2. reviews the published papers per metal forming areas. This
classification is also optional, and can be made in a different way. Main contents are
divided into: display of MF status and further development, materials formability in MF
procedures, sheet metal forming processes and bulk hot and cold forming. The areas
392
About papers in the area of metal forming presented at DEMI conferences held so far
of micro-forming, composites, machines and tribology are separate. Most papers were
published in the area of cold and hot bulk forming (23), sheet metal forming (14) and
formability (11) etc.
Table 2 Review of published papers per metal forming areas
No.
1.
2.
3.
4.
5.
6.
7.
8.
MF area to which the published

paper is related
Development of processes and
new technologies of MF
Materials formability in MF
processes
2.1. Sheet metal forming by deep
drawing
2.2. Bulk forming
Sheet metal forming processes
3.1. Cutting through/punching
3.2. Bending
3.3. Deep drawing
Processes for hot and cold bulk
forming
4.1. Pipes rolling and forming
4.2. Profiles pressing and
hardening
4.3. Cold extrusion and
embedding, rings rolling
4.4. Hot forging
Micro-forming, Size-effect
Composites forming
Machines for MF
Tribology and ecology in MF
processes
Papers number
6, 33
Total of
papers
2
7, 16, 27, 30, 37, 38, 39, 53
29, 44, 54
59
9, 19, 24, 36
5, 21, 23, 40, 41, 43, 47,
50, 57
1
4
9
2, 3, 4, 10, 13, 18, 20, 22,

35, 52
1, 8, 15, 25
10
4
12, 28, 55
11, 32, 42, 48, 56, 61

34, 46,
51, 58
14, 17, 45
26, 31, 49, 60, 62
6
2
2
3
5
According to their contents, the published papers mainly followed the ongoing
trends of worldwide MF development. The published results reflect the capability of
research institutions for scientific work and the presence of young researchers, so it is
obvious that there are no satisfactory results in particular areas of science and
technology. When comparing the contents of DEMI papers with contents of papers
presented at eminent conferences worldwide, related to metal forming area, e.g. [1],
[2], [3], [4], it is obvious that the particular scientific areas are missing. This is primarily
related to the introduction of modern materials into technological processes
(composites, bio-materials, alloys Al, etc.) and development and application of
innovative technologies (micro-forming, laser application, semi-hot forming,
incremental forming, high-speed forming, super-plasticity etc.). Similar conclusions can
be reached after the analysis of papers published at conferences in the area of
393
production engineering in the Republic of Serbia. Quality changes in the research

areas, i.e. adjustment to modern trends of MF development, can happen with the
further state support, increase of investments into science and inevitable integration of
economy and research institutions. In the Republic of Serbia, a very important factor is
a new possibility for inclusion of young researchers and students attending doctoral
studies into constantly financed work on state projects in the technological
development area.
4.
CONCLUSION
At previous 9 DEMI conferences, 62 papers were published in the area of

metal forming. Only three papers were published by experts from the industry area, 2
were published by research institutes and the others belong to the faculties and
colleges. The lack of authors from the economy, i.e. factory-related institutes, is
obvious. Due to the privatisation of public companies, many factory-related institutes
were closed down, which ruined significantly the possibilities for the transfer of
knowledge on relation university-industry and further economic progress of the society
as a whole.
According to their connection to the MF area, the published papers were
roughly divided into: status reviews and further MF development, materials formability
in MF procedures, sheet metal forming processes and bulk hot and cold forming. Areas
of micro-forming, composites, machines and tribology are separate. Most papers were
published in the area of cold and hot bulk forming (23), sheet metal forming (14) and
formability (11) etc. According to their contents, the published papers mainly followed
the ongoing trends of worldwide MF development.
The published results reflect the capability of research institutions for scientific
work and the presence of young researchers, so it is obvious that there are no
satisfactory results in particular areas of science and technology. This is mainly related
to introduction of modern materials into technological processes (composites, biomaterials, Al alloys etc.) and development and application of new technologies (microforming, laser application, semi-hot forming, high-speed forming, super-plasticity etc.).
The presented materials can also be used for more extensive and specific analyses of
the position and role of scientific and research work in the economic development of
particular countries.
LITERATURE
[1]
[2]
[3]
[4]
394
10th ICTP 2011, International Conference on Technology of Plasticity, Aachen,

Germany, 2011,www.ictp2011.com
The 8th international ESAFORM Conference on Metal Forming, Cluj-Napoca,
Romania, 2005, http://conference.utcluj.ro/esaform2005/
50th IDDRG Conference" Graz , Austria, 2010,
http://www.iddrg.com/pub/iddrg/central/index.html
The 3rd International Conference on Manufacturing Engineering, (ICMEN)
Chalkidiki, 2005., http://ithaki.meng.auth.gr/index-3.asp
Enclosure:
The list of authors and papers in MF area published at DEMI conferences from 1989 to
2009:
1998.
1. M. ljivi, O. Mileti, . Babi, Possibilities of the Technology of Halfhot
Extrusion, pp. 21-25.
2. Z. Boikovi, Computer Aidid Manufacturing of Polygonal Tubes on CNCHydraulic tandem Press, pp. 26-31.
3. O. Mileti, M. ljivi, . Babi, Phenomena of Outstripping and Remaining in
Rolling Process of Thin Sheet Metals, pp. 32-39.
4. B. Mii, I. ui, M. ljivi, Horizontal Solution of Strip Accumulation on the
Rolling Lines of Welded Tubes, pp. 45-49.
5. . Babi, Measurement Deep-Drawing Process Force, pp. 50-55.
1999.
6. M. ljivi, O. Mileti, . Babi, New Techolology -THIXOFORMING, pp. 66-71.
7. M. Stefanovi, S. Aleksandrovi, M. Milovanovi, M. Samardi, New Materials
and Technologies in Car Body Manufacturing , pp. 61-65.
8. O. Mileti, M. ljivi, . Babi, Pressing of Aluminium alloy by backward
extrusion method, pp. 76-81.
9. Z. Boikovi, Additional analysing on tools problem and possibility for
manufacturing protective rail-bumper by bending process, pp. 107-109.
10. B. Mii, M.ljivi, Contribution to the plasticity analysis in pipes rolling,
pp.110-113.
11. V. Mandi, M. Stefanovi, Physical simulation of hot forming process by
applying plasticine as modelling material, pp. 72-75.
2000.
12. Miroslav Planak, Dragia Viloti, ore upkovi, An Contribution to the
Investigation of Double Inverted Extrusion, str.18-21.
13. Ostoja Mileti, Milan ljivi, Drago Blagojevi, ivko Babi, High-speed
Expand Process Applied on Tube Joints in Heat Exchangers, pp. 78-81.
14. Ostoja Mileti, Vid Jovievi, Mladen Todi, Tihomir Latinovi, Concept of
crank press with rotary tools for hot pressing, pp.102-105.
15. Mladen Todi, Ostoja Mileti, Tihomir Latinovi, Normal and tangential
stresses on contact surface at axis-symmetrical profiles drawing, pg. 106-109.
16. M. Stefanovi, S. Aleksandrovi, M. Milovanovi, M. Samardi, Development
and Application of Al-Alloys in Manufacture of Carbody Elements of Passenger
Cars, pp.37-40.
395
2001.
17. .V. Jovievi, P. Daki, O. Mileti, M. Todi, Some Aspects of Research of
Heavy Presses Production Possibilities, pp. 45-50.
18. O. Mileti, M. Todi, M. urevi, Enlargings of Tubs Adapters for Bushing, pp.
75-80.
19. O. Mileti, P. Daki, M. Todi, T. Latinovi, Analysis of process of thin-waled
work-piece in die bending tool and by roll bending, pp. 93-102.
20. M. Todi, O. Mileti, V. Jovievi, T. Latinovi, Balance of Energy by Pulling
Axis of Symmetry Profilies, pp. 103-108.
21. . Babi, M. ljivi, D. Adamovi, Influence of Tribological Conditions on Limit
Drawing Ratio, pp. 116-120.
22. B. Mii, Parameters for Value of Phases Number Configuring Aright-angle
Tubes, pp. 143-148.
23. S. Aleksandrovi, M. Stefanovi, Deep Drawing of Aluminium Alloy AlMb4,5Mn
Thin Sheet in Nonmonotonous Forming with Variable Blank Holder Force, pp.
69-74.
2002.
24. O. Mileti, M. Todi, Work - hardening angle at contour roll bending, pg. 65-70.
25. M. Todi, O. Mileti, V. Jovievi, T. Latinov, Significance of parameters in
drawing process related to mechanical properties of axis-symmetrical profile,
pg. 91-96.
26. . upkovi, M. Planak, D. Viloti, Influence of Initial Disc Dimensions on
Diagram for Determination of Coefficient of Friction in Metal Forming, pp. 97102.
27. . Babi, M. ljivi, Application and Development Tailored Blanks in
Manufacture of Body Cars, pp. 121-126.
28. M. Milutinovi, D. Viloti, . upkovi, Strain analysis at manufacture of
conical punch by embedding method, pg. 127-133.
29. M. Krainik, Determining of formability of materials at cylinder compression by
spherical dies, pg. 159-165.
30. M. Stefanovi, S. Aleksandrovi, . Babi, T. Vujinovi, Formability of Thin
Sheet Metals at Deep Drawing in New Technologies Conditions and
Applications of Modern Materials, pp. 43-54.
31. D. Adamovi, M. Stefanovi, V. Lazi, Investigation of parameters which
influence the ironing process, pg. 166-172.
32. V. Mandi, M. Stefanovi, Significance of physical-numerical modelling in
designing hot forging technology, pg. 147-152.
2003.
33. M. ljivi, Achievements and Perspectives in Development of Production
Technologies, pp.11-22.
34. M. Planak, D. Viloti, J. Jevremov, Size Effect in Metal Forming, pp. 51-54.
35. B. Mii, S. Baji, Z. Boikovi, Applicability of manufacture method for
tubular elements with complex longitudinal and cross section, pp. 101-106.
396
36. O. Mileti, M. Todi, Strain flow at bending at 1800, pp. 125-134.

37. M. Milovanovi, M. Stefanovi, V. Milovanovi, Analysis of Possibility for
Introduction of New Materials and Technologies for Manufacturing of Car Body
Parts, pp. 557-562.
38. S. Aleksandrovi, M. Stefanovi, V. Mandi, T. Vujinovi, Application
Perspective and Current Formability Problems of High Strength Steel Sheets,
pp. 579-84.
39. M. Stefanovi, V. Mandi, M. ivkovi, S. Aleksandrovi, Physical and
Numerical Modeling of Deflexion Appearances at Thin Sheet Metals-Yoshida
test, pp. 45-50.
2005.
40. . Babi, M. ljivi, New Technologies in Manufacture of a Carbody, pp.183190.
41. L. Coman, Technological Considerations Regarding Deep Drawing, pp.203208.
42. V. Mandi, Elimination defects in Metal Forming Processes by Using Modelling
and FEM Simulation, pp. 153-158.
43. S. Aleksandrovi, M. Stefanovi, T. Vujinovi, Control of Blank Holding Force
in Deep Drawing Proces-Significance and Limitations, pp.159-164.
44. M. Krainik, D. Viloti, M. Planak, M. Stefanovi, Material Formability in the
Process of Upsetting of Cylinder by Holoww Dies, str. 225-230.
2007.
45. O. Mileti, M. Todi, Examination of Carrying Structure of Hydraulic Press, pp.
159-168.
46. M. Stefanovi, S. Aleksandrovi, M. Samardi, Micro Forming Possibilietis
and Limitations pp.181-186.
47. S. Aleksandrovi, M. Stefanovi, T. Vujinovi, Possibilities of Deep Drawing
Process Control pp.187-192.
48. M. Stefanovi, Z. Guliija, V. Mandi, Specific Properties of Forging of High
Strength Al-alloys pp.193-198.
49. D. Adamovi, M. Stefanovi, M. ivkovi, Experimental Methods for
Determination of Friction Coefficient During Ironing, pp.199-206.
50. M. Samardi, M. Milovanovi, M. Stefanovi, Forming Simulation of Deep
Drawing in Concurrent Concepe the Car Body Development Process, pp. 207212.
51. M. Todi, O. Mileti, Influence of Position of Layer of Two-Layer Composite at
Banding on Stability of Process, 221-226.
52. R. Miji, Minimal Radius of R Bending of Pipes by Cool Deformation,
pp.251-258.
53. . Babi, M. ljivi, Parameters of Laser Welding Blanks Forming, pp. 213-220.
54. Z. Guliija, A. Patari, M. Stefanovi, Effect of Electromagnetic Field on
Mechanical Properties of Continualy Casted AI EN AW 2024 Alloy, pp. 277-282.
55. I. Trbojevi, Factors which Impact Manufacturing Technology for Ring-Shaped
Parts, pp. 283-290.
397
2009.
56. V. Marinkovi, Analysis of the Influential Factors on the Flow Stress in the
Warm Forging Processes by Applying Latin Square Design, pp. 163-168.
57. . Babi, S. Aleksandrovi, M. ljivi, Influence of a Weld-Line Location in
Deep Drawing of Tailored Blanks, pp.181-188.
58. M. Todi, O. Mileti, D. Mari, Hardness in Function of Stress-Strain State of
Two-Layeer Composite, pp.189-194.
59. D. Mari, M. Todi, O. Mileti, Influence of Clearance of the Separation of
Closed Contours, pp. 195-200.
60. M. Stefanovi, M. orovi-Stanojevi, S. Aleksandrovi, Ecologically Aspeccts
of Development Metal Forming Technologies, pp. 335-338.
61. M. Stefanovi, Z. Guliija, V. Mandi, Modern Procedures for Hot Forging of
Orthapaedic Implants, pp. 151-156.
62. D. Adamovi, M. Stefanovi, S. Aleksandrovi, Modelling of Tribological
processes in Ironing, pp. 151-156.
398
POSITION OF THE NEUTRAL SURFACE DEFORMATION AT

BENDING TWO LAYER COMPOSITES
Todic Mladen1, Miletic Ostoja2
Summary: The position of the neutral surface deformation, at bending two layer
composite is in function: of layers thicknes, mechanical characteristics of the material
layers, the position of the base and plakiranog layer relative to the center of bending
and degrees of the deformation. However, the mechanical properties of layers and
their thickness have a dominant influence. During the process of forming the position of
neutral surface is nemonoton, and nemonotost increases with increasing degree of
deformation.
Key words: strain condition, composite, explosion welding, neutral surface of deformation
1.
INTRODUCTION
Layered metal laminar composites are growing in the application

in science and technology. Commercial production of these composites
is possible with the plating technologies and plaquing. Their technology
proces intermediate present and products require reliable processes
that do not lead to bifurcation of the boundary layers and layers
themselves. As a consequence of defects can be caused microcracks
which become initial parameter for the degradation of the applied
composite products. Before the technological process of layered
composites it is very important to know which layer composita can be in
the application on the inside, and wich on the outside. Due to the basis
of these findings, access technology which processes the composite,
will not lead to stratification on the boundary layer and thus in layers of
composites, the degradation of the layer thickness and the formation of
microcracks. The position of the neutral surface can be described
analytically and experimentally.
1
2
Dr Mladen Todic, Banjaluka, Faculty of Mechanical Engineering, (mladentodic@unibl.rs)

Dr Ostoja Miletic, Banjaluka, Faculty of Mechanical Engineering, (miletic@urc.rs.ba)
399
Todic Mladen, Miletic Ostoja
2.
DEFINITION OF POSITION BY ANALYTICAL WAY

For determation of radius of neutral surface deformation
UM ,
is needed to
know the size of an external ( Rs ) and internal ( Ru ) bend radius, on this base we can
determine the radius of neutral layer surface of deformation
UM
Rs2 Ru2
2su
(1)
where :
su thickness two-layer composite of pripremka

After determing the radius of neutral layer of deformation, knowing and the size
of the boundary layer of liquid metals, primary and plakiranog (V cU i V U ) can be
determined the border between on the elastic and plastic zones
Intensity of deformation the cross-term cross-section:
x
on the outer surface: U Rs ,
M is
x
2
R
ln s
3 UM
on the inner surface:
Ru ,
(3)
on the border of clad and the basic layer
R
2
ln g
3 UM
M ig
x
U
2
ln M
3 Ru
M iu
x
(2)
the neutral layer stress
U
2
ln M .
3 UV
M iU V
Rg ,
(4)
U UV ,
(5)
The previous equation leads to the geometric size of the intersection of twolayer materials:
3
Rs
UM e 2
M is
3
M iu
2
Ru
UM e
Rg
UM e 2
UV
400
UM e
(6)
(7)
M ig
3
M iUV
2
(8)
.
(9)
Position of the neutral surface deformations at bending two layer composites
Radius of neutral surface of deformation we determine from the condition of

stability of volume (ignoring and elastic changes in volume),
Rs2 Ru2
2su
where su thickness of pripremka.
UM
(10)
If the expression (10) includes quantity given by the equations (6) and (7),
obtaines,
UM
su
3M is
e
(11)
3M iu
Substituting expressions (11) in the equality (5) and (6) obtaines,
Rs
su
3
M is
2
2e
e
3M is
e
(12)
3M iu
Ru
2e 3M iu
su e 3M is e 3M iu
Rs Ru s current thickness of workpiece.
(13)
Equation (13) according to the equation (12) ,
s
su
2 e
3
M is
2
3M is
e
e
3
M iu
2
3M iu
(14)
From the condition of stability of volume the basic layer of the workpiece, we get the
radius of a neutral layer of deformation
UM
Rg2 Ru2
2s'
where s ' thickness of the the basic layer of pripremka.
(15)
Using expressions (6) and (7) we have,
UM
s'
2
e
3M ig
e
(16)
3M iu
that is,
Rg
s'
Ru
s'
2e
e
3M ig
2e
e
3M ig
3
M ig
2
e

(17)
3M iu
3
M iu
2
e
3M iu
(18)
401
The term of the current relative thickness of the basic layer is obtaines from equation
(18) and (17)
s
s'
2
3
M ig
2
e
3
H iu
2
(19)
As with previous volumes in terms of stability of clad layer of the workpiece is obtained
UM
Rs2 Rg2
2 s"
(20)
where s" thickness of clad layer pripremka
UM
s"
Rs
s"
2
e
3M is
3
M is
2
2e
e
Rg
(21)
3M in
e
3M is
2e
(21)
3M ig
e
3
M ig
2
(22)
3M
s" e
e ig
su , Rs Rg s"
3M is
s'
s"
2
e
3
M is
2
e
3
M ig
2
(23)
Current thickness of workpiece makes current thickness of primary and clad

layers
s
3.
s' s"
EXPERIMENTS TO DEFINE THE POSITION OF NEUTRAL SURFACE

DEFORMATION BY METHODS OF PARALLEL LINES
In order for this method to fin positions of neutral lines of deformation and
positions of neutral line of stress and strain should be conducted lines on lateral
surface of pripremka. Lines can be applied using the laser and microscope tool. Lines
are perpendicular to the outer surface weight, distance between the line is about one
millimeter and may also be lower. Through three-dimensional and two-dimensional
camera in a continuous deformation are, identifyed the current status of inflicted
parallel lines until they are straight lines (unless they are making no significant
influence of shearing stress to the position of neutral line of plastic deformation).
Processing the recorded position leads to the position of neutral axis of deformation at
more layer and single-layer. Strain neutral layer does not change its length of
deformation during bending, axcept changes to the curvature which is the result of
bending radius. On the basis of recorded geometric quantities calculated position of
402
Position of the neutral surface deformations at bending two layer composites
the neutral line, in relation to inner workpiece surface. Mathematical dependence can
be expressed by the following equation
snM
(l l ' )
s
(l ' 'l ' )
(24)
where:
s nM , mm distance from the neutral surface deformation of inner surface of workpiece,
l, mm the initial gap in parallel lines on the sample,
l , mm spacing parallel protected line on the inside surface of the work,
l, mm space between parallel lines painted on the outside surface of the workpiece,
s, mm thickness of workpiece.
Display of size of the position and distance of painted lines on the lateral
surface pripremka and workpiece during bending double-layer material (composite),
Figure 1.
l''
snV
so
l'
a.
b.
Fig. 1 Geometrical size of the position and distance of parallel nanasenih lines on
lateral surface of a. pripremka and b. workpiece
Display of pripremka and workpiece when plakirani layeris on the inside surface,
and the outer surface with to inflicted lateral parallel lines at distance one millimeter,
figure 2 and 3.
Fig. 2 Parallel lines applied lateral to the dual-layer preform

(base layer on the inside side)
Fig. 3 Parallel lines lateral on a dual-layer preform

(base layer on the outside)
403
This method can be effectively used to determine neutral position of

deformation at bending layered materials of total thickness greater than two
millimeters. Below two millimeters application is hampered, and that is greater error
them depends on the accuracy of nano parallel lines and measurement of geometrical
size of the workpieces. With increasing thickness of clad formulation plakiranog layers:
neutral layer deformation shifts to the side of clad layer, the strain induced by this
increasing on the inner surface and is more increasing due to shifting of neutral layer
deformation to the side of clad layers, relative clad layer thinning is reduced and
thickened base layer increasing, but strengthing of clad layers, it slightly deformes and
causes great strain of base layer so that the total thickening of base layer exceeds
volue of clad thinning as result we get increased strengthening of the wall thickness of
workpiece. With increasing strain by strengthening (reduction Ru/s) : neutral layer
deformation moves to the side of the center of curvature of bending of greater clad
layer thickness neutral layer of the beginning of process of deformation can be found in
the clad layer and with more intense deformation neutral layer of deformation gradually
changes to the basic layer, the result is to increase strain on the inner surface more
rapidly than the outside, clad layer thinning and thickening of the base layer growth, if
Ru/su<2) is thickening of the base layer, then it is gradually increasing when the
intensity of stress crosses second boundary condition starts thinning in the layer.
CONCLUSION
The results of theoretical and experimental determination of the position of
neutral surface of strains differ in acceptable limits. It is essential before any process of
technological bending of layered composite to knowledge of mechanical properties,
geometric size of composite layers and the position of layers in the application.
Knowing these parameters leads to the stability of technological processes of
composite and its applications. If we do not analyze these parameters before the
technology process starts stability can not the predicted. Thus, can not be given a
guarantee that it will not occur and emergence of micro-cracks that may affect the
application of the product which is made from layered composites.
LITERATURE
[1] Todic M. (2006), Position of neutral surface stress and strain in composite
laminates, PhD thesis, Faculty of Mechanical Engineering Banja Luka.
[2] Miletic O. (2002). Konstitutivne relacije procesa neprekidnog profilisanje
kompozitne ljuske i ploce, Univerzitet u Banjoj Luci Masinski fakultet.
[3] Miletic O.,Todic M.Uticaj triboloskih e3fekata kontaktnih povrsina na fenomen
ojacavanja, Orocedings 2008, p.179-183.
[4] Todic M., Miletic O., Maric D. Tvrdoca u funkciji naponsko-deformacionog stanja
dvoslojnog kompozita, DEMI 2009, p. 189-195.
[5] Todic M., Miletic O., Maric D. Uticaj polozaja sloja dvoslojnog kompozita pri
savijanju na stabilnost procesa, DEMI 2007, p. 221-227.
404
POSSIBILITY TO USE RHEOCASTING PROCESS FOR

MANUFACTURING PARTS IN AUTOMOTIVE INDUSTRY
Milena Cosi1, Zagorka Acimovi-Pavlovi2, Zvonko Guliija3, Mirjana
5
Stojanovi4, Zoran Janjuevi
Summary: In this work, the results of research concerning application of rheocasting
process for piston alloys structure and properties improvement have been shown. The
procedure of stirring the melt by a plate stirrer at different speeds (rpm): 500, 1000 and
1500 was used. The influence of mechanical stirring effect on morphology and
distribution of primary phases (D- Al i Si) was tested in the hypereutectic Al-18wt%Si
alloy. Test results showed a positive influence of melt stirring speed on the structural
change of the tested alloy in comparison with the structure of samples obtained by a
conventional casting procedure. The effects of a structural change are optimised at the
melt stirring speed of 500 rpm.
Keywords: piston alloy, rheocasting process, castings quality.
1.
INTRODUCTION
Aluminium-silicon alloys are a part of the group of important alloys for

automotive industry. Piston alloys, with reference to application conditions, must show
good thermal stability, good heat conductivity, corrosion resistance and satisfactory
tribological properties. Within the exploitation life time, these alloys are subject to
mechanical and heat stresses and to agressive environmental application conditions.
Piston alloy properties greatly depend on a proper choice of a melting technology, melt
treatment, casting process and thermal treatment of castings.
A basic limit factor affecting application of these alloys is the presence of
coarse and brittle Si crystals in a soft Al base. One of important parameters influencing
piston alloys properties is a modification procedure, i.e. the choice of type and quantity
1
M. Sc., Milena Cosi, Tehnikum Taurunum, College of Mechanical Engineering, 11080 Belgrade, N. Dimic
4, Serbia, cosic.milena@gmail.com.
Ph. D., Zagorka Aimovi, Belgrade, Universtiy of Belgrade, Faculty of Technology and Metallurgy, 11120
Belgrade Karnegijeva 4, Serbia zagorka@tmf.bg.ac.rs
3
Ph. D., Zvonko Guliija, Institute for Technology of Nuclear and Other Mineral Raw MaterialsITNMS
Franchet dEsperey street, 11000 Belgrade, Serbia z.gulisija@itnms.ac.rs
4
Ph. D., Mirjana Stojanovi, Institute for Technology of Nuclear and Other Mineral Raw MaterialsITNMS
Franchet dEsperey street, 11000 Belgrade, Serbia m.stojanovic@itnms.ac.rs
5
Ph. D., Zoran Janjuevi, Institute for Technology of Nuclear and Other Mineral Raw MaterialsITNMS
Franchet dEsperey street, 11000 Belgrade, Serbia z.janjusevic@itnms.ac.rs
2
405
Milena Cosi, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi, Zoran Janjuevi
of modification agent. With Al-18wt%Si alloy, a modification process is mostly limited to

modifying primary Si crystals. These procedures are aimed at a break and proper
distribution of Si particles in a soft aluminium base. As a modifier, phosphorus is
commonly used. A basic imperfection concerning application of this modifier is the fact
that eutecticum remains needle-shaped and coarse, which is manifested by poor
mechanical properties. Usage of Sr as a modifier significantly improves alloy properties
and this modifier was used in the tests shown.>1-3@
One of the ways to change the structure is to apply new processing
procedures in the area of a semi-solid state, representing an important concept for
development of new casting technologies giving way to the possibility to change the
dendrite structure in a non-dendrite one. These procedures include rheocasting and
tixocasting procedures. Rheocasting referrs to the alloy stirring procedures during
solidification in the area between a solid and a completely liquid state , i.e. in a two
phase region, followed by the casting process. >4@
This work shows the results of microstructural changes made by mechanically
stirring a semi-solid melt at different stirring speeds.
2.
EXPERIMENT
Table 1. shows the chemical composition of the alloy tested. The hypereutectic
Al-18wt%Si alloy was obtained by conventional (series I) and rheocasting procedure
o
-1
(series II) at the cooling rate of 0,16r0,05 Cs . The samples of rheocasting alloy were
obtained at three rotary melt stirring speeds (rpm): 500 (marked II 1),1000 (marked II2) i
1500 (marked II3).
Table1. Chemical composition of Al-18Si alloy (wt. %)
Element
Wt.%
Si
18
Cu
0.8-1.5
Mg
0.8-1.3
Ni
0.8-1.3
Fe
0.7
Zn
0.2
Mn
0.2
Ti
0.2
Al
Balance
For liquid cast preparation, NaCl-based, KCl-based and Na3AlF6-based salt

was used. Cast degassing was carried out with hexachloretane (C6H6), and
modification of Al10%Sr was carried out in the quantity of 0.05%Sr. Series I samples
were casted into metal molds. Samples II were obtained by rheocasting process. In
each experiment, the alloy was heated up to 720oC. After that, the alloy melt was
cooled at the cooling rate of 0,16r0,05oCs-1.
Apparatus used to obtain rheocast samoles consists of a crucible placed in an
electric resistance furnace. The temperature was controlled by a thermocouple
accommodated in the crucible inner side wall. The temperature was recorded each 30s
in the given temperature interval. For experimental purposes, a specially designed
plate-shaped stirrer with the dimensions of (38x2x100) mm was used; it was made out
of a steel sheet and an aluminium oxide was applied onto it by the plasma technique.
The stirrer was connected to an electrical motor with the power of 250 W and
maximum number of revolutions of up to 2000 rpm.
For microstructural tests, all samples were prepared through a standard
metalographic procedure of polishing and etching of samples (by Kelers reagent).
Microstructural tests were carried out applying the optical microscope REICHERT
JANG.
406
Possibility to use rheocasting process for manufacturing parts in automotive industry
3.
RESULTS AND DISCUSSIONS
Figure 1. shows microstructures od series I and series II samples.

Microstructure of samples I, (fig.1.a.) consists of polygonal-shaped primary Si crystals
accommodated in the eutectic base consisting of (D- Al i Si). In the area around
primary Si crystals, there is an D - phase with dendrite morphology. Figure 1.b-d shows
microstructures of series II samples created during rheocasting process.
a) Samples, series I
b) Samples, series II 1
c) Samples, series II 2
d) Samples, series II 3
Fig. 1 Microstructure of series I-II sample

Particle distribution of primary D- Al phases in the area around the stirrer was
analysed. During subcooling created by introducing the stirrer into the metal melt, the
first D- Al particles in a globural shape are locally formed around the stirrer. Stirring
speed increment influences the increment of the number of nucleation germs resulting
in reduction of the D- Al particle size. Increasing the stirring speed, the D-Al particle
shape is also changed. With the samples II 1, rosette-shaped particles are noticed (fig.
1.b), and with increment of the stirring speed (samples II 2 and samples II 3 ), an
increment of the roundness degree of these particles is noticed, fig. 1.c-d.
The tests showed that congregating rate of primary Si particles is increased
with an increment of the melt stirring speed. Gathering rate of Si particles is not
noticable at lower stirring speeds, i.e. at the stirring speed of 500 rpm, (fig.1.b), while
segragation of Si particles is more noticable at higher stirring speeds. The process of
congregating primary Si crystals in clusters-blocks, consisting of a number of individual
Si particles mutually connected, is a consequence of shearing forces appearing during
melt stirring, fig. 1.c,d.
A basic imperfection of a conventional casting process for the piston alloys is a
distinctive notion of primary Si particles segregation affecting mechanical properties of
alloys. When applying rheocasting process, it was noticed that Si particle segregation
is least expressed at the stirring speed of 500 rpm. Bearing in mind the importance of
407
Milena Cosi, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi, Zoran Janjuevi
the triad: technology-structure-properties in

the development of contemporary
materials, such as piston alloys, it may be expected that the structural changes
mentioned above, provoked by the change of rheocasting process parameters, will
have an influence on the properties of the alloy obtained, which should be the subject
of a further reasearch within this field.
4.
CONCLUSIONS
Test results of stirring speed influence on the microstructure of rheocast

samples of the Al-18wt%Si alloy in comparison with the microstructures of
conventionally casted samples show the following:
-By stirring the melt, the morphology of D - Al phase is changed from the
dendrite into a non-dendrite one. At 500 rpm, morphology of D - Al phase is rosettelike, while the roundness degree of these particles is increased with an increment of
the melt stirring speed. Increment of the stirring speed leads to a greater
fragmentation of dendrite morphology of the D - Al phase, resulting in smaller sizes of
these particles.
-Under the activity of shearing forces created as a result of mechanical melt
stirring, primary Si crystals are segregated. At the stirring speed of 500 rpm , a
segregation extent is least expressed; with an increment of the melt stirring speed,
(1000 and 1500 rpm) primary Si particles tend to form blocks. As the clusters of Si
particles deteriorate the properties of piston alloys, it is recommended to apply the
melt stirring speed of 500 rpm for this type of alloy.
These are preliminary test results showing that varying the rheocast process
parameters may significantly influence the change of the phase morhology in the
piston alloy structures, representing a fundamental dependance of structure from
technology. For a detailed analysis of influence of the rheocasting process parameters
on the change of structure and properties, it is necessary to conduct a detailed.
REFERENCES
[1] H. Mirzadeh, B. Niroumand,: Effects of rheocasting parameters on the
microstructure of rheo centrifuged cast Al-7.1wt%Si alloy, Journal of Alloys and
Compounds, 474 (2009) 257 - 263
[2] M. Reisi and B. Niroumand, Effects of stirring parameters on rheocast structure of
Al-7,1%wt.Si alloy, Journal of Alloys and Compounds, Volume 470,(2009) 413-419
[3] Delong Lu, Yehua Jiang, Guisheny Guan, Rongfeng Zhou, Zhenhua Li, Rong
Zhou: Refinement of primary Si in hypereutectic Al-Si alloy by electromagnetic
stirring, Journal of Materials Processing Technology 189 (2007) 13-18
[4] Z. Fan, J. Y. Chen, Modeling of rheological behaviour of semisolid metal slurries,
part 4 effects of particles morphology, Mater.Sci.Technol 18 (2002) 258-267
ACKNOWLEDGE
These investigations were conducted under the Project 34002 funded by the Ministry of Science
and Technological Development of the Republic Serbia.
408
PROPERTIES OF THE WROUGHT Al ALLOY 7075 OBTAINED BY

ELECTROMAGNETIC CASTING PROCESS
Zvonko Guliija1 Marija Mihailovi2, Aleksandra Patari3, Zoran Janjuevi4
Summary: The advantages of electromagnetic casting process (EMC) comparing to
the conventional continuous casting process of Al alloys are presented here. Besides
the better ingot surface obtained as the result of the reduced metal-mould contact
pressure, the uniform fine-grained microstructure, and hence the mechanical
properties of the alloy can be achieved. Following the correlation microstructuremechanical properties, the preferred mechanical properties can be assured by the
adequate microstructural properties obtaining. The microstructural properties
quantification is the object of this investigation. Finally, as the consequence of
improved as-cast microstructure, the casting defects are avoided, the material and
energy savings are achieved.
Key words: Al alloy 7075, electromagnetic casting
1. INTRODUCTION
The adequate combination of the main operating parameters of the
electromagnetic casting process (EMC), such as frequency and strength of
electromagnetic field, ensures the efficient elimination of the casting defects such as:
surface imperfections, grain boundary segregation, non-uniform grain size and porosity
of as cast ingots.
In this paper the special attention was paid to the effect of electromagnetic field
on the quantitative microstructural parameters: morphology, size, volume fraction and
distribution of phases. Knowing the microstructure-mechanical properties correlation, it
is possible to obtain preferred ingot properties by controlling the main operating
parameters and at the same time to avoid needs for additional operations, such as
machining and prolonged homogenization. This research is aimed to obtain better
quality of ingots at the beginning of the production process through the proper
1
Ph.D., Zvonko Guliija, Institute for Technology of Nuclear and other Mineral Raw Materials - ITNMS,
2
M.Sc., Marija Mihailovi, Belgrade, Institute for Technology of Nuclear and other Mineral Raw Materials ITNMS, m.mihailovic@itnms.ac.rs
3
M.Sc., Aleksandra Patari, Institute for Technology of Nuclear and other Mineral Raw Materials - ITNMS,
a.pataric@itnms.ac.rs
4
Ph.D., Zoran Janjuevi, Institute for Technology of Nuclear and other Mineral Raw Materials - ITNMS,
z.janjusevic@itnms.ac.rs
409
Zvonko Guliija, Marija Mihailovi, Aleksandra Patari, Zoran Janjuevi
combination of the main operating parameters. In this way, the great savings in energy
and time can be achieved.
2. EXPERIMENTAL
The EN AW 7075 is heat treatable high strength wrought alloy with application
in military industry, but it is characterized by a number of defects that occur during the
solidification process: porosity, hot cracks, non-uniformal grain size and crystal
segregation. The chemical composition of the used ENAW 7075 alloy is presented in
Table 1.
Table 1. Chemical composition of the alloy EN AW 7075
Element
Content
%
Zn
Mg
Cu
Mn
Cr
Fe
5.51
2.29
1.45
0.13
0.19
0.14
The experimental equipment consists of medium frequency induction furnace

with 100 kg capacity. There is a drainpipe, at the bottom of the furnace, with graphite
crystallizer that is intensively cooled with water. The low frequency magnetic field is
placed around the crystallizer itself. The testing samples were taken out of ingots with
a diameter of 80 mm, obtained by vertical continual casting. The casting temperature
was in the range of 710-720C and the average casting speed was 1,5 mm/s. The
frequency of electromagnetic field was 30 Hz, because both, our previous experience
shows that this is the optimal frequency [1], and literature data indicate the same [2-8].
The current intensity was 200 A.
The microstructure was examined by optical microscopy using the image
analysis device Leica Q500MC. Samples were metallographically prepared and etched
in Kellers reagent (revealing morphology of Al segregation-solid solution and intermetallic phases). The content of key alloying elements, Zn and Mg, was determined by
chemical analysis of the samples taken from the cross section of both ingots, obtained
with and without the electromagnetic field influence.
3
The samples of series 1 were obtained from the ingot casted without the
presence of electromagnetic field. Samples of series 2, were obtained from the ingot
casted in the presence of electromagnetic field with the frequency of 30 Hz. Their
comparison enabled the evaluation of the field effect on the microstructure.
The characteristic microstructure appearance with cellular/dendritic morphology
is the result of Al segregation from the solid solution during the solidification process.
The morphology of the samples casted without the electromagnetic field effect is more
dendritic, in comparison with distinctive cells obtained with the electromagnetic field
influence. The interdendritic type of porosity is reduced in samples of series 2,
obtained with electromagnetic field presence.
The microstructure was quantified by measurement of microstructural
parameters: dendrite arm spacing (DAS), interdendritic space width (LIMP) where
intermetallic phases and eutecticum were separated, as well as their volume fraction,
410
Properties of the wrought Al alloy 7075 obtained by electromagnetic casting process
VVIMP. They were acquired using linear method These parameters describe the
structure dispersity and they are the result of the solidification conditions and are
presented in Table 2.
Table 2 Measured microstructural parameters
Parameter
DAS (m)
Sample
1
2
min
7.86
0.79
max
268.4
172.5
LIMP (m)
average
45.8
31.8
min
0.76
0.38
max
12.10
26.40
VVIMP
average
3.40
2.90
VVIMP (%)
5.80
9.10
The decrease of both parameters, DAS and LIMP, from sample 1, obtained
without the electromagnetic field effect, towards the sample 2, casted with the
electromagnetic filed presence, can be observed. It is evident that the decrease is a
consequence of the electromagnetic field effect.
3. CONCLUSION
The investigations results of alloy EN AW 7075 ingots obtained with and
without the influence of electromagnetic field, undoubtedly show its positive effect on
the microstructure and mechanical characteristics. It is proven that the application of
low frequency electromagnetic field (of 30Hz) has the positive effects. Comparing the
samples casted with and without the electromagnetic field influence, the grain size
decreases, as it can be registered through the finer microstructure. Namely, sample 2
which is obtained by electromagnetic casting with a frequency of 30 Hz, has finer and
more homogenous microstructure than sample 1, which is obtained without
electromagnetic field influence. This was confirmed by measuring the quantitative
microstructural parameters DAS and LIMP. The sample 2, obtained by electromagnetic
casting, has lower values of dendrite arm spacing and lower interdendritic space width
than sample 1.
Better ingot quality contributes surface machining elimination, and uniform and
finer microstructure gives the possibility to shorten the heat treating time, thus the
material and energy savings are achieved.
ACKNOWLEDGEMENT
The authors wish to acknowledge the financial support from the Ministry of Science
and Technological Development of the Republic of Serbia through the project 34002.
LITERATURE
[1]
Ptari, A., Guliija, Z., Markovi, S., Microstructure Examination of

Electromagnetic Casting 2024 Aluminum Alloy Ingots. Practical Metallography, 44
(2007) 6, p.290-298
411
Zvonko Guliija, Marija Mihailovi, Aleksandra Patari, Zoran Janjuevi
[2]
[3]
[4]
[5]
[6]
[7]
[8]
412
Stefanovi, M., Guliija, Z., Jordovi, B., Patari, A., Hot Forging of High Strenght
Al Alloys. Journal for Technology of Plasticity, vol 32 (2007), No 1-2, p.29-38
Patari, A., Guliija, Z., Jordovi, B., Effect of Electromagnetic Field on the
Microstructure of Continual Casting Al 2024 Alloy Ingots, 3rd International
Conference Deformation Processing and Structure of Materials, September 2022, 2007, Belgrade, 141-149.
Patari, A., Guliija, Z., Mihailovi, M. (2009), The possiblity for application the
new production process for casting aluminum alloys, Machine Design 2009, Ed.
Sinia Kuzmanovi, University of Novi Sad, Faculty of Technical Sciences, p.431434
Vives C., Ricou, R. Experimental Study of Continuous Electromagnetic Casting of
Aluminium Alloys; Metallurgical Transactions B, vol 16, 1995, p 373.
Zhihao, Z., Jianzhong, C., Dong, J., Beijiang, Z., Effect of low-frequency
magnetic field on microstructures and macrosegregation of horizontal direct chill
casting 7075 aluminum alloy, Journal of Materials Processing Technology 182
(2007) 185190
Zhiqiang, C. Fei, J., Xingguo Z., Hai H., Junze J., Microstructures and mechanical
characteristics of electromagnetic casting and direct-chill 2024 aluminum alloys,
Materials Science and Engineering A237 (2002) p.133-137
Dong, J., Cui, J.Z., Le, Q.C., Liquidus semi-continuous casting, reheating and
thixoforming of a wrought aluminum alloy 7075, Materials Science and
Engineering A345 (2003) p.234-242
RAPID TECHNOLOGY IN DENTAL BIOMECHANICS

Vrtanoski Gligorce1, Andonovic Vladan2
Summary: The main purpose of the article is overview the advantages of rapid
technology in dental biomechanics, introduction and implementation in medical and
dental field and finally to view current and future status. In this field of science, complex
geometrical forms are ever presented and 3D data have to be processed where main
problem is how to collect data and how to design images. Rapid technology in dental
application such as restorative dentistry, implant dentistry and orthodontics will be
described as basic way in every process of additive manufacturing. However, dental
CAD/CAM systems are very important from the beginning of the process particular with
their graphical visualization and simulation. Some other details from this point of view
will be explained and general process introduction will be provided. Worldwide known
applicative software used in everyday rapid technology practice will be of great interest
offering state of the art overview in the mentioned area. Finally, an example will be
presented, with tendency to show the process for best understanding of above given
theoretically steps.
Key words: rapid technology, dental medicine, CAD/CAM technology, computer
tomography
1. INTRODUCTION
Rapid Prototyping (RP) is new technology, mostly developing in the design and
manufacturing industry. This technology provides an great opportunity to receive
physical details and the model without machining and tooling or molding manufacture,
by transforming data from CAD-system. Rapid prototyping is the automatic
construction of physical objects using additive manufacturing technology [1,2]. The first
techniques for rapid prototyping became available in the late 1980s and were used to
produce models and prototype parts, in the mechanical engineering field, in order to
evaluate the ease of assembly and manufacture of designed products ahead of actual
production. Today, they are used for a much wider range of applications and are even
used to manufacture production-quality parts in relatively small numbers. Building a
prototype is usually based on solid models from CAD-systems or models with closed
1
PhD, Vrtanoski Gligorce, Skopje R. Macedonia, University of Skopje - Faculty of Mechanical Engineering,
(gliso@mf.edu.mk)
2
PhD student, Andonovic Vladan, Stip R. Macedonia, University of Stip - Faculty of Mechanical
Engineering, (vladan.andonovik@ugd.edu.mk)
413
Vrtanoski Gligorce, Andonovic Vladan
surface contours. This model is divided into thin layers in cross section with a special
software, and the thickness of each layer is equal to the resolution of the equipment on
perpendicular coordinate [2,3]. Today, there are numerous terms for rapid prototyping,
including: freeform fabrication, solid freeform fabrication, autofab, automated freeform
fabrication, digital fabrication, 3D printing, laser prototyping, layer based
manufacturing, additive manufacturing, near net-shape form freeform fabrication,
layered manufacturing, solid imaging [5]. As overview of rapid prototyping technology,
it can be mentioned following: the manufacturing of the model is taking place layer by
layer from the bottom upwards which is based on the solid CAD file, production of
highly customised products is with no need of tooling, manufacturing of real models is
using directly solid 3D CAD files, and rapid manufacturing methods can result in time
reduction compared with the conventional manufacturing methods. Prototyping is a
mandatory step in the development of any new product creating a quality prototype,
similar to the future product. The device of rapid prototyping builds a model material by
layers sequentially, one after another, until getting the complete three-dimensional
model. Thus, the time of production the model does not depend on the complexity of
the geometry, but is determined only by the size of the prototype. This is one of the
major advantages of rapid prototyping systems compared with traditional technologies
such as such as machining or casting [6]. It is also necessary to note that for the
production model from RP it is not required expensive equipment, for casting, or
building programs for CNC machine tools in CAM systems, as occurs in machining.
These advantages are especially important in the manufacture of prototypes, when it is
made single products, but not a big series [6]. Unlike traditional technologies, such as
machining or molding, all the RP systems are set for the layer additive synthesis of the
models. The beginning of every RP system is a three-dimensional solid-state computer
model of the product, created in any 3D CAD program stored in the standard STL file
format. The process of RP can be shown on figure 1 as follows:
Fig. 1 RP process as a whole [2]

414
Rapid technology in dental biomechanics
A large number of competing RP technologies are available in the marketplace,

such as:
STL Stereolithography,
SGC Solid Ground Curing,
FDM Fused Deposition Modeling,
BPM Ballistic Particle Manufacturing,
SLS Selective Laser Sintering,
LOM Laminated Object Modeling,
MJM Multi Jet Modeling, etc.
As all mentioned belong to additive technologies, their main differences are
found in the way layers are built to create parts. Some are melting or softening material
to produce the layers (SLS, FDM) where others are laying liquid materials thermosets
that are cured with different technologies. In the case of lamination systems, thin layers
are cut to shape and joined together [6].
2. PROCES CYCLE DESCRIPTION
Since each process may use different materials and different techniques for
building the layers of a part, each process employs the same basic steps, even in
dental medicine, listed below [7].
Create CAD model - for all additive processes, the designer must first use
Computer-Aided Design (CAD) software to create a 3-D model of the part.
Convert CAD model into STL model - Each form of CAD software saves the
geometric data representing the 3-D model in different ways. However, the
STL format (initially developed for Stereolithography) has become the
standard file format for additive processes. Therefore, CAD files must be
converted to this file format. The STL format represents the surfaces of the
3-D model as a set of triangles, storing the coordinates for the vertices and
normal directions for each triangle.
Slice STL model into layers - Using specialized software, the user prepares
the STL file to be built, first designating the location and orientation of the
part in the machine. Part orientation impacts several parameters, including
build time, part strength, and accuracy. The software then slices the STL
model into very thin layers along the X-Y plane. Each layer will be built upon
the previous layer, moving upward in the perpendicular direction to X-Y
plane.
Build part one layer at a time - The machine builds the part from the STL
model by sequentially forming layers of material on top of previously formed
layers. The technique used to build each layer differs greatly amongst the
additive process, as does the material being used. Additive processes can
use paper, polymers, powdered metals, or metal composites, depending
upon the process.
Post-processing of part - After being built, the part and any supports are
removed from the machine. If the part was fabricated from a photosensitive
material, it must be cured to attain full strength. Minor cleaning and surface
finishing, such as sanding, coating, or painting, can be performed to
improve the part's appearance and durability.
415
3. RAPID PROTOTYPING IN DENTAL MEDICINE

Rapid prototyping technology is also useful in dental medicine for producing
medical models to assist diagnosis and simulate surgery [7]. In dental biomechanics,
rapid prototyping is a technology used to create models of any complexity directly from
digital data such as computerized tomography (CT) and magnetic resonance (MR).
Complex 2D images are converted into 3D models that can be easily manipulated for a
complete understanding of the anatomy and pathology [1]. Roadmap of rapid
prototyping with every step from start to finish used in dental medicine is given by
figure 2.
Fig. 2 Roadmap of rapid prototyping [1]

Complex 2D images are converted into 3D models that can be easily
manipulated for a complete understanding of the anatomy and pathologies [2,7]. This
process is shown on figure 3. This method consists of the following two steps:
Step 1: virtual modelling, see figure 3; this consists of using imaging
processing tools to create a virtual 3D model from DICOM made by
computer tomography (CT) or magnet resonace imaging (MRI) files.
416
Rapid technology in dental biomechanics
Step 2: physical modelling, see figure 3; the virtual prototype created

previously is built layer by layer through deposition of plaster powder.
Fig. 3 Plane images are converted into 3D [2]

When we talk about RP in dental medicine, the process can be shown in three
phases on figure 4, as follows.
Phase I: Digital geometry processing
Phase II: 3D digitalization
Phase III: 3D RP model

Fig. 4 Dental implants - the process in phases [2]
As a result of above mentioned steps in RP process, on figure 5 is shown end
products of that which are of great interests in dental medicine.
417
Fig. 5 Possible end products of RP technology [2]

4. CONCLUSION
This paper is supposed to show actual state in rapid prototyping technology
used in denatal biomechanics. Doubtless, the last years, modeling by computer is
widely used day-to-day practices increasing application in medicine especially in dental
biomechanics. However, as already mentioned all along in this paper, the main issue
concerns additive technology with all phases of the rapid prototyping process in dental
medicine. In conclusion, it can be said that rapid prototyping technology is very usefull
in oral and maxillo facial surgery. Every future step in RP developing will be of great
interest for dental medicine.
LITERATURE
[1] Daskalaki, A. (2009). Dental Computing and Applications. Max Planck Institute for
Molecular Genetics Germany.
[2] Andonovic, V.,Vrtanoski, G. (2009). New way of rapid technology used in dental
medicine. International Congress Machines, Technologies, Materials. Paper
No.50, Sofia.
[3] Bernard, G. T., Karunakaran, K.P. (2009). Evolutions of rapid product development
with rapid manufacturing: concepts and applications. International Journal Rapid
Manufacturing, vol. 1. Issue 1, p. 3-18.
[4] Elghany, K.A. (2005). Rapid prototyping and Additive Manufacturing in Egypt.
Proceedings of Egyptian Association for Additive Manufacturing EAAM. pp. 1-8.
[5] Bourell, D.L. (2009). Roadmap for additive manufacturing: Identifying the future of
freeform processing. The University of Texas Austin.
[6] Cohen, A. (2008). Vacuum Forming Applications Using Rapid Prototyping
Technology. White Paper, Objet Geometries.
[7] Chow, B.K.C. (2009). Applications of office-based three-dimensional technologies
including cone-beam computed tomography and rapid prototyping in the
management of maxillofacial traumaliterature review and a case report. Hong
Kong Dental Journal. vol. 6. p. 93-7.
418
SELECTION OF VARIANT FOR MATERIAL FLOW TYPE IN

CONDITIONS OF GROUP APPROACH USING THE SOFTWARE
SYSTEM TECNOMATIX PLANT SIMULATION
Stevo Borojevi1, Vid Jovievi2, Gordana Globoki Laki3, ore ia4,
Branislav Sredanovi5, Marko Radisavljevi6
Summary: This paper presents a model of material flows in terms of group approach
with the characteristics of the variability of operating groups along the materials flow in
the production system. It was presented the structure of this model and its mode of
functioning in the more intensive changes in environment and needs for development
of production systems. It was accentuate the importance for simulation of processes
planning using software system Tecnomatix Plant Simulation. Effects of the simulation
are given on concrete examples of determining flow type, in terms of group
approaches, in the process of design of materials flow in the production system.
Key words: simulation, modeling, variant of material flow type
1. INTRODUCTION
Simulation and modeling of process planning are executed using the following
steps:
Assessment and collection of data from real production processes, which
are essential for designing of simulation models,
Determination of target for simulation studies and the creation of a
simulation model in accordance with defined objectives,
Running experiments in the simulation model, to perform process of
simulation. It makes a number of results, such as how often the machines
are in state of failure, how often are blocked, which set the machine to use,
what kind of utilization of machines they have, etc.,
Interpretation of simulation data,
MsC. Stevo Borojevi, Banja Luka,Fac. of Mech. Engineering,+38751468320 stevoborojevic@hotmail.com

PhD. Vid Jovievi, Banja Luka, Faculty of Mechanical Engineering, +38751468320 vid.jovisevic@blic.net
3
PhD. Gordana Globoki Laki, Banja Luka,Faculty of Mech. Engineering,+38751468320 globocki@blic.net
4
PhD. ore ia, Banja Luka, Faculty of Mechanical Engineering, +38751468320 djordje@urc.rs.ba
5
Assistant Branislav Sredanovi, Univ. Eng., Banja Luka, Faculty of Mechanical Engineering, +38751468320
sredanovic@gmail.com
6
Marko Radisavljevi, Univ.Eng., Banja Luka, Megastil d.o.o., +38751456555
markoradisavljevic@yahoo.com
2
419
Borojevi S., Jovievi V., Globoki Laki G., ia ., Sredanovi B., Radisavljevi M.
In the process of defining tasks and goals of simulation studies, it is usually

necessary to ask the following questions:
What kind of throughput and output from process model we can expect?
What is the optimal number of resources (machines, workers, tools)?
Where is necessary to put a buffer?
What is the optimal size of buffer?
What the optimum number of pieces can be processed?
What strategies are most appropriate for the task?
How some or all of these factors combine and interact to produce different
results?
Simulations of process planning are applied in the case of: planning a new
production system, improving the existing production system and introduced a plan
that is defined in practice.
2. GROUP APPROACH IN DESIGN OF MATERIAL FLOW
In designing the material flow in manufacturing systems there are two border
approach, such as: individual and group, which significantly affect on the efficiency of
production systems [1].
Individual approach to the design of material flow in manufacturing systems is
based on the process planning for all parts and assemblies from the program
production.
Group approach in the design of material flow in manufacturing systems is
developed based on knowledge of the effects of application of group technology, which
basically aims to create a volume production on the principle of similarity. Increasing
the amount is based on the workpiece grouping with similar characteristics based on
the classification system, which contains criteria for classifying subjects into groups of
similar technological characteristics (Figure 1).
Fig. 1 Integration of workpieces, parts in the processing and components of the

assembly, in the operating group
420
Selection of variant for material flow type in conditions of group approach using the software
system Tecnomatix Plant Simulation
Thus integration of workpieces, parts in the processing and components of the

assembly, in the operating group (Figure 1), as the basic unit of observation in the
development of material flow, is performed using the procedure of grouping based on
similarity of the characteristic features of workpieces, according to the criteria defined
by the system of classification. Thus, the basic criteria of the ability for creating processing and assembling a group of workpieces in the one material flow.
Operations group Qopi belongs to a certain material flow - t in the production
system. Number of workpieces k in the operating group and therefore its size, in
general, are changing along the material flow in accordance with the terms determinate
with a system of classification and with a function of the degree of technological
complexity of workpieces and the parameters of technological system, which allows set
up a group model of material flow in production systems (Figure 2).
Model according to Figure 2 shows that in accordance with the terms of a
system of classification and function of the degree of technological complexity of the
workpieces and the parameters of technological systems, occurring, on the different
operations in material flow - t, different operating groups of parts - k, or their quantities,
on the manner as given by Figure 3. As the output value it is obtained a relationship
load/capacity for each operational group (Figure 4).
Fig. 2 Group model of material flows in production systems
Fig. 3 Operating groups with different number of parts

421
Fig. 4 Ratio of load/capacity for each operational group

The general model of material flow in production systems (Figure 5) is
determined based on the criteria given with degree of series (batches), which
represents the ratio of the total load-fund of the time required to perform design work
operations in a given section of the system and the effective capacity of the relevant
structures of the system, according to the formula (1):
(1)
Fig. 5 The general model of material flow in production systems

3
MODEL OF SIMULATION FOR SELECTION OF VARIANT MATERIAL TYPE

FLOW IN CONDITIONS OF GROUP APPROACH
Tecnomatix Plant Simulation, as software system, is designed for modeling,

simulation and optimization of production process planning. Simulation and modeling
of process planning by using this software system is based on time-oriented simulation
and of event-oriented simulation.
422
Time-oriented simulation takes into account the wide range of different types of
production time, and of event-oriented simulation takes into account only those points
in time at which events have an impact in the simulation model.
This paper, based on previously-developed type process planning and
generating of the simulation results, as also using appropriate tools as diagrams,
shown a model of selecting material flow type variant in terms of group approach in the
design of material flow.
Fig. 6 Variant 1.1
Fig. 7 Variant 1.2
Fig. 8 Variant 2.1
Fig. 9 Variant 2.2
Variant of material flow type for a constant effective capacity of production

system is determined on the basis of known conditions and criteria that are shown at
Figure 5. With application of software systems Tecnomatix Plant Simulation in
simulation process and displaying the results from the model of type process planning
in form of a diagram, may be defined for each operational group affiliation of variant
material flow type according to Figure 5, as follows:
Variant 1.1 of material flow type - basic group flow, which features a load
section which is for each operation less than effective capacity (Fig. 6);
Variant 1.2 of material flow type - a group flow of process type, where in a
number of operations load section is less than the effective capacity (Fig. 7);
Variant 2.1 of material flow type - a group flow of multy-workpieces type,
where in a number of operations load section is greater than or equal the
effective capacity (Fig. 8);
Variant 2.2 of material flow type - single-workpieces flow, which features a
load section which is in all operations greater than or equal to the effective
capacity (Fig. 9);
423
Model of simulation of type process planning presented in this paper and

automated generation of relationships LOAD / CAPACITY, contributes to the
automation of procedures for selection of material flow type variant in production
systems.
4.
SIMULATION RESULTS OF SELECTION MATERIAL FLOW TYPE VARIANT

IN CONDITIONS OF GROUP APPROACH
The subject of simulation in this paper is a type process planning, which is

designed for three operating groups of workpieces with similar geometric and
technological characteristics (Fig. 10). With application of software systems
Tecnomatix Plant Simulation, it was created a model of type process planning (Fig. 11)
for a given operating groups.
OPERATIONAL GROUP A
OPERATIONAL GROUP B
OPERATIONAL GROUP C
Fig. 10 Operational groups of workpieces
Simulation results of model for type process planning are presented in forms of
graphs that show the relationship between the load / capacity for the different
manufacturing operations. Load section for each operation is specified as multiplying a
424
mean value of time duration of operations for each operational group and the optimum
quantity of workpieces. Effective capacity is constant size and its value is Ke = 125
days/year (1 day=24 hours).
Fig. 11 The model of type process planning in 2D environment of TPS

For each operational group it is shown a diagram of material flow type variant
selection (Fig. 12). From the diagram it can be seen, that for operational group A and C
the optimal variant of material flow type is variant 1.2. a group flow of process type,
and for operational group B the optimal variant of material flow type is variant 2.1 - a
group flow of multy-workpieces type.
a) Diagram of variant 1.2 flow type

for the Group A
b) Diagram of variant 2.1 flow type

for the Group B
425
c) Diagram of variant 1.2 flow type for the Group C

Fig. 12 Diagrams for selecting variant of material flow type for observed
operational groups of workpieces
Presented methodology for material flow type variant selection, with simulation
of type process planning, can be applied to any operational group of workpieces, and
thus determine the material flow type variant according to the general model of
material flow in production systems.
5.
CONCLUSION
Application of software systems Tecnomatix Plant Simulation in terms of group

approaches in the design of materials flow contributes to the automation of procedures
for material flow type selection in production systems. Criteria for selection of flow type
based on the relationships load/capacity is modeled and variantly simulated with
graphic output, which allows selection of the best structure for the executing of
production processes. The model includes four basic variants of material flow type in
production ant these are: a basic group flow, group flow of process type, group flow of
multy-workpieces type and single-workpiece flow.
LITERATURE
[1]
[2]
[3]
[4]
[5]
426
Borojevi, S., Jovievi,V., Jokanovi, S. (2009): Modeling, simulation and

optimization of process planning, Journal of Production Engineering, vol. 12, no.1,
p. 87-90.
Jovievi, V. (2005): Projekovanje tehnolokih procesa, Mainski fakultet
Banjaluka.
Zelenovi, D. (2003): Projekovanje proizvodnih sistema, FTN, Novi Sad
Borojevi S., Jovievi V. (2010): Proizvodne strukture bazirane na simulaciji
tipskih tehnolokih procesa, 36. JUPITER Konferencija, Mainski fakultet
Beograd,
UGS Corporation - Tecnomatix Plant Simulation 8.2., Plano, TX, USA
TECHNOLOGICAL BASIS FOR THE DEVELOPMENT AND

IMPLEMENTATION OF FLEXIBLE MANUFACTURING SYSTEMS
Velimir Todi1, Dejan Luki 2, Mijodrag Miloevi3, Jovan Vukman4
Summary: Flexible manufacturing systems (FMS) characterized by high price and
costs exploitation, therefore must take account of all stages of lifecycle, focusing on the
required and possible techno-economic effects of their application. Defining the
necessary technological and economic basis for the development and implementation
of FMS, requires very complex and reliable-economic analysis that will show the
reasons for selecting the most favorable variant FMS and appropirate elements.
In this paper are show the technological basis for the development and implementation
of FMS, for example, group of shaft manufacturing.
Key words: flexible manufacturing systems, technological basis, group technology
1.
INTRODUCTION
Considering to complexity of the problem economical exploitations of flexible

manufacturing systems, the solutions which are very expensive in terms of winning
andmanufacturing, on the one side, and which must provide reliable effects on the
market or in exploitations, on the other side, efforts related to versatile technical and
economic analysis in their design and implementation are justified. [1]
By applying the concept of group technology (GT) it's possible to create the
necessary basis for determining the elements of FMS and which will be applied in the
process of designing an appropriate range, in reference to family of parts and products.
These requirements, it's possible to created based on analysis of all the necessary
parts from the observed range, or based on a appropriate analysis of representative
items. On machining modules of FMS performs the corresponding manufacturing
processes of individual parts which make up a family of parts. It is certain that
constructive-technological solutions of individual modules influence the effect of
manufacturing the appropriate process, such as technology solutions of manufacturing
process for individual modules to influence those effects [2 ,3 ,4].
In this present case for formed a part families, are shows the methodology for
defining technological basis for the development and implementation of FMS, for
example turning process of shaft group.
1
Prof. dr Velimir Todi, Novi Sad, Faculty of Technical Science, DPM, (e-mail: todvel@uns.ac.rs)
Mr Dejan Luki, Novi Sad, Faculty of Technical Science, DPM, (e-mail: lukicd@uns.ac.rs)
3
Mr Mijodrag Miloevi, Novi Sad, Faculty of Technical Science, DPM, (e-mail: mido@uns.ac.rs)
4
MSc Jovan Vukman, Novi Sad, Faculty of Technical Science, DPM, (e-mail: vukman@uns.ac.rs)
2
427
Velimir Todi, Dejan Luki, Mijodrag Miloevi, Jovan Vukman
2.
METHODOLOGY FOR DEFINING TECHNOLOGICAL BASIS
The proposed approach to development and application of FMS is composed of

three basic phases: Defining the technological basis for the development and
implementation of FMS, Evaluation and selecting elements of FMS and Modeling and
simulation of FMS.
The methodology for defining the technological basis for the development of FMS
is composed of the following tasks:
Analysis of the production program,
Forming part families,
Design and selecting complex parts,
GT process planning,
Specify the process plan and the determination of time of individual process,
Determining the level of complexity and number of elements FMS,
Evaluation of application effects of FMS.
3.
CASE STUDY-DEFINING TECHNOLOGICAL BASIS
The main task set refers to the development and implementation of FMS to
produce formed group of shaft by type of serial production (Table 1).
Table 1. Basic information about the group of shaft
Order Part
Number
Shaft 1
Shaft 2
Shaft 3
Shaft 4
Shaft 5
Shaft 6
Shaft 7
Shaft 8
Identification
code of part
100.011
100.012
100.013
100.014
100.015
100.016
100.017
100.018
Volume of production
(pcs/yr)
2000
3000
4000
2000
4000
2000
6000
5000
Mass of part
(kg)
2,1
2
1,9
1,8
2,1
2,1
2,2
2
Price
(EUR)
50
65
65
75
75
80
80
80
Based on geometric and technological characteristics from the group of parts,

is projected the imaginary complex part, with the appropirate features (Fig.1).
Fig.1 Complex part of technological group of shafts

428
Technological basis for the development and implementation of flexible manufacturing systems
On the basis of the projected complex part, volume of production and other
necessary information, is projected GT process plan. In this process plan is a key
turning process (op30), that is implemented on a CNC lathe.
3.1 Determination of cycles time for group machining process
Group manufacturing process plan and the corresponding processes allow
precise definition of the process plan for any part, including the representatives parts of
the appropriate part families. [3 ,4 ,5]
In order to make rational and rapid determination cycle time of machining
process is necessary to precise machining process for corresponding representatives
part, in depending on the method used to determine this time. Contents of the
developed methodology will be achieve on example of group turning process of shaft
group.
3.1.1
Graph-analytical method
To determine the necessary cycle time of machining process during the

observed parts group by graph-analytical methods, must be extracted using
technologically simplest and most complex part of that group and based on the group
process designed to achieve precise process. As a criterion for the selection of the
simplest and most complex part of the group commonly used procedure in the
observed number of machining operations.
Total cycle times of machining process during one process groups of products
or parts, can be determined from the expression:
Tc
(Q t
i 1
i ci
(1)
where are:
Qi- volume of production individual parts of group
tci- cycle time of machining process for certain parts of group
Cycle time of machining process (tci), which, as it's known, include the basic
(main) and auxilary time, are determined from a diagram similar. In the observed case
diagram cycles times a group of shafts on the turning process on a CNC lathe, has the
look as in Fig. 2. The diagram shows the dependence cycles times and number of
operations, the simplest part has 7 operations and cycles time of 4.8 min/piece, and
the most complex part has 14 operations and cycles time of 8.7 min/piece.
Cycles times of machining process within individual parts of group can be
defined graphically, using the diagram in Fig. 2 or analytically, with the prior
determination of the coefficient that describes the direction of change cycles times
depending on the number of operations (z).
Based on the pictures, we see that the coefficient of direction k:
t c 2 t c 1 8.7 4.8
(2)
k
0.557
14 7
z 2 z1
For example, the shaft 3 which in this process has 10 operations, cycles time of
machining process is calculated as follows:
(3)
t c3 t c1 k ( z3 z1 ) 4,8 0 ,557 ( 10 7 ) 6 ,471 | 6,5 (min./pc.)
429
Fig. 2 Dependence cycles times of turning process on the number of operations

In table 2, there are data on cycle time of turning process some parts of the
group, obtained using methods observed.
Table 2. Data for cycles times of turning process
Designation of
part (shaft)
Cycles time
(min/pc)
100.011
100.012
100.013
100.014
100.015
100.016
100.017
100.018
4,8
5,9
6,5
8,1
7,6
8,1
8,7
According to the following information is provided total cycles times of turning

process observed technological group by (1), which is:
Tc
2000 4,8 3000 5,9 4000 6,5 4000 7 2000 7,6 (2000 6000) 8,1 5000 8,7
Tc
204800(min./god.)
3.1.2. Method based on a representative parts of technological group

For the observed group should choose a part representative of applying ABC
analysis, which specifies the machining process the related GT process plans, and
then determine the cycle time of machining process representative parts (tkp).
On the basis of volume, mass, value of the ABC analysis, according to Table 3,
the part representative regarding shaft groups, the selected is shaft 7. Figures 3a and
3b shows the quantitative and mass ABC analysis, where can see the dominance of
the 7th shaft.
Table 3 Data for ABC analysis
Mass of part
m (kg)
Value
V (EUR)
Name of
part
Volume of
production
Q (pcs/yr)
kg/pcs
kg/yr
EUR/pc
EUR/yr
Shaft 1
Shaft 2
Shaft 3
Shaft 4
Shaft 5
Shaft 6
Shaft 7
Shaft 8
2000
3000
4000
2000
4000
2000
6000
5000
2,1
2
1,9
1,8
2,1
2,1
2,2
2
4200
6000
7600
3600
8400
4200
13200
10000
50
65
65
75
75
80
80
80
100000
195000
260000
150000
300000
160000
480000
400000
430
Q
(%)
m
(%)
V
(%)
7,14
10,71
14,29
7,14
14,29
7,14
21,43
17,86
7,34
10,49
13,29
6,29
14,69
7,34
23,08
17,48
4,89
9,54
12,71
7,33
14,67
7,82
23,47
19,56
Technological basis for the development and implementation of flexible manufacturing systems
100%
100%
90%
80%
80%
70%
60%
60%
50%
40%
40%
30%
20%
20%
10%
0%
0%
Os7
Os8
Os5
Os3
Os2
Os6
Os4
Os1
Os7
Fig. 3a Quantity ABC analysis
Os8
Os5
Os3
Os2
Os6
Os4
Os1
Fig. 3b Value ABC analysis
To determine the cycle time of machining process observed group, is

determined by the reduced quantity of parts of the group (Qr). Reduced quantity of
parts are determined by the expression [6]:
(4)
Qri Qi ri
where are:
Qi number of different parts of the group
ri reduction degree of the ith part
The reduction degree of a certain part of the observed technological group (ri)
includes the reduction of mass (rm) and complexity (rs). Degree of technological
complexity of certain parts of the group is determined by the ratio of the machining
operations of individual parts and the number of machining operations appropriate
representative parts. Reduced quantity of all parts of the groups determined by the
expression (5), and presented in Table 4.
k
Qr
Q r Q
i i
i 1
(5)
ri
i 1
Total cycle time of turning process of observed technology group of shafts

during the processing on a CNC lathe is:
(6)
Tc Qr tkp 24140 8,4 202776 (min./god.)
Table 4. Reduced quantity for group of shaft
Product
Qi
(pcs/year)
Shaft 1
Shaft 2
Shaft 3
Shaft 4
Shaft 5
Shaft 6
Shaft 7
Shaft 8
2000
3000
4000
2000
4000
2000
6000
5000
mi
(kg/pcs)
2,1
2
1,9
1,8
2,1
2,1
2,2
2
rm
rs
ri
Qri
0,955
0,909
0,864
0,818
0,955
0,955
1,000
0,909
0,700
0,833
0,867
0,967
0,933
0,967
1,000
1,000
0,669
0,758
0,749
0,791
0,891
0,923
1,000
0,909
1340
2280
3000
1580
3560
1840
6000
4540
ANNUAL QUANTITIES Qr=
24140
3.2 Calculation of FMS application effects

For the case when is doing to products that belong to one group, the total time
431
of occupation machining module is determined by the following expression:

T TC ns Tpz (min./yr.)
(7)
If we assume that the observed production system to work me = 250 days/year.

in two shifts of the day se = 2 shifts/day, with ne = 7.5 hour/shift and the level of
utilization e = 0.8, its effective capacity according to [6] will be Ke = 180,000 min./yr.
Based on previously calculated data for cycles times for the two considered
methods, and taking that ns = 10 ser/yr., can be determined time occupation, the
number and level of utilization machining modules to be observed the turning process.
Based on the budget of machining systems N adopts the Nu like as an integer.
T 204800 10 40 205200 (min./yr.)
Ti 205200
1,14
N
1,14 , Nu=1 machine (CNC lathe), with K
114 %
1
Q 180000
4.
CONCLUSION
GT process plan for the technology groups allow precise manufacturing

processes of their representatives, thus creating high-quality technological base for the
development and implementation of FMS. By defining the content of group process
plan and the corresponding manufacturing processes and operation on the basis of
real or imaginary complex parts, creating a technological basis for the selection of
types and levels of certain elements of the FMS for individual processes.
Calculation of the necessary capacity, or number of FMS for certain processes
into group process plan of manufacturing certain technological groups, based on a
specified cycle time of machining process of representatives, reduced quantities of a
group or a diagram of similarity, and this provides a solution, complex tasks in a
rational, sufficiently reliable way. The developed methodology was verified on the
example shown technological group of shafts.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
432
Shivanand, H. K., Benal, M. M., Koti, V. (2006). Flexible Manufacturing System.

New age International (P) Limited, New Delphi.
Andres, C., Albarracin, J. M., Tormo, G., Vicens, E., Garcia-Sabater, J.P. (2005).
Group technology in a hybrid flowshop environment, A case study. European
Journal of Operational Research. vol. 167, no. 1, pp. 272281.
Ham, I., Hitomi, K., and Yoshida, T. (1985). Group Technology. Applications to
Production Management, Kluwer-Nijhoff , Boston.
Mitrofanov, S. P. (1996). The Scientific Principles of Group Technology. National
Landing Library Translation, Yorks, UK, Boston Spa.
Borisov, V.B., Borisov, E.I., Vasilyev, V.N., et al. (1985). Spravochnik tekhnologamashinostroitelya (Handbook of Machine Building Production Engineer).
Mashinostroenie, Moscow.
Zelenovi, D. (2003). Projektovanje proizvodnih sistema, Fakultet tehnikih
nauka, Novi Sad.
THE PROCESS OF MEASURING EQUIPMENT MANAGEMENT AND

ITS AUTOMATIZATION
Goran Janji1, Predrag Nagraisalovi2, Zorana Tanasi3,
Miroslav Bobrek4, ivko Pejainovi5
Summary: Organizations that want to have valid data on the characteristics of
processes, products and systems, need an efficient and effective management system
of measuring equipment throughout all phases of its life cycle. Otherwise, the results of
measuring or testing in the system become unreliable, which adversely affects the
quality of the product and services, customers satisfaction and decision making based
on irrelevant data and information. The measuring equipment management is also the
requirement of certain standardized management systems (ISO 9001, ISO 14001, ISO
10012, etc.), which obliges organizations that want to certify their management system
in relation to given standards, to provide evidence that they manage it in an efficient
and effective manner. In the process of measuring equipment management, a great
deal of information emerges about its characteristics and status, its exploitation,
verification and servicing which makes management process very demanding and
complex. An application created for the purpose of automatization and simplifying the
process of measuring equipment management is presented in this paper.
Key words: measuring equipment, measuring equipment management, automatization
0 INTRODUCTION
One of more important elements of business system management is
monitoring and measurement of important characteristics of products, processes and
systems. Values of important characteristics in the process of analysis are translated
into information that helps management in the process of decision making.
Measurement of important characteristics in business system is mainly performed with

1
Goran Janji, M.Sc., Banja Luka, University of Banja Luka, Faculty of Mechanical Engineering, Department
for Industrial Engineering and Management; e-mail: gjanjic@urc.rs.ba
2
Predrag Nagraisalovi, M.Sc, Banja Luka, Ministry of interior of the Republic of Srpska, npredrag@gmail.com
3
Zorana Tanasi, Ph.D., Banja Luka, University of Banja Luka, Faculty of Mechanical Engineering,
Department for Industrial Engineering and Management; e-mail: tami@blic.net
4
Miroslav Bobrek, Ph.D. Professor, Banja Luka, Ministry of Education and Culture of the Republic of Srpska;
e-mail: m.bobrek@mp.vladars.net
5
ivko Pejainovi, Ph.D. Assistant Professor, Banja Luka, University of Banja Luka, Faculty of Mechanical
Engineering, Department for Industrial Engineering and Management; e-mail: zivkop@teol.net
433
Goran Janji, Predrag Nagraisalovi, Zorana Tanasi, Miroslav Bobrek, ivko Pejainovi
help of measuring equipment. Quality of information presented to the managers, and

consequently the quality of made decisions, depends on the accuracy of the measuring
equipment. Also, unreliable information from the measurement process can have
negative influence on the quality of the very product and service which causes
decrease of customers satisfaction. In order to have the results of measurement valid,
within business system, there must be effective and efficient system of measuring
equipment management through all phases of its life cycle. Necessity of measuring
equipment management is recognized even within development of standardized
management systems and it is included in the requirements of standards ISO 9001,
ISO14001, ISO10012 etc. /1/. Also, the reasons for metrological confirmation, as one
of activities of measuring equipment management, are conditioned by ageing and
changes that occur on measuring equipment due to outer or inner influences
(atmosphere, mechanical impacts, frequent use, etc.). In the process of measuring
equipment management a lot of information on characteristics and status of measuring
equipment, its exploitation, verification and servicing occurs, which makes
management process very demanding and complex /2/. Considering this, in case of
complex business systems that are placed in several locations and which have many
measuring tools within its business processes, integrated measuring equipment
management is almost impossible without its automatization. This paper presents one
Web oriented application solution created with a purpose of simplifying the process of
management of measuring equipment.
1. THE PROCESS OF MANAGEMENT OF MEASURING EQUIPMENT
The process model of measuring equipment management (Fig. 1) consists of
the following activities:
Purchase and receiving. The need for new measuring equipment occurs in
the process of selection of important measuring characteristics and in the
process of technology projecting. When choosing measuring equipment one
should look for standard equipment and have in mind the characteristics of
the product or process to be measured. Purchased equipment is examined,
compared with declared properties in the producers documentation and
tested for use in the prescribed processes and conditions, after which
receiving is performed.
Recording and labelling. Recording measuring equipment is performed by
opening of Carton of measuring equipment and inserting all prescribed data,
which enables its purposeful use, maintaining and metrological confirmation
in prescribed intervals. Labelling of measuring equipment is performed by
assigning classification, identification and inventory number (according to the
adopted labelling system) which are inserted in Carton of measuring
equipment, and inventory number is also permanently written in the appropriate place on the equipment (engraving, permanent sticker, fixing tile, etc.)
Exploitation and withdrawal. Exploitation of measuring equipment is
performed in a manner that enables preserving metrological and functional
characteristics during use and its timely exploitation and withdrawal in case
of losing the above mentioned characteristics.
434
The Process of Measuring Equipment Management and its Automatization
Metrological validation of measuring equipment. It is performed by

authorized metrological laboratory in prescribed time intervals. Plan of
equipment for metrological validation is performed based on the data in
Carton of measuring equipment. In the given carton one inserts all
performed data changes, and data on performed metrological validation. [3].
Fig. 1 Decompositional diagram of measuring equipment management

2. APPLICATION DISPLAY
Application is meant to work in network environment (in intranet network).
Internet technology and threelayered client/server architecture was used in
development of the application. Based on the performed calculations of data quantity,
Oracle XE was chosen as the best solution for data storage. Namely, this is a free
version of Oracle Database Server with space limitation up to 4 GB for data storage,
which satisfies the needs of most of organizational units. The application was created
in PHP program language in combination with HTML. Considering the fact that
application works in intranet network and allows no access from public network
(internet), concept of username and password was chosen as the best for user identity
check, whereas one way MD5 algorithm is used for password encrypt. Administrator
performs management of user accounts.
2.1 Measurement Equipment data records
Measurement Equipment Data and its status records during exploitation are
run through the form of Measuring Equipment Records shown on the Fig. 2.
Previously, through the Module of Records of Business Elements, measuring
equipment had to be given a name, classification and identification number and if
necessary important characteristics of the given group of measurement equipment,
whilst in this phase every measurement is given inventory number which determines it
435
uniquely, because within one identification number there can be greater number of
identical measurement instruments. This form offers possibility of search of inserted
data based on serial and inventory number, and name of measuring equipment.
Fig. 2 Form for insertion of main data on measuring equipment
In the same form, Record of Measuring Equipment (only in a different display),

for each item of measuring equipment it was made possible to insert its verification data,
servicing, debit, termination of debit (Fig. 3). During the insertion of measuring equipment
after purchase or entering data on its verification, it is obligatory to enter date of the
following verification based on which the application forms a list of measuring equipment
whose verification period is due in 15 days in the form of Sheet 15 (Fig. 4).
Fig. 3 Form for data entering on verification, debit and termination of debit of
measuring equipment
436
Fig. 4 Form ''Sheet 15''

In case when organization owns its own measuring equipment verification
laboratory (Fig. 5.), verification operation flow is as follows:
1. operator that runs records gets a warning on measuring equipment that has
to be verified on Sheet 15,
2. operator sends request through the application to internal measuring
equipment verification laboratory and forwards it physically in the internal
laboratory,
3. after completion of verification, laboratory man enters data on verification in
the application from the laboratory, and confirms that delivered request was
completed and at the same time measuring equipment is being physically
returned to operator,
4. after receiving measuring equipment, operator enters other important data
and closes the request for verification.
Fig. 5 Display of measuring equipment verification in internal laboratory

437
In case if measuring equipment verification is performed in external laboratory,

person running verification does not have access to application and all data regarding
verification are entered by operator (Fig. 6). Work flow of verification in external
laboratory is the following:
1. operator that runs records gets on Sheet 15 a warning on measuring
equipment that needs verification,
2. operator fills in a form with data on measuring equipment being sent to
verification through application and physically sends it to laboratory that
performs verification,
3. after measuring equipment was returned from verification, operator enters
other important data on verification and closes request for verification.
Fig. 6 Display of measuring equipment verification in external laboratory

2.2 Display of measuring equipment report
Application offers possibility to generate reports in the form of Portable
Document Format (PDF). All information on characteristics and status of measuring
equipment, its exploitation, servicing and metrological confirmation can be obtained by
creating Cartoon of measuring equipment that is in a form of report from the
application (Fig. 7).
438
Fig. 7 Cartoon of measuring equipment
Based on entered data on measuring equipment verification and date of next

verification, Measuring Equipment Verification Plan report can be generated (Fig. 8),
with review of measuring equipment that needs to be verified in given time.
Fig. 8 Measuring Equipment Verification Plan
439
3. CONCLUSION
Automatization of the process of measuring equipment verification can greatly
help business systems in the processes of gathering and analysis of values of
important characteristics of product, process and system, especially in complex
business systems in several far apart locations that use great number of measuring
instruments. The application presented in this paper supports the processes necessary
for efficient and effective measuring equipment management, and it was successfully
implemented in many service and manufacturing companies.
LITERATURE
[1]
[2]
[3]
440
SRPS ISO 9001:2008, Quality Management Systems Requests,

Standardization Institute, Beograd
Vulanovi, V. and others (2003). Quality System - ISO 9001:2000, Faculty of
Technical Science Institute for industrial Engineering and Management, Novi
Sad
Janji, G. (2004). Development of Information Systems for Integrated System
management with emphasis on QMS: Masters Thesis, Mechanical Engineering
Faculty, Banja Luka
VIRTUAL ENVIRONMENT PLATFORM FOR INDUSTRIAL

OPERATION AND MAINTENANCE
S. Mazzola1, P. Pedrazzoli2, G. Dal Maso3, C. R. Bor4
Summary: Virtual Environments are a key enabler towards radical evolution in
training, maintenance and repair operations of industrial plants and manufacturing
facilities. Indeed, the complete 3D simulation of products, processes and plants fosters
a new approach to technical documentation and assistance, where the user can be
easily guided event through the most complex and critical operations. This paper
present the work done in relation to the design and implementation of a SW platform
integrating the very latest in 3D modeling, 3D animation and simulation, and humanmachine interface technologies. The platform finds a vast number of application:
industrial system design, assembling, operation, training and maintenance. The paper
also reveals how virtualization of the maintenance environment allows off-site
collaborators to monitor and assist with repairs.
1.
3D ENVIRONMENTS FOR TRAINING AND MAINTENANCE
Most companies already define their products and plants digitally and manage
those information by product data management systems. Those digital representations
are mostly used within the early phases of design and prototyping, but their true
potential along the entire life cycle of products and plants is still unexploited. The work
hereinafter presented will demonstrate the effective utilization of 3D environments over
both the training and maintenance phases of products and plants.
The potential of 3D environments lies first in the training of skilled personnel:
virtual training solutions offer the potentiality of real world training without the cost,
delays, and inconvenience of physical mock-ups and classroom training [1]. Within a
virtual environment, the trainee can experience complex training operations and wide
variety of everyday and failure scenarios and develop a set of skills he could hardly
acquire in real scenarios. Moreover Virtual Training offer a number of tangible benefits
over traditional training methods:
Teach with hands-on participation. 3D training replicates the real world
experience and increases trainee participation and retention.
1
Eng., Stefano Mazzola, Italy, Technology Transfer System s.r.l.,

Eng., Paolo Pedrazzoli, Italy, University of Applied Science of Southern Switzerland
Eng., Giovanni Dal Maso, Italy, Technology Transfer System s.r.l.
4
Eng., Claudio R. Bor, Switzerland, University of Applied Science of Southern Switzerland
2
3
441
S. Mazzola, P. Pedrazzoli, G. Dal Maso, C. R. Bor
Trainees can explore solutions to instructor-specified problems in a riskfree, realistic environment that encourages learning and experimentation.
Accelerated learning with less instructor time adds up to reduced overall
training costs.
Virtual Training demonstrates interrelations and mechanics that cannot be
viewed by the human eye, such as internal part interaction.
Training is the first step towards a larger use of 3D technologies within plants:
since virtual environments are easily accessible outside the classroom, and their
functionality can be triggered by specific events, this kind of tools become the natural
means to support maintenance in the field. The approach hereinafter presented
integrated 3d models, technical manuals, photographs, audios and videos to enhance
an extremely accurate 3D representations of products and plants for training, everyday and extra-ordinary maintenance.
The market has not been deaf to the potential expressed by new technologies.
The development of 3D training and maintenance environments has been undertaken,
with assorted outcomes, by, for instance, Intelliact AG [7], Dassault Systems [8],
Parallel Graphics [9], to quote some of the main actors.
2.
THE SYSTEM ARCHITECTURE AND ITS INNOVATION
Within the Framework we hereby present, the animation within the 3D

environment, where the user can interact, learn and draw instructions, is the result of
the combination of three main elements: 1) The virtual world itself; 2) The animation
instructions; 3) The visualization engine
The virtual world is the common space where the animation instructions and
the visualization speak together. It allows the description of a complex scene graph,
including machine kinematics, plant layout, multimedia contents. The animation
instructions operate on the objects of the virtual world. The visualization engine listens
to changes to the virtual world and visualizes them.
The animation is created combining a series of instructions (that encode the
knowledge related to the real world maintenance procedure) into a scene graph. This
innovative approach, that differs from most key-frame/timeline based software (Adobe
Flash, Dassault Systemes 3DVIA) allows both a quick modification of the movements
and the creation of complex sequences. Furthermore, given that the animation
instruction operates directly on the state of the objects, it is possible to reuse subgraphs just with copy & paste. For example a complex maintenance operation can just
be copied straightforward to a different piece (if the steps are the same or similar) and
then just linked to the new objects. On the other hand, key-frame based software
requires reproducing on the new objects all the steps that lead to the final frame.
The Visualization engine is based on OpenGL: it is a highly optimized engine
that can render millions of triangles at sustained cinema-like frame rates (30fps). The
technology that allows this result, and that differentiates the developed engine from
most popular 3D engines, is based on a static geometry analyzer that takes the inputs
triangles and generates compact meshes optimized for rendering on GPUs (Graphical
Processing Units). The meshes generated preserve an exact visual representation and
the ability to animate the model. These algorithms try to improve data layout [3], first
minimizing the number of meshes to enhance spatial locality (requested data is
442
Virtual Environment Platform for Industrial Operation and Maintenance
physically stored close to data that has already been used), and then grouping objects
based on their access patterns to improve temporal locality (requested data has been
already used recently). These algorithms are also optimized to be executed quickly
(with low CPU usage) and with minimal memory overhead. As a result, it is possible to
apply these optimizations automatically during the loading of the meshes without
requiring explicit manual user intervention.
The animation and associated assets are stored in a compressed archive
optimized for streaming. File are stored in a single compressed stream one behind the
other, this means that the file content can be read as soon as the data arrives, without
requiring the whole archive to be downloaded. The files are stored in an optimized
order, in such a way that the files needed are already present when requested [4].
The approach used to develop the software is based on a modular architecture
[6]: we have different specialized applications (World Builder, and various Players)
build on internally-developed and third-party libraries. The structure is based on three
layers: on the top we have the applications, in the middle we have the supporting and
third-party libraries (mainly used for user interface components and services) and on
the very bottom we have the fundamental libraries. These fundamental libraries
implements the functionalities described in the previous section: the Core library deals
with the virtual world objects and events, the Animation library provides the instructions
and sequence flow management, and the Viewer provides the OpenGL visualization
engine. On top of these fundamental libraries there is the World Builder module, which
utilizes the Core and Viewer libraries and provides the user interface and tools required
to create and edit virtual worlds, and the Player Runtime, which utilize the Animation
library and provides components and basic services to play the animations. At the
libraries level we also have the NetBeans Platform, a third-party module providing
supporting functionality to develop full-featured desktop applications, supporting
features such as a customizable menu and toolbars, a docking framework, multilanguage support, auto update services, etc. The choices made at architectural level,
and the innovation introduced in the 3D representation are key enabler for some of the
distinguishing features we can thus achieve:
Quick animation creation and reuse, thanks to the instruction scene graph.
Visualization of huge worlds (millions of triangles): thanks to the
unparalleled optimization algorithms and together with traditional 3D engine
techniques (VBO, clipping) it is possible to visualize complex worlds.
Visualization through the Internet [5]: using Java Applet (Windows) and
Java Web Start (Mac, Linux) is it possible to view the animation from the
web browser, without any software preinstalled. The custom file format
allows small file size enabling minimal download times.
3. DEMO CASE 1: COMPLEX PLANT FOR HEAVY CARPENTRY PRODUCTION
Heavy Carpentry (i.e. the iron beams and plates that constitute the back-bone
of large buildings, bridges, and many structures) is produced in large plants, whose
set-up and maintenance is a complicated task. In this section we will describe a
demonstration scenario referring to a complex plant for heavy carpentry production.
Our framework provides a 3D environment with the virtual model of the whole factory
layout in an high detail level, enriched with a 2d layout layer. The framework
443
performance empowers the quickly loading of very heavy models and to navigate in a
scene with many millions of triangles with high speed and fluidity in order to have a
complete overview of the factory plant, as shown in Figure 1.
Fig. 1
As the whole virtual representation is constantly linked to the real world, Its
possible to read the status of each machine directly in the environment, and zoom to a
desired one in order to see the details and investigate the state. For example, a
machine can report the need to substitute an internal component, and this is shown in
pop up window message directly in the scene.
If the observed machine needs a specific maintenance procedure or setup
operation, its possible to recover the relative 3D simulation to guide the operator in his
actions (Fig.2). In this case the operator chose to visualize the procedure to replace
the blocking wedge (the elements that needs repair in our example) and he can
follow the simulation of the single steps to do with the help of a textual description.
Its important to point out that our maintenance framework provides all the tools
required to realize the virtual environment and to use it. In the case just shown, the first
operation would be to realize the 3d model for each machine, thanks to the integrated
Model Builder and then it would be necessary to create the animation of the operation
to be simulate, thanks to the Instruction Builder. All this component, models and
animations, must be stored in the framework repository to be recalled. In this Heavy
Carpentry scenario its possible to identify some operative cases that involve many
different users and framework functionalities:
Case 1: On board operator needs to do standard procedures like installation,
setup or normal maintenance and searches for the proper simulation from the
animation stored locally regarding the machine on which is working by a wireless
connection on his portable device. The NC machine could recall automatically the
procedure and sends it to the operator.
Case 2: A predictive maintenance system or a supervisor decides to execute a
standard operation on a machine. The technical office personnel recalls the suitable
444
Virtual Environment Platform for Industrial Operation and Maintenance
procedure and sends it to the on board operator with the relative machine id.
Case 3: The operator or the technical office personnel need information
regarding a non ordinary, non standard operation not present in the animations
repository. The customer service, in force by the provider of the involved machine,
realizes immediately the new 3d instructions and supplies the user with them on web in
order to guide the on board operator in the necessary procedure.
Fig. 2
4.
DEMO CASE: WOOD AND WOOD-COMPOSITES MATERIALS FINISHING

MACHINE
Within the second demo case we illustrate the maintenance framework

application for the wood panels finishing industry. The wood finishing machine (Fig 3)
is a spraying system specifically designed to coat shaped panels and apply glue.
Fig. 3
445
The main maintenance procedure, corresponding to the belt conveyor

replacement, needs to execute complex and resource consuming operations, and to
use an external device such as a forklift. The difficulty to explain the instructions to the
operator leads to the need to support him with a 3D description enabled by the
Framework application: indeed only a 3D visualization is capable to demonstrate how
the new belt conveyor, a single looped piece, can be handled to be properly
positioned.
This application was very succesful and induced the machine manufacturer to
develop, for this machine model, a complete 3D maintenance manual directly linked to
the digital documentation and distribute to the machine customers.
5.
CONCLUSIONS
Virtual Environments are revolutionizing training, maintenance and repair

operations of industrial plants and manufacturing facilities. This paper is meant to
present the work done in relation to the design and implementation of a SW
framework, integrating the very latest in 3D modeling, 3D animation and simulation,
capable to fully support training and maintenance operations. The platform finds a vast
number of application: industrial system assembling, operation, training and
maintenance. Two applications have been presented and demonstrated.
LITERATURE
Journal Papers:
[1] Gerbaud, S., Mollet, N., Ganier, F., Arnaldi, B., Tisseau, J., (2008). GVT: a
platform to create virtual environments for procedural training. Virtual Reality
Conference VR '08 IEEE, p.225-232.
[2] Yoon, S., Lindstrom, P., Pascucci, V., Manocha, D. (2005). Cache-oblivious mesh
layouts. ACM SIGGRAPH 2005 Papers (SIGGRAPH '05), Markus Gross (Ed.),
p.886-893
[3] Shao-Shin Hung, Damon Shing-Min Liu. (2007). Efficient reduction of access
latency through object correlations in virtual environments. EURASIP Journal on
Applied Signal Processing, Volume 2007, Issue 1, p.178-178.
Books:
[4] Carter, C., Rhalibi, A., Merabti, M., Price, M. (2009). Networking Middleware and
Online-Deployment Mechanisms for Java-Based Games. Transactions on
Edutainment II, Zhigeng Pan, Adrian David Cheok, Wolfgang, Abdennour El
Rhalibi (Eds.), Lecture Notes In Computer Science, Vol. 5660, Springer-Verlag,
Berlin, Heidelberg.
Standards:
[5] OSGi Release 4.2 (R4.2)-2009. OSGi service platform, core specification. OSGI
Alliance.
WWW pages:
[6] Intelliact AG, Instructor, from www.instructor3d.com, accessed on 2011-03
[7] Dassault Systmes, 3DVIA Composer, from www.3ds.com, accessed on 2011-03
[8] Parallel Graphics, Cortona3d, from www.cortona3d.com, accessed on 2011-03
446
C. TERMOTECHNIQUE AND ENERGETICS

BASIC ASPECTS OF DEFINING MECHANICAL-TECHNOLOGICAL

SOLUTIONS FOR THE PRODUCTION OF BIOGAS FROM LIQUID
MANURE
Nataa Soldat1, Mirjana Radii2
Summary: Defining mechanical-technological solutions for the production of biogas
requires defining of single devices and their functional integration into a whole. This
paper defines mechanical and technological solutions for the production of biogas from
liquid manure, the choice of material, the method, the possibility of adaptation of
existing devices and the removal of hydrogen sulphide. The research and the results
thereof are based on years of research done on existing facilities for the production of
biogas.
Key words: liquid manure, biogas, facilities
1. INTRODUCTION
The production of biogas from liquid manure is carried out using anaerobic
fermentation in a facility called reactor, where this process is carried out in multiple
phases. For each phase a functional and dedicated group of devices is used.
This paper shows the research that was carried out with regard to finding
mechanical-technological solutions for the production of biogas fro liquid manure, the
selection of a facility for anaerobic fermentation, the method for their construction, and
also the possibility of adjusting existing devices and purifying biogas.
The objective of this paper is to identify the most suitable reactors for the
production of biogas fromliquid manure by studying the mentioned mechanicaltechnological solutions, along with the possibility of adjusting devices and purifying
biogas.
2.
MECHANICAL-TECHNOLOGICAL SOLUTION FOR THE PRODUCTION OF

BIOGAS
Defining a mechanical and technological solution for the production of biogas

from liquid manure is based on defining the single devices and putting them together
1
Dipl. ma. in.-master Nataa Soldat, Belgrade, Serbia, Faculty of Mechanical Engineering,
University of Belgrade (nsoldat2000@yahoo.com)
2
Prof. dr Mirjana Radii, Zrenjanin, Serbia, Technical College of Appplied Sciences in
Zrenjanin, (mirjana_radisic@yahoo.com)
449
Nataa Soldat, Mirjana Radii
into one functional unit.

The production of biogas comprises a number of phases [1]:
preparation of the substrate,
fermentation,
capturing the gas,
using biogas.
To every single phase a functional and dedicated group of devices is assigned.
A group of devices for maceration and separation of coarse particles is used for the
process of collecting and preparation of the substrate. Actually, polypropylene meshes
and centrifugal fecal sledge pumps, usually with two buckets, are used for this'.
The main phase of the biogas production process is the anaerobic
fermentation which is a multistage biochemical process used on a number of different
types of organic substances. The technological process of anaerobic fermentation
depends on a number of conditions, which, when met, produce a result showing a high
degradation level for organic matter together with an acceptable quality and amount of
biogas.
Fig. 1. Vertical reactor made of concrete

3.
CHOOSING A ANAEROBIC FERMENTATION FACILITY
The facility where the anaerobic fermentation takes place is called reactor. Its
solution is a basic condition for a good performance of the whole installation. Globally,
as well as in our country, there is a number of reactors that are different in structure
and the material they are made of [2, 3].
Reactors most often are discontinuous and usually there are at least two or
three in series. Continuous reactors are more rare. Both reactor types are supplied
through a hydraulic sealing system, which, for now, is a practical and simple solution.
As already mentioned, during the process of anaerobic fermentation the matter
is biologically degraded. Besides the absence of oxygen, this process also needs
constant temperature. The degradation of matter is most efficient on a temperature of
15C (psychrophile), 35C (mesophile) i 55C (termophile process). The mesophile
process is used most frequently, and in the summer season also the termophile
process [2].
450
Basic Aspects of Defining a Mechanical and Technological Solution for The Production of Biogas
from Liquid Manure
Reactors are classified by shape, size, type, by the material they are made of,
the mixing system or substrate heating used.
Depending on the material they are made of reactors are from: steel, concrete
or plastics. Very rarely they are made of stainless steel.
Figure 1. shows a vertical reactor made of concrete, one that is most widely
used [3].
Due to the corrosion of steel and the porosity of concrete, it is necessary to
plastify the first two types.
Due to its chemical stability, easy use and low plastification cost and
unsaturated polyester is used as a basic reinforcing material. In order to achieve
mechanical reinforcement, the polyester layer is reinforced with glass fiber, and to
enhance the shock strength, a bitumen component is added to polyester in form of a
50% bitumen stirene or toluene solution. An ortho- or isonaphtal polyester resin is used
for this. Such reinforced protection containing approx. 30% of alcohol-free glass fiber,
draws us towards the idea that most of the vital device parts, such as the ractor and
the bell, can entirely be made of polyester laminate [2].
Due to gas permeability and water absorption, and also the inhibition of
enzymate reactions, especially saponification under the influence of different types of
estherases, the hydrophility of the laminate is reduced by adding 3% of a five percent
solution pf paraffin in styrene. Gas permeability is reduced by foliar multilayer
lamination.
Mechanical and other properties of laminate are shown in Table 1.
Table 1. Mechanical and other properties of laminate
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
4.
Tensile strength
Bending strength
E-module from bending
Pressure strength
Shock strength
Max. Stretch till break
Content of glass
Density
Coefficient of thermic conductivity
Resistance to temperature change
Linear expansion coefficient
Absorption of water (24 h on 20C)
100-140 N/mm
120-160 N/mm2
6000-8000 N/mm2
240-300 N/mm2
70-90 N/mm2
2%
30 %
1450 kg/m3
0.20 W/mK
-40 - +120 0C
3x10-5 K-1
max 0,2 %
POSSIBILITY OF REACTOR ADJUSTMENT AND PURIFICATION OF BIOGAS
The adjustment of both steel and concrete reactors is easy. On concrete

facilities, non-permeability of gas is also achieved using a penetration layer and foliar
lamination.
Purification of biogas, i.e. removing hydrogen sulfate and water prevents the
installations from corroding, and the removal of CO2 increases the caloric capacity of
biogas.
451
Nataa Soldat, Mirjana Radii
The level and method for the purification of biogas depends on the method of
use, purpose and other factors. When biogas is compressed it has to be dried. Usually
it is dried through absorption, i.e. using agents that are binding water, as are calcium
hydroxide or calcium chloride
The deficit of this procedure is that also a part of CO2, CaO and CaCl is
absorbed, increasing the consumption of lime. When biogas passes through the
granular layer, the granules of CaO and CaCl2 adhere to each other, binding water. In
that way the absorbent cloddes and prevents the flow of biogas.
5.
CONCLUSION
Studying mechanical-technological solutions for the production of biogas from

liquid manure, we can conclude that due to their characteristics, concrete, steel and
plastic reactors are most frequently used.
Due to the corrosion of steel reactors and the porosity of concrete reactors,
they need to be plastified. Polyester is used as a basic reinforcment material, due to its
chemical stability, easy use and low plastification cost, in form of a polyester coating,
the reinforcement is achieved through glass fiber and the shock strength is enhanced
by adding a bitumen component in shape of a 50% bitumen-styrene or toluene
solution.
The adjustment of steel or concrete reactors is carried out in using a
penetration layer and foliar lamination for concrete facilities, achieving gas nonpermeability at the same time.
Purification of biogas, i.e. the removal of hydrogen sulfate and water prevents
the installations from corroding, and the removal of CO2 increases the caloric capacity
of biogas.
LITERATURE
[1] ulbi, M. (1986). Biogas (dobijanje, korenje i gradnja ureaja), Novinskoizdavaka radna organizacija Tehnika knjiga, Beograd.
[2] Radii, M. (2006). Proizvodnja i primena biogasa, Via tehnika kola, Zrenjanin.
[3] Soldat, N. (2010). Biogas-mogunosti proizvodnje i primene(diplomski rad),
Mainski fakultet Univerziteta u Beogradu, Beograd.
452
DETERMINATION OF PRESSURE DROP TWO-PHASE FLOW OIL

AND GAS FOR ISOTHERMAL FLOW IN HORIZONTAL PIPELINE
Crnojevi C. 1, Lei M. 2
Summary: In this paper, an onedimensional isothermal two-phase flow of oil; and gas
mixture in the straight horizontal pipeline was analyzed. The flows of the stated fluids
are considered as one-dimesional and the pressure distrubution in the pipe is
determinated for each of the fluid. In this case, condition of equal pressure drop of twophase mixture and the particular pressure drops of one-phase flows was used. As the
result of presented analize, the new formula for pressure drop determination of twophase flow in hydraulically smooth and rough pipes was obtained. The new formula
shows that the pressure drop two-phase mixture is a complex function of flow regime,
density and viscosity of oil and gas, the mass concentration of components and the
parameter X which appears in the method of Lockhart-Martinelli.
Key words: two-phase flow, horizontal pipeline, preassure drop, isothermal flow
1. INTRODUCTION
Two-phase flow accurs in a number of application such as oil and gas flow,
water vaporation or condensation in thermal plants, in the process of cavitation, etc. In
the oil industry, liquid-gas, liquid-liquid (oil-water), liquid- solid particles and gas-solid
particles are the posible types of two-phase flow. Beside the two-phase flow, the flow
can apear as three-phase flow of oil-gas-water and presents more complex case for
study. For determination pressure drop in the straight circular horizontal pipes through
which the two-phase mixture flow of liquid and gas is commonly used well-known
Lockhart-Martinelli method for incompressible flow involves mixture of both phases.
The two phase flow is decomposed in two one-phase flows, using the equality of
pressure drops. Lockhart-Martinelli method is the basic method of one-dimensional
two-phase flow calculation, and there are a number of its amendments, which are also
based on a model incompressible flows of both phases of the mixture. For the
calculation of incompressible two-phase flow in inclined pipelines as a standard
method is used Beggs-Brill [1].
However, in the transport of oil, natural gas or two-phase mixture of oil and
1
Prof. dr Cvetko Crnojevi, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade, Serbia, (e-mail:
ccrnojevic@mas.bg.ac.rs)
Prof. dr Milan Lei, Faculty of Mechanical Engineering, Kraljice Marije 16, Belgrade, Serbia, (e-mail:
mlecic@mas.bg.ac.rs)
453
Crnojevi C., Lei M.
natural gas, it is often necessary to consider the flow of the gas in the pipeline as
compressible, as in this paper is done.
2. PROBLEM STATEMENT AND GOVERNING EQUATIONS
m,
Consider isothermal two-phase flow of oil and gas mixture with mass flux m
in the horizontal straight pipeline (lenght L and diameter D), Fig. 1. This flux consists of
the oil mass flux and gas mass flux, and continuity equation and expression for the
gas mass concentration can be applied, sequenced:
m
m
L m
g ;
m
ck
g /m
L ,
m
(1),(2)
The arbitrarely taken cross section S-S (Fig. 1a), with cross sectional area A,
can be decomposed in two parts: the first one with liquid flow AL and the second one
with gas flow Ag (Fig. 1b), where applicable: A A L Ag .
1
L, D
Ag Dg
A, D
Ag
Ug
AL
Ul
m g
L
m
b)
a)
m m
AL DL
c)
Fig. 1 Decomposition of two-phase flow in two single-phase flow.

These surfaces can be represented as a circular, determined by the hydraulic
diameter Di , such as:
Ai
D i Di2 S / 4 ,
(i=L,g)
(3)
where D i is a corection coeficient of the area due to the use of hydraulic diameter.
Formed one-phase flows through the pipes with diameter Di (i=L,g) are considered as
flow in hydraulically smooth pipes and the friction factor can be determined by Blasius
expression:
ag
aL
, Og
,
(4),(5)
OL
nL
n
ReL
Re g g
Where Rei v i Di / Q i are Reynolds numbers estimeted by average flow
i / Ui A i in the cross sections, and coefficients ai and ni (i=L,g) define
velocities vi m
the flow regime, and their characteristic values are given in Table 1 [2,4,5].
454
Determination of Pressure drop Two-phase Flow Oil and Gas for Isothermal Flow in Horizontal
Pipeline
Table1 Parameter values ai and ni (i=L,g)

Fluid
Luquid + gas
Luquid + gas
Laminar flow
Blasius formula
Langs formula
Panhandle A formula
Panhandle B formula
a
64
0,3164
1,7
0,0847
0,01471
Natural gas
Natural gas
n
1
1/4
1/2
0,1461
0,03922
For further analysis it is useful to specify the parameters of the method

Lockhart-Martinelli
)i2
n 2
Di i
D 5 ni
)
, (i=L,g) ;
Di
)g
)L
(6),(7)
Observed two-phase flow of oil and gas is decomposed into two single-phase
isothermal flows, except that the flow of oil is incompressible and the flow of the gas is
compressible.
2.1 Isothermal oil flow
The basic assumption of Lockhart-Martinelli method is that the pressure drop is
the same for both phases, that is 'pm 'pL 'pg . Pressure drop in a straight-line
pipes for single-phase isothermal flow of oil determines the Darcy formula:
'pm
p1 p2
U L OL
L vL2
,
DL 2
(8)
that with expresions: (3), (4) and (6) can be transformed as following:
'pL
aL 4 2nL
L
( )
U LQ LnL 5n D LnL 2 VL2nL
2 S
DL L
aL 4 2nL K LnL L
( )
m L2 nL ) 2L
5 nL
2 S
UL D
aL 4 2nL K LnL
L
( )
D nL 2 m L2nL
5 nL L
2 S
U L DL
'poLD LnL 2 (
D 5nL
)
DL
'poL ) 2L
(9)
where 'poL is the pressure drop arrisen if throw the pipe with diameter D flows the oil
L .
only with the same mass m
2.2 Isothermal gas flow
One-dimensional compressible isothermal flow of gas in an imaginary pipe of
diameter Dg is described by the equations of motion:
dv g
dp
O
NM 2

NM 2dL
p
vg
2D
(10)
455
Crnojevi C., Lei M.
g
state p ZmUg RT and continuity m
Ug A g v g , where M is Mach number, N is
isotrope coefficient and Z m is a gas compressibility factor determined with a mean

pressure [4]. In classical fluid mechanics [2] the differential equation (10) has a solution
as follows:
p12 p22
ZmRT (
g
m
Ag
) 2 (O g
L
p
2 ln 1 ) .
Dg
p2
(11)
It is common [4,5], that in the long pipeline the member 2 ln p1 / p2 , the lower
order of magnitude, neglect in relation to a member O g L / Dg , so that (11) becomes:
p12 p22
ZmRT (
g
m
Ag
)2 Om
L
,
Dg
(12)
This expression can be transformed in a way that the right side of equation has
the form of the right side of Darcy formula:
p12 p22
2 p1
Ug O g
2
L vg
.
Dg 2
(13)
Expression (13) defines the pressure drop in the isothermal gas flow. Using
expressions (3), (5) and (6) expression (13) becomes:
p12 p22
2 p1
ag 4 2 ng Kg g
L
g2 ng ) g2 .
( )
m
5 ng
2 S
Ug ,1 Dg
(14)
The resulting solution (14), valid for single phase gas flow in the tube defined
by hydraulic diameter, can be written as:
p12 p22
2 p1
p12 p22 2
) ,
2p g
1 o
(15)
where
p12 p22
2p
1 o
ag 4 2 ng Kg g
L
g2 n g
( )
m
5 ng
Ug ,1 D
2 S
(16)
is the pressure drop if in the tube with diameter D is the isothermal gas flows with the
g .
same mass flow rate m
2.3 Two-phase isothermal flow
Individual pressure drops for one-phase flow of oil and gas are defined by (9)
and (14), respectively. Merging these single-phase flows, under the condition of the
pressure drop equality, and using expressions (2) and (7), result in pressure change of
one-dimensional flow of two-phase mixture:
456
Determination of Pressure drop Two-phase Flow Oil and Gas for Isothermal Flow in Horizontal
Pipeline
p2
p1
ag 4 n n U K g g n n
2 n
n n
2 ( ) L g L n D L g X 2 ck g m LL g 1 ,
aL S
U g ,1 K LL
n
(17)
or in the form of pressure drop:

'p
2 p1
p1 p2
2 p1
1
ag 4 nL n g UL Kg g nL ng 2 2 n g nL n g
L
( )
D
X ck
m
.
aL S
Ug ,1 KLnL
(18)
The resulting solution (17) or (18) shows that the pressure drop of two-phase
flow of oil and gas under high pressure is a complex function of: the flow regime,
physical properties of oil and gas - density and viscosity, the mass concentration of gas
ck and Lochart Martinelli X parameter. In the case where the oil has no natural gas,
ck o 0 , the solution (17) or (18) is not reduced to a known expression, because for
2 ng
X o f in article X 2 c k
is not defined limit.
Furthermore, we can analyze some special cases: namely, when the friction
coefficients of liquid and gas phase can be represented by the same expression, when
applicable: ag aL a and ng nL n , the following expression for the pressure drop
is obtained:
'p
2 p1
1
UL Kg
Ug ,1 KL
X 2 ck2 n ,
(19)
which in the case of both laminar flow - oil and gas ( a 64 , n 1 ) reduces to:
'p
2 p1
1
UL Kg 2
X ck .
Ug ,1 KL
(20)
Expression (18) which defines the pressure drop of two-phase flow in

hydraulically smooth pipes can be used for the case of the flow of both phases of a
mixture in a regime of hydraulic fully rough pipes. Namely, if O L aL , O g ag ,
nL
ng
0 , from (18) follows the expression:

'p
2 p1
p1 p2
2 p1
1
O g UL
X 2 ck2 ,
O L Ug ,1
(21)
which is valid for hydraulically completely rough pipes.

3. CONCLUSION
The proposed model of isothermal two-phase flow of oil and gas under high
pressure follows a well-known Lochart Martinelli method, but unlike it, in which the
liquid and gas phases are considered incompressible fluids, in this paper the flow of
crude oil is considered incompressible and the gas flow compressible. In this way, the
new formula, (18), was obtained, for determination of the pressure drop of two-phase
457
Crnojevi C., Lei M.
flow of oil and gas. This new formula shows that the pressure drop is a complex
function of: the physical properties of oil and gas - the density and dynamic viscosity,
flow regimes, which are covered by coefficients aL , ag , nL i ng , and mass
g ), the mass flow rate of oil
concentration of gas ck (or mass flow rate of gas m
L and the X Lochart-Martinelli parameter. New expression (18) is valid for
m
hydraulically smooth pipes, and its shape (21) for hydraulically rough pipes completely.
Acknowledgements The authors thank the Ministry of Science and Technology of

Serbia, which is the project MNT 35046 helped the realization of this research.
LITERATURE
[1]
[2]
[3]
[4]
[5]
458
Beggs H. D., Brill J. P. (1973). A study of Two-Phase Flow in Inclined Pipes. J.

Pet. Tech. p.607-617.
Djordjevi D. V., (2000): One-dimensional fluid dynamics (in Serbian), Faculty of
Lochart R., Martinelli R. (1949). Proposed Correlation of Data for Isothermal Twophase, Two-component Flow in Pipes. Chemical Engineering Progress, Vol. 45,
No 1 , p. 39-48.
Menon E.S., (2005): Gas pipeline hydraulics, CRC Press.
ai M., (1987): Pipeline transportation of fluids (in Serbian), Faculty of
DEVELOPMENT AND APPLICATION OF REGULATORY

DISPERSION MODEL FOR AIR POLLUTION ASSESSMENT
Igor Andreevski1, Gligor Kaneve2, Ljubica Kaneve3, Aleksandar Markoski4,
Sevde Stavreva5
Summary: Today, we are witnessing increased environmental pollution, including
ambient air, on local, regional and global scale. The existing air pollution from industrial
facilities or thermo plants, as well as potential pollution scenarios in normal or accident
conditions can be simulated in a relatively simple and efficient way using computer
dispersion models. Dispersion models used today are different in many aspects. In this
paper, the dispersion models categories suitable for application in conditions and in
solving problems of local air pollution in Macedonia are considered. Also regulatory
model MADAM development is presented.
Key words: air pollution, regulatory dispersion model, dispersion simulation.
1.
INTRODUCTION
Dispersion models are used to calculate pollutants ground level concentrations

based on the information and data about the emission source characteristics, the
pollutants nature, structure of the surrounding environment and meteorological
conditions. Dispersion models are widely used to simulate different pollution scenarios,
in a current moment, in the past and future. Besides the application for scientific and
research purposes, many dispersion models are commonly used for regulatory
requirements, air quality assessments and in order to establish critical pollution
scenarios.
2.
DISPERSION MODELS CATEGORIZATION
Today, there are number of dispersion models which are developed and used
worldwide. They all have certain characteristics such as: the hypothesis on which are
based, the minimum amount of data necessary for their application, the coverage
degree of atmospheric and dispersion processes, calculation accuracy, the application
area etc.
1
Asist.Prof.,PhD, Igor Andreevski, Macedonia, Faculty of Technical Sciences, igor.andreevski@uklo.edu.mk

Academici, Gligor Kaneve, Macedonia,Macedonian Academi of Science and Art, kanevce@sonet.com.mk
3
Prof., PhD, Ljubica Kaneve, Macedonia, Faculty of Technical Sciences, kanevce@sonet.com.mk
4
Prof.,PhD,Aleksandar Markoski,Macedonia,Fac. of Technical Sciences,aleksandar.markoski@ uklo.edu.mk
5
Ass. MSc., Sevde Stavreva, Macedonia, Faculty of Technical Sciences, sevde.stavreva@uklo.edu.mk
2
459
Igor Andreevski, Gligor Kaneve, Ljubica Kaneve, Aleksandar Markoski, Sevde Stavreva
Knowing the models characteristics, algorithms used and the results obtained
by applying them is important in terms of determining which algorithms can be
modified, or which parts of some model are better than those from other models and
may be adapted for use in other models with sufficient accuracy.
A number of authors and agencies offer different ways for dispersion models
categorization, but due to the wide range of features, every model can simultaneously
belong to different groups.
U.S. Environmental Protection Agency categorizes dispersion models on
several grounds. They can be divided into four groups: Gaussian, Numerical, Statistical
and Physical models. Also, according to USEPA, there are models with special
purposes such as models for ozone, carbon monoxide, nitrogen dioxide, particulate
matter and others.
Lately, models are more commonly categorized as models from old and new
generation ("new generation models") [1]. This categorization is made according to the
period in which they were formed, but also based on the period when theoretical
foundations on which they are based were established.
Dispersion models are used for different purposes, but likely they have major
application in regulatory processes. The models for legislation needs are used to
achieve two main goals, as tool for predicting air pollution condition and to provide
detailed interpolation concentrations maps between the measured data using
receptors monitoring stations.
The regulatory models results are used in making decision for allowable
emissions levels, in the design phase for new industrial buildings and roads to assess
their impact on ambient air quality.
Generally, regulatory models should provide information about the spatial
distribution of pollutants concentrations when high episodic concentrations occurred,
and data on the expected mean concentrations for a longer time period.
As regulatory dispersion models are usually used classical and modified
Gauss-ian models.
The models of Gaussian type are widely used, easy to apply and all their
positive and negative characteristics are well defined. They are based on the
assumption that the pollutants dispersion at various distances from the source is based
on the Gaussian distribution in horizontal and vertical direction and formulation is
derived assuming stationary conditions.
Older generation Gaussian models who have found largest application are
R91, ISCST3 and others, while as a modified Gaussian models or new generation
models are commonly used ADMS, AERMOD, OML and others. Several Gaussian
dispersion models are developed by the individual activities of different research
groups, among which are included: HPDM, RTARC, Safe-Air, IFDM, etc.
Particularly interesting for our analysis, in terms of applying some algorithms
from its original source code for present conditions in the Republic of Macedonia, is
Danish regulatory model OML.
OML model is a new generation Gaussian dispersion model that can be
applied in rural or urban conditions, within the local spatial scale up to 20 [km] from the
emission source. The latest official version of this model was published in 2001 [2],
while during 2007 revised research version was issued [3], which is still subject to
analysis and is not proposed for official use. The new generation models, as OML,
continuously descript atmosphere stability using quantities such as heat flux sensitive
460
Development and application of regulatory dispersion model for air pollution assessment
HSF, Monin-Obukhov length L, the mixing layer height hm, and in a quite complex and
detailed way calculate dispersion coefficients Vy and Vz . Also these models are using
contemporary features such as plume partial penetration.
3.
REGULATORY DISPERSION MODEL DAM
Model MADAM (MAcedonian Dispersion Atmospheric Model) belongs to

Gaussian models, and is developed to be adopted as a regulatory model in the
Republic of Macedonia [4], [5], [6].
It is widely used in determining the existing level of air pollution, in the case of
emission from thermal power plants and other industrial sources in our country. Also,
this model is used for simulation of various air pollution scenarios, in normal and
accident operating conditions and various external influences (meteorological, terrain
configuration, etc.).
Topol ~ani
20000
[m]
Kanat l arci
700
Budakovo
10000
Radobor
Mogi l a
5000
SO2
[Pg/m3]
Dobromi ri
15000
600
Dedebal ci
Kukure~ani
500
Agl arci
400
D. Ori zari
Bi t ol a
REKBI TOLA
Novaci
300
Logovar di
Ri barci
200
-5000
Bukovo
100
Gneot i no
Bi str i ca
-10000
Brod
Kr emeni ca
La` ec
-15000
-15000
-10000
Ger mi jan
-5000
5000
[m]
Fig. 1 First highest short-term (1 hour) SO2 [g/m3] ground level concentrations
461
We can see on figures 1 and 2 two examples of calculated ground level

concentrations in the case of emission from Thermal power plant Bitola, which satisfies
over 70% of the Macedonia electricity needs. The first picture shows the SO2
concentration during normal power plant operation, while second picture simulates
scenario of possible contamination with particular matters in accident conditions, when
ES filter of plant units Bitola 1 and 2 works with 50% of its capacity.
In recent years MADAM was subject to a numerous analysis and corrections in
its algorithms, in order to achieve better performance under the specific conditions of
its application in Macedonia.
Topol ~ani
20000
Kanat l arci
[m]
estice
(PM)
[Pg/m3]
Dobromi ri
15000
1500
Budakovo
10000
1000
Radobor
Mogi l a
5000
Dedebal ci
500
Kukur e~ani
250
Agl arci
150
D. Ori zari
Bi t ol a
REK BI TOLA
Novaci
100
Logovardi
Ri barci
-5000
70
Bukovo
50
30
Gneot i no
Bi st ri ca
-10000
Brod
Kr emeni ca
La` ec
-15000
-15000
-10000
Ger mi jan
-5000
[m]
5000
Fig. 2 Highest short-term (1 hour) particular matter (PM) [g/m3] ground level
concentrations when ES filter of units Bitola 1 and 2 works with 50% of its capacity
Thus the following corrections in its algorithms were made:
x
462
The model MADAM algorithm is adopted to handle cases when wind speed is
lower than 1 [m/s] [7], unlike other Gaussian models who cannot give a realistic
picture of pollution in low wind speeds conditions.
Development and application of regulatory dispersion model for air pollution assessment
Atmosphere conditions and classification from the model OML is applied, which
removes shortcomings from in-continuous description using stability classes [6].
Unlike the previous calculating method of the dispersion coefficients Vy and Vz,
based on the Pasquill-Gifford equations, the new MADAM variant solutions are
using expressions from the model OML [6].
The algorithm for plume partial penetration from model OML is applied in model
MADAM variant solution v.3, [6].
Inverse calculations methods are applied in several variant solutions of

regulatory model MADAM. Determining new values of certain influential
parameters in variant solutions v.1 and v.3 of model MADAM was analyzed.
Researches were conducted using experimental databases from MVKit (Model
Validation Kit) [8]. The latest available model MADAM version v.4 was
developed as a result of new defined values of some specific analyzed
parameters.
The indicators derived on the basis of the implemented evaluation procedures

(BOOT procedure, [9]), of the characteristics of the original model variant v.1 and the
final version v.4, together with data from comparison of these results with the
evaluations results of some world-known and widely used models, are given in Table
1.
Table 1 Model evaluation results using BOOT procedure database Kincaid SF6, Q3
Measurment
MADAM v.1
MADAM v.4
HPDM
IFDM
INPUFF
OML
AERPOL
Safe-Air
ISCST3
AERMOD
ADMS3
ADMS4
Sigma
54.34
35
Bias
NMSE
FAC2
FB
40.25
0.00
0.00
1.00
1.00
0.00
45.15
19.54
1.42
0.374
0.476
0.438
44
44.84
47.54
38.55
9.92
9.50
0.94
0.75
0.444
0.44
0.536
0.57
0.201
0.19
29.42
26.03
24.92
2.00
-0.13
0.42
0.60
34.61
26.76
19.72
1.29
0.14
0.50
0.44
47.45
45.48
6.89
1.24
0.15
0.55
0.14
42.05
31.90
11.64
1.09
0.13
0.57
0.24
34.89
31.50
19.45
1.52
0.04
0.39
0.44
30.00
60.00
24.30
2.80
0.26
0.28
0.58
21.80
21.80
32.60
2.10
0.40
0.29
0.86
51.70
34.70
2.70
0.57
0.45
0.67
0.05
48.50
31.50
5.90
0.60
0.45
0.68
0.11
These data provide a realistic basis to determine that due to the innovation of
the model algorithms, by introducing new solutions and implementing the results of
applied inverse procedures, we have made progress in the model MADAM
performances, so it can be included among the group of models with solid performance
according the criteria defined by [10].
463
Namely, from these verifications data can be seen that the model reaches
values for the data fraction within a factor of two compared to the measured
concentrations greater than 50%, FAC2> 0.5, and fractional deviation ~FB~<0.3, while
random, stochastic data scatter represented by the value of the normalized
intermediate-square error NMSE is less than 1.5, [10].
4.
CONCLUSION
Development and improvement of dispersion models performances are goals

that inspire all scientific and research groups in this area.
Generally, experiences regarding the introduction of new algorithms and
methods for the calculation of certain influential factors and the inverse procedure
application in dispersion models are positive.
Namely, the results from the application of these modifications in the regulatory
model MADAM showed that it is possible to use them as an option for improving the
model performances. This statement is evident from the results of the evaluation
procedure, but level of success depends on more subjective and objective
circumstances.
LITERATURE
[1] Schnelle K. B. and Dey P. R. (1999): Atmospheric Dispersion Modeling
Compliance Guide. McGraw-Hill, New York.
[2] Berkowicz R., Olesen H.R. and Lfstrm P. (2001): OML: An Operational
Atmospheric Dispersion Model. NERI, January.
[3] Olesen H. R., Berkowicz R. and Lfstrm P. (2007): OML: Review of model
formulation. National Environmental Research Institute, University of Aarhus. NERI Technical Report 609, 130 pp. (electronic).
[4] Kanevce L., Kanevce G., Markoski A. and Trombev G. (2000): Intercomparison of
Different Models for Estimation of the Bitola Power Plant Air Quality Impact.
International Conference of Balkan Environmental Association (BENA) on
Transboundary Pollution, Bucharest, November 23-26.
[5] Kanevce G. H. and Kanevce L. P. (2006): Dispersion modeling for regulatory
applications. Thermal Science, Vol. 10, No.2, pp.141 154.
[6] ndreevski I. (2008): Inverse concept application in dispersion models
development and verification, PhD thesis, Faculty of Technical Sciences - Bitola,
Bitola, Macedonia, July.
[7] ndreevski I., (2002): Planetary boundary layer modeling and pollutants dispersion
modeling at small wind velocities, Master thesis, Faculty of Technical Sciences Bitola, Bitola, Macedonia, December.
[8] Model Validation Kit. HARMO, MVKit, from http://www.harmo.de/kit, accessed on
2009-10-06.
[9] Chang J. C. and Hanna S. R. (2005): Technical Descriptions and User's Guide for
the BOOT Statistical Model Evaluation Software Package, Version 2.0. July 10th.
[10] Chang J. C. and Hanna S. R. (2004): Air quality model performance evaluation.
Meteorol. and Atmos. Phys., 87, 167-196.
464
DIMENSIONING OF THE COOLING SYSTEM OF A BUILDING FOR

PREDETERMINED RELIABILITY
Gordana Tica1, Veljko urikovi2, Petar Gvero3
Summary: A very important question is put before the cooling systems engineers:
How to dimension a plant for a climate area in the summer regim, if parametars are
known for that environment?
This paper shows an approach for analyzing this significant engineering problem.
Suggested method for defining the project parametar, which should be relevant for
dimensioning of cooling plant, as well as which would have significant advantage in
relation to all the methods used nowdays for calculation of cooling balance of a
building.
Cooling balance is given in this paper, as a coincidental value. The biggest mean daily
values of cooling balance in the specific environment are determined, as well as
empric and theoretic curves of reliability.
Key words: dimensioning for cooling system object, summer regime of work, reliability.
1. INTRODUCTION
Cooling balance of building is changed in the course of the year, even in the
extended period of time, which is particularly notable in the second half of the last
century. These changes are particularly found in areas with continental climate, such
as, for example, in Banja Luka. Therefore, the study of these changes is still very
important, especially for dimensioning of cooling systems, forecasting of energy
consumption for cooling and development of the energy system of the state. For these
reasons, it presents an approach to determining the likely change of the balance of the
cooling facility, according to data from several years of measuring the parameters of
the external environment.
The basic characteristic of changes of the mean daily values of cooling the
balance of buildings is the law of distribution of probability of their occurrences. The
law of probability distribution can be expressed by a reliability curve. When it comes to
reliability of the cooling balance, this means probability of performance in a number of
years, in a multi-year time interval, in which the value of the cooling balance equals to
Doc. Gordana Tica Ph.D, B&H, RS, Faculty of Mechanical Engineering, Banja Luka, tica@urc.rs.ba
Prof. Veljko urikovi Ph.D, B&H, RS, TermotehnikaInvest doo, Banja Luka, termotehnika-invest@blic.net
3
Prof. Petar Gvero Ph.D, B&H, RS, Faculty of Mechanical Engineering, Banja Luka, pero@urc.rs.ba
2
465
Gordana Tica, Veljko urikovi, Petar Gvero
or is greater than the default. For simplicity, the time interval is expressed in
percentages.
Method of determining the likely mean values of average daily cooling balance
of the building is given for the month of July of the summer operation regime, values q i
q' (with a calculated value of heat accumulation in the outer wall).
The month of July was chosen because it has the highest average
environmental sampling in the value of q = 1962,92.43 kJ/m2dan (q '= 2927.19
kJ/m2dan). The data are presented in Table 1. Parameters are sorted in a descending
order for the corresponding year.
Table 1 Mean daily cooling balance of the building q in the Banja Luka climate region,
for Julys summer workregime of the device for cooling the building, (kJ/m2dan).
Cooling
balance
Year
Frequency
Falling
values
q (kJ/m2dan)
No.
Interval
q
(kJ/m2dan)
Year
Reliability
Year
1958
2370.38
2370.38
14
2400-2350
4.545
4.545
1959
1769.23
2277.09
13
2300-2250
9.09
13.635
1960
1654.79
2252.04
12
2250-2200
4.545
18.18
1961
1788.83
2229.47
11
2200-2150
4.545
22.725
1962
1707.3
2173.4
10
2150-2100
4.545
27.27
1963
1798.54
2138.21
2100-2050
4.545
31.815
1964
1860.48
2069.21
2050-2000
4.545
36.36
1965
2252.04
2004.84
2000-1950
4.545
40.905
1966
1939.92
1969.2
1950-1900
9.09
11
49.995
1967
1897.12
1939.92
1900-1850
13.635
14
63.63
1968
2069.21
1900.07
1850-1800
4.545
15
68.175
1969
1735.75
1897.12
1800-1750
18.18
19
86.355
1970
1817.05
1860.48
1750-1700
9.09
21
95.445
1971
2229.47
1857.49
1700-1650
4.545
22
99.99
1972
1765.13
1817.05
22
99.99
1973
1969.20
1798.54
1974
2173.4
1788.83
1975
1900.07
1769.23
1976
1765.13
1977
1857.49
1735.75
1978
2277.09
1707.3
1979
2138.21
1654.79
1980
2004.84
22 god.
1962.92
42974.54
466
Dimension object's cooling system for predetermined known reliability

Table 1a. Mean daily cooling balance of the building q' in Banja Luka climate region,
for Julys summer workregime of the device for cooling the building, (kJ/m2dan).
Cooling
balance
Year
Falling
value
q' (kJ/m2dan)
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
3638.15
2654.28
2420.23
2769.56
2592.34
2839.07
2506.32
3436.08
2896.72
2818.05
3193.45
2453.36
2654.25
3521.15
2805.65
2914.04
3393.32
2629.64
3638.15
3521.15
3436.08
3393.32
3262.5
3221.9
3193.45
2914.04
2896.72
2866
2839.07
2818.05
2805.65
2769.56
2718.6
2654.28
2654.25
2629.64
2592.34
2506.32
2453.36
2420.23
22 god.
64204. 66
2718.62
3221.93
3262.45
2865.96
2927.19
No.
12
11
10
9
8
7
6
5
4
3
2
1
Frequency
Interval
q'
(kJ/m2dan)
Year %
3650-3550
3550-3450
3450-3350
3350-3250
3250-3150
2950-2850
2850-2750
2750-2650
2650-2550
2550-2450
2450-2350
1
1
2
1
2
3
4
3
2
2
1
22
4.545
4.545
9.09
4.545
9.09
13.635
18.18
13.635
9.09
9.09
4.545
Reliability
Year %
6
1
2
4
5
7
10
14
17
19
21
22
7
4.545
9.09
18.18
22.725
31.815
45.45
63.63
77.265
86.355
95.445
99.99
Eg. reliability of the value of the cooling balance of 1850 kJ/m2dan (2850
kJ/m dan) is equal to 15 (14) years, which means that in 15 (14) years interval, in a
series of 22 years, the value of balance of the cooling building q (q') in the observed
climate region can not be lower than this value.
The curve in Fig. 1 (index a) is called the curve of distributionand it is
constructed using the mean values of the interval. The curve in Fig. 1 (index b) is
called a false confidence.
2
467
Fig. 1 Course of distribution (index a) i reliabiliy (index b) for object's cooling balance q
i qc in Banja Luka's area, for the month of July of the summer work regime.
Within the limits of a given period of system observation, the using the curve of
distribution, the value of the cooling balance, witharbitrarily given reliability, can be
determined.
In solving practical problems, an interval of observation is usually short, and is
therefore not possible to construct a secure distribution curve. Due to relatively short
time interval of observation of 22 years, the influence of chance is very strong, which is
whythe reliability curve is unreliable. This is evident in Fig. 1 It is therefore necessary to
choose a suitable theoretical distribution curve and use it to determine the value of rare
balance cooling frequency, which goes beyond the observation interval.
2. APPLICATION OF DISCRETE BINOMIAL DISTRIBUTION
Discrete binomial distribution [3, 4, 5] is often used in meteorology and
hydrology. Because of the obvious influence of meteorological conditions on the
cooling system of balance, it is logical to expect that this distribution could be applied in
analyzing the balance of the cooling system.
Binomial confidence curve is determined by parameters in Table 2.
468

Table 2 Characterizes parametars by binomial confidence curve
Middle row of the samplethe mean cooling balance
in the observed time
interval n
Coefficient of variation
cv
Coefficient of asymmetry
cv ,q
0, 235
22 1
( k c 1)
cv ,qc
mq
1962,92 kJ/m 2 dan
mqc
2927,19 kJ/m 2 dan
For n 30 cv
n
n
k c
0,106
, k J /m 2 d a n
k c 1
cs
Modular ratio
qi
i 1
mq
(k c 1)
n 1
cs 2 cv
k 1
( n 1) c v3
q
m
i
q
0,5614
22 1
0,16
cs ,q
2 cv ,q
0, 212
cs ,qc
2 cv ,q '
0,32
Calculation of ordinates of theoretical reliability curve are done using the

Rybkins table, which is given in [1] for the known parameters of binomial distribution.
This table gives relative removals from the mid-ordinates of the integral curves for cv =
1.00 and various percentages of reliability p%. Ordinate values M are taken according
to the table Rybkin [1], the calculated values of the coefficient of skewness.
3.
DETERMINATION OF THEORETICAL CURVES OF RELIABILITY COOLING

BALANCE
Binomial distribution of parameters is needed to determine the parameters of

binomial distribution (ovaj prvi dio recenice mi nije jasan, doslovno kaze: Binominalna
raspodjela parametara je potrebna da se odrede parametri binominalne raspodjele),
determined by dividing the cooling balance classify in a descending order. Modular
ratio k 'is determined after that. The setting of a value (k ' -1) and (k ' -1) 2 For checking
of the calculation, it should be taken into account that the sum of the values must be
equal to the number of rows, and the sum (k ' -1) must be equal 0. Table 2, can be
calculated by a formula often used in meteorology and hydrology:
m 0,3
100, %
n 0, 4
where the parameters in the formula are: m - the number of observations (column 3),
and n - total number of years of observation.
469
Table 3 The parameters of the empirical curve reliability of mean daily value of object q
of the cooling balance in the Banja Luka area, for the month of July.
Year
1
1958
q,
2
kJ/m dan
No
Falling value
2
q, kJ/m dan
(k-1)
(k-1)
Reliability p,
%
2370.38
1958
2370.38
1.20758
0.20758
0.04309
3.12
1959
1769.23
1978
2277.09
1.16005
0.16005
0.02562
7.58
1960
1654.79
1965
2252.04
1.14729
0.14729
0.02169
12.05
1961
1788.83
1979
2229.47
1.13579
0.13579
0.01844
16.51
1962
1707.3
1971
2173.4
1.10723
0.10723
0.0115
20.98
1963
1798.54
1974
2138.21
1.0893
0.0893
0.00797
25.44
1964
1860.48
1968
2069.21
1.05415
0.05415
0.00293
29.91
1965
2252.04
1967
2004.84
1.02136
0.02136
0.00046
34.37
1966
1939.92
1980
1969.2
1.0032
0.0032
0.00005
38.83
1967
1897.12
10
1961
1939.92
0.98828
-0.0117
0.00014
43.30
1968
2069.21
11
1973
1900.07
0.96798
-0.032
0.00103
47.76
1969
1735.75
12
1962
1897.12
0.96648
-0.0335
0.00112
52.23
1970
1817.05
13
1966
1860.48
0.94781
-0.0522
0.00272
56.69
1971
2229.47
14
1975
1857.49
0.94629
-0.0537
0.00288
61.16
1972
1765.13
15
1970
1817.05
0.92569
-0.0743
0.00552
65.62
1973
1969.20
16
1964
1798.54
0.91626
-0.0837
0.00701
70.08
1974
2173.4
17
1977
1788.83
0.91131
-0.0887
0.00787
74.55
1975
1900.07
18
1972
1769.23
0.90133
-0.0987
0.00974
79.01
1976
19
1969
1765.13
0.89924
-0.1008
0.01015
83.48
1977
1857.49
20
1963
1735.75
0.88427
-0.1157
0.01339
87.94
1978
2277.09
21
1960
1707.3
0.86978
-0.1302
0.01696
92.41
1979
2138.21
22
1959
1654.79
0.84302
-0.157
0.02464
96.87
1980
2004.84
23
22
42975.5
21.8937
-0.1063
0.23493
1962.92
Table 3a. The parameters of the empirical curve reliability of daily mean value of object
q' of the cooling balance in the Banja Luka area, for the month of July
Year
qc,
No
kJ/m2dan
Falling value
q, kJ/m2dan
(k-1)
(k-1)2
Reliability
p, %
1958
1959
1960
1961
3638.2
3521.2
3436.1
3393.3
1
2
3
4
1958
1978
1965
1979
3638.15
3521.15
3436.08
3393.32
1.24
1.2
1.17
1.16
0.24
0.2
0.17
0.16
0.0576
0.04
0.0289
0.0256
3.12
7.58
12.05
16.51
470

1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
3262.5
3221.9
3193.5
2914
2896.7
2839.1
2818.1
2805.7
2769.6
2718.6
2654.3
2654.3
2629.6
2592.3
2506.3
2453.4
2420.2
3638.2
64204.66
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
1971
1974
1968
1967
1980
1961
1973
1962
1966
1975
1970
1964
1977
1972
1969
1963
1960
1959
3262.5
3221.9
3193.45
2914.04
2896.72
2866
2839.07
2818.05
2805.65
2769.56
2718.6
2654.28
2654.25
2629.64
2592.34
2506.32
2453.36
2420.23
1.11
1.1
1.09
0.99
0.99
0.97
0.96
0.96
0.95
0.94
0.93
0.91
0.90
0.89
0.88
0.85
0.84
0.83
0.11
0.1
0.01
-0.01
-0.01
-0.03
-0.04
-0.04
-0.05
-0.06
-0.07
-0.09
-0.01
-0.11
-0.12
-0.15
-0.16
-0.17
0.0121
0.01
0.0001
0.0001
0.0001
0.0009
0.0016
0.0016
0.0025
0.0036
0.0049
0.0081
0.0001
0.0121
0.0144
0.0225
0.0256
0.289
22
2927.19
21.47
0.5614
20.98
25.44
29.91
34.37
38.83
43.30
47.76
52.23
56.69
61.16
65.62
70.08
74.55
79.01
83.48
87.94
92.41
96.87
Table 4 Calculation of theoretical curves reliability value of average daily balance of the
cooling object q for July at predetermined values of the parameters of distribution
p, %
Ms
cv ,q
3
kc
M s 1
4
q(q') = k'mq(q')
5
471
Fig 2 Diagram of theoretical and empirical calculations of the amount of heat q i q'
depending on the reliability, by Rybkin
Diagram with a linear scale on the abscissa in Fig. 3. is not suitable for
displaying false confidence. Using the curve is more difficult for small and large values
of reliability, it is very steep. Given that these are very interesting areas for
consideration, it is a curve plotting the reliability of the logarithmic scale on the
abscissa where the "corrected" (Fig. 4). Thus increases the accuracy of
readings. These are similar to Hasena diagram, which is made according to the table
of Rybkin, given in [1]. Basic characteristic of this diagram consists of making it a
normal Gauss curve (when CS = 0) that takes the form of direction.
472
Fig. 4 The theoretical curve of reliability with middle of July the value of the balance of
the cooling facility in Banja Luka q i q' climatic region, drawn in a diagram using semialgorithm Hasen using the table of Rybkin.
3
CONCLUSION
Diagram drawn in this manner can be used to compare empirical data and
theoretical curves of reliability, and to graphically display the empirical curves and
reliability of the extrapolation of these curves outside the observations (in areas of very
low and very high reliability).
LITERATURE
[1]
[2]
[3]
[4]
[5]
Tica, G. (2010). Prilog izuavanju meusobnog djelovanja hleenog objekta i

njegove okoline, Disertacija, Banja Luka.
urikovi V. (1984). Toplifikacioni sistem kao sloeni vjerovatnosni sistem,
Doktorska disertacija. Banjaluka.
De Coursey, W. J.: Statistic and proability for Engineering Aplications, Elsevier
Science, 2003.
Feller V.: Vedeniev teoriju verojatnostej i jejo priloenija. M., Mir, 1997.
Andrejanov V. G.: Gidrologieskie raseti pri projektirovanii malih i srednji
gidroelektrostancij., Gidrometeoizdat, 1996.
473
[6]
474
Is the transfer function method reliable in a European building context? A

theoretical analysis and a case study in the south of Italy, Giorg, Applied Thermal
Engineering, Volume 25, Issues 2-3, Pages 341-357, February 2005.
EXPERIMENTAL INVESTIGATION OF THERMAL AND FLOW

PROCESSES IN SHELL AND TUBE HEAT EHCHANGERS
Mia Vuki 1, Velimir Stefanovi 2, Predrag ivkovi 3, Mirko Dobrnjac 4
Summary: Experimental investigations were done to identify influence of thermal and
flow quantities and shell side geometry to STHEs heat exchange intensity. In this
paper special attention was made to number of segmental baffles (with baffle cut of
22%) influence on apparatus efficiency investigation.
Key words: Shell and tube heat exchanger, experimental research, number of baffles
1.
INTRODUCTION REMARKS
Shell and tube heat exchangers (STHEs) are apparatuses in which the heat
exchange between hoter and colder fluid is performing. Fluid flowing through tubes is
called tube fluid, and fluid flowing around tube bundle is called shell side fluid.
Baffles, placed in shell side space, are providing the cross flow direction of shell side
fluid and so the more intensive heat exchange between fluids could be realized.
Besides, baffles are carriers of tube bundle, which helps them to decrease the
deflection in horizontal and vibrations in horizontal and vertical apparatuses. STHEs
usually have combined fluid flow, which means that there is parallel in one, and
counterflow in other part of the exchanger. This apparatuses are usually denoted as mn STHEs, where m is the number of fluid passes through the shell, and n is the
number of fluid passes through the tube bundle. If the STHE is with so called full tube
bundle, the shell side fluid flows through baffle cuts along the tubes. On the shellside,
there is not just one stream, beside a main cross-flow stream the four leakage or
bypass streams exist as a result of design type baffle to tubes, baffle to shell and
tube bundle to shell gaps (tube to baffle hole leakage stream, bundle bypass
stream, pass partition bypass stream and baffle to shell leakage stream).
One of STHE manufacturers main causes is to improve their exploitation
reliability and efficiency. Two approaches in STHE design improve are possible:
experimental investigation, which is very expensive and long lasting, because of shell
side complex geometry, and numerical investigations. Numerical simulations can be
1
Prof. dr Mia Vuki, Ni, University of Ni, Faculty of Mechanical Engineering in Ni,
(micav@masfak.ni.ac.rs)
2
Prof. dr Velimir Stefanovi, Ni, University of Ni, Faculty of Mechanical Engineering in Ni
3
Dr Predrag ivkovi, Ni, University of Ni, Faculty of Mechanical Engineering in Ni
4
Prof. dr Mirko Dobrnjac, University of Banja Luka, Faculty of Mechanical Engineering
475
Mia Vuki, Velimir Stefanovi, Predrag ivkovi, Mirko Dobrnjac
used to check the old and to develop new, more efficient STHE design. Shellside flow
is almost always turbulent since tube bundle and baffles are very nice turbulent
promoters. Character of flow around some tube rows in tube bundle is strongly
influenced by tube layout (square, rotated square, triangular, rotated triangular,
circular).
Character of flow around the tubes has a direct influence on heat exchange
between fluids. Problem is more complex if heat transfer is simultaneous with phase
change. Since the detailed measurements of turbulent characteristics of shellside fluid
flow are almost impossible, the calculated fields of pressure, velocity, temperature as
well as turbulent characteristics are of great significance in explaining very complex
thermal and flow processes in STHEs.
Basically, one can conclude that heat transfer between fluids in STHEs is
highly influenced not only by thermal and flow quantities, such as inlet temperatures
and velocities, but also with baffle cut size, baffle spacing, size of inlet and outlet zones
and number of baffles [1 - 8]. To investigate influence of mentioned parameters,
thermal, flow and geometric, with other words, to find the apparatus responseto
thermal and fluid quantities and shell sidegeometry, in steady regime, by experimental
and numerical methods, it was necessary to concept one compact experimental STHE.
In this paper, special attention was made to experimental investigation of
segmental baffles number (with baffle cut of 22%) to STHE efficiency.
2.
EXPERIMENTAL INVESTIGATION
Experimental STHE, tupe 1-2U, was projected and manufactured in

cooperation with firm MIN Inzinjering from Nis. Measuring place was formed in the
boiler house of the Mechanical Faculty in Nis. Experimental installation is shown on
Photograph 1. Basic part of the installation is STHE, type 1-2U, which is shown on the
Photograph 2. Tube bundle is made of copper U-pipes, 15/13 mm, with rotated
triangular tube layout and tube pitch of 21mm. There are 48 tubes in the tube bundle.
2
STHEs active area for heat exchange is 1.8 m . Shell is made of carbon steel,
193.7/182.9 mm. STHEs full length is 1217 mm. Taps for pressure drop
measurements are set up on the shell, along with openings for thermocouple carriers.
Three packages of segmental baffles with baffle cuts of 22%, 27% and 32% are
located in the shell. There are 5 segmental baffles in every package. Segmental baffles
and horizontal baffle are made of copper. Two measuring paths with impurity catchers,
measuring diaphragms and TA-STAD valves are made. There are long enough
rectilinear pipe sections before and after the diaphragms in order to stabilize the water
flow. Water flow regulation is done with valves. Heating of water up to certain
temperature is done with four electrical boilers, with 30kW strength. Water temperature
adjustment is provided with calotrons, as well as with one of the boilers heaters
regulation of strength. The heating fluid is warm water, and the heated is cooling water
from the local water supply. Flowing of the heating fluid through the shell is done by
circulation pump.
476
Experimental Investigation of Thermal and Flow Processes in Shell And Tube Heat Ehchangers
Photograph 1
Photograph 2
Experimental installation
Experimental shell and tube heat exchanger
Experimental investigation is done in conditions of the summer ambient

conditions in the period May-September. In all experiments constant heating fluid flow
(9m3/h), as well as inlet STHEs heated fluid (15C) temperature was maintained.
477
Heated fluid flow rate (3m3/h, 4m3/h, 5m3/h) and temperature (50C, 60C) were varied
at the inlet. At the beginning series of experiments without baffles in the shell were
done. After that, baffles were set into the shell (one, two, three, four and five baffles).
Six experiments were done for every geometry change.
3.
MEASUREMENTS RESULTS
Heating and heated fluid flow rate measurements were done with standard
measuring diaphragms, as well as with pressure drop measurements on TA-STAD
valves with CBI acquisition system (Computerized Balancing Instrument).
Pressure in the closed circulation heating fluid circle (primer) was measured
with laboratory mechanical manometer. In all experiments manometer indicatons was
in the range 3 3.5 bar.
Pressure drop in the STHEs tube bundle and on the measuring diaphragms
were done with the hydrostatic manometer (U-pipe with mercury). In the
measurements done average tube bundle pressure drop was 3710 Pa.
STHEs shell side pressure drops were measured by specially made system
consisting of taps set up on the shell, transparent rubber hoses, collector with 16
connexions, valves, shanks, millimeter partitioned scale and hand pump.System is
shown on the photography 2.1. All taps from the STHEs shell are connected by
transparent rubber hoses into a collector (totally 15 branches). One side of the
collector is closed, and the other side is connected to the compressed air source
through a valve. After desired heated fluid flow compressed air is was insufflated into
the collector, which caused forming of adequate high water columns. This way shell
side pressure drop, as well as pressure drop along STHEs shell, i.e. pressure drop
between two baffles and pressure drop in the baffle cut.
During each experiment, after accomplishing steady state heat exchange
regime (which was accomplished after 25-30 min. in expreiments done), fluid
temperature are measured with 16 previously calibrated chromel-allumel
thermocouples of 0.2 mm diameter, using the Hewlett-Packard acquisition system.
Cold sides of all thermocouple were sunk into insulated tank filled with 1:2 water and
ice mixture. All thermocouples were set into so called thermocouple movable
platforms. Apparatus inlet and outlet, heating and heated fluid temperatures (five
thermocouples), as the heating fluid temperatures in the central shells plane and 5 mm
from the shell wall, on eleven formerly defined locations along the STHE were
measured. Heated fluid inlet temperatures in the beginning, and outlet temperatures in
the end of the experiment were controlled with a mercury thermometer. This
thermometer was also used for ambient temperature measurements.
Measuring results are showed in the form of diagrams. On the x-axis
measuring place position is shown, and temperature value on the y-axis.
Figure 1 shows heating fluid temperature change along the STHEs axis and on
5mm from the shell wall for one work regime. Figure 2 shows heating fluid temperature
change in the STHEs axis depending on the flow rate. Figure 3 show heating fluid
temperature change along the STHEs axis depending on the segmental baffles
number. Figure 4 shows the apparatus heat efficiency increase depending on the
segmental baffles number, comparing with the case with no baffles.
478
Experimental Investigation of Thermal and Flow Processes in Shell And Tube Heat Ehchangers
Fig. 1 Heating fluid temperature change

along the STHEs axis and on 5 mm from
the shell wall for: Vt, = 4 m3/h, Np = 1,
BC = 22%, t1t = 60oC

along the STHEs axis depending on the
flow rate for: Np = 1, BC = 22%, t1t = 60oC

along the STHEs axis depending on the
segmental baffles number for:
Vt, = 5 m3/h, BC = 22%, t1t = 60oC
Fig. 4 Apparatus heat efficiency

increase depending on the segmental
baffles number, comparing with the case
with no baffles, for t1t = 50oC
4.
DISCUSSION OF THE RESULTS
Results of the experiments done shows that STHEs heat excange strongly
depends of the shell side geometry (number of segmental baffles, baffle cut size, baffle
distance, the first and the last baffle position to inlet and outlet nozzle, respectively,
size of the constructive clearances).
With the experimental results analysis the following can be concluded:
479
x
x
x
x
x
x
x
x
x
Shell side baffle strongly influence the fluid flow characteristics and cause vortex
appearance (downstream behind the baffle);
Downstream fluid temperatures are significantly lower (up to 100C behind the first
baffle) which is caused by intensive heat exchange in the cross flow ahead and
before the baffle, and by axial flow in the baffle cut. Certainly, this is also caused
by baffle to shell leaking, and by cooling of the fluid through the shell wall;
Increasing of heating fluid flow rate decreases temperature drops behind the baffle;
Increasing of the heating fluid inlet temperature, and/or flow rate maintains
temperature change character in the axis and near the shell wall;
The most intensive heat exchange is in the STHEs inlet zone (to the first baffle)
Heating fluid temperature near the shell wall is lower, compared to the heating fluid
temperature in the STHEs axis, up to the first baffle, and after the first baffle
situation is reversed, which is consequence of the baffle to shell leaking, because it
doesnt take place in the heat exchange;
Influence of this leaking increases downstream the first baffle;
Increase of segmental baffles number has higher influence to the STHEs heat
efficiency than the increase of the heating fluid flow rate;
3
o
In this case (Vt = 5 m /h, t1t = 50 C), with five segmental baffles STHEs heat
efficiency increased for 9.4%.
LITERATURA
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
480
Aicher, T., Kim, W., Experimental Investigation of the Influence of the Cross Flow
in Nozzle Region on the Shell Side Heat Transfer in Double-Pipe Heat
Exchangers, Int. Comm. Heat and Mass Transfer, Vol. 25, No. 1, pp. 43-58, 1998.
Gaddis, E., Gnielinski, V., Pressure Drop on the Shell Side of Shell-and-Tube
Heat Exchangers with Segmental Baffles, Chemical Engineering and Processing
36, pp. 149-159, 1997.
Gay, B., Mackley, N., Jenkins, J., Shell-Side Heat Transfer Coefficients in
Cylindrical Heat Exchangers the Influence of Geometrical Factors. II The
Leakage Case, Letters in Heat and Mass Transfer, Vol. 8, Iss. 6 , pp. 437-452,
1981.
Ili, G., Vuki, M., Vukovi, G., Parameters Influencing the Effective and Optimal
Design of Shell and Tube Heat Exchangers, International DAAD Seminar, 2729.11.2000.god., Sofia, Bulgaria.
Li, H., Kottke, V., Effect of Baffle Spacing on Pressure Drop and Local Heat
Transfer in Shell-and-Tube Heat Exchangers for Staggered Tube Arrangement,
International Journal of Heat and Mass Transfer, Vol. 41, No. 10, pp. 1303-1311,
1998.
Mukherjee, R., Effectively Design Shell and Tube Heat Exchangers, Chemical
Engineering Progress, 1998.
Vuki, M., Eksperimentalno i numeriko istraivanje termo-strujnih procesa u
doboastim razmenjivaima toplote, Doktorska disertacija, Mainski fakultet
Univerziteta u Niu, Ni, 2004.
Sparrow, E. M.; Reifschneider, L. G., Effect of Interbaffle Spacing on Heat
Transfer and Pressure Drop in a Shell-and-Tube Heat Exchanger, International
Journal of Heat and Mass Transfer Vol. 29, Iss. 11, pp. 1617-1628, 1986.
INVESTIGATION OF THE ENERGY CONSUMPTION IN

REGULATING THE FLOW RATE OF PUMP SYSTEMS
Popov G.1, Klimentov Kl.2, Kostov B.3
Summary: The energy consumption involved in transporting fluids (liquids and gas)
represents a large part of the total energy demand in the world. The problems
associated with the energy analysis and a systems efficiency improvement when they
transport fluids is very topical and important, because of the wide use of these
systems. In this paper, the results of investigating the energy consumption of systems
operating with centrifugal and axial pumps when their flow rate is regulated are
presented. The impact of both the characteristics of a pump type and a pipe system on
the energy consumed in transporting a given volume of liquid is also investigated.
Key words: pump system, energy efficiency, regulating the flow rate
1. INTRODUCTION
The literature categorically argues that the frequency method is the most
effective way for the flow rate to be regulated. About the other existing standard
methods throttle and bypass, it is well known that they are less effective in terms of
energy efficiency. However, some literature sources argue that the last two methods of
regulation can be used in some particular situations to improve the systems efficiency,
but there is no information abot how big this improvement is.
In this paper the numerical results for the specific energy consumption used to
transport a given volume of liquid when you regulate the flow rate using the three
methods frequency, throttle and bypass, are presented.
2. METHODOLOGY AND CONDITIONS OF THE INVESTIGATION
To estimate the efficiency of the different ways to regulate the flow rate the
methods in [5] are used. For the purpose of this investigation different type of pumps
are used:
Centrifugal pumps (single stage and two side sucking) produced in Bulgaria
by VIPOM, company [6]. The equations fo the characteristics for these
pumps are given in [4].
1
Prof. Popov G., Rousse, Bulgaria, Angel Kanchev University, gspopov@uni-ruse.bg

Assist. Prof. Klimentov Kl., Rousse, Bulgaria, Angel Kanchev University, kklimentov@uni-ruse.bg
3
PhD student Kostov B., Rousse, Bulgaria, Angel Kanchev University, bkostov@uni-ruse.bg
2
481
Popov G., Klimentov Kl, Kostov B.
Axial pump SJP 48PO 1s

data from the companys
characteristic H f (Q) and
determined by the methods
determined by the interval
equations are:
produced by Sulzer company. Based on the

catalogs [7] the coefficients of the head
the characterictic of efficiency K f (Q) are
in [4]. For the work part of the characteristic,
Q (3000 y 5000 )l / s , the coefficients of the
Table 1
A
M
3,7565
(m.s/l)
0,0034
m(s/l)2
-7.10-7
- 1,483
s/l
0,0671
(s/l)2
- 7.10-6
In establishing an outcome work regime a pumps nominal work regime where

it works with a maximum coefficient of efficiency is selected. The specific energy
consumption e V for the outcome work regime can be determined by the following
equation:
g a bQ cQ 2
(1)
,
eV
3 600 d eQ fQ 2
where a, b, c are the coefficients for the head equation; d, e, f - coefficients of the
equation of efficiency, g 9,81m / s 2 - gravity acceleration.

The energy consumption e V1 for each of the methods used to ensure a lower
flow rate Q1 can be found by the following equations:
Throttle method of regulating the flow rate:
g H1
g a bQ1 cQ12
(2)
,;
e V1, dr
3 600 K1 3 600 d eQ1 fQ12
Frequency method of regulating the flow rate:

2
H
g
H
Hst Q1
st
, kWh / m3
(3)
1
e V1,n

H Q
3600 d e QN f QN2 H
Bypass, method of regulating the flow rate:

2
H
g
H
Hst Q1 Qp
st
(4)
, kWh / m3
1
e V1,b

H Q Q1
3600 d e Qp f Qp2 H
where H, Q are respectively the pump's head and the flow rate for its outcome work
regime; QN - the flow rate of the pump's similar work regime, when the speed of
rotation is nominal and the frequency method is used for regulation; Hst - the static
head of the system; QP - the pump's flow rate when bypass method is used for
regulation.
The analysis is accomplished by the relative values:
Relative specific energy consumption - e V,1 / e V ;
Relative flow rate - Q1 / Q .
482
Investigation of the energy consumption in regulating the flow rate of pump
The investigation was done for three different values of the systems static head,
whose values referred to the nominal head of the pump:
Hst
Hst
Hst
0,8 ;
0,6 ;
0,4 .
H
H
H
The impact of the static head for systems operating with axial pumps is
investigated in two different cases:
Hst
Hst 0 ,
0,5 .
H
3. RESULTS AND ANALYSING THEM
The change of the relative specific energy consumption when the flow rate has
been regulated from Q1 to 0,8Q (the outcome systems flow rate) is shown in fig.1. A
centrifugal pump 50E50 is used in this system. [6]. The character of the curves for all
other investigated pumps is analogical for them both.
Fig. 1
Looking at fig.1 it is easy to conclude that the frequency method of regulation
is the most energy efficient method in terms of the energy efficieny of the system.
When the relative flow rate decreases the efficiency of using this method increases
compared with the other two. The specific energy consumption decreases with the
decreasing of the flow rate, independently, that the liquid volume given to the user for
unit of time also decreases, and somewhat the coefficient of efficiency decline too. The
previous fact is due to the lower head which the pump produces when it works with
small flow rates. When the relative static head Hst / H decreases, according to (3), the
systems energy consumption also decreases. Another influence on the previous fact is
the smaller change of the pumps coefficient of efficiency to the outcome work regime
in systems with lower static head. The fact is that a given flow rate of a system with
lower static head can be provided by deeper regulating (a bigger reduction of the
speed of rotation) to a system with higher static head - Hst . In particular cases it can
lead to a significant decrease of the motors coefficient of efficiency.
483
Using the throttle method the specific energy consumption increases gradually
with the decreasing of the relative flow rate, as for Q1 0,8 Q the rise is about 15%. In
this case the static head of the system doesnt influence the energy efficiency of the
system. The efficiency decrease, as it is well known, is due to the fact that a smaller
flow rate Q1 is produced when the pump works with higher head that the system
needs and the unnecessary head (energy) is absorbed by the control device.
The bypass method of regulation is the least energy efficient method. The
increase of the energy consumption with the decreasing of the relative flow rate can be
seen in fig.1. In this case, it is the same as when the frequency method is used, a
pump with lower head produces lower systems flow rate - Q1 , but the difference is that
the pumps flow rate is bigger than the systems output flow rate. Its clear that as the
smaller the flow rate Q1 is, the pumps flow rate QP and the ratio Qp / Q1 in (4) will be
bigger. For this method of regulation when the static head decreases the specific
energy consumption significantly increases. For a system with lower static head the
pumps head which is necessary to provide a given flow rate Q1 also will be also
smaller and then the difference between the flow rates of the pump and the system will
be larger.
Allowing an outcome work regime to be different than the nominal regime does
not influence the nature of the change in the specific energy consumption if the relative
flow rate is reduced.
The change of the specific energy consumption for the investigated system
when the original (primary) flow rate is reduced by 20% and the outcome flow rate is
also bigger than the nominal Qn (in this case Q 1,2 Qn ) is given in fig.2.
Fig. 2
As can be seen, with the decrease of the relative flow rate the specific energy
consumption increases for both the frequency and the bypass methods of
regulation. Nevertheless, when the flow rate is decreased by the throttle method of
regulation the coefficient of efficiency is improved, as for Q1 / Q 1 it has its maximum
Kp Kmax , and the relative energy consumption is bigger than the outcome energy
484
consumption. The literature argues that the cases when a throttle method of regulation
is used (reducing the flow rate) and the pumps work regime has a higher coefficient of
efficiency, then this method can be defined as effective. The analysis of fig.2 clearly
indicates that from the point of the systems energy efficiency to use the throttle
method of regulation is not effective and it is also associated with increasing the
specific energy consumption.
When pumps work with bigger flow rates than the nominal flow rate, they
produce lower values of specific energy consumption. Nevertheless, the pumps
coefficient of efficiency decreases for these work regimes and the lower head,
according to (4), leads to lower values of e V . Figure 3 shows the changes of the
relative energy consumption for different flow rates related to the energy consumption
when the pump works with its nominal work regime - e V, n . These curves are
characteristics for single-stage centrifugal pumps whose specific speeds of rotation are
different - nq . It can be seen, that when the flow rate increases the specific energy
consumption gradually decreases. The inclination of both curves is different, as when
nq decreases the lines start to be more slopping. This is due to the different head
characteristics of each pump. This typical curve for pumps with a lower specific speed
of rotation becomes more slopping and when the flow rate increases the pumps head
changes also becoming more sloping.
Fig. 3
An interesting fact is the change of the energy consumption for transporting a
unit volume liquid when the system operates with an axial pump. A well known fact
about these pumps is that when their power decreases the flow rate increases.
Therefore, many literature sources (for instance - [1], [2] and [3]) argues that in this
case the bypass method of regulating the flow rate is the most effective that can be
used.
Figure 4 shows the change of the specific energy consumption for a given
pump system operating with an axial pump SJP 48PO 1s (produced by Sluzer
company). The comparison between fig.4 and fig.1 shows, that for the throttle method
of regulation the flow rate is significantly less efficient when axial pumps are used
485
instead of centrifugal pumps. It can be seen, that the relative energy consumption
increases at a steeper rate than the energy consumption when the pump works with its
nominal work regime. This fully corresponds with the consumed power increasing and
flow rate decreasing for this type of pump. Another interesting fact is the impact
(change) of the specific energy consumption when the bypass method of regulation is
used. For an axial pump system this method also is less efficient. The energy used in
sluicing a part of the total flow rate (the unnecessary flow rate) is significant, even
though that when the flow rate increases the consumed power decreases. In this case,
the systems static head has less influence on the energy consumption. This is due to
the relation between the pump's power consumption and the flow rate (fig. 4). When
the systems static head is lower the pump will produce a bigger flow rate to provide
the required users flow rate, consequently less power will be consumed. This is some
compensation for the energy lose when the flow rate is reduced and the unnecessary
volume of liquid is sluiced (taken).
Fig. 4
4. CONCLUSION
This numerical investigation of the impact of the specific energy consumption in

regulating the flow rate of systems operating with turbo pumps shows:
The pump type doesn't influence the fact that the bypass method is always
associated with the highest value of energy consumed in transporting a unit
volume of liquid. The specific energy consumption decreases when the flow
rate also decreases, only in cases where the frequency method is used.
The work regimes where the flow rate is overloaded are characterized by a
decrease in the specific energy consumption. Reducing the flow rate by
using the throttle method, until the best efficiency work point is found, does
not improve the system's efficiency. The head characterictic, determined by
the specific speed of rotation, directly influences the rate of change in
energy consumption.
The systems static heads influence is different if the frequency method to
486
regulate the flow rate is used rather than the bypass method. The relative
static head decreases with a decreasing of the energy consumption when a
frequency method of regulation is used. If the bypass method is used, then
the specific energy consumption increases when the relative static head
decreases. For systems operating with axial pumps this increase is
significantly smaller than if centrifugal pumps are used.
Special thanks to: EN-The study was supported by contract BG051PO0013.3.04/28, "Support forthe ScientificStaff Development in the Field of Engineering,
Research andInnovation. The project is funded with support from the Operational
Programme "Human Resources Development" 2007-2013, financed by theEuropean
Social Fund of the European Union.
LITERATURE
Journal Papers:
[1]. Klimentov, Kl., Popov G., Tujarov Kr., (2008). Equation of parameters of rotary
pums, Energetika 6-7, Sofia, p. 60-63.
Books:
[2]. Grozev, G., Stoyanov, St., Guzhgulov G.,(1990). Hydraulic and pneumatic
machines and drive systems. Tehnika, Sofia
[3]. Gulich J. F. Centrifugal pumps, Springer, (2010) (Second edition)
[4]. Karassik Igor J., W. C. Krutzsch, W. H. Fraser, J. P. Messina, Handbook pumps.
McGraw-Hill book company (Second edition)
Proceedings Papers:
[5]. Popov G., Klimentov Kl., Kostov B., (2011). Methods to estimate the energy
consumption in regulating the flow rate of pump systems, Scientific conference
Banja Luka, Bosnia, May, 26-27 (printed in this book)
WWW pages:
[6]. www.vipom.ru
[7]. www.sulzer.com
487
LDA, CLASSICAL PROBES AND FLOW VISUALIZATION IN

EXPERIMENTAL INVESTIGATION OF TURBULENT SWIRL FLOW
ore S. antrak1, Slavica S. Risti2 and Novica Z. Jankovi3
Summary: Investigation of turbulent swirl flow occupies attention of many researchers
as it has great significance on processes in nature, engineering applications, medicine
and elsewhere. Here are presented experimental results of research of turbulent swirl
flow generated on the axial fan pressure side. These results are obtained with Laser
Doppler Anemometry (LDA), classical probes with precise micro-manometers and flow
visualization with usage of Nd:Yag laser in circular pipe for Reynolds number 1.5105.
Flow integral characteristics obtained from LDA results have shown good agreement
with ones calculated on the basis of classical probe measurements, except in the
central region, where radial, the third velocity component, is present. Flow visualization
reveals vortex dynamics.
Key words: turbulence, swirl flow, LDA, classical probes, flow visualization.
1. INTRODUCTION
Turbulent swirl flow occurs in space, nature and industry, and it has been
occupying researchers for years [1-7. Main researcher's aim is defining more complete
picture about physical essence of extremely complex processes which occur in
nonhomogeneous flow and definition of adequate turbulent transport models. The aim
is modelling and solving system of equations which describe processes in pipe
turbulent swirl flow. Basic transport equations and influence of the centrifugal force on
turbulence statistical characteristics and turbulent exchange processes are tested.
Joint theoretical and experimental researches lead to analysis of the process of nonlocal turbulence transport and non-gradient turbulent diffusion in turbulent swirl flow.
Turbulent swirl flow in technical systems can be generated in numerous ways,
by guided vanes [2], swirl generators with various geometries [6], as well in various
hydraulic machines pumps, turbines, fans [1-5], combustors, engines, separators and
etc. Here are presented results of turbulent swirl flow investigation behind the axial fan
with laser based experimental methods.
Dipl.Ing., Teaching and Research Assistant, Belgrade, Faculty of Mech. Eng., (djcantrak@mas.bg.ac.rs )
Ph.D., Principal Research Fellow, Belgrade, Institute Goa, (slavica.ristic@institutgosa.rs )
3
Dipl.Ing., PhD student, Belgrade, Faculty of Mech. Eng., (njankovic@mas.bg.ac.rs )
2
489
. antrak, S. Risti and N. Jankovi
2. TEST RIG
Turbulent pipe swirl flow was studied in the test rig presented in Fig.1a where:
1- LDA-3 measuring section and flow visualization, 2-Nd:Yag laser for visualization, 3classical probe measuring section and angle probe, 4-smoke generator probe position,
5-LDA-1 measuring section, 6-profiled free bell-mouth inlet, 7-axial fan (swirl
generator), 8-DC motor. Axial fan, swirl generator with blade position R=22, is
positioned at the pipe inlet, after the aerodynamically profiled free bell-mouth inlet for
providing uniform flow at the fan inlet. Turbulent swirl flow is studied on fans pressure
side. Pipe has inner diameter D=0.4m. It has two transparent sections, one, at the inlet,
1.5m long, and one at the outlet 1.38m long. Two plastic, nontransparent, pipe sections
are 4m each long. Therefore, total length of the installation is L=27.2D. Test rig
configuration can be varied depending on the measurements needs. Fan rotation
speed was regulated by a fully automated thyristor bridge with error up to 0.5rpm,
where fan shaft speed has been measured by photo cell. It was, for all measurements,
n=1000rpm, which generated Re=1.5105, calculated for the volumetric mean axial
velocity. Axial fan is equipped with DC electric motor, with 5kW power.
Fig. 1 a) Experimental facility; b) LDA measurements of circumferential velocity

component; c) Determination of total and static pressure.
3. MEASUREMENTS AND ANALYSIS
3.1 LDA measurements
LDA measurements were performed along the vertical diameter in sections
x=3.26D, measuring section LDA-1, and x=26.1D, measuring section LDA-3 (Fig. 1.)
in points on 10mm distance each. Commercially available one-component laser
anemometer, Dantec Flow Explorer Mini LDA with BSA F30 signal processor was
used. It has measurement distance at 285mm, laser power 35mW, measurement
volume d=0.1mm, l=1mm and maximum velocity 27m/s. It works in backscattered
mode. Velocity is measured with uncertainty lower than 0.1%.
LDA measurements of circumferential velocity component are presented in
Fig.1b, where 1-laser, 2-computer controlled linear guide, 3-profiled bell-mouth inlet, 4swirl generator, 5-seedings. Seeding, naturally sucked in the test rig by the axial fan,
was obtained by fog generator Hurricane 1700, manufactured by Chauvet. Used liquid,
HI TECH, by JBSYSTEMS, is water based. The main chemical constituent of this fluid
is eucalyptus extract. This was the way to obtain enough seeding for needed sampling
rate of more then 6 kHz. With this generator it was possible to obtain, in some
positions, more than 15 kHz. Distribution of time averaged values of axial and
490
LDA, Classical Probes and Flow Visualization in Exp. Investigation of Turbulent Swirl Flow
circumferential velocity components are presented in Fig.2, where index denotes LDA
measuring section.
Fig. 2 Distribution of the time averaged a) axial (U), b) circumferential (W) and c) radial
(V) velocity components along vertical diameter in sections LDA-1 and LDA-3.
Behind this axial fan is generated swirl flow with four distinguished zones. First,
in central region is named vortex core where fluid rotates by the solid body law
W=rconst. Zone situated between the swirl core and the sound flow region is called
viscous zone, where is the big influence of inner viscosity caused by velocity gradients
in radial direction and dominant turbulence intensity [2]. Next zone is sound flow
region. Here fluids rotates by optional law W=f(r), while here is W=const/r, what was a
design rule for this axial fan. In this zone axial velocity distribution is almost uniform,
what is more evident in the downstream section LDA-3 (Fig. 2a). Boundary layer zone
is the fourth one. It is evident that healthy region (Fig. 2a) is wider in the downstream
cross-section. Volume flow rate and the averaged circulation over whole flow mass are
calculated as follows:
1
k UWdk
2
2S R 2 kUdk ,
0
2S R 0
kUdk
0
where k=r/R is dimensionless radius. Values of volume flow rates calculated for
measuring sections LDA-1 and LDA-3 are, respectively, Q1=0.796m3/s and
Q3=0.7695m3/s. Relative difference is less than 3.33%. Averaged axial velocities
Um,1=6.33 m/s and Um,3=6.12 m/s are calculated from volume flow rates.
The averaged circulations are 1=3.67m2/s and 3=3.31m2/s. Decrement of
averaged circulation along the pipe proves swirl decay process [2,7].
From these velocity profiles additional integral values could be obtained like
non-dimensional swirl parameter (), introduced by Strscheletzky, as follows:
=Q/(R). It has intensities 1=1.08 and 3=1.16, what also agrees with swirl decay
processes. Various definitions of swirl parameters exist and they are of interest for
characterization of swirl flow.
Radial velocity distribution (Fig. 2c) shows significant values in the core region
and very small ones in other regions. Also, it has significant gradient. These
distributions almost follow symmetrical distributions of axial velocities.
The level of turbulence is calculated for all three velocity components in both
sections (Fig. 3). The highest intensity is in the core region. This is also related to the
491
swirl transformation and decay process, as well as vortex core dynamics. The
turbulence doesnt go under 10% in the section LDA-1, and 5% in section LDA-3.
Fig. 3 Turbulence intensity in sections a) LDA-1 and b) LDA-3.

Measurements were done within the acquisition time AT=10s, set as stop
criteria. The total number of data collected at points along the vertical diameter ranged
from very few up to 56694. The data validation during test was 90-100%. Sensitivity
was between 1200-1400V. The 95% confidence limits for the Mean and RMS values
are computed on line during acquisition. The equations between the uncertainty of the
mean velocity, the RMS velocity, the turbulence intensity, and the user specified
relative uncertainty , determine the required number of independent samples N. The
uncertainty is =1.3% [8,9].
3.2 MEASUREMENTS WITH ORIGINALLY DESIGNED CLASSICAL PROBES
New, original designed, classical probes were previously developed and
described in [3]. Measurements with these probes were performed in section x=21.4D
(Fig. 1a). Set up for determination of total and static pressure is presented in Fig. 1c: 1combined Prandtl probe, 2-micromanometer; 3-data acquisition, 4-differential pressure
transmitter (p=100Pa), 5-differential pressure transmitter (p=10hPa), 6-data logger,
7- temperature and relative humidity measurement. The flow is considered to be twodimensional with neglecting radial velocity component. Measuring procedure with
classical probes is, consisted of three steps. The time-averaged angle between the
velocity vector and pipe axis x is measured by using the classical angle probe, which is
more sensitive to the angle variation than in the case of multi-hole probes. Following
this step, the head of the combined Prandtl probe is positioned in a determined angle,
and the total pressure is measured in this way. Static pressure is measured by using
the same probe but with attached sleeve to the probe head, which has a forepart,
rounded on the top and completely closed.
In Fig. 4a are presented total and static pressure distributions, while in Fig. 4b
are presented axial and circumferential velocity components distribution, calculated on
the basis of pressure distributions. Volumetric flow rate differs around 8% from value
calculated on the basis of LDA measurements. Also, this circumferential velocity
distribution resulted in average circulation smaller then in the downward section LDA-3,
2
2=2.68m /s, what is physically unacceptable. Comparison of circumferential velocity
profiles obtained by LDA measurements shows difference in the core region. As radial
velocity has significant values in this region, it means that assumption of quasi two-
492
LDA, Classical Probes and Flow Visualization in Exp. Investigation of Turbulent Swirl Flow
dimensional flow is unacceptable in core region even in sections on a finite distance

from swirl generator.
Fig. 4 Measurements with classical probes: a) Total and static pressure distribution; b)
Calculated axial and circumferential velocity components.
3.3 Flow visualization
Flow visualization was performed in section LDA-3, in the pipe cross section
and in the meridian plane. For this reason was employed 15Hz dual Nd:Yag laser
(New Wave Research, Solo PIV, max power: 30mJ/pulse, wavelength 532nm, Fig. 1),
with mounted cylindrical lens f=25mm and one spherical lens f=500mm was.
Smoke generator (Elven, Precision Limited) probe with Pitot-tube geometry,
was positioned upstream the measuring section in position x=17.1D (Fig.1). Flow
visualization was performed in pipes cross-section and in meridian section r-x plane
for n=1000rpm (Fig. 5). Used camera was SONY, DSC-H3, 10X Optical Zoom, F/3.54.4, 8.1MP with 30 fps.
Fig. 5 Flow visualization in section LDA-3: a) cross-section and b) meridian plane.

4. CONCLUSION
Measurements in two separated sections have proved a model of swirl decay
in swirl pipe, with average circulation decreasing. Relative difference in measuring
volume flow rate was 3.33%, what is satisfying. Also, the averaged circulation
measured in section 1, was smaller than in LDA-3 section. The circumferential velocity
distribution, measured with classical probes differed from previous ones with LDA.
Relative difference in volumetric flow rate was around 8%. Previous results [2]
493
recognized problem of non-repeatability in central region with classical probe. Another

fact enforced this conclusion. Distribution and values of radial velocity in central region
proved concept of non-planar fluid flow in vortex core.
Flow visualization qualitatively proved total and static pressure distribution in
vortex core. Non-uniform seeding distribution is reported, but needs further
investigation in this field. Optical methods, very useful for this study, reveal three
dimensional flow character. More profound results are expected with three component
LDA systems. It should be also mentioned that one should take care of precise
traversing and clear optical access to the measuring point. It is of great significance for
acquiring reliable results.
ACKNOWLEDGMENT
This work was funded by grants from the Ministry of Science and Technological
Development, Republic of Serbia, Projects No. TR 35046 and No. IP 391-0000027/2009-02/164. The authors would like to express gratitude, also, to colleagues
from the Faculty of Technical Sciences, Department of Fluid Mechanics, Novi Sad,
Serbia for use of LDA system.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
494
Proti, Z. (1970). Determination of the Dead Water Diameter in Swirl Turbulent

Flow in Straight Circular Pipes, (in Serbian language), 10th Yugoslav Congress of
Rational and Applied Mechanics. Conference Proceedings, p. 429-442.
antrak, S., Beniek, M., Nedeljkovi., M., Lei M. (2001). Problems of NonLocal Turbulent Transfer Modelling. ZAMM, vol. 81, s4, p. 913-914.
Beniek, M., Lei, M., Ili, D., antrak, . (2010). Application of New Classical
Probes in Swirl Fluid Flow Measurements. Experimental Techniques, vol. 34,
issue 3, p. 74-81.
Proti, Z., Nedeljkovi, M., antrak, ., Jankovi, N. (2010). Novel Methods for
Axial Fan Impeller Geometry Analysis and Experimental Investigations of the
Generated Swirl Turbulent Flow. Thermal Science, vol. 14, p. 125-139.
antrak, ., Ili, J., Hyde, M., antrak, S., oi, A., Lei, M. (2008). PIV
Measurements and Statistical Analysis of the Turbulent Swirl Flow. ISFV 13 13th
International Symposium on Flow Visualization, Conference Proceedings, CDROM, ID 183-0804203
Volchkov, E., Lebedev, V., Lukashov, V. (2004). The LDA study of flow gasdynamics in a vortex chamber. International Journal of Heat and Mass Transfer,
vol. 47(1), p. 3542.
Steenbergen, W., Voskamp, J. (1998). The Rate of Decay of Swirl in Turbulent
Pipe Flow. Flow Measurement and Instrumentation, vol. 9, issue 2, p. 67-78.
Risti, S. (2007). Laser Doppler Anemometry and its Application in Wind Tunnel
Tests. Scientific Technical Review,vol.LVII, no.3-4, p. 64-76.
Ili, J., Risti, S., Srekovi, M. (2001). The Detection Angle Choice in
Characterization of Aerosols and Hydrosols by LDA. International Conference on
Laser 2000, Conference Proceedings, p. 620-627.
METHODS TO ESTIMATE THE ENERGY CONSUMPTION IN

REGULATING THE FLOW RATE OF PUMP SYSTEMS
Popov G.1, Klimentov Kl.2, Kostov B.3
Summary: The energy consumption involved in transporting fluids (liquids and gas)
represents a large part of the total energy demand in the world. The problems
associated with the energy analysis and a systems efficiency improvement when they
transport fluids is very topical and important, because of the wide use of these
systems. This paper presents a method to determine the specific energy consumption
used in regulating the flow rate of pump systems whose purpose is to ensure the
transport of liquids to the users. The methods presented in this work ensure a
comperative analysis of he efficiency by using different ways to regulate the systems
flow rate to be done.
Key words: pump system, energy efficiency, regulating the flow rate
1. INTRODUCTION
Pump systems are mostly used to transport fluids, mainly for water supply in
buildings, for industrial use and for irrigation systems in farming, etc. Because they are
widely used, it is especially important to solve the problems associated with their
efficient working performance enabling them to save energy. To ensure the better
efficiency of these systems an optimal work regime should be provided [2]. The work
regime depends on the agreement between the characteristics of the pump head and
the pipe system. When the pump motor is started the transitional processes in this time
also influence the work regime of the system (turbo pump head pipe water turbine)
[3],[4] When the pumps flow rate is changed to achieve a new and more effective work
regime, different methods for flow rate regulation can be used. It is well-known fact that
the energy efficiency of the system depends on the selected method of regulation. By
the existing standard methods for changing the pumps flow rate the most appropriate
regulation method seems to be the frequency method. The effective use of energy into
pump systems significantly depends on their water supply schedule. An important
factor in determining the level of efficiency for the particular work regime of a pump
system is the price of the final product, i.e. what is the energy coast to transport 1m3
1
Prof. Popov Gencho, Rousse, Bulgaria, Angel Kanchev University, gspopov@uni-ruse.bg

Assist. Prof. Klimentov Kliment, Rousse, Bulgaria, Angel Kanchev University, kklimentov@uni-ruse.bg
3
PhD student Kostov Boris, Rousse, Bulgaria, Angel Kanchev University, bkostov@uni-ruse.bg
2
495
Popov G., Klimentov Kl., Kostov B.
water to the user. The coast can be varied by using different methods to regulate the
flow rate. Therefore, it is necessary to establish a criterion to estimate the specific
energy consumption for the method to be used to regulate the flow rate. In this paper a
method to determine the values of this criterion when three different types of regulation
have been used the frequency, the throttle and the bypass methods, is presented.
2. SPECIFIC ENERGY CONSUMPTION IN TRANSPORTING OF FLUIDS
As a criterion to estimate the liquid transports efficiency for a given pump
system the invested energy in transporting of a unit volume of fluid can be used:
E kWh
,
,
eV
(1)
V m3
where e V is the specific energy consumption; E the invested input energy of the
system, usually electrical, which the pump systems need to get started; V the
transported fluids volume, most measured in [m3 ] .
This equation could be also presented by the following equation:
P
E 1
(2)
,
eV
t V Qh
t
where P is the consumed input power of the system, [kW ] ; Qh the transported
fluids volume, [m3 / h] .
Reading the power balance of the pump system (motor, gears and pump) the input
power of the system can be given by the following equation:
PP
U gQH
, [kW ]
(3)
P
KE KT KP 1000 KE KT KP
where PP is the actual useful pumps power; H the pumps head; U the liquids
density (for cold water U
1000, [kg / m 3 ] ); g the gravity acceleration; KE the
motors coefficient of efficiency; KT the transmissions coefficient of efficiency

(usually KT | 1 ) for transmissions which connect the motors shaft to the pumps shaft);
KP the pumps coefficient of efficiency.
After substituting (3) in (2) and taking into account, that the flow rate Q
Qh
the pump systems specific energy
measured in [m3 / h] is equal to
3600
consumption can be expressed by the following equation:
g H
eV
, kWh / m3 ,
(4)
3 600 K
where K is the total pump systems coefficient of efficiency ( K KE .KT .KP ).

By this coefficient e V an analysis of the pump systems efficiency can be
accomplished, when different methods of regulating the flow rate were used. For a
given work regime each pump produces a concrete head H and has also its coefficient
496
Estimation of the energy consumption in regulating of the flow rate of pump systems
of efficiency KP therefore, using (4) the specific energy consumption e V of the system
can be estimated.
To investigate the impact of the efficiency when the work regimes have been
changed (i.e. in regulating of the pump systems flow rate) it is necessary to know
about the equation of head H f (Q) and the equation of efficiency KP f (Q) for the
particular pump. For centrifugal pumps these equations usually can be expressed:
(5)
H a bQ cQ2 ;
KP d eQ fQ 2 ,
(6)
where a, b, c are the coefficients for the head equation; d, e, f - coefficients of the
equation of efficiency.
When the flow rate Q is known, using the upper equations the coefficient e V
could be presented:
g a bQ cQ2
(7)
.
eV
3 600 d eQ fQ2
The most used and well known methods to regulate the flow rate of centrifugal
pump systems are the throttle method and frequency method and in some cases a
method where a part of the flow is sluiced by the outflow pipe (known as bypass
method) also can be used. The specific energy consumption for a new and smaller
flow rate can be determined in different ways for the different methods of regulation.
2.1 Specific energy consumption when the throttle method is used
When using the throttle method to determine the work regime the work points
start to move left on the graph when the regulating device is fully opened. In this case,
the head increases and the flow rate decreases. The pumps coefficient of efficiency
also changes its value. For the new and smaller flow rate Q1 (after throttle regulation)
the specific energy consumption of the pump system can be given:
g H1
g a bQ1 cQ12
(8)
.
e V1, dr
3 600 K1 3 600 d eQ1 fQ12
2.2 Specific energy consumption when the frequency method is used
Using the frequency method we can reduce the speed of rotation which will
decrease both the flow rate and and the pump head and the work points will move
along the pipe systems characteristic. For the new and smaller flow rate Q1 the
pumps head can be found by using the pipe systems characteristic:
H1 Hst kQ12 ,
(9)
where Hst is the static head of the pipe system; k coefficient of the resistant
characteristic of the pipe system.
The coefficient k can be estimated by the pumps indicators when the pump
works with its nominal speed of rotation i.e. by using the values of the output flow
rate Q and head H too:
497
(10)
H Hst
.
Q2
Using (9) and (10) for the head of the new work regime a conclusion can be
k
found:
2
H
H Q
H1 H st 1 st 1 .
H Q
H
The value of the coefficient of efficiency KP1 for the new work regime which also
has a smaller speed of rotation n1 can be estimated by the equations of similarity. For
this purpose the similar work regime with nominal speed of rotation nN should be
found, when the equation Kp f Q is previously known. The flow rate of this work
(11)
regime is considered to be - QN . The parabola of similarity connecting the work points

to the flow rates Q1 and QN can be given by the following equation:
H1 m 1Q12 ,
(12)
where m1 is the parabolas coefficient for both similar work regimes which at first
approximation the pump works with same coefficient of efficiency KP1 . When Q1 and
H1 are known, the following equation can be given:
(13)
m1
H1
Q12
For the work point of the characteristic, where the pump has flow rate QN and
nominal speed of rotation, the following equations can be used:
HN a bQN cQN2 ;
(14)
HN m1 QN2
(15)
Using the two previous equations a conclusion can be found:
1
b 2 4 a m c b .
(16)
QN
2 m1 c
The pumps coefficient of efficiency when the work regime has flow rate Q1
and head H1 is:
(17)
Kp1
d eQN fQN2 .
The specific energy consumption of the new flow rate obtained by the using the
frequency method can be determined by the pump systems indicators, which are
estimated by (11) and (16):
2
H
g
H
Hst Q1
st
, kWh / m3 .

e V1,n
1
(18)
3600 d e QN f QN2 H
H Q
Using the previous equation it can be concluded, that the specific energy
consumption depends on the new flow rate Q1 , but it also depends on the systems
head Hst .
498
Estimation of the energy consumption in regulating of the flow rate of pump systems
2.3 Specific energy consumption when bypass method is used
Using this method of regulation, the work point starts to move right on the graph
towards the outcome work point whose indicators when the control device in the sluice
pipe is fully closed are: flow rate Q and head H . The systems flow rate Q1 which
is given to the user can be determined by the pipe systems characteristic:
The pumps head H1 with the new flow rate Q1 can be estimated by equation
(11) and the pumps flow rate QP by the equation (5):
1 2
b 4 c a H1 b .
2c
With the estimated flow rate given by the previous equation the pumps
coefficient of efficiency can be given:
Kp1 d eQp f Qp2
(20)
(19)
Qp
The specific energy consumption in this method of regulation can be determined

indicating the fact that the flow rate Q1 which is given to the user is just a part of the
total pumps flow rate QP . Thus (2) the following conclusion can be given:
(21)
e V,b
P
Q1
U g Qp H1
3600 1000 Kp1 Q1
g H1 Qp
.
3600 Kp1 Q1
Substituting of (11) in (21) can be expressed by the following equation:

2
H
g
H
Hst Q1 Qp
st
e V1,b
1

(22)
, kWh / m3
3600 d e Qp f Qp2 H
H Q Q1
When using the bypass or the frequency method to regulate the flow rate the
specific energy consumption depends on the systems head Hst . Thus (22) it can be
concluded that the ratio of the actual used pumps flow rate to the total produced
pumps frlow rate has a significant impact on the energy consumption. When the value
of this ratio increases the energy used to transport a unite volume of fluid also
increases proportional to it.
3. CONCLUSION
The methods presented in this work ensure a relatively quick and easy way to
estimate the specific energy consumption of pump systems operating with turbo
pumps. For this purpose it is necessary to know the equations of a pump's head
chasracterictic and the characteristic of efficiency. To analyse how effective a particular
method used to regulate the flow rate will be, it is necessary for its specific enrgy
consumption to be estimated by some of the equations (7), (8), (18) and (22),
depending on which method is chosed exactly. For a given flow rate the specific
method to regulate the flow rate that procuces the lowest energy consumption is the
most effectiv method in terms of energy efficiency.
Special thanks to: EN-The study was supported by contract BG051PO0013.3.04/28, "Support forthe ScientificStaff Development in the Field of Engineering,
Research andInnovation. The project is funded with support from the Operational
499
Programme "Human Resources Development" 2007-2013, financed by theEuropean

Social Fund of the European Union.
LITERATURE
Journal Papers:
[1]. Klimentov, Kl., G. Popov, Kr. Tujarov, (2008). Eqution of parameters of rotary
pums, Energetika 6-7, Sofia, p. 60-63.
Proceedings Papers:
[2]. Krasteva A., (2004) Eenrgy efficiency of water supplying systems.// Scientific. work
of the University of Rousse , 41, 3.1, p. 1519
[3]. Ormandzhiev K., S. Yordanov, P. Ivanov, (2003) Experimental Study of Dynamical
Processes in Pressure Turbo-Sysem in Real Time, 3rd International Conference
Research and Development in Mechanical Industry RaDMI 2003, Herceg Novi,
Serbia and Montenegro, September 1923, p. 1861-1865
[4]. Ormandzhiev K. (2004), Study on Dynamic Processes in Automatic System for
Controlling the Cross-Flow Water Turbine, International Scientific Conference
Uniteh04, Gabrovo, November 1819, vol. II, p. 521526.
500
NUMERICAL INVESTIGATION OF HEAT TRANSFER

ENHANCEMENT IN NATURAL CONVECTION AND FORCE
CONVECTION IN A FLUID SATURATED VARIABLE POROSITY
MEDIUM
Majid Soleimaninia1
Summary: The present paper investigates the numerical simulation of steady laminar
incompressible natural convection and force convection heat transfer in an enclosed
cavity and channel that is filled with a fluid-saturated porous medium. In this study, the
governing equation for the flow in the cavity and channel filled with porous media is
developed, and the appropriate formulation which includes both single phase flow and
the flow in porous medium field is derived. SIMPLE algorithm is applied to solve this
governing equation which QUICK scheme is utilized to descritize the convection term.
To have a better solution especially near the boundary layer region, clustered grid is
used. The code is validated through the comparison with valid references for two test
cases.
Key words: Heat transfer, Porous media, Force convection, Natural convection
1. INTRODUCTION
Flows through porous media occur in a wide variety of applications such as
geothermal energy modeling, thermal insulation material, cooling of electronic devices
and solar receivers to name a few. In many applications, the Darcys law is
inapplicable where the fluids flow in porous media bounded by an impermeable
boundary. For these cases, the study of convective heat transfer should include inertia
and boundary effects. A number of recent studies incorporated these effects by using
the general flow model known as the BrinkmanForschheimer-extended Darcy model.
For example, Kaviany [1] used the Brinkman- extended Darcy model to obtain a
numerical solution of laminar flow in a porous channel bounded by isothermal parallel
plates. Vafai and Kim [2], using this model, studied forced convection for thermally fully
developed flow between flat plates while Amiri and Vafai [3] reported a numerical study
on the thermally developing condition. Also, for flow in parallel plate channels, Lee and
Vafai [4] used a two-equation model to study the effect of local thermal non-equilibrium
condition. Alazmi and Vafai [5] extended the earlier work by investigating different
1
M.Sc, Majid Soleimaninia, Famagusta, North Cyprus, Eastern Meditteranean University,

(majid.soleimaninia@emu.edu.tr)
501
Majid Soleimaninia
NOMENCLATURE
C
Specific heat, (J/kg.K)
cF
Dimensionless form-drag coefficient

Darcy number
Gravitational acceleration (m/s2)
Da
g
Gr
Grashof number
Convection heat transfer coefficient
2
(W/m C)
h
k
Thermal conductivity, (W/m.K)
Permeability ( m2 )
Nu
Nusselt number
Pressure (Pa)
Pe
Peclet number
Prandtl number
Time (s)
Temperature (K)
Velocity (m/s)
V
Greek Symbols
D
Thermal diffusivity
Porosity
Viscosity coefficient
T
U
\
*
Nondimensional temperature
Density
Incidence angle
Diffusion coefficient
qcc
Heat production W/ m
Subscripts
Fluid
f
Ra
Re
Rayleigh number
Reynolds number
m
s
Pr
Effective
Solid
porous media transport models. An extensive study of flow in porous media is available
in ([8], [9], [10]). Angirasa [6] discusses the history of development of transport
equations in porous media and finite difference simulations. Nield et al. [7] presented
the effect of local thermal non-equilibrium on thermally developing forced convection in
a porous medium. This paper is organized in five sections. In Section 2, mathematical
formulation based FVM is described in some detail. In Section 3, the numerical method
is described in detail. In Section 4, results are presented. Concluding remarks are
made in Section 5.
2. Mathematical formulation
The Governing equations for laminar incompressible flow are summarized as
follow,
I
wUf &
&
. Uf v 0
wt
&
& & &
wv
Uf I1 I2 v. v
wt
Uc m
502
Continuity
equation
&
& && P &
& &
P Pv v c F K 1 2Uf v v
K
&
&
wT
&&
Uc p v.T . k m T qccm
f
wt
Momentum
equation
Energy
equation
(1)
(2)
(3)
Numerical Investigation of Heat Transfer Enhancement in Natural Convection and Force

Convection in a Fluid Saturated Variable Porosity Medium
Where;
Uc m 1 I Uc s I Ucp f
(4)
km
(5)
qccm
1 I k s Ik f
1 I qccs Iqccf
(6)
2.1 Nondimensionalizing
These equations can be nondimensionalized in two ways. The first sets of
equations are appropriate for natural convection problem, whilst the other one is used
in forced convection problem.
To nondimensionalized the equations for natural convection problem, the
following variables are used:
X
x
,Y
L ref
km
U c P f
uL ref
,V
D
y
,U
L ref
vL ref
,P
D
pL2ref
,T
Uf D 2
T TC
, Ra
'T
gE'TL3ref
, Pr
XD
Q
, Da
D
K
,
L2ref
(7)
Where 'T is the temperature difference between the cold wall and hot wall,
TC is the cold wall temperature, and L ref is the reference length which is the height of
the cavity.
The nondimensionl form of the governing equations for steady, two
dimensional laminar flow without any source of heat, becomes:
wU wV

wX wY
Continuity equation
1 wUU wUV

2
wY
I wX
U2 V2
wP Pr w 2 U w 2 U Pr U
2

cF U
2
Da
wX I wX wY Da
1 wUV wVV
wP Pr w 2 V w 2 V Pr V
U2 V2

2 2
cF V
Ra Pr T

2
wY
wY I wX wY Da
Da
I wX
2
2
wUT wVT w T w T

2 2
wY wX wY
wX
(8)
X-Momentum
equation
(9)
Y-Momentum
equation
(10)
Energy equation
(11)
We can generalize these set of equations to use them both for single phase
flow and flow field in porous medium by introducing a factor F which is equal to 0 and
set I 1 , when single phase flow is considered and set F equals to 1 and applying
appropriate I when flow in porous medium is considered. Here is the form of
equations for both cases.
wU wV
Continuity
(12)

0
equation
wX wY
1 wUU wUV

2
wY
I wX
1 wUV wVV

2
wY
I wX
Pr V
wP Pr w 2 V w 2 V
U2 V2

2
F
cF V
Ra Pr T
2
Da
Da
wY I wX wY
Pr U
wP Pr w 2 U w 2 U
U2 V2
2
F
cF U
2
wX I wX
wY
Da
Da
XMomentum
equation
YMomentum
equation
(13)
(14)
503
Majid Soleimaninia
2
2
wUT wVT w T w T
Energy

(15)
2 2
equation
X
Y
X
Y
w
w
w
w
On the other hand, to nondimensionalized the equation for forced convection,

the following terms are used:
X
Da
x
,Y
L ref
y
,U
L ref
K
,D
L2ref
u
,V
U ref
km
, Re
UcP f
v
,P
U ref
p
,T
2
Uf U ref
T TC
, Ra
'T
gE'TL3ref
, Pr
XD
Uf U ref L ref
Pf
Q
D
(16)
Where 'T is the temperature difference between the wall and inlet
temperature, TC is the cold or inlet temperature, U ref is the reference velocity which is
the inlet velocity, L ref is the reference length which is the width of the channel.
By applying these set of variables, the governing equations have the following
form:
wU wV

wX wY
Continuity
equation
1 wUU wUV

2
wY
I wX
1 wUV wVV

2
wY
I wX
wUT wVT

wY
wX
1 w2U w2U
U
U2 V2
wP

cF U
2
2
wX I Re wX wY Re Da
Da
wP
1 w 2V w 2V
V
U2 V2
RaT

cFV

2
2
wY I Re wX
wY Re Da
Da
Pr Re 2
1 w 2T w 2T

Re Pr wX 2 wY 2
XMomentum
equation
YMomentum
equation
Energy
equation
(17)
(18)
(19)
(20)
Again by introducing factor F these equations can be used both for single
phase flow and flow in porous media.
wU wV

wX wY
1 wUU wUV

2
wY
I wX
U
1 w2U w2U
U2 V2
wP

F
cF U
2
2
Re Da
Da
wX I Re wX wY
1 wUV wVV

2
wY
I wX
1 w 2V w 2V V
U2 V2
wP

cFV
2
2
Da
wY I Re wX
wY Re Da
Continuity
equation
(21)
X-Momentum
equation
(22)
RaT

Pr Re 2 Y-Momentum
equation
(23)
wUT wVT

wY
wX
1 w 2T w 2T

Re Pr wX 2 wY 2
Energy
equation
(24)
To omit the effect of Forchheimer term c F should be set to zero, else the
relation for c F is:
cF
Re
504
1.75
(25)
150I3 2
By comparing these two sets of equations, one can realize that by setting
1 Pr in the second form, the first set of equations can be obtained.

3. Numerical Modeling
The solution of these two sets of equations (12-15 & 21-24) presents us two
problems:
The convection terms of the momentum equations contain non-linear
quantities.
The first three equations in each set of equations (continuity, x-momentum and
y-momentum equations) are intricately coupled because every velocity
component appears in each momentum equation and the continuity equation.
The most complex issue to resolve is the role played by the pressure. It
appears in both momentum equations, but there is evidently no (transport or
other) equation for pressure.
If the pressure gradient is known, the process of obtaining discretised
equations for velocities from the momentum equations is simple. In general purposes
in flow computations we also wish to calculate the pressure field as part of the solution.
So its gradient is not normally known beforehand. For the incompressible flow,
coupling between pressure and velocity introduces a constraint on the solution of the
flow field: if the correct pressure field is applied in the momentum equations the
resulting velocity field should satisfy the continuity equation.
Both the problems associated with the non-linearities in the equation set and
the pressure-velocity linkage can be resolved by adopting an iterative solution strategy
such as the SIMPLE algorithm. In this algorithm the convective fluxes per unit mass F
through cell faces are evaluated from so-called guessed velocity components.
Furthermore, a guessed pressure field is used to solve the momentum equations and a
pressure correction equation, deduced from the continuity equation, is solved to obtain
a pressure correction field which is in turn used to update the velocity and pressure
fields. To start the iteration process we use the initial guesses for the velocity and
pressure fields. As the algorithm proceeds our aim must be progressively improving
these guessed fields. The process is iterated until convergence of the velocity and
pressure fields.
4. Result and discussion
To validate the code, two test cases are solved and results are compared with
validated references.
Natural convection in a cavity saturated with porous media
Forced convection in a duct saturated with porous media
4.1. Natural convection in a cavity saturated with porous media
The geometry of the problem and the boundary condition which is used is
shown below:
505
Majid Soleimaninia
Insulated
Porous media
inside
T 1
T 0
Insulated
Fig 1: Problem geometry and the boundary conditions for natural convection in cavity
saturated with porous medium
The mesh that is used for this problem is presented in Fig 2.
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.5
Fig 2: Grid system for natural convection in cavity saturated with porous medium
The goal of this problem is to find the average Nusselt number for different
Rayleigh number, Darcy number, and porosity and study the effect of these
parameters on the average Nusselt number. To compute average Nusselt number
based on the numerical results for this problem, the appropriate equation should be
derived. The definition of convection heat transfer coefficient for flow in the cavity is as
follow:
h
wT
k fluid
wx x
'T
(26)
It is also common practice to nondimensionalize the heat transfer coefficient h

with the Nusselt number, defined as:
Nu
hL c
k
(27)
Where k is the thermal conductivity of the fluid and L c is the characteristic

length. Now by combining these two equations, the final relationship between Nu and
the nondimensional temperature can be find as follow:
506

wT
k fluid
wx x
'T
hL c
k fluid
Nu
wT
k fluid 'T
wX x
L c 'T
wT
wX x
k fluid wT
Lc wX x
(28)
So the value of Nusselt number at each point along Y axis and the average
Nusselt number at the hot wall can be found based on the following formula:
N
T1, j T2, j
Nu j
X1, j X 2, j
Nu
, Nu
(29)
j 1
N
where N is the number of division in Y direction. Nusselt number at the hot
surface for all the conditions is presented in Fig 3 and Fig 4:
Da = 0.01
Pr = 1.0
Da = 1E-6
Pr = 1.0
12
7
Phi =
Phi =
Phi =
Phi =
Phi =
Phi =
Phi =
Phi =
Phi =
11
6
10
Phi = 0.4, Ra =
Phi = 0.6, Ra =
Phi = 0.9, Ra =
Phi = 0.4, Ra =
Phi = 0.6, Ra =
Phi = 0.9, Ra =
NU
8
7
1E7
1E7
1E7
1E8
1E8
1E8
NU
6
5
0.4, Ra =
0.6, Ra =
0.9, Ra =
0.4, Ra =
0.6, Ra =
0.9, Ra =
0.4, Ra =
0.6, Ra =
0.9, Ra =
1E3
1E3
1E3
1E4
1E4
1E4
1E5
1E5
1E5
4
2
3
2
1
0
0.25
0.5
0.75
Fig 3: Nusselt number at the hot wall for

flow in cavity saturated with porous
medium, Da = 10 6 and Pr = 1.0
0.25
0.5
0.75
Fig 4: Nusselt number at the hot wall for

flow in cavity saturated with porous
medium, Da = 10 2 and Pr = 1.0
The value of average Nusselt number for these cases which is obtained in this
study and those presented in [8] is presented in
Table 1.
Moreover, the following trends are seen from the obtaining data:
For low Darcy-Rayleigh numbers ( Ra u Da ) the average Nusselt number is
almost independent of the individual values of Ra, Da, and I .
For a given Rayleigh number and porosity, the average Nusselt number
increases with Darcy number, due to the higher permeability of the medium which
results in larger flow velocity.
For a given Rayleigh number and Darcy number, the average Nusselt number
is seen to be increased with porosity of the medium. The reason for this can be
understood from the form of momentum equation. It is clear that for higher porosity,
the inertial and non-linear drag terms are less significant, leading to higher
dimensionless flow velocities and higher values of average Nusselt number.
507
Majid Soleimaninia
Table 1 Comparison of average Nusselt number between present study and those
presented in [8].
Ref. [8]
Darcy
Rayleigh
Present study
number
number
I = 0.4
I =0.6
I =0.9
I = 0.4
I =0.6
I =0.9
10 6
107
1.0816
1.0818
1.082
1.079
1.079
1.08
6
10
3.1892
3.2071
3.22
2.97
2.997
3.0
102
10
1.0076
1.012
1.017
1.01
1.015
1.023
102
10 4
1.3574
1.488
1.628
1.408
1.53
1.64
102
2.988
3.439
3.926
2.983
3.555
3.91
10
10
As it is shown in
Table 1, it is observed that calculated results are in good agreement with
benchmark results.
4.2 Forced convection in a duct saturated with porous media
In this section, the forced convection heat transfer in a duct packed with a
porous material is considered. Geometry of the problem and boundary conditions are
specified in Fig. 5.
U 0, V 0, Ts 1
in
V
in
in
Saturated with Porous media
U 0, V 0, T 1
wU wX
wV wX
wT wX
Fig. 5 Problem geometry and the boundary conditions for forced convection in a duct
packed with porous material
Grids are clustered in Y direction near both ends, but clustered just near the
inlet, in X direction. Sample grids that is used, is presented in Fig. 6:
508

1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Fig. 6 System of grid utilized for the forced convection problem

In this section forced convection through a duct is studied. Likewise previous
sections, the most important parameter for this study is Nusselt number. To derive the
appropriate equation for Nusselt number, we start with the definition of convection heat
transfer coefficient.
h
wT
k
wy y
Ts Tm
(30)
Where Ts , Tm , and H are the surface temperature, mean temperature (or bulk
temperature), and height of the channel, respectively. The mean temperature for
internal flow can be calculated from the following relation:
y
Tm
uTdy
udy
0
(31)
This equation should be nondimensionalized.
UU T'T T d YH T
UU d YH
1
Tm
Tm 'T Tin
ref
Tm 'T Tin
0
1
UdY
'T Tin
in
in
Tin Tm
Nu
ref
'T UTdY
U T'T T dY UT'TdY T UdY

UdY
UdY
UdY
1
in
UTdY
UdY
(32)
hH
k
(33)
Now by combining these two equations, the final relationship between Nu and
the nondimensional temperature can be find as follow:
h
Nu
wT
k
wy y H
Ts Tm
hH
k
wT
k
wy y H
Ts Tin Tin Tm
1 wT
1 Tm wY y
wT
k
wy y H
, Ts
Ts 'T Tm 'T
1 h
k'T wT
H wY y 1
'T 1 Tm
k wT
H wY y
1 Tm
(34)
The governing equations are those presented in Eqs. (17), (18), (19), and (20).
For this problem as it is discussed earlier, the Forchheimier term is neglected, because
the effect of this term is very small, so c F is set to zero for this problem. The Prandtl
509
Majid Soleimaninia
number that is used for this problem is 1.05252 u103 which is the effective Prandtl
number when foam metal is applied. I set to 0.92, and Darcy number is 4.252 u 10 5 .
The program is run for different Peclet number. Since the Nusselt number at the region
of fully developed flow is desired to compare with those presented in [8], aspect ratio of
the duct should be chosen so that this goal can be achieved. Based on the
approximate relation which is presented in [8], aspect ratio is set equal to Peclet
number for all cases.In the following figures, isothermal contours for different Peclet
number is presented.
1.8
0.9
0.95
0.6
5
5
0.4 0.3
0.3
0.8
1.4
0.5
5
0.4
0.7
0.25
0.8
0.2
0.4
0.5
0.4
0.25
0.4
0.25
0.5
0.75
0.7
0.5
0.6
5
0.9
0.7
0.3
500..32
0.2
0.5
0.95
0.5
0.7
0.6
0.2
0.95
0.9
0.8
0.65
0.45
0.35
0.3
0.25
0.1
0.05
0.15
0.4 00.3
5 .35
0.4
0.9
0.85
0.8
.75
0.9
0.7
5
0.5
0.7
0.6
5
0.8
0.85
6
0 75
0.9
5
0.3
0.1
0
0.6
0.8
0.15
0.1
0.2
0.05
0.3
0.6
0.8
0.3
0.2
0.4
0.95
0.9
0.8
5
0.75
0.50.6
55
0.65
0.15
0.75
0.5
0.5
0.8
5
0.8
0.6
0.05
0.6
1.2
0.1
0.7
0.7
0.9
0.85
0.45
0.8
1.6
0.95
0.5
5
0.2 0.1
0.7
0.6
1 0.85
0.9
0.9
0.85
0.95
0.25
0.5
0.75
Fig. 7 Isothermal contours, Pe = 1
1.25
1.5
1.75
1.5
1.5
0.05
0.9
0.5
0.1
0.95
0.2
0.35
0.45
0.2
0.65
0.85
0.9
5
0.3
0.5
0.05
0.10
.3
0.8
0.8
0.75
0.5
0.9
0.6
0.6
5
05 .7
0.695
0
0.8
0.7
0.4
0.7
5 0.0.25
2
0.95
0.3
5
0.5
0.45
0.4
0.9
0.7
5
0.5
5
0.6
5
0.15
0.5
0.95
0.9
0.
0.6 8
0 050.15
0.7
5
00.31
0.85
0.7
10

2
1.5
1.5
0.7
0.8
5
0.4
0.25
0.15
0.65
0.5
5
0.05
0.1
0.3
0.45
0.5
0.9
0.75
0.5
0.85
0.2
0.25
0.3
0.4
0.9
0.8
5
0.3
0.9
0.8 0.75
0.15
0.5
0.5
0.9
0.5
0.6
0.6
05.6 0
.7
0.2
.45
0 350 0.10.05
5
0.9
10
510
15
20
0.2
5
0.6
0.7
0.9
0.800.40.3
.65 5
0.8
0.7
0.85
0.9
0.9
10
15
20
25
30
35
40
45
50

As it is expected, since the Prandtl number is very small, so the thermal

boundary layer is developed at near the inlet.
The variation of Nusselt number at the hot top wall is presented in Fig. 13. The
value of the Nusselt number at the region of fully developed flow is 4.85 which is very
close to 4.93 which is presented in [8] resulting from analytical solution of reduced
governing equations.
PR = 1.05252E-3
EPS = 0.92
DA = 4.252E-5
100
PE
PE
PE
PE
PE
PE
90
80
=1
=2
=5
= 10
= 20
= 50
Nu
70
60
50
40
30
20
10
NU = 4.85
10-2
10-1
100
X / PE
Fig. 13 Variation of Nusselt number with respect to Peclet number for forced
convection in a duct packed with porous media
4.3. Grid sensitivity analysis
One of the most important things in CFD problem is grid sensitivity analysis. In
this section it is shown that the number of grids that is selected for all the problem
(18044) is appropriate, since increasing the number of grid does not change the
results significantly, while decreasing the number of grid has some influences on the
results. To analysis grid there are two other grid numbers which are selected; one
14032 and the other one is 25064. Nusselt number for Pe = 2 is selected as a main
target to analyze the grid sensitivity.
Here is the Nusselt number variation with respect to number of grid.
100
90
80
140*32
250*64
180*44
70
NU
60
50
40
30
20
10
10-2
10-1
100
X / Pe
Fig. 14 Grid sensitivity analysis

It is cleared that the results are not dependent on the grid.
511
Majid Soleimaninia
CONCLUSION
The present paper investigates the numerical simulation of steady
laminar incompressible natural convection and force convection heat transfer
in an enclosed cavity and channel that is filled with a fluid-saturated porous
medium. In case of Natural convection in Cavity it is concluded that:
For low Darcy-Rayleigh numbers ( Ra u Da ) the average Nusselt number is
almost independent of the individual values of Ra, Da, and I.
For a given Rayleigh number and porosity, the average Nusselt number
increases with Darcy number, due to the higher permeability of the medium which
results in larger flow velocity.
For a given Rayleigh number and Darcy number, the average Nusselt number
is seen to be increased with porosity of the medium. The reason for this can be
understood from the form of momentum equation. It is clear that for higher porosity, the
inertial and non-linear drag terms are less significant, leading to higher dimensionless
flow velocities and higher values of average Nusselt number.
In case of force convection in a channel it is concluded that:
The value of the developed Nusselt number for the channel saturated with
porous media is independent from the Peclet number and has the value of 4.85 for all
the cases.
Forchheimer term has very little effect on the solution. This term can be omitted
for most industrial application to reduce the time of processing, however when the
accuracy of the solution is very important, the effect of this term cant be ignored.
To have the effect of natural convection, it is better to have high Rayleigh
number, and the only parameter in hand for this goal is the width of the channel. By
increasing the width of the channel, the value of Rayleigh number is increased, so the
natural convection part has some contribution on Nusselt number and makes it greater.
LITERATURE
Journal Papers:
[1]
[2]
[3]
[4]
[5]
[6]
[7]
512
M. Kaviany, (1985), Laminar flow through a porous channel bounded by

isothermal parallel plates, Int. J. Heat Mass Transfer 28 (4) 851858.
K. Vafai, S. Kim, (1989), Forced convection in a channel filled with porous
medium: an exact solution, ASME J. Heat Transfer 111 (4) 11031106.
A. Amiri, K. Vafai, (1994), Analysis of dispersion effects and non- thermal
equilibrium, non-Darcian, variable porosity incompressible flow through
porous media, Int. J. Heat Mass Transfer 37 (6) 939954
D.-Y. Lee, K. Vafai, (1999), Analytical characterization and conceptual
assessment of solid and fluid temperature differentials in porous media,
Int. J. Heat Mass Transfer 42 (3) 423435
B. Alazmi, K. Vafai, (2000), Analysis of variants within the porous media
transport models, ASME J. Heat Transfer 122 (2) 303326.
D. Angirasa, (2002), Forced convective heat transfer in metallic fibrous
materials, ASME J. Heat Transfer 124 (4) 739745.
D.A. Nield, A.V. Kuznetsov, M. Xiong, (2002), Effect of local thermal nonequilibrium on thermally developing forced convection in a porous medium,
Int. J. Heat Mass Transfer 45 (25) 49494955.

Books:
[8] D.A. Nield, A. Bejan, (1999), Convection in Porous Media, second ed.,
Springer-Verlag, New York.
[9] K. Kaviany, (1991), Principles of Heat Transfer in Porous Media, SpringerVerlag, New York.
[10] K. Vafai (Ed.), (2000), Handbook of Porous Media, Marcel Dekker, NewYork.
513
OBJECTIVE FUNCTION DEFINITION FOR PRIMARY HVAC SYSTEM TOTAL

COSTS MINIMIZATION
Milica Grahovac1
Summary: One of the major advantages of the computer simulated HVAC system
models is the convenience to optimize them using an appropriate optimization
algorithm. The purpose of the model is to help the architects in the conceptual design
phase. The objective is to minimize the annuity of the total costs. An annuity cost
calculation method for thermal systems is presented. The constraints are replaced by
penalty functions, which penalize the objective in case the constraints have been
breached.
As an example, Global bounded Nelder-Mead algorithm and discrete runs have been
used to find optimal dimensions of the energy-balance based model and the results are
presented. The model consisting of a biomass on-off boiler, thermal storage tank and
solar collectors is coupled to a whole year hourly system simulation. The model needs
to cover a given hourly load and insure a solar ratio of 30%.
Key words: Optimization, Objective, Penalty, Costs, Annuity, Model
1. INTRODUCTION
Optimal design of a thermal system can be based on several objectives, such
as minimal costs, minimal energy consumption and emissions, maximal utilisation of
renewables. The economical aspect is the most important decision-making factor in the
building sector [1]. We need to compare the prices for alternatives in advance. Thus,
the modelling and simulation in thermal system design are gaining importance. It is
possible to optimize the system while evaluating its performance in an hourly resolution
through the period of one typical year.
Until all the details about the plant design have been defined, we can only
make cost estimations. As long as all the estimation assumptions are well explained,
we are able to judge the quality of the cost calculation result, or even improve it. The
required precision of the estimation depends on the purpose of the result. The model
used in this paper for illustrating the optimization function definition is to be used by
architects during the conceptual planning phase. Thus, it is a simplified energy balance
based model.
After showing the basic mathematical formulation of an optimization problem,
the cost calculation is presented in detail, followed by further objective function
1
Dipl.-Ing. Milica Grahovac, Munich, Technische Universitt Mnchen, mgrahovac@tum.de
515
Milica Grahovac
definition utilizing the penalty function. Finally, the optimization results for an example
system model are shown. The formulation of the model itself is not an objective of this
paper and will be published separately.
2. OPTIMIZATION IN THERMAL ENGINEERING
The general formulation of an optimization problem is:
(1)
minn f x
xR
such that
gi x d 0;
h j x 0;
i 1,..., p
j 1,..., q
where x is a vector of optimization variables, f(x) the objective function, gi(x) and hj(x)
are inequality and equality constraints, respectively.
The characteristics of the objective and the constraints determine the
optimization algorithm suitable for solving the problem. There is an increasing trend in
using computer modelling and simulation to describe and analyse the performance of
thermal systems. In that case, the objective and constraints depend on the results of
the simulation. These functions are often neither linear nor differentiable. Additionally,
optimization problems in thermal engineering are almost always constrained. In spite of
that, algorithms for unconstrained problems can be applied if the constrained problem
is reformulated into an unconstrained. This is carried out by adding a penalty function
to the solutions that violate any of the constraints, which relocates the optimum to the
feasible region. The unconstrained problem utilising the penalty function equivalent to
the formulation in (1) would be:
minn I x where I x f x P i G g i x P j H h j x
(2)
xR
Pi and Pj are the penalty coefficient vectors, G and H are the functions of the
constraints. To find out more about the definition of penalty functions, see [2]
Optimization algorithms should be robust, efficient and accurate. One of the
algorithms that perform well if applied to a black box function, which is how the
simulation defined objectives are sometimes called, is the Global Bounded NelderMead algorithm [3]. It searches for the global minimum of the non-linear discontinuous
function. Evolutionary algorithms also perform well applied to such problems.
Apart from the algorithm efficiency, the simulation runtime can significantly
influence the optimization speed, since the simulation is performed at the each iteration
step of the optimization algorithm. Imagine dimensioning a primary HVAC plant.
Components are made only in several different standardized sizes, meaning the
optimization parameters can only take values from the predefined finite set of
elements. Discrete optimization, also referred to as integer programming, is an
expanding area dealing with this kind of problems. Nevertheless, if the number of
elements in the Cartesian product of sets of optimization parameters (discrete values)
is not too large and the simulation runtime short enough, the definitive descrete global
minimum can be obtained by evaluating the objective function in all the points.
3. OPTIMAL DIMENSIONS OF A HEATING SUPPLY SYSTEM
In the example presented in this paper, the objective is to minimize the costs.
Yearly simulation of the model, consisting of a biomass boiler, thermal storage and
516
Objective Function Deffinition for Primary HVAC System Total Cost Minimization
solar collectors, is performed in hourly resolution. The goal is to identify the component
dimension combination that yields the minimal annuity cost, while fulfilling the supply
task and insuring the minimal yearly solar ratio. The objective function consists of a
cost function, which is a result of the cost calculation, and two penalty functions.
f x ACt x P1 x P2 x
(3)
4. COST CALCULATION
The basic purpose here is to compare alternative plant designs, as in many
other engineering applications. Hence, the calculation method must provide
equivalence between costs calculated for different plants. For comparing alternatives,
DIN EN 15459 [4] provides two methods of cost calculation: global cost, where all the
costs are brought to a present value, and annuity cost, where all the costs are
represented by a uniform annuity value.
An annuity cost calculation method based on DIN EN 15459 [4] and VDI 2067
[5] has been implemented. The goal is to obtain a yearly value, which accounts for all
the costs we consider, allowing us to compare alternative designs over the same
calculation period.
Fundamentally, total costs consist of investment costs and annual costs.
Total investment costs consist of initial investment costs, installation costs and
replacement costs.
Time Value of Money
The same amount of money does not have todays value at a certain point in
time. The interest rates that rule the dynamics of the cost are inflation rate Ri, market
interest rate R, and price development rate Rp, which takes the component price
evolution into account. Inflation rate and market interest rate define the real interest
rate RR:
(4)
RR
R Ri
1 Ri
Based on the point in time of an amount we can project it to any moment

before or after using the following factors. The present value of a payment due in the
future is obtained by multiplying the payment amount by the present value factor:
P 1 RR
(5)
T represents the number of years in the calculation period. Its inverse is the future
value, F
T
F 1 RR
(6)
which is used in calculating the value of an amount in the future.
Annuity factor is used to calculate equal annual portions (annuities) to be paid
over a given period of time:
T
(7)
a( T )
RR
T
1 1 RR
It is the inverse of the sums of the finite power series, where each addend
517
Milica Grahovac
represents the future value for one of the yearly amounts.

Annuity of the total investment costs
Initial investment, Ci, is the price we pay for the component itself at the moment
of purchase. Installation costs can be easily assessed using the installation factor, finst,
provided in [3], which represent the percentage of initial investment needed for the
installation. Now we defined the costs at the beginning of the calculation period:
C0 Ci f inst Ci
(8)
If the calculation period goes beyond the component life duration, the
component will be replaced. The component price at that time will differ from the initial.
This change is embodied in Rp. The new price is then brought to the present moment
using RR. Replacement costs are the sum of all the necessary component
replacements during the calculation period:
T
Tlife
(9)
Cr
C 1 R 1 R
Tlife j
Tlife j
j 1
year
2T
l i fe
where Tlife is the component life duration and T is the calculation period.
ar
ye
SR
1
RR
Ci
FR
1
year
lif
1

ar
ye
1
V
F
F
Fig. 1 Present value of the final value and replacement costs. Example: two
replacements, the life duration of the second one exceeds the calculation time limit.
Total investment cost: Ci(1+finst)+Cr1+Cr2-V.
If the end of the calculation period is not coincident with the end of the last
installed component life duration, that component still has a certain value at the end of
the calculation period. This is a final value of the component and it needs to be
subtracted from the previously calculated costs in order to assess the actual
investment costs occurring during the calculation period. Illustrated at, final value is
determined using linear depreciation of the price for the last replacement, taking the
price development into account. The value obtained at the end of the calculation period
is then discounted to yield its present value. This represents the final value.
518
(10)
Ci 1 R p life Tlife
T
Tlife T
Tlife
1 RR T
Now we have all the prerequisites to calculate the annuity of the total
investment costs [5]:
ACi a( T )C0 C r V
(11)
Annuity of the running costs
Running costs include energy, Ce, and maintenance costs, Cm. Since we
cannot expect fuel and electrical energy prices to stay constant over the calculation
period, dynamic present value factor is introduced:
1 R px
R Ri
1
1 R 1 Ri
T
R R px
1 Ri
1
1 R px
1 R
T
(12)
Ex
Rpx is a price development rate for a given good.

The average yearly running costs are:
ACr Ce E e C m E m
(13)
Annuity of the total costs
The total costs due in yearly intervals, ACt, are:
(14)
ACt
ACi ACr
The annual cost defined in [4] is thus the average yearly cost from the present
value perspective.
The annual cost defined in this way is the average yearly cost in the year 0 [4].
Fig. 2 Schematic representation of cost classification and calculation.

Assumptions adopted
Interest rate and other dynamical cost calculation parameters are considered to
stay constant during the calculation period. Risks and uncertainties of the future costs
have been neglected. Since the models are focused on components for energy
production, conversion and storage, only these components are taken into account
while calculating investment, installation and maintenance costs. We assume the costs
519
Milica Grahovac
for piping, valves and other armatures, pumps etc. are equal for all alternatives and
thus do not play the role in the comparisons. Taxes and insurances have also been
neglected. Nonetheless, the models can easily be extended to comprehend more
detailed cost categories. Again, this is only possible in the advanced planning phases,
where all the dimensions and parameters of the building are determined.
Cost data
Manufacturer data have been used to estimate the purchase cost. Costs for the
solar collector and thermal storage tank are linear, with 300 /m2 and 800 /m3
respectively. Logarithmic fit is utilized for biomass boilers:
C iBoiler 2244.6 ln Pd 3136.8 ,
(15)
where Pd is the boiler design power. Data for pellet boilers used. Pellet price
taken is 0.228 /kg. Market rate is 0.04, inflation rate 0.02. Component life durations,
installation and maintenance cost factors and price development factors are taken from
[5] and [4].
5. PENALTY FUNCTIONS
Constraints are replaced by the penalty functions.
The first one, called the supply penalty, penalizes the objective in case of the
insufficient covering of the building load. This penalty has a rather soft definition, since
the model allows the uncovered load from the previous hour to be covered within the
next hour. In case this load procrastination happens too often and we accumulate the
unsatisfied load, the penalty function becomes active and increases the value of the
objective, which will be illustrated in the following chapters. It is directly proportional to
the sum of uncovered loads. This sum is the result of iteration process of the
simulation.
P1 P1
unsatisfied Load
(16)
The second one ensures we achieve the desired yearly solar ratio by
penalizing the objective with:
1
P2 P 2
(17)
solar Ratio
Both penalty functions are nonsmooth, since their penalty parameters are 0
until the constraint has been violated.
6. RESULTS
A load profile for a building located in Moscow, Fig. 3, together with the
appropriate weather file has been imposed to the system. The goal of the optimization
is to calculate the optimization parameters that minimize the total annuity costs and still
perform satisfactory in terms of covering the load and achieving the solar ratio of 0.3.
The optimization parameters are the boiler design power, the storage volume, the
storage discharge rate, the roof area that is to be covered by solar collector panels and
the collector temperature. They have been discretized in steps of 5 kW, 0.5 m3, 20%,
5 m2, and 5C, respectively.
520
Heating Load, kW
50
40
30
20
10
0
0
1000
2000
3000
4000
5000
6000
7000
8000
hh
Fig. 3 Load profile for a building located in Moscow after peak load shaving.
Table 1 and Table 2 show the optimization results obtained utilizing a discrete
runs and continuous Global Bounded Nelder-Mead (GBNM) algorithm.
Table 1 Optimal parameters.
Discrete
GBNM
Boiler Power,
kW
Storage
3
Volume, m
Storage
Discharge
Rate, %/h
Roof area for

Solar
2
Collectors, m
Mean Collector
Fluid Temp.,
C
40
40
2.5
2.751
40
79.2
60
55.495
60
60
Table 2 Objective function, costs and characteristic values of the optimum.
Discrete
GBNM
Objective
Function
Invest.
Costs,
/a
Running
Costs, /a
Total
Costs,
/a
Supply
Penalty
Solar
Penalty
Supply
Task, %
Solar
Ratio
3235.5
3204.5
1457
1423.3
1778.5
1781.2
3235.5
3204.5
0
0
0
0
100
1
32.4
30
The optimization time for the discrete runs is 6006 s, for GBNM 4200 s.
To view the objective function in 3D, three parameters need to be set to
constant. In the following plots storage volume and discharge rate, as well as collector
temperature are set to their optimal values. The value of the objective function
increases as both boiler power and roof area decrease, Fig. 4 left. This is caused by
the penalty functions increase, Fig. 5. Total costs, naturally, grow with the component
size increase.
x 10
4
0
3400
3200
3000
2800
2600
50
40
Total Costs,
Objective Function
3600
35
30
Boiler Power, kW
25
20
100
Roof Area, m2
100
50
Roof Area, m2
20
25
30
35
40
Boiler Power, kW
Fig. 4 Objective function, left, and total costs, right, as a function of the boiler power
and the roof area, on which the solar collectors are mounted. Storage volume, its
discharge rate and collector temperature are constant values from Table 1. Objective
function has a minimum at 40 kW and 60 m2.
521
Milica Grahovac
x 10
4
4
Solar Penalty
Supply Penalty
x 10
2
1
100
0
20
80
60
25
40
30
20
35
Boiler Power, kW
40
Roof Area, m2
3
2
1
0
40
30
Boiler Power, kW
20
100
80
60
40
20
Roof Area, m
Fig. 5 Supply penalty, left, and solar penalty, right, as a function of the boiler power
and the roof area, on which the solar collectors are mounted. Storage volume, its
discharge rate and collector temperature are constant values from Table 1. As the
boiler design power decreases, it cannot fulfil the supply task and the penalty
increases. Similarly, as long as the roof area for collectors is less than 60 m2, the
penalty is active and forces the roof area to increase.
7. OUTLOOK AND CONCLUSION
In the case study the costs minimization problem has been approached by
summing the annuity of the total costs with the penalty function replacing the
constraints. Global Bounded Nelder-Mead algorithm identifies a global minimum and
similar optimal parameters are obtained by discretized runs, which is a confident but a
slow solution. As expected, the objective function grows with penalty violations,
excluding the cheeper solutions, that are not able to satisfy the constraints. Further
work includes optimizing different system models for building loads in several climates.
Modelling and simulation methods coupled with the optimization are on their
way to become a standard design procedure in the future. For that to happen, a faster
and more robust procedures (models, algorithms) need to be developed or identified.
LITERATURE
[1] Jaluria, Yogesh, Design and optimization of thermal systems, CRC Press, 2008
[2] Nocedal, J., Wright, S., Numerical Optimization, Springer, 2006
[3] Luersen, M. A., Le Riche, R. Globalized Nelder-Mead method for engineering
Optimization, In ICECT03: Proceedings of the third international conference on
Engineering computational technology, pages 165166, Edinburgh, 2004
[4] EN 15459:2007 Energieeffizienz von Gebuden - Wirtschaftlichkeitsberechnungen
fr Energieanlagen in Gebuden (Energy performance of buildings - Economic
evaluation procedure for energy systems in buildings)
[5] VDI 2067:2010 Blatt 1 Economical efficiency of building installations
Fundamentals and economic calculations (Wirtschaftlichkeit gebudetechnischer
Anlagen- Grundlagen und Kostenberechnung)
522
PARALLEL SOLUTIONS TO ACCELERATE MATHEMATICAL

ALGORITHMS IN HYDRODYNAMIC STABILITY PROBLEMS
Diana Alina Bistrian1, Manuela Pnoiu2, Tihomir Latinovi3, Marcel Topor4
Summary: This paper presents two different solutions to accelerate numerical
algorithms used in fluid stability problems by means of parallel computing. Recent
advances in computational technology using the computational power offered by
modern graphic processing units (GPUs) are assessed and tested throgouly. In
comparison with the GPU computing a parallel processing using distributed
computation on a cluster platform was involved for numerical investigation of the
stability of swirling flow in a hydropower Francis turbine. This two different solutions are
evaluated and compared in order to reveal the main features and particularities from
the point of view of a scientific computation user. The benefits and the drawbacks of
this new technology are pointed out.
Key words: accelerating algorithms, GPU computing, parallel computation
1.
INTRODUCTION
As computational fluid dynamics (CFD) matures so rise the expectations of

what it can or should deliver. Practitioners and designers are no longer content with
qualitative statements on trends, but judge the utility and value of CFD by its ability to
provide quantitatively accurate predictions for property fields and engineering
parameters derived therefrom. In the extreme, theoretical fluid dynamicists expect fullyresolved simulations of turbulence to provide fundamental information on turbulence
mechanics of greater accuracy and detail than can be derived from the most
sophisticated experimental techniques.
In computational fluid dynamics, several software packages solving either the
Euler or the Navier-Stokes equations around complex geometries have been
developed and are currently used by aircraft or engine manufacturers. The use of
unstructured meshes is one of the most efficient ways to solve complex problems,
1
Dr. Mat. Diana Alina Bistrian, Romania, Politehnica University of Timisoara, Faculty of Engineering of
Hunedoara, diana.bistrian@fih.upt.ro
2
Asistant Prof. Dr. Eng. Manuela Pnoiu, Romania, Politehnica University of Timisoara, Faculty of
Engineering of Hunedoara, manuela.panoiu@fih.upt.ro
3
Prof. Dr. Eng. Tihomir Latinovi, Republic of Srpska BIH, Banja Luka University, Faculty of Mechanical
Engineering, drtiho@gmail.com
4
Drd. Eng. Marcel Topor, Politehnica University of Timisoara, Faculty of Engineering of Hunedoara,
marcel.topor@fih.upt.ro
523
Diana Alina Bistrian, Manuela Pnoiu, Tihomir Latinovi, Marcel Topor
because they allow high flexibility in specifying a geometry. The possibility of using the
modern parallel machines even for unstructured mesh based codes implies the
necessity of dealing with two problems: the first one lies on the strategies and
algorithms which can be used for the partitioning of the grid and the mapping of the
subdomains among the processors; the second one is related to the structure of the
code, which has to be designed is such a way that the features of the modern parallel
machines are fully exploited.
The first efficient implementation of a 2D unstructured flow solver on MIMD
parallel computers was presented by Venkatakrishnan et al [1] in 1992: they
demonstrated that a good supercomputer performance can be reached and that a
careful implementation of the message passing routines is a critical point, even for an
explicit code. In 1995, Lanteri [2] developed a parallel version of an industrial code
based on a mixed finite element/finite volume method. The parallelization strategy
combines mesh partitioning and message passing programming model: basically, the
same old serial code is going to be executed within every subdomain. Modifications
occurred in the main loop in order to take into account the assembly phases of the
subdomain results.
The literature on these topics can be considered exhaustive for twodimensional applications and parallel machines of the old generation. The same
cannot be said for three-dimensional complex flows and for the new machines in terms
of optimization of the performances and assessment of algorithms and techniques (i.e.
[3-6]).
This paper presents an evaluation of the parallel processing for meshless
algorithm for solving Navier-Stokes problems in fluid dynamics. The fluid dynamics
existing solvers take use of finite element methods which are cumbersome and very
exhaustive in computational resources. Meshless algorithms uses a different
approaches using high level eigenvalue solvers. In order to evaluate the parallel
processing performance, we present recent advances in computational technology
using the computational power offered by modern graphic processing units (GPUs) in
the same time with the parallel and distributed investigation of the hydrodynamic
stabiliy problem using Matlab Distributed Computing Server on a Windows operating
system cluster.
2.
SOLUTIONS FOR USING GPUs IN SCIENTIFIC COMPUTING
Finding parallelism in a program has been a challenge for at least the past
twenty years of research in parallelism, and is still one today. But with the arrival of
GPU chips capable of executing many instructions in parallel, available to average
users, it become even more important to be able to utilize this power so that the endusers of computers can benefit from it and not only elite or research members.
Languages should evolve and low-level languages like C are not well suited for
programming complex parallel tasks. It is a necessity to give to programmers new tools
to effectively develop parallel program while reducing the need of a full parallelism
comprehension and good knowledge of the underlying architecture. Matlab [7] is a
high-level language with a powerful matrix-based system. It is often used for easy
prototyping in early development stages because it relieves the programmer of many
details like memory management and has an impressive library of built-in many matrix
524
Parallel Solutions to Accelerate Mathematical Algorithms in Hydrodynamic Stability Problems
functions. Hence, it is a useful tool for many simulation domains like fluid mechanics,
biology or mechanical engineering, especially for people whose first domain of
competence is not computer science and are programmers by necessity.
There already exist different ways to speed up the computation of Matlab
programs using GPUs, including several third-party toolboxes. Some options available,
all built on top of the Compute Unified Device Architecture (CUDA) developed and
provided by NVIDIA to access its GPUs will be presented in forthcoming sections.
2.1 GPUmat library
GPUmat is a freeware library, in an advanced state so memory management
on the GPU is automated, even if data movements still need to be explicit. This is done
through the addition of a new type of variables, GPUsingle, which makes use of GPU
memory transparent for users. Deletion of unused variables is taken care of by a
garbage collector so as not to clobber the GPU memory. This is important as data
processed by GPUs tend to be very big since computations are based on dataparallelism. The use of this new type allows Matlab functions to be overloaded so that
either the GPUmat or the original Matlab function is called depending on the type of the
arguments.
GPUmat enables Matlab code to run on the Graphical Processing Unit. The
following is a summary of GPUmat most important features:
GPU computational power can be easily accessed from Matlab without any
GPU knowledge.
Matlab code is directly executed on the GPU. The execution is transparent
to the user.
GPUmat speeds up Matlab functions by using the GPU multi-processor
architecture.
Existing Matlab code can be ported and executed on GPUs with few
modifications.
GPU resources are accessed using Matlab scripting language. The fast
code prying capability of the scripting language is combined with the fast
code execution on the GPU.
GPUmat can be used as a Source Development Kit to create new functions
and extend the library functionality.
GPU operations can be easily recorded into new functions using the
GPUmat compiler.
2.2 Jacket platform for gpu computing with matlab
Jacket is a software platform designed for engineers, scientists, and analysts
who want maximum application performance, minimum programming difficulty and
maximum leverage of available technical computing resources, including laptops,
desktops, servers, clusters and the Cloud.
Jacket provides a middleware approach to GPU programming, with MATLAB
as the fronted point of interaction for the user [8]. MATLAB with millions of users
worldwide is the platform of choice for engineers and scientists alike, for rapid
algorithm prototyping. MATLAB is an extensible interactive programming environment
for numerical analysis built on a vector language called M. The M-language, like other
525
vector languages, provides users with a high-level interface at which operations may
be specified over large sets of data at once making the expression of data-parallel
algorithms natural. M is also dynamically typed, adheres to pass-by-value semantics,
and is integrated into a well developed interpreted environment. With these
characteristics, M has proven to be a powerful, user-friendly language. Using Jacket,
the M-language and MATLAB transparently adapt to GPU computing. Unlike other
GPU solutions, Jacket provides GPU computation and graphics ability from a language
which is inherently parallel and interpreted, thereby providing a standard, extensible,
and simple method of programming for the GPU in an already proven rapid prototyping
environment. Jacket adds few GPU-specific data types to MATLAB with overloaded
operators and entire CPU-bound MATLAB programs can be converted into GPUenabled programs through as little as adding a 'g' prefix onto memory allocation
commands. Otherwise, the user interacts with MATLAB as they normally would either
from the command line or when running scripts.
Changing the data type allows the user to tap into the GPUs tremendous
computational power. All standard data types are supported, including real and
complex versions of singles, doubles, and integars. CUDA-capable GPUs of all types
are supported. For instance, NVIDIA's Fermi-based Tesla C2050 contains 448 cores
that are now fully accessible with Jacket, but those with lesser budget can also run
Jacket on GeForce and Quadro GPUs.
Performance tests were performed on a desktop configuration of dual core
CPU and Windows 2010 64 bite operating environment, using Matlab 2010a with
Jacket v1.4. The GPU used is CUDA enabled GeForce 9500 GT with CUDA drivers.
2.3 Comparative numerical results
The following example presents the computational solution of the wave
equation using the discrete Fast Fourier Transform. Consider the variable coefficient
wave equation
du
du
0, c x 0.3 cos3 x 1 ,
c x
(1)
dt
dx
for x >0,2S @,t ! 0 , with periodic boundary conditions. As an initial condition we take
u x,0
exp 100 x 1
. This function is not mathematically periodic, but it is so
close to zero at the ends of the interval that it can be regarded as periodic in practice.
In the first case the solving is executed using the GPU unit when the fft function
was realized using GPU instructions with the help of GPUmat library, while in the
second case a modified Jacket code was developed for solving the wave equation (1)
using the Fast Fourier Transform on Graphic Processing Unit. The computation using
the GPUmat fft function introduced the occurrence of spurious dispersion. In Figure 1a
can be noticed that the algorithm loses its stability with oscillations that will grow
exponentially and swamp the solution. The code requires another corrections to
increase the accuracy. Figure 1b shows solution of the wave equation (1) computed on
GPU with Jacket v1.4.
Some reasons for which Jacket is not suitable for development of
hydrodynamic stability algorithms are the facts that Jacket are a rather limited functions
526
for matrix operations (for example diag(M) diagonal scaling is not supported), there is
no eigenvalue solver supported by Jacket (as qz factorization for generalized
eigenvalue problems is not currently supported, eigs largest eigenvalues and
eigenvectors of matrix are unsupported), except some trials, these being insufficient for
a generalized eigenvalue problem, at the time being.
Fig. 1 a. Solution of the wave equation (1) computed using GPUmat- algorithm loses
its stability
b. Solution of the wave equation (1) computed on GPU with Jacket v1.4.
That is the reason that most applications will use both CPUs and GPUs,
executing the sequential parts on the CPU and numeric intensive parts on the GPUs.
This induces a significant overhead of memory transfers between the host CPU and
the GPU. In general, the overhead of time spent in sending data to the GPU and
bringing it back neutralizes any performance benefit obtained by computing on the
GPU.
3.
PARALLEL AND DISTRIBUTED INVESTIGATION ON CLUSTER PLATFORM
Parallel programming is an important aspect of high-performance scientific

computing but it used to be a niche within the entire field of hardware and software
products. However, more recently parallel programming has left this niche and will
become the mainstream of software development techniques due to a radical change
in hardware technology.
Jeremy Kepner [9] introduces the theory, algorithmic notation, and an under
the hood" view of distributed array programming. He discusses metrics for evaluating
performance and coding of a parallel program. A selected survey of parallel application
analysis techniques are presented in this book, with a particular emphasis on how the
examples used in the book relate to many wider application domains. Rauber and
Runger [10] take up the new development in processor architecture by giving a
detailed description of important parallel programming techniques that are necessary
for developing efficient programs for multicore processors as well as for parallel cluster
systems or supercomputers.
In this paper we have build and tested a small scale cluster based on a Matlab
Parallel Processing Toolbox. Using the internal cluster manager from Matlab we were
able to evaluate the algorithm behaviour using a distributed process. In this situations
527
we have perform the profiling and we have noticed a speed increase of the algorithm
compared to a single computer run.
3.1 The hydrodynamic stability problem in a hydropower turbine
The helical vortex breakdown (also known as precessing vortex rope) is a selfinduced instability of a swirling flow, encountered in the draft tube cone of hydraulic
Francis turbines operated far from the best efficiency. We consider the vortex rope
model developed in [11] for our stability analysis processed on a cluster platform.
Let us define the eigenvalue problem governing the hydrodynamic stability in
operator formulation
k L> @ ZL> @ L u
Z
0, u
G H P
(2)
where
1 dr
m
0
r 0 0 0
0 0 0 0
0
0 U 0 0
0 1 0 0
0
mW / r
2W / r dr
k
Z
(3)
L> @
, L> @
, L
0 0 rU 0
0 0 r 0
0
W rdW / dr mW m
dU/ dr
0
0
U 0 0 1
1 0 0 0
mW / r
For the case of the flow downstream a Francis turbine runner, the physical
condition that the radial amplitude of the velocity perturbation at the wall vanishes, i.e.
G(rwall ) 0 , is valid.
Four boundary conditions must be added to define the mechanical equilibrium
of the fluid in the Francis turbine runner.
They have been deduced in axis and at the wall boundary and depending on
mode number are listed below
F r
m ! 1,
G r
F r
r1,
G r
m
G r
rwall
0, H r
rwall
H r
0,
(4)
0.
0 r G r
0, P r
0,
(5)
0.
dr F r 0 0, G r
0,
G r rwall 0.
H r
0,
(6)
The numerical investigation employed both the positive and the negative
modes. We denote by ZCR (critical frequency) the temporal frequency corresponding
to a maximum growth rate GRmax for a given mode number. The numerical results are
summarized in Table 1.
Table 1 The critical frequency and the maximum growth rates obtained for the
investigated modes
Mode m
3
ZCR
0.4
GRmax
528
0.1379
2
0
1
Spatial stability results
0.4
0.4
0.3
0.1255
0.0603
0.0345
0.3
0.4
0.4
0.9223
0.8102
0.8224
Fig. 2 depicts the amplitudes of the disturbances extracted at several locations

on the wall boundary. Amplitudes peak and decay as time increases, observation that
leads to the conclusion that the flow exhibit pressure fluctuations at the wall.
Absolute instability of the perturbed flow is depicted in Fig. 3. The growing
evolution of the perturbation in pressure is observed as it evolves in time. As time
increases the disturbances grow in magnitude, their leading edges move toward the
wall boundary along the draft tube.
Fig. 2 Fluctuating pressure as function of time, at the wall boundary, extracted at

several locations on the draft tube: non-dimensional axial coordinates
z 1.71,2.43,3.30,4.12,4.25 .
Fig. 3 Absolute instability of the flow in hydraulic turbine draft tube after perturbing
flow. The curves are extracted at several non-dimensional time units
t 10,15,18,22,25 at the wall boundary.
The dimensionless rope frequency is defined as Zrope
:rope / : . i.e. it
represents the ratio between the rope precession angular speed and the runner speed.
We found that the uniform core frequency yields Zrope 0.305 , the same as in the
experiment of Ciocan et al. [12].
529
3.2 Evaluation of the parallel algorithm performance

Fig. 4 presents the parallel processing time on four cluster configurations, in
comparison with sequential elapsed time. It is noticeable that the optimal processing
time is reached when the numerical experiments are running on four or six parallel
processors instead of the other configuration.Thus the conclusion that the cluster has
no need to be extended over six parallel workers for numerical processing of our
stability investigation problem.
Fig. 4 Elapsed time for mode m
1, on four cluster configurations.
In parallel computing, speedup refers to how much a parallel algorithm is faster

than the corresponding sequential algorithm. Efficiency is a performance metric
typically between zero and one, estimating how well-utilized the processors are in
solving the problem, compared to how much effort is wasted in communication and
synchronization
The speedup of the parallel algorithm computed for three values of the spectral
parameter into the convergence interval N ^83,61,46` is depicted in Fig. 5a as
function of the processors number. A maximum speedup is obtained when the
algorithm is processed on four parallel processors with spectral parameter N 83 for
which the convergence is reached. It is noticeable that Sp p for all cluster
configurations investigated here, thus we reach a sublinear speedup.

The efficiency of the parallel algorithm computed for the same values of the
spectral parameter located the convergence interval N ^83,61,46` is depicted in Fig.
5b as function of the processors number.
530
b
Fig. 5 a. Parallel algorithm speedup; b. Parallel algorithm efficiency
It is confirmed again that the best efficiency of the spectral algorithm presented
in this chapter is reached when the cluster is configured with four parallel processors,
setting the spectral parameter N 83 collocation nodes.
4.
CONCLUSION
Two different solutions to accelerate numerical algorithms used in fluid stability

problems by means of parallel computing were presented in this paper.
Concerning the GPU computing, we conclude that Jacket platform provides an
extensive list of Matlab functions with their state of support, ranging from fully
supported, meaning that they should work in any situation authorized by the original
Matlab function, to not supported, which cannot be used with gsingle arguments and
there is no plan to support them in the immediate future. The number of supported
functions is much bigger in Jacket than in GPUmat at the time of this writing.
Because of no dynamic matrices with Jacket, hence the assignment
data = [v; zeros(nplots,N)];
is not valid as computed on GPU, it was necessary a two-way directional transfer of
variables between the GPU and the CPU. This increases the elapsed time, despite of
the performance improvement obtained by using the GPU processing. In Table 2 a
comparison between the CPU computation time of the wave equation (1) and the GPU
computation time of the same is presented.
Table 2 Comparison between the computation time of the wave equation, on CPU and
GPU.
Elapsed time of computation on CPU
0.883485
Elapsed time of computation on GPU

1.539880
Since our GPU works only with single precision floats, hence gives a less
precise calculations, the conclusion is that this technology is not mature enough to be
used effortlessly by the average user.
The cluster technology, used as the second accelerating solution, represents
the state of the art in computational technology. This extends the limit imposed by the
single computer analysis, reducing in a large manner the amount of time required for
531
solving complex mathematical models. Using a rather usual hardware (desktop PCs) it
is possible to obtain a significant acceleration of the computation process. However,
the acceleration depends on the algorithm structure and it is necessary to perform a
profiling of the solving process (that means to analyze carefully the behavior of the
computational process, the number of steps required, the mathematical operations
involved, the size of matrix variables) in order to avoid the eventual bottlenecks in the
algorithm. The numerical algorithm was validated with existing results in the literature
[12] and valuable stability results have been computed.
LITERATURE
[1]
Venkatakrishnan, V., Simon, H. D., Barth, T. (1992). A MIMD implementation of a

parallel Euler solver for unstructured grids. The J. of Supercomputing 6, pp. 117137 .
[2] Lanteri, S. (1995). Parallel solutions of Three-Dimensional compressible flows.
INRIA, Rapport de recherche n. 2594.
[3] Diurno, W. G. (1995). Higher Order Solution of the Compressible Viscous Flows
Arising in Aerothermodynamics Using a Finite Element Method. Proc, AIDAA
Congress, Roma (Italy).
[4] Barth, T. J. (1992). Aspects of Unstructured Grids and Finite-Volume Solvers for
the Euler and Navier Stokes Equations. VKI - AGARD R 787.
[5] Peraire, J., Peiro, J., Morgan, K. (1993).
Multigrid Solution of the 3-D
Compressible Euler Equations on Unstructured Tetrahedral grids. Int. J. Num.
Meth. in Engin. 36.
[6] Bucchignani, E., Diurno, W. G. (1998). Parallel computation of inviscid 3D flows
with unstructured domain partitioning: performances on SGI-Power Challenge
Supercomputer. Parallel Computing: Fundamentals, Applications and New
Directions Elsevier Science B.V.
[7] Matlab from http://www.mathworks.com
[8] Jacket v1.4 Getting Started Guide, http://www.accelereyes.com
[9] Kepner, J. (2009). Parallel MATLAB for Multicore and Multinode Computers.
SIAM, Philadelphia, ISBN 978-0-898716-73-3.
[10] Rauber, T., Runger, G. (2010). Parallel Programming for Multicore and Cluster
Systems. Springer-Verlag Berlin Heidelberg, ISBN 978-3-642-04817-3.
[11] Bistrian, D. A., Dragomirescu, I., Muntean, S., Topor, M. (2009). Numerical
Methods for Convective Hydrodynamic Stability of Swirling Flows. Recent
Advances in Systems, Proceedings of the 13th WSEAS International Conference
on Systems, Rodos, 22-24 iulie Grecia, pp. 283-288, 978-960-474-097-0.
[12] Ciocan, G. D., Iliescu, M. S., Vu, T. C., Nennemann, B., Avellan, F. (2007).
Experimental Study and Numerical Simulation of the FLINDT Draft Tube Rotating
Vortex, J. Fluids Engineering, Vol. 129, pp.146-158.
532
REDUCING CONCUMPTION OF ENERGY OF DATA CENTERS

Sevde Stavreva1, Marko Serafimov, Igor Andreevski2
Summary: Great energy consumption and cost incurred in data centers are a huge
incentive to find ways of reducing them. Increased energy consumption is driven by
the increased capacity of computer equipment and need for its cooling, or acquiring
and maintaining the prescribed criteria for temperature, humidity and cleanliness in the
premises of such buildings. Estimates of annual energy cost in U.S data centers are
now at about 3,3 billion dollars and studies show that substantial improvements are
possible in the efficiency of data centers as well as to save energy. Yet even more than
in other systems,care should be taken to reduce energy consumption,which should not
be at the cost of reduced reliability and functionality of datacom equipment. Trends and
methods of calculating loads give insights how to establish criteria for the designing of
the centers that would deal with current and future loads and thereby to find ways to
save energy.
Key words: data centers, energy, consumption, cooling, load, efficiency, equpment,
design
1. WAYS TO SAVE ENERGY
Placing of Datacom equipment and its efficient cooling. Datacom equipment is
mostly set upon racks, ordered in rows. The front part of the racks (i.e the side through
which the air enters), facing an aisle(cold aisle), while the rear part wherein the plugs
and sockets are located, is facing the other aisle(hot aisle). Such way of placing the
equipment is most efficient, because there is no mixing of cold inlet air with hot outlet
air, and the possibility of the hot air which leaves the equipment, to re-enter it , is
minimized thereby. Thus a better cooling of the datacom equipment is achieved, and
an energy saving, too. If we apply the concept of hot / cold aisles properly, and we
supply enough cold air to the cold aisles, thus preventing recycling of the hot air back
to the cold aisles , we will maintain a constant temperature of the inlet air. Most often
the cold air in these centres is unnecessarily supplied to the cold aisles having a
temperature between 13 16C dry bulb, while the recommended teperature of the
cold air in the inlet to the equipment amounts to 20C, or more. By suplying the air to
the center with the reccomended temperature, the need of supplying the cooling water
1
M-r Sevde Stavreva, Ivo lola Ribar bb Bitola, Tehniki fakultet, sevde.stavreva@uklo.edu.mk
D-r Marko Serafimov, Skopje, Mainski fakultet, marsersk@ukim.edu.mk. D-r Igor Andreevski, Bitola,
Tehnicki fakultet, igor.andreevski@uklo.edu.mk
533
Sevde Stavreva, Marko Serafimov, Igor Andreevski
with lower temperature, is reduced, and more efficient cooling is provided.

In these centers the equipment is most often placed over a raised floor with
built-in perforated tiles through which the cold air enters the room with datacom
equipment. The perforated tiles on the raised floor are placed in the cold aisles, and
through them the cold air flows towards the inlets to the equipment. The analyses have
indicated that a significant part of the cold air leaks through openings in the raised floor
which are supposed to be tight. Such leak-outs may amoun up to 10 50 % of the total
air supplied into the raised floor. To reduce energy looses , i.e the leak-outs of the cold
air through openings not intended for this purpose (cord openings, cracks in the floor
and alike), they should be dealt with and reduced to minimum. This reduces the total
energy costs, without any investnments in new technologies. The computational fluid
dynamics (CFD) model can be used to discover the inefficiences and for optimizing
the airflow in the data centers.
Fig.1 Raised floor implementation most commonly found in data centers

2. SEPARAION OF COLD AND WARM AISLES
With the conventional cooling system (cold/hot aisles) of the datacom
equipment with high density loads, often a certain warming up of the cold air occurs,
while it is being supplied to the equipment placed on higher level racks (the height of
the racks is usually 1,98 m), as well as at the ends of the cold aisles. This stems from
the diffusion of the warm aircoming from the warm aisle. The increased flow of the cold
air through the cold aisles is helpful in such cases, but in the same time increases the
consumption of the energy and reduces the energy efficiency. A part of the cold air
surplus bypasses the equipment , and directly blends with the hot return air from the
racks, reducing the temperature of the return air which recirculates towards the airconditioning equipment. Instead of supplying higher amount of cold airflow, we can
meet the needs for efficient cooling in a simple manner, by separating the hot from the
cold aisles using physical barriers. The screen which separates the cold from the hot
air is necessary in order to achieve maximum degree of separation. Three approaches
are feasible, screening of the cold aisle,screening of the hot aisle, or encasing the
racks. When screening the cold aisles, the supplied cold air is retained within the
aisles, the mixing with hot air is prevented, and a constant temperature of the cold air
at the inlets to the equipment is provided. Thus the temperature of the supplied cold air
can be increased, because the inlet air temperature to the equipment equals the
534
Reducing concumption of energy of data centers
temperature of the air supplied from the plenum of the raised floor. The higher
temperature of the supplied air yields a higher efficiency of the cooling equipment, and
increases the periods of free cooling. Also, the required amount of airflow can be
reduced, resulting in lower energy consumption for recirculating of air. It also allows
use of fans with variable speed, and enables adjusting of the fan speed to the airflow
quantities necessary for cooling the equipment. With the screening of the hot aisles,
the hot air is kept within the hot aisles, and the rest of the data center is exposed to the
impact of the cold air from the cold aisles. The retaining of the hot air offers the same
advantages as the retaining of the cold air. The encasing of the racks is similar to the
previous case, but it is implemented within individual racks. Such enclosures have
distinct inlet and outlet pathways for the cold/ warm air preventing their blending. The
encasing of the racks has its advantages, because the rack is the smallest unit in the
center which can be isolated and located everywhere. The capacity as to the loads per
rack, can be further increased, since there is no further mixing of cold and hot air. The
benefits are considerable and allow for profitability of the upgrading of the mechanical
system.
Fig. 2 Potential recirculation from hot aisle to cold aisle

3. HIGH - DENSITY SUPPLEMENTARY COOLING
The praxis to date has shown that the air- conditioning systems used in the
data centers, are efficient in providing the required enviroment for IT equipment. But in
actual data centers the energy densities steadily grow and often surpass the densities
of 30 kW. This makes dificult the removal of the heat generated by the equipment ,
requiring more eficient energy utilization. The problem is additionally complicated by
the increased requirements for cold airflow to the equipment, as the load densities
grow. To overcome the effects of the mixing of cold/warm air, the cold airflow is
significantly higher than the necessary one, which requires further increase of the
energy comsumption. Using additional cooling equipment placed over or alongside the
equipment racks, the warm air is directly sucked into the warm aisle, and cold air is
535
allowed to enter the cold aisle. The employed additional equipment enables significant
energy savings. These savings are achieved by bringing the cooling closer to the heat
source, and reducing the fan power necesary to move the air. The closer placing of the
coolinq equipment near the heat loads , alows much shorter pathway for bringing cold
air and removal of heat. Due to the proximity of the cooling equipment to the
equipment racks, the problems with the inappropriate distribution of the cold airflow
and its mixing with the warm air, are eliminated.
4. USE OF THE FREE COOLING
The cooling equipment is the biggest energy consumer of the entire HVAC
equipment in the centers. Significant savings of the HVAC energy can be achieved by
reducing the required energy for the cooling eguipment. On one hand, an equipment
should be chosen which has greater energy efficiency, but on the other hand a
solution is to be found to reduce the working hours of the cooling equipment. The
centers run 24 hours a day, that is , 9760 hours a year continually, and through the
bigger part of the year, with lower outdoor temperatures, a free cooling can be
implemented. The free cooling can be carried out by outdoor air or water. Considering
that there are many periods during the year when the outdoor air temperature is within
o
o
the range of 13-16 C ( i.e 16-20 C if the recommendations for higher temperature of
the supply air to the IT equipment, are accepted ), then the use of free cooling with air
contributes to the significant savings of energy. At these temperatures and within
certain ratio of mixing the outdoor and the recirculated air, the cooling can be provided
without activating the cooling compressors. When the temperatures are higher then the
temperature of the supplied air, the cooling must be done by partially activating the
compressors. Newest studies have determined insignificantly increased particle
concentrations within the data centers, using free cooling with outdoor air, compared
with the centers using 100 % recirculated air, but yet the particle concentration is
considerably above the most stringent standards. The control and regulation of the
relative humidity can be successfully managed in the centers using free cooling. If a
bigger data center uses so much energy as a little town, the savings achieved by using
free coling will equal the energy consumption of a smaller municipality of such town.
5. ADJUSTING THE POWER IN DEPENDANCE OF THE LOADING
Data centers are designed for extreme conditions, which arise seldom during
the work. The working efficiency during maximum loading is frequently not a good
indicator as to the real working efficiency. Applying new technologies with the air
conditioning equipment, allows achieving higher efficiency during partial loading.
Usually with data centers, the daily requirements increase progressively from 5.00 h to
11.00 h a.m, and then drops beginning from 5.00h p.m. The energy consumption
remains relatively high, although the loading of the servers increases. A center working
with only 20% of its capacity, can consume as much as 80% of the energy which the
same center would consume when working with 100% capacity. The server processors
may have the ability to regulate the power depending on the loading level. Most often
this ability is not used, and these decisions should be reevaluated because of the
savings such regulations would yield.
536
Reducing concumption of energy of data centers
6. MONITORING AND OPTIMIZATION

One of the consequences of the energy density increment within the
equipment, is the increased diversity within a data center. The energy densities of the
electronic equipment racks in the center, are mostly unevenly distributed, which
causes cooling inefficiency, if monitoring and optimization are not implemented. The air
conditioning equipment on one side of the room may humidify the air, starting from the
conditions prevailing locally, while the equipment of the opposite side of the center,
may dehumidify it. Therefore a control system is needed to monitor the conditions
throughout whole data center and coordinate the activities of the entire air conditioning
equipment, to prevent the counteracting effects in its work and to reinforce the
cooperation. Thus, energy saving is achieved by means of coordinated work of the air
conditioning equipment, while implementing the control system throughout the data
center.
7. PROCESSOR POWER
Lacking the real standard measure for the prosessor efficiency, the thermal
design power (TDP) serves as indicator for the server energy consumption. The typical
thermal design power with the processors in use today, is between 80 and 103 W. Due
to the price drop, the processor manufacturers offer models with lower voltage,
consuming averagely 30 W less than the standard processors. Independent studies
have shown that such lower voltage processors provide the same computer
performances like the models with higher voltage, while diminishing the energy
consumption.
8. POWER SUPPLY
The power supplies of the servers used in today data centrs, are mostly with
lesser efficiency than the optimum. Nowdays, there are power supplies which provide
90% efficiency. The use of such supplies reduces the energy consumption in data
centers. As with other systems, the data centers also show variable supplies' efficiency
of the servers, depending on the loading. Other possibility to reduce the energy
consumption is to size the supplies as close to the real server loading as possible. The
server manufactures size the supplies according to the maximum available server
configuration, whilst the servers may be shipped out with considerably lower energy
requirements. Most of the servers cannot be upgraded upto the maximum
configuration, thus the supplies work with lesser partial loadings and smaller efficiency.
The end users should ask from the server manufacturers to size their power supplies
to match the configuration they intend to use.
9. CONCLUSION
Although safe working is priority with most data centers, considerable energy
reductions may be achieved without disturbing the datacom equipment work.The
proper placing of the datacom equipment, the optimization of cold air pathways, the
optimum sizing of the airflow, considerably contribute to the energy reduction. The use
of free cooling with air or water, the aditional cooling equipment or equipment with
537
higher energy efficiency, justify the investments many times. The above said suggests
that using optimum strategies with data centres and air conditioning equipment, may
result in considerable energy savings and may improve the energy efficiency.
LITERATURE
Journal Papers:
[1] Martin, M., Khattar, M., Kljajin, M., Germagian, M. (2007). High-Density Heat
Containment. ASHRAE Journal, vol. 49, no. 12, p. 38-43.
[2] Judge, J., Pouchet, J., Ekbote, A., Dixit, S. (2008). Reducing Data Center Energy
Consumption. ASHRAE Journal, vol. 50, no. 11, p. 14-26.
[3] Beaty, D., Davidson, T. (2003). New guideline for Data center cooling. ASHRAE
Journal, vol. 45, no. 12, p. 28-34.
Books:
[4] ASHRAE Technical Committee 9.9. (2006). Design Considerations for Datacom
Equipment Centers
Chapters in Books:
[5] ASHRAE Technical Committee 9.9. (2005). Air Cooling of Computer Equipment.
ASHRAE, Datacom Equipment Power Trends and Cooling Application, p. 29-39
[6] ASHRAE Handbook.(2007). Data Processing and Ellectronic Office Areas.
ASHRAE, HVAC Applications, chapter 17.
538
RESULTS OF EXPERIMENTAL INVESTIGATION OF PELLET STOVE

ACCORDING TO EN 14875
N. Mani1, V. Jovanovi2, D. Stojiljkovi3
Summary: The results of investigation of small scale pellet stove (8.5 kW) for
combustion of wood pellets used for household heating are presented in this paper.
Experiments were performed on the test installation, designed and assembled in the
Fuel and Combustion Lab at the University of Belgrade, Faculty of Mechanical
Engineering. Design and manufacturing of test installation, as well as the entire testing
procedure were carried out in accordance with the requirements of the applicable
European regulation EN 14785 for ovens of this type. Tests were conducted for three
thermal loads of the stove (minimum, medium and maximum) and three types of test
fuel. Pellets of different quality (made of beech wood, but with different proportions of
bark and various technological parameters of the production process) were used as
test fuels. Recalculation of test results in accordance with the requirements of a
standard provided determination of thermal and environmental characteristics of the
tested stove and its classification according to the criteria defined by standard EN
14785.
1.
INTRODUCTION
Low energy density is the main disadvantage of raw biomass for energy use.
Pelletising as a densification process of this kind of renewable energy source is
possible solution for improving this characteristic. Nowdays, pellets made from
biomass especially from wood are used in all areas for energy production from small
scale stoves and boilers up to large scales energy plants. Automatic feeding, comfort
use and relatively low maintenance costs with high efficiency of combustion process
made pellets highly recomended for thermal energy production especially in
households.
Basic aim of the experimental investigations described in this paper was to
obtain influence of pellet production parameters (die dimensions and temperature) on
combustion process in smale scale pellet stove. During the realization of the
experiments three types of wood pellets were used and thermal and environmental
characteristics of investigated pellet stove were tested.
1
Teaching Assistant, University of Belgrade Faculty of Mechanical Engineering, nmanic@mas.bg.ac.rs

Teaching Assistant, University of Belgrade Faculty of Mechanical Engineering, vjovanovic@mas.bg.ac.rs
3
Full Professor, University of Belgrade Faculty of Mechanical Engineering, dstojiljkovic@mas.bg.ac.rs
2
539
N. Mani, V. Jovanovi, D. Stojiljkovi
2.
TESTS OF THE SMALL SCALE PELLET STOVE
The tests were based on definition of test program and choice of different fuels.
Definition of listed parameters was necessary to provide the same test conditions for
different pellets used in same pellet stove.
2.1 Construction of investigated pellet stove
Stove used for experimental investigations in this paper is small scale pellet
stove used for heating air in households with declared thermal output 8.5 kW.
Schematic view of vertical cross section of investigated pellet stove is presented on
Figure 1.
flue gas
heated air
fuel
combustion air
Fig. 1 Schematic view of investigated pellet stove

Pellets from the built-in reservoir (1) with screw conveyor (2) are inserted from
above into the combustion chamber (3) and fall into the burning pot (4), which is
perforated from the bottom. Fan for hot gasses suck the flue gas from the combustion
chamber providing slight underpressure (5 - 10 Pa) in it. This underpressure provide
draft for airflow required for the combustion process, from the bottom of the burning
pot, through a circular holes. Flue gasses from the combustion chamber are extracted
by exhaust fan through the flue duct (5) to output section of the stove. Second fan
introduce the air for room heating through the air duct (6) over heat exchanger and
counterflow from flue gasses. This allows heat exchange between hot flue gasses and
heated air for room heating.
2.2 Test procedure and installation
The tests of pellet stove were made according to the demands of regulation EN
14785. Typical test cycle for determination of nominal heat output is foreseen by this
standard, that comprise:
startup and pre-test for reaching the steady state;
minimum 3 hours is used for determination:
- total efficiency
540
Results of experimental investigation of pellet stove according to EN 14785
- heat output.
time required to reach the steady state.
Experimental tests were conducted for nine experimental regimes of
investigated pellet stove. Three test fuels (marked as P_1, P_2 and P_3) and three
thermal loads of the pellet stove (minumum, medium and maximum - marked as TL_1,
TL_3 and TL_5 respectively) were tested.
The test installation was designed and constructed in Fuel and Combustion
Lab at the University of Belgrade, Faculty of Mechanical Engineering, according to the
demands of regulation EN 14785 (Figure 2). It consists of investigated appliance,
weight scale and measurement section for flue gas composition, temperature and flow.
The investigated pellet stove prepared for the test is shown on Figure 3.
Fig. 2 Test installation used for

investigations [1]
(1-exaust to atmosphere, 2-fan, 3-adjustable
dumper, 4-adjustable gather, 5-flue gas
connector, 6-measurement section, 7trihedron side wall, 8-flue gas adaptor - bend,
9-pellet stove, 10-trihedron side wall, 11trihedron test hearth, 12-platform scale)
Fig. 3 Pellet stove on the test installation
2.3 Test FUELS

Three different wood pellets were used as test fuels for the experimental
investigations. All pellets are produced from same raw material (beech wood) and with
same type of pellet press (presses with vertical die), but with different production
parameters (die diameter, height and temperature). Pelletising production process
parameters for each test fuel are given in Table 1.
541
Table 1 Production process parameters for test fuels

Production parameter
Die dimensions DxW [mm]
Die hole dimensions DxL [mm]
Die temperature [C]
P_1
1100 x 350
6 x 24
125-135
P_2
1010 x 300
6 x 22
110-130
P_3
1010 x 300
6 x 28
134-147
Proximate and ultimate analysis of all test fuels were made according to the
regulation EN 14961.
3.
TEST RESULTS
Results obtained during the experiments were classified in 4 groups:
fuel analysis,
combustion process,
thermal characteristics,
environmental characteristics.
3.1 Fuel analysis

Analysis of all test fuels were made according to the regulation EN 14961.
Proximate analysis was made in Fuel and Combustion Lab (University of Belgrade,
Faculty of Mechanical Engineering) and ultimate analysis in the Institute for chemistry,
technology and metallurgy Center for chemistry. The results of proximate and
ultimate analysis (only for as received mass) for all test fuels are given in the Table 2.
Table 2 Proximate and ultimate analysis of test fuels (as received)
P_1
PROXIMATE ANALYSIS
Total moisture (% m/m)
5.17
Ash (% m/m)
1.82
Combustibles (% m/m)
93.0
Volatiles (% m/m)
77.4
Fixed carbon (% m/m)
15.6
High heating value (kJ/kg)
18399
Low heating value (kJ/kg)
16866
ULTIMATE ANALYSIS
Carbon (% m/m)
45.2
Hydrogen (% m/m)
6.24
Nitrogen (% m/m)
0.18
Sulphur (% m/m)
Oxygen, as the difference (% m/m)
41.5
P_2
P_3
6.86
1.20
92.0
77.1
14.9
18359
16801
7.66
1.45
90.9
77.5
13.4
17403
15832
44.6
6.16
0.17
41.0
44.1
6.13
0.17
40.5
3.2 Combustion process

Results of combustion process obtained during experiments and used for
determination thermal and environmental characteristics of investigated stove were:
542
fuel consumption,
oxygen (O2) content in flue gas,
carbon dioxide (CO2) content in flue gas,
flue gas temperature.
Results of fuel consumption for each experimental regimes are given in Table 3
Table 3 Fuel consumption in kg/h for all experimental regimes
P_1
0.45
0.95
1.60
TL_1
TL_3
TL_5
P_2
0.60
1.00
1.70
P_3
0.41
0.86
1.47
Mean values of oxygen (O2) and carbon dioxide (CO2) content in flue gas and
flue gas temperature during experimental tests are presented in Figure 4, Figure 5 and
Figure 6 respectively.
TL_1
TL_3
TL_5
25
O2[%v/v]
20
15
10
18,77
19,3
18,33
17,18
16,89
14,75
18,23
16,50
13,96
P_1
P_2
P_3
Fig. 4 Oxygen content in flue gas for all experimental regimes
543
TL_1
TL_3
TL_5
CO2[%v/v]
4
6,67
5,92
2,63
2,53
2,11
4,29
3,89
3,62
1,62
P_1
P_2
P_3
Fig. 5 Carbon dioxide content in flue gas for all experimental regimes
TL_1
TL_3
TL_5
300
Temperature[C]
250
200
150
253,2
240,7
100
188,6
185,3
180,9
148,2
141,4
126,2
50
98,8
P_1
P_2
Fig. 6 Flue gas temperature for all experimental regimes
544
P_3
3.3 Thermal characteristics

For all experimental regimes, thermal characteristics that were determined
during the tests were:
obtained heat output;
total efficiency of the pellet stove.
All calculations of listed thermal characteristics were conducted according
procedures defined in EN 14785 and based on obtained experimental results for each
firing regime. Results of thermal characteristics from the tests of investigated pellet
stove for different test fuels and different thermal loads are given on Figure 7 (obtained
heat output) and Figure 8 (total efficiency).
Regulation EN 14785 doesnt have strict limits for the nominal heat output of
pellet stove. However, total efficiency is defined and according to this regulation shall
be at least 75% at nominal heat output and 70% at reduced heat output.
TL_1
TL_3
TL_5
9
8
7
Heatoutput[kW]
6
5
4
3
5,56
5,13
2
1
3,58
3,09
2,88
1,78
1,33
1,79
0,83
P_1
P_2
P_3
Fig. 7 Obtained heat outputs for all experimental regimes
545
TL_1
TL_3
TL_5
100
90
LowerlitimaccordingtoEN14785
fornominalheatoutput
80
LowerlitimaccordingtoEN14785
forreducedheatoutput
TotalEfficency[%]
70
60
50
40
66,97
68,87
72,90
67,58
70,32
74,36
30
55,36
45,87
20
47,22
10
0
P_1
P_2
P_3
Fig. 8 Total efficiency for all experimental regimes

3.4 Environmental characteristics
Under environmental characteristics only carbon monoxide (CO) emission level
is defined by the regulation EN 14785. According this standard, during testing,
maximum permitted mean carbon monoxide concetration based on 13% of oxygen
(O2) content in flue gas can be 0.04% v/v (500 mg/m3) at nominal heat output and
0,06% v/v (750 mg/m3) at reduced heat output.
Results of CO emissions (calculated on 13 %v/v O2 content in flue gas) from
the tests of investigated pellet stove for different fuel types and different thermal loads
are given on Figure 9.
546
240000
TL_1
TL_3
TL_5
200000
200671
160000
CO mg/m
120000
142931
Upper limit according to EN 14785
for reduced heat output
80000
69001
Upper limit according to EN 14785

for nominal heat output
4000
3428
3000
2814
2680
2000
1772
1000
799
393
P_1
P_2
P_3
Test fuel
Fig. 9 CO emission calculated at 13 %v/v O2 content in flue gas for all firing regimes
4.
CONCLUSIONS
Heating appliances fired on biomass pellet could be a part of the future,

especially having on mind that biomass is renewable energy source with zero CO2
balance. The importance of this type of heating appliances is emphasized by the all
advantages using pellets especially for household heating. European regulation EN
14785 has brought clear test procedures and strict limits for heating appliances fired
with this type of solid biofuel.
The results of combustion process for investigated small scale pellet stove
obtained from the tests made completely according to the demands of EN 14785, are
presented in this paper. Based on these results it can be concluded that:
total efficiency required by EN 14785 for nominal heat output (more than
75%) was not achieved for neither experimental regime;
total efficiency required by EN 14785 for reduced heat output (more than
70%) was achieved for maximum thermal load (TL_5) with test fuels (P_1)
and (P_2) and for medium thermal load (TL_3) with test fuel (P_2);
total efficency for other thermal loads for this two test fuels was slightly
below this limit, except for the test fuel (P_3). For this test fuel for all thermal
loads total efficency was between 45,87% to 55,36%;
3
CO emission limit for the nominal heat output (less then 500 mg/m at 13%
O2) was achieved only for maximum thermal load (TL_5) with test fuel
(P_2). With other test fuels and for other thermal loads this demand of EN
14785 was not fulfilled;
CO emission for test fuel (P_3) was extremly high and for all thermal loads it
has varied from 69001 mg/m3 to 200671 mg/m3.
Extremly high CO emission and low total efficency for test fuel (P_3) proves
that pelletising production parameters have significant impact on combustion process.
547
Also according to obtained results for thermal output and total efficency it is possible to
make an improvment of investigted stove construction and combustion process. These
facts will be the subject of further investigations of the authors.
LITERATURE
[1]
[2]
[3]
548
ONORM EN 14785:2006. Residential space heating appliances fired by wood

pellets Requirements and test methods. Osterreichisches Normunginstitut.
Wien
Stojiljkovic D. Jovanovic V. Radovanovic M. Manic N. Radulovic I. Perisic S.
(2005). Investigations of combustion process in stove fired on biomass. Strojniski
vestnik-Journal of mechanical engineering, vol. 51. br. 7-8. pp. 426-430.
V.K. Verma , S. Bram, G. Gauthier, J. De Ruyck (2011). Performance of a
domestic pellet boiler as a function. Biomass and bioenergy. vol35. pp 272-279.
SOME EXPERIENCES IN THE PRODUCTION OF BIOGAS FROM

LIQUID MANURE
Mirjana Radii1, Nataa Soldat2
Summary: The production and application of biogas on the basis of organic remains is
an industrial procedure, with which experts globally intend to contribute to finding a
solution to the energy crisis, to environmental protection and the production of high
quality natural fertilizer. Following global trends, as well as the needs of the Republic of
Serbia, biogas-related activities are becoming visible and gaining more and more
supporters. The work of the small number of industrial-scale facilities is insufficiently
monitored, and the results are not open to the public. This paper presents the results of
a research concerning the functioning of an industrial system for the production of gas.
Key words: liquid manure, biogas, bio-fertilizer.
1.
INTRODUCTION
Production and use of biogas obtained from organic waste, is a new technology
and industrial procedure, which contributes to the solution of following issues:
energy crisis,
production of high quality quality manure,
environmental protection.
Following global trends, the activities related to the production of biogas are
becoming very popular in the countries of this region, too. Research is carried out in
labs and on devices in operation, although the number of reported research results is
very modest. They mostly refer to the fermentation process carried out in laboratory
conditions, using a wide range of substrates.
The operation of a small number of facilities erected on an industrial scale is
insufficiently observed, and research results are mainly unavailable to the wider public.
The construction of a device based on liquid manure in the Republic of Serbia dates
back to the early 1980's. Up to date, a few high capacity facilities have been built,
which differ in concept and design of equipment, for practically every phase of the
process.
Prof. dr Mirjana Radii, Zrenjanin, Serbia, Technical College of Appplied Sciences in Zrenjanin,
(mirjana_radisic@yahoo.com)
Dipl. ma. in.-master Nataa Soldat, Belgrade, Serbia, Faculty of Mechanical Engineering, University of
Belgrade (nsoldat2000@yahoo.com)
2
549
Mirjana Radii, Nataa Soldat
2.
GENERAL CONSIDERATIONS
The degradation of material, that takes place in these facilities, is most effective
at a temperature of 15C (psychrophilic), 35C (mesophilic) and 55C (thermophilic
process). Most often used is the mesophilic, and during summer also the thermophilic
process.
The main phases of the process are [1,2]:
Collecting and homogenizing of substrate,
fermentation,
purifying and storing of biogas,
processing and distribution of the over-fermented rest.
According to incomplete available data, there were 7 biogas facilities built in the
mid 1980s in the Republic of Serbia. These are located on pig farms in Surin, Vizelj,
Banatski Karlovac, Vladimirovac, Negotin, Varvarin and Seanj. None of them has
been completely finished, and has not worked successfully for even a year. The main
reason were various breakdowns and serious technical deficiencies, as well as a lack
of market for their product.
As for the process of obtaining biogas, for which the basic inputs are provided
on the farm itself, it should be emphasized that farms with biogas production devices
were built according to different technologies, with a different number of buildings,
different length and slope of the manure extraction canals and so on.
3.
SUBSTRATE PROPERTIES
Chopped up raw material, which on agricultural farms refers to biomass, and

mixed with water in a certain proportion, is the substrate that is used to obtain the
amount required for the realisation of a biogas production process in a biogas facility.
Table 1. The amount of excrement from a pig farm with a capacity of 3000 pcs./yr.
1.
Number of pigs at a
time, by type and
age
2.
Breeding sows (200220 kg)

Breeding boars
(200-250 kg)
Piglets for feeding
(6-27 kg)
Female breeding
pigs (100-120 kg)
Fatlings (27-103 kg)
Total:
3.
4.
5.
6.
Number
of pigs
(pcs.)
Amount of dung
and urine per pig
(kg/day)(kg/day)
Total amount of
excrement per pig
(kg/day)(kg/day)
Content of
excrement dry
matter
(kg/day)
1600
4.78
6.97
11.75
18.80
2.68
24
3.98
5.14
9.12
219
11
6000
0.36
0.51
0.87
5220
540
175
10500
18300
1.85
2.37
-
2.35
2.61
-
4.20
4.98
-
735
52290
77264
63
4612
7908
Substrate is obtained by systematic flushing of manure from the canal to the

collecting pit, where it is prepared for fermentation (homogenisation, heating). On
farms with similar a capacity, the number of pigs at a time by age and mass is similar.
550
Some Experiences in The Production of Biogas from Liquid Manure
The differences are small, as are the chemical composition of the food, as well as the
preventive health protection measures applied through food. Disinfectants for the
objects are used depending on habit and the actual situation on the market. Ususally
the water from washing must not reach the fermentation device.
The dry matter in excrement is about 10.2%. The water for the flushing of the
canal is roughly 50.000-70.000 liters/day. Based on this data, liquid manure should
contain 5-6% of dry matter.
The research of the author [1], conducted between 1985-1987, point out the
fact that water consumption is not taken into account on farms, and therefore the
content of dry matter in manure varies from 0,9-7%.
Table 1. shows the amount of excrement from a pig farm with a capacity of
3000 fatlings per year [1].
4.
ABOUT FERMENTATION DIGESTION DEVICES
The facility for the production of biogas consists of a complex installation

comprising a great number of parts. The appearance of the facility depends on the type
and amount of raw material to be used for the production of biogas.
The technological process of biogas production is based on anaerobic, methanic
fermentation taking place in a basic part of the equipment known as digestion
fermentation device. The use of heterogenous raw materials (by origin, composition,
consistency, etc.) in the production of biogas, and the insufficient knowledge on
biological, biochemical and physico-chemical basics of anaerobic fermentation, are the
reason why we cant say for sure which type of digestion device, and therefore
technological process of biogas production would be the best. Due to lack of
experience in the production of biogas in large facilities, a variety of equipment and
technological procedures is used [3].
The digestion device, a core part of a biogas facility, is a hermetically closed
3
device with a volume ranging from a few cubic meters up to 2.500 m . The walls of the
digestion device can be made of steel sheets, concrete, reinforced concrete, concrete
combined with plastics or steel sheets, bricks or plastic mass reinforced with glass fiber
etc..
Digestion devices are classified by shape, size, type and the material they are
made of, the mixing system used or substrate heating system.
The digestion device should meet the following requirements [2, 3]:
it must not be gas or liquid permeable-therefore well sealed,
it must endure a large static load,
it has to be corrosion-resistant,
it must allow gas or liquid to get out.
Heating of mass and maintaining temperature in fermentation devices is
carried out in a number of ways: directly-using low pressure steam injected into the
substrate, indirectly-through heating rings on the outer side of the digestion ndevice
shell, using heat exchangers located aside the fermentation device where manure
circulates through the pipes, and heating is provided thorugh warm water. Data tells us
that only the first two systems performed in a satisfiying way.
551
5.
PURIFICATION AND STORAGE OF BIOGAS
Biogas purification procedures vary, they are insufficiently tested in practice

and mostly incomplete. It should be emphasized that the need for purification of this
gas varies and depends on the location, purpose and method of use.
The following devices are used:
for separation of solid particles,
for separation of water,
for absorption of CO2,
for absorption of H2S.
The removal of H2S and water prevents the installation from corroding, and the
ramoval of CO2 increases the caloric power of biogasa. The removal of H2S can be
carried out by using various methods: active coal, oxides of iron, iron chips, copper
chips, molecule meshes, lime and solutions (natrium hydroxide, potassium hydroxide,
lime, iron sulfates, etc.). Hydrogen sulphides are most often removed using iron
oxides.
Purifiers are made as a rectangle-shaped or circular reactor, of cast iron,
stainless steel or concrete coated with aluminium sheets. The reactor is filled with
chips or a mixture of iron sulfates, chips and lime. The flow rate of biogas through the
purifier is 30 - 60 m3/m3/h.
Biogas is stored under low or high pressure. Low pressure tanks are of the ,,bell
on water type, and for high pressure steel tanks are used.
On most existing facilities the distribution of the over-fermented rest has not
been solved in a satisfying way.
6.
USE OF BIOGAS
6.1. Preparation of biogas for use

Biogas can be used for various energy purposes, depending on the nature of the
source and the demand for this kind of energy. However, the quality level obtained
through the production process does not entirely meet specific types of its use.
If however, biogas is used as fuel for internal combustion piston engines, the
usual quality of raw biogas most often does not meet the requirements and it has to be
subjected to a modification process in order to be adjusted for specific types of use.
Biogas is mainly used for the production of thermic energy through durect
combustion, as a fuel, and the for the production of thermal and electrical energy in
cogeneration facilities, as fuel for internal combustion enginesand in the chemical
industry for the production of grime, acetylene and dry ice [1,2, 3].
6.2. Biogas for the production of thermal energy
The easiest and most widely spread use of biogas is its direct combustion in
boilers and burners. This type of use is typical for biogas produced in small family-type
digestion devices.
Direct combustion is used in burners designed for natural gas. To produce heat,
biogas can be burned at the production site or transported to the consumers thorugh
gaspipes.
552
Some Experiences in The Production of Biogas from Liquid Manure
In that regard, biogas is primarily used as a replacement for other energy

sources and it behaves like natural gas.
Water or air heating devices can successfully use biogas as a replacement for
propane-buthane. The same burners are used for this purpose, and the pressure of
biogas at the output should be as close as possible to the lower limit of the operating
pressure of that device.
The advantages of using biogas as fuel for these devices over solid and liquid
fossil fuels is that it burns entirely, without soot or ashes. The disadvantage of biogas
compared to natural gas lies in its slightly worse combustion capacity and less stable
flame, so that the flame intensity control is more difficult, especially at low flame.
6.3. Use of biogas as fuel
Biogas is used more than 50 years to drive cars or tractors. For that purpose,
biogas compressed in containers is used, which significantly lowers the economy of
use. The investements into a biogas compressing and filling facility is also significant.
Standardn machines operating according to the Otto cycle are easily modified
for using biogas and usually it needs only the carburetor to be replaced. The use with
diesel engines is achieved by replacing 10 15 % of the total amount of fuel with
gasified diesel fuel and bringing it into the cylinder for ignition, because diesel fuel
ignites easier than a misture of biogas and air. The specific consumption of biogas
3
(60% methane) in internal combustion engines is approx.. 0,65 m /kWh. To work
3
3
continuously, a 50 kW engine needs 32,5 m /h (1 m of biogas equals 0,5 kg diesel
fuel). According to practical data, the yearly technical maintenance and service of gas
engines costs approx. 1% of the engine price [2].
The use of biogas for the production of electrical energy becomes more
interesting and economically sustainable due to the development of the equipment
used for that purpose. These devices have become very economical due to the use of
waste heat, which makes about 70% of the energy losses of biogas. There are a
number of aggregates for the production of electrical energy with power ratings from 80
kW 350 kW. The existing experience has shown that the lower limit for the
economical use of biogas for producing electrical energy is an enginer power of 100
kW (1 m3 of biogas, depending on the efficiency coefficient for the device, yields 1, 6 1,9 kWh of electrical energy).
The benefits from electrical energy produces with biogas an from the waste heat
can be better visualized when you take into account that the biogas production costs
consist of 1/3 maintenance costs, 1/3 energy costs i 1/3 workforce cost.
Biogas produced on agricultural facilities (manure, agricultiural waste) besides
heating the digestion device, is also used for covering the energy needs of the facility
itself. Excessive biogas, which can be used for heating of rooms, drying of hay, crops
and vegetables, depends on a series of factors like the capacity of the facility, type of
raw material being subjected to anaerobic fermentation, season, weather conditions,
etc. To take all these into account it is necessary to make an annual production and
consumption plan for the facility.
553
6.4. Other uses of biogas

The cogeneration production of heat and electrical energy is a very good way
of using biogas. Electrical energy produced from biogas can be used for the
operationiof electrical devices like pumps, control systems and mixers. The
heatproduced is partly used to heat the digestion device, and two third of the total
energy produced can be used for other purposes.
The intesive development of methane chemistry and its penetration into the
chemical industry as one of the basic raw materials has also opened wide possibilities
for using biogas. There is a great number of procedures for the production of
chemicals or chemical products based on methane [3].
7.
CONCLUSION
The production and use of biogas play its part in resolving the issues of the
energy crisis, production of high quality natural fertilizer, environmental protection etc.
The operation of the small number of industry scale facilities is insufficiently
observed, and research results are mainly unavailable to the wider public.
The anaerobic fermentation in the digestion device fermentation device i
carried out in multiple phases.
Digestion devices are classified by shape, size, type, the material they are
made of, the mixing system used or substrate heating system, and therefore they have
to meet a number of technological requirements.
Biogas purification methods vary, they are unsufficiently tested and mostly
incomplete. Biogas is stored in low and high pressure conditions.
Biogas is most often used in the production of heat by direct combustion, as
fuel, and also for the production of heat and electrical energy in cogeneration facilities,
as fuel for internal combustion engines and in the chemical industry for the production
of soot, acetylene an dry ice.
Obviously further studies on this subject and new research should inspire
researchers, especially the younger ones, to research further keeping in mind the
significance of biogas as a new type of energy source.
LITERATURE
[1] Radii, M. (2006). Proizvodnja i primena biogasa, Via tehnika kola, Zrenjanin.
[2] ulbi, M. (1986). Biogas (dobijanje, korenje i gradnja ureaja), Novinskoizdavaka radna organizacija Tehnika knjiga, Beograd.
[3] Soldat, N. (2010). Biogas-mogunosti proizvodnje i primene(diplomski rad),
Mainski fakultet Univerziteta u Beogradu, Beograd.
554
TEMPERATURE CHARACTERISTICS OF PHOTOVOLTAIC MODULE

Aleksandar Stjepanovi1, Slaana Stjepanovi2, Ferid Softi3, Zlatko Bundalo3
Summary: This paper presents a comparative analysis of the temperature
characteristics of photovoltaic modules using the software packages PSpice and
PVsyst. Considered are the influence of radiation intensity on the temperature of solar
cells and thus on its output characteristics. Described are the changes that occur on
open circuit voltage and short circuit current of photovoltaic modules under the
influence of different radiation intensities and consequently change the temperature of
module.
Key words: Photovoltaic module, temperature characteristics, Pvsyst software, Pspice
software, Temperature coefficient, Open circuit voltage, Short circuit current.
1
INTRODUCTION
Photovoltaic module is an element that the principle of photovoltaic conversion

is converting incident sunlight into electricity. Solar cells absorb photons from sunlight
and use their energy to create electron-hole pairs. When the solar cell or photovoltaic
module is exposed to sun rays is a change of temperature solar cells and thus its
output characteristics. Part of the power committed by the photon is converted into
electrical energy and part into heat. Temperature influence on the features of the
photovoltaic modules can be analyzed using models of photovoltaic modules that take
into consideration the effects of ambient temperature and module temperature
changes under the influence of sunlight [1].
2
TEMPERATURE CHARACTERISTICS OF SOLAR CELLS
Effect of temperature on the characteristics of solar cells can be analyzed by

the impact of each size that characterizes the solar cell. Effect of temperature on the
short circuit current of solar cells can be approximated by the expression [1]:
I sc
I0 e
EG 0
k B T
qVoc
k B T
(1)
Faculty of Transport and Traffic Engineering, University of Eastern Sarajevo, Vojvode Miia 52, 74000
Doboj, Bosnia and Herzegovina, aco_stjepanovic@yahoo.com
2
Telekom Srpske, Kneza Lazara 4, 74000 Doboj, Bosnia and Herzegovina
3
Faculty of Electrical Engineering, University of Banjaluka, Patre 5, 78000 Banjaluka, Bosnia and
Herzegovina
555
Aleksandar Stjepanovic, Sladjana Stjepanovic, Ferid Softic, Zlatko Bundalo
where: kB is Boltzmann's constant, temperature independent, EG width of the energy

gap, T is temperature, q the electron charge, I0 inverse current saturation diffusion
diode solar cells.
Besides the temperature effect on the short circuit current of solar cells, it can
be shown that there is a specific effect of temperature on the open circuit voltage,
which can be calculated from the expression [1]:
Voc T
B T[
1
k T
EG 0 B
ln
I sc
q
q
(2)
where: EG is width of the energy gap, B constant, kB Boltzmann's constant, constant,

is covered by the influence of temperature on the current value I0.
3.
SIMULATION OF THE EFFECTS OF TEMPERATURE ON CURRENTVOLTAGE CHARACTERISTICS OF SOLAR CELLS
Using the basic model of the solar cell shown in Figure 1. and using the
software package PSpice to simulate the effect of temperature on IV characteristics of
solar cells [2].
Fig. 1 The basic model of the solar cell

When the solar cell exposed to sunlight leads to changes in temperature of
solar cells. Voltage current characteristic Ic = f (Vc) for silicon solar cell can be
represented by the expression:
Ic
qVc
I ph I 0 (T ) e nk T 1
(3)
where: Ic is output current from solar cells, light-generated current Iph, I0(T) the inverse
temperature-dependent saturation current, q is electric charge, k Boltzmann constant,
n the correction factor, Vc voltage solar cells, T is temperature.
Photocurrent Iph is proportional to the intensity of solar radiation G, where the constants
of proportionality. Current solar cell can be expressed through form [3]:
Ic
qVc
K E G I 0 e nkT 1
(4)
From here, the voltage solar cell can be calculated from:

Vc
556
K G Ic
nkT
ln E
1
q
I0
(6)
Temperature Characteristic of Photovoltaic Module
The optimal working point is very important for solar cells. Power from a solar cell is
the maximum for the condition that d(VcIc)/dIc = 0 Value of factor n is obtained from the
condition that the dPC / dVC = 0 and short-circuit solar cell is equal Isc = Iph, when the
voltage Vc = Vc,Opt. Taking the Isc>> I0 is obtained:
I 0 exp qVc,opt / nkT

n
K nkT I sc I 0 / nkT qVc,opt
I
q
Vc,opt 1 sc
kT
Ic
(7)
(8)
Figure 2. presents the simulation results of temperature effect on IV characteristics of

the radiation intensity of 1000W/m2, done in PSPICE software package for the value of
temperature (T = 5, 10, 27,45, 55 and 800C). An example is given for the solar cell
used in photovoltaic module type H250 produced by "Solaris" Novi Grad, Croatia that
was used for experimental measurements. Dimensions of solar cells 6.8cm x 6.8cm,
area 46.24 cm2, the current density Jsc = 34.3 mA/cm2, and Jo = 10-11A/cm2, the value
of parallel resistance RSH = 1000.
Fig. 2 The effect of temperature on the characteristics of solar cells

With graphic shows that the value of circuit voltage Voc of solar cells at a temperature
of 270C is equal to 570mV. At a temperature of 800C circuit voltage is equal to the 450
mV. Taking the difference of temperature and voltage can be calculated that the
temperature coefficient of solar cells equal to -2.3 mV/0C.
4
TEMPERATURE CHARACTERISTICS OF PHOTOVOLTAIC MODULES
The model is based on information provided by manufacturers under standard

test conditions (STC Standard Test Condition) [4], which usually differ from the
fluctuating conditions in which the photovoltaic module is [4]. The lack of such a model
can be solved using the model of photovoltaic module that takes into account the effect
of radiation intensity on the temperature of the module and thus the voltage and
current of photovoltaic modules. The current PV module depends on the intensity of
incident radiation and the temperature of photovoltaic modules [1].
557
I scM
I scMr
dI
G scM Tc Tr
1000
dT
(9)
where: ISCM is short circuit photovoltaic modules, the G intensity of incident radiation on
the surface of photovoltaic modules, IscMr short circuit at the reference temperature, Tr
reference temperature, Tc temperature of solar cells and modules.
The temperature dependence of circuit voltage can be approximated as over
the form [1]:
I
wV
VocM | VocMr ocM Tc Tr VT ln scM

I scMr
wT G
(10)
where: VocM is circuit voltage photovoltaic modules, VocMr circuit voltage of photovoltaic
modules at the reference temperature, Tr reference temperature (270C), Tc temperature
of photovoltaic modules, VT temperature potentials.
Figure 3. presents the simulation results using PSPICE software package, the
effects of different intensities of radiation on the output characteristics of photovoltaic
module H250 produced by "Solaris" Novi Grad, Croatia.
The basic technical data for the photovoltaic module H250 [5]: Number of
series-connected solar photovoltaic cells that make up the module Ns = 42, Pmax =
25W, ISCM = 1.56 A and VocM = 21.5 V (standard test conditions STC, T = 250C, AM1 ,
5G, 1000W/m2). Simulation of the influence of radiation intensity with change in
temperature of the module due to the effects the sun has been done in PSPICE
package for various intensities of incident radiation 200, 400, 600, 800 and 1000W/m2.
Fig. 3 The influence of radiation intensity on the characteristics of photovoltaic modules

Table I shows the values of temperature of solar cells for different intensities of
radiation:
558
Temperature Characteristic of Photovoltaic Module
Table I. Temperature Solar Cells With Different Intensity Of Radiation

Intensity of radiation
H250
Tc
200 W/m
32C
400 W/m2
39C
600 W/m2
45C
800 W/m2
53C
1000 W/m2
60C
The comparative analysis used the software package PVSYST 5.0. for analysis
and synthesis of photovoltaic systems.
Figure 4. presents the simulation of the influence of radiation intensity taking
into account the effect of radiation intensity on the temperature module. For each IV
curve shown in Figure 4. seen the value of incident radiation and module temperature
calculated at the same intensity of radiation. The simulation was done for the ambient
temperature of T = 25C.
Fig. 4 The influence of radiation intensity with temperature changes

Table II presents the results obtained in the simulation software package
PSPICE Figure 3. and the results obtained by simulation in a package PVSYST 5.0.
Figure 4.
Table II. Comparative simulation results pspice-pvsyst
H250-module
PSpice
PVsyst
PSpice
PVsyst
IscM
1,568A
1,62A
0,312A
0,320A
Parameters
Intensity of radiation
VocM
17,792V
1000W/m2
18,8V
20,544V
200W/m2
18,60V
559
Previous analysis of simulation results using two package PSPICE and

PVSYST, shows that there are differences obtained values of circuit voltage and short
circuit current of photovoltaic modules. Comparing the results of simulation with data
obtained from the manufacturer, the value of radiation 1000W/m2, shows that the
software package PSpice obtained value of short circuit in an amount of 1.568 A,
which corresponds to the manufacturer. The simulation results PVSYST package for
the short circuit current is 1.62 A, which deviates by 0.6 A value of data obtained from
the manufacturer module. On the other hand the value of circuit voltage for simulation
in PSPICE is 17.792 V, while the simulation PVSYST package is 18.8 V. The vendor
says the fact that the circuit voltage value for the intensity of radiation from 1000W/m2
equal to 21.5 V.
5.
CONCLUSION
The paper is demonstration that there are deviations of the results obtained
from simulation data obtained from the manufacturer. Data on photovoltaic modules,
provided by module manufacturers give a partial picture of the behavior of modules
under real operating conditions. These data are usually given for standard conditions
of testing are required for this type of device.
In order to get results for real conditions in which the photovoltaic module is
necessary to take into account the real conditions in which the photovoltaic modules in
order. Under that means the effect of different intensities of radiation on the
photovoltaic module, variable temperature, and changes in temperature of photovoltaic
modules due to effects of radiation on the module.
LITERATURE
[1]
[2]
[3]
[4]
[5]
560
Luis Castaner and Santiago Silvestre, Modelling Photovoltaic Systems using

PSpice, John Wiley & Sons Ltd, The Atrium, Southern Gate,Chichester West
Sussex, England, 2002
Aleksandar Stjepanovi, Ferid Softi, Analiza i modelovanje solarne elije
koritenjem programa PSpice, INFOTEH Jahorina, Vol. 8, E-I-21, p 441-444,
mart 2009.
F. Softi, Z. Bundalo, A. Stjepanovi, M. Gaanovi, M. Knei, . Ivanovi:
Modeling and analysis of temperature characteristics of solar cells and
photovoltaic module, Proceedings of Second International ReCIMiCo Workshop,
pp. 93-98, Novi Sad, Serbia, March 2010.
A. Stjepanovi, F. Softi, Z. Bundalo, S. Stjepanovi: Solar Tracking System and
Modelling of PV Module, Proceedings of International Conference MIPRO,
Opatija, Croatia, May 2010., pp.131-135.
Photovoltaic module H250, Solaris Novigrad, Croatia.
WIND POTENTIALS ASSESMENT IN COMPLEX TERRAIN

Predrag ivkovi1, Gradimir Ili2, Mirko Dobrnjac3, Mladen Tomi4, ana
Stevanovi5,
Summary: Wind power assessment in complex terrain is more demanding task than
site assessment in flat terrain. Modeling wind conditions with standard linear models
such as WAsP (Wind Atlas Analysis and Application Progam) does not sufficiently
reproduce wind conditions in complex terrain, ecpecially on leeward sides of terrain
slopes. Even anemometer measurements directly at a wind turbines site do not
necessarily deliver the results necessary for prediction calculations, as extrapolations
of wind speed to hub height is tricky.
A more complex model, based on Reynolds averaged Navier-Stokes equations and
modified two-equations k-H turbulence model for neutral atmospheric boundary layer
conditios, written in general curvelinear nonorthogonal coordinate system, has been
used. The sets of full conservation equations are numericaly solved by Computational
Fluid Dynamics (CFD) technique.
A comparison of the application of linear model WAsP and CFD model WindSim on
complex terrain is presented. Siting of mezzo and micro models of Selievica
(southeasterly Serbia, nearby the City of Ni) has been investigated. Considerable
discrepancies of wind speed have been obtained using linear and CFD models.
Statistics of annual electricity production varies up to 30% for Selievica site.
This paper is concerned by the National Program of Energy Efficiency, project number:
TR33036, funded by the Government of Republic of Serbia.
Key words: wind power assesment, complex terrain, CFD, WAsP, WindSim
1.
INTRODUCTION
The most assesments of the classical fuel resources, mostly fossil, clearly
marks the fact that their resources, especially for oil, are close to the end. From the
other side, the global heating and pollution problem, mostly caused by large emissions
of flue gasses from power plants and engines, arises constantly. Needs for energy
constantly rises, so even the reachest states experience the energy problems (as the
1
PhD, Predrag ivkovi, Ni, University of Ni, Faculty of Mechanic. Engineering, pzivkovic@masfak.ni.ac.rs
PhD, Gradimir Ili, Ni, University of Ni, Faculty of Mechanical Engineering, gradei@masfak.ni.ac.rs
3
PhD, Mirko Dobrnjac, Banjaluka, University of Banjaluka, Faculty of Mechanical Engineering,
dobrnjac.mirko@gmail.com
4
BSc, Mladen Tomi, Ni, University of Ni, Faculty of Mechanical Engineering, mladeninenad@yahoo.com
5
BSc, ana Stevanovi, Belgrade, Institute for nuclear sciences Vina, szana@sbb.co.rs
2
561
Predrag ivkovi, Gradimir Ili, Mirko Dobrnjac
two-day California energy system collapse). All this facts points to the necessity of
transition to the sustainable development, especially to the usage of renewable energy
sources, among which the wind energy clearly takes its place, considering its large
potentials, purity and availability. The present constrains are mostly of financial nature.
Having all that in mind, the most important task is the siting of wind turbines. For
that purpose, the wind atlas method is developed, which became the best for use with
the fast development of computers. The task itself is comparativelly simple when the
terrain considered is flat. But, in terrains with complex orography, situation is much
more complex. With change of wind turbine position of only few dozens of meters, their
potentials for energy production can be drastically altered. It can be shown that simple,
linear models (as the one used for WAsP) cant estimate correctly the wind energy
potentials in the terrain where the ruggedness index exceeds 0.3.
Terrains with complex orography can have larger potentials than flat terrains.
This is caused with the speed-up effect, which also depends on the main wind
directions. Considering all above mentioned, the necesity of using more complex, CFD
software, which is solving the complete set of conservation equatios, is obvious. The
present practice is that the change of wind turbine location is justified if the potentials
increases at least for 1%.
2.
COMMON BASES FOR ALL SOFTWARE PACKAGES
All software packages needs more or less the same input data, i.e. it is
necesary to have:
digital model of heights, with resolution up to 100 m (desirable 25 m)
roughness model (with basic roughness classes, at least)
obstacle model in the vicinity of the anemometer and the wind turbine sites
wind roses, with resolution of an hour, desired 10 minutes
wind turbine data (for the type that will be instaled)
enough computer power for simulations.
3.
MATHEMATICAL MODELLING
3.1 General model prognostic

This models involves the full set of the differential conservation equations. They
are solved using the adequate existing software packages, or by developing new ones.
On the right side of the momentum equation some source terms can be noticed. They
are: D pressure forces term, E buoancy forces term and F Coriolis force term.
Adequate terms can be find in the rest of the conservation equations.
This model was used as the base for the WindSim software package
wU
w
U U i 0

wt wx i
wU
wU j
wU i w U U i U j
w
Peff i

wx j wx i
wt
wx j
wx j
wP
U g i 2 eijk : j U k
x i ,

w
,
E
F
D
562
Wind potentials assesment in complex terrain
w U jk
wx j
w U jH
wx j
Pk
CH1
w
wx j
Q T wk
Q
V k wx j
w
wx j
Q T wH
Q
V H wx j
wU i wU j

wx j wx i
QT
CH 2
Pk H
H
k
c H 1 Pk c H 2 H
wU i
,i z j
wx j
N2
C PV H
3.2 Kinematic model

Simplifying the above set of equations by neglecting and removing some terms,
finally we came to the kinematic model. This model, except the continuity equation,
consists of the boundary conditions on the lateral sides of the domain in the logarithmic
wind profiles form, with using of the Weibull distribution for acquision of the wind speed
and direction data.
w
U U i 0
wx i
Uz
T
U* z
ln
\
N z0
T0 a z
f U
k U

AA
k 1
U k
exp
A
U k
FU exp
A
, k distribution parameters
4.
COMPARATIVE ANALYSIS OF WIND POTENTIALS ASSESMENT
Wind energy potentials on a site can be compared in many ways. Parameters

for comparison can be mean wind speed, wind power density, annual energy
production on a site etc. But, generally, all of them depends on the mean wind speed,
so the mean wind speed field is taken as the basis for comparison.
Comparison of the results will be shown on the mezzo model of Selievica,
nearby Ni. On the chosen site, farm of 20 wind turbines is sited. Wind turbine tupe is
Enercon E-48, with unit power of 800 kW. Considered mezzo model was choosen by
former simulation on the bigger model, from which, using the nesting technique, named
mezzo model is obtained.
For the turbine siting the method of wake loss minimization and maximal annual
energy production was used. Also, the recommendations about distance between wind
turbines for the siting were as follows: in the wind direction minimally 7D (D rotor
diameter) and in the normal direction 4D.
Wind turbine arrangement was the same for the simulations in both software
packages.
563
Fig. 1 Digital macro model of the terrain and the chosen mezzo model with wind rose
On the basis of the data about the terrain slopes and roughnesses, obtained
from the digital map, and the data about the wind speed and direction represented with
the long term wind rose on the site of the main meteorological station Ni, the
simulations were performed in both software packages: WAsP and WindSim. The
result are the following mean wind speed fields.
Fig. 2 Mean wind speed fields simulated in WAsP (left) and WindSim (right)
It can be noticed that the tendencies are similar, i.e., minimums and maximums
are on the same positions, but the difference in intensities is considerable, as shown
on figure 2. The range of the mean wind speed is 7.5915.54 in WAsP, while it is
4.960612.639 [m/s] in WindSim.
564
Wind potentials assesment in complex terrain
Fig. 3 Comparison of vertical wind speed profiles obtained numerically, computationally

and experimentally for wind from North (left) and the South (right). The profiles were
fitted at 30 m on the site Gtsch, nearby Andermatt [1]
This difference is the consequence of the way the software is working. On the
left side of the figure 3 it can be noticed that there are no considerable differences
when uniform wind profile flows over a wind turbine in flat terrain. The differences
starts when flow in a complex terrain. The reason is speed-up from the hills, because
WAsP does not recognize slow-up effects on the downwind side of the hill. This is
caused by the mathematical model, since this effect shows up in the second
derivatives of the conservation equations, which are not being solved by WAsP, while
WindSim solves the full set of the conservation equations.
Nevertheless, it should be kept in mind that kinematical models needs mostly
few hours for their work, while CFD models needs even few days.
Considering all above mentioned, it seems that the best solution is the
combination of these two methodologies: firstly, on the bigger model, we can find the
areas with the best potentials using the fast kinematical model and then, using the
2
nesting technique, separate the micro models in the order of few dozens km . After
that, on a smaller model for less time and higher accuracy, we can obtain more reliable
results using CFD software package.
5.
CONCLUSION
In the similar way the other fields can be compared, of which the annual energy
production field (table 1) is the most important. The big difference in numerical values
can be noticed, of about 29%, which is in agreement with the data that can be found in
the available literature.
This confirms that this specific methodology can completely be applied in our
country, which makes place for making the Wind Atlas of Serbia, and, after that, for
utilization of the potentials which our country has.
565
Table 1 Estimated annual energy production range obtained by using WasP and WindSim
AEP free = 129.546
AEP = 94.4437
[GWh/year]
[GWh/year]
frequency table
WAsP
AEP park = 128.517

[GWh/year]
WindSim
AEP = 89.0043
[GWh/god]
Weibull distribution
Wake loss = 0.79 [%]
Discrepancy = 5.76 [%]
LITERATURE
[1]
[2]
[3]
566
Schaffner B., Wind Modeling in Mountains: Intercomparison and Validation of

Models, study for METEOTEST, Bern, Switzerland and VECTOR AS, Tonsberg, N.
Stevanovi, M.. (2008). Numeriki aspekti turbulentnog prenoenja impulsa i
toplote. Univerzitet u Niu, Mainski fakultet, Ni.
ivkovi, M.P. (2006). Procena energije vetra na terenima kompleksne orografije
uporedna analiza metodologija. Univerzitet u Niu, Mainski fakultet, Ni

WIND POWER RESOURCES IN THE REPUBLIC OF SRPSKA
Ljubo Glamoi1
Summary: In addition to water potential, as the most important renewable energy
source, Republic of Srpska has significant potential and other forms of renewable
energy sources such as wind power, biomass, solar and geothermal energy. For a
quality selection of location for setting up and installing wind turbines it is necessary to
make quality wind atlas. Based on the wind atlas, it is possible to define the most
promising locations on which setting up the measurement columns and measurement
of the wind potential and fundamental characteristics is performed. Based on the
results and the preparation of the economic feasibility study, the investment decisions
on the construction of wind turbines are made.
Key words: renewable energy sources, wind energy, wind atlas,
1.
INTRODUCTION
We live in times of energy. Of course, our time can be called the time of
information systems, Internet, science or the age of climate change and global crises.
Our time carries a general term: the age of globalization. In general, we can say that in
its energy hi
story world has reached the end of an epoch, when oil cannot be the basis for
planning the future development and when the question is - what next? The answer
lies in the introduction of new energy sources as the urgent need and in a call to
intensify work on improving the usage and finding new - alternative energy sources.
There are many controversies, exaggerated marketing messages and
sometimes unrealistic expectations in making the policy of using the renewable energy
sources. Basic characteristics of renewable energy sources are:
Renewable energy sources are not cheap energy sources compared to
fossil fuels,
Certain technologies require significant space (solar energy and biomass)
There is potential for conflict of interest in food and energy production
Using renewable energy sources has a significant multiplier role in the
countries whose industry is able to produce the equipment for the use of
renewable sources Basis of European energy policy has three main
objectives, out of which the most important is sustainable development,
Ph.D. Ljubo Glamoi, Banja Luka, Ministry of Industry, Energy and Mining (lj.glamocic@mier.vladars.net)
567
Ljubo Glamoi
which includes primarily the development of renewable energy sources by

competitive conventional and, at the same time, reduced harmful emissions.
Pursuant to the new Directive 2009/28/EC of 23 April 2009 "the energy from
renewable sources means energy from renewable non-fossil sources, i.e. wind power,
solar, aerothermal, geothermal, hydrothermal and ocean energy, hydropower,
biomass, landfill gas, gas from the plants for waste treatment and biogas. Starting from
the above strategic objectives and commitments of EU Member States and the fact
that Bosnia and Herzegovina signed the Accession Agreement as well as the Treaty
on Energy Community for Southeast Europe, BiH i.e. Republic of Srpska is obliged to
transpose all the proclaimed principles and EU acquis into its legislation.
2.
THE MOST PROMISING LOCATIONS IN THE REPUBLIC OF

SRPSKA
Republic of Srpska, because of its natural features (varied relief, relatively rich
in rainfall, well-developed hydrographic network), is one of the regions with significant
potential of renewable energy sources.
Analyzing the energy potential of the Republic of Srpska it is evident that a very
small part is used for electricity generation. Republic of Srpska with 30% of use of
hydropower is one of the economic areas that use this form of energy the least. Other
sources of renewable energy such as wind power, solar, aerothermal, geothermal and
other forms are practically not used or used only sporadically.
In recent years efforts have been intensified to identify the potential for
electricity generation from renewable sources that Republic of Srpska has, including
the use of wind potential.
In this context, in co-operation with international financial institutions, a wind
atlas for the wider area of the Republic of Srpska and BiH, a part of Serbia,
Montenegro and Croatia has been done.
Analyzing the wind atlas for the entire territory of the Republic of Srpska it can
be concluded that there are areas within its territory with very promising locations for
the construction of wind parks. Geographically, southern and southeastern parts of the
Republic of Srpska are favorable for the construction of wind parks, while the western
and northwestern parts are less suitable and economically justified for this kind of
investment.
During the analysis and disposition of the data provided by wind atlas, it is
important to point out a few facts that need to be kept in mind when making investment
decisions, as follows:
Change in wind speed only in one day is very high
Additional capacity for secondary regulation is required
Additional capacity for tertiary regulation and energy balancing is required
Need for conventional power plants spare capacity due to the inability to
guarantee power of wind turbines
Only 1/3 of wind power is turned into electricity,
Life expectancy of wind turbine is 20-25 years
Specific investment is quite high and amounts to 1300-1700 Euro/kW
Review and analysis of the most promising locations in the Republic of Srpska,
where there is significant wind potential, is based on wind atlas. In the coastal
568
Wind Power Resources in The Republic of Srpska
countries along the coast or in mountain passes near the sea, the winds blow
constantly, so it is not difficult to select the location. In continental countries, notably in
mountainous areas, on only a hundred meters, leeward and fire-swept areas alternate
where the wind blows strongly. Important parameters for electricity generation are:
wind speed, decisive direction, frequency of speed, frequency of silence, air density. A
typical variation of wind is usually described by the Weibull's distribution. By summing
each speed multiplied by the probability of its occurrence we reach the average speed
in the period. According to the potential and the geographical configuration, the
Republic of Srpska's territory can be divided into several characteristic areas. It is
known that the southern part of the Republic of Srpska is located on about 20-40
kilometers from the sea, i.e. mountain massif extends from the central part to the
southeastern part that separates continental from the coastal part, so in this region
there is continental as well as Mediterranean climate. In the continental part given the
configuration there are a few mountains, the basic elements and indicators of wind
potential in those areas are analyzed.
According to this atlas, the most promising locations for additional
measurements and tests to investigate the wind potential and its use for electricity
generation, are located in the southeastern part of the Republic of Srpska. Wind
speeds in some areas of this region have an average value (observed in the period
1978 to 2007) and to 7.7 m/s (at an altitude of 50 meters), with increasing altitude
above the ground. Wind speed is slightly higher at an altitude of 120 m from the ground
and reaches an average value of up to 8.2 m/s for the observed area. Most promising
sites are located in the massif of the Kalinovik towards Avatovac and Gacko. In
continental part, areas of mountains Manjaa, Vlasi, Kozara, Ljubi Majevica and
Romanija were analyzed. Figure 1 presents the given area, where you can clearly see
the area with significant wind potential.
Fig.1 Most promising locations in the southeastern area of the Republic of Srpska
569
Ljubo Glamoi
Based on this figure, five locations that have the highest wind speed have been
selected. Location 1 is situated west of Kalinovik and Dobro Polje towards mountain
Treskavica. Part (west of 1815 '- approximately). Location 2 encompasses the area
south of Kalinovik towards the mountain Zelengori. Location 3 includes the area
northeast of Nevesinje towards the mountain Cvrnj. Site 4 includes the area south of
emerno towards the mountain Lebrnik. It is important to note that there is national
park Sutjeska around this area. The area south of Tjentite and east of emerno,
towards the mountain Magli, where there is also considerable wind potential, is mainly
located in Montenegro (east of 1840 '- approximately). Location 5 is located east of
Berkovi towards the mountain Bjelasnica. Presentation of these locations is shown in
the following figure.
The following figures present the location of Kalinovik and Dobrop polje
municipalities, for the altitude of 50m, 80m and 120m, respectively (approximately
cover the area S 43.55 - 43.62 I 18.35 - 18.45 - location 1, and S43, 35 - 43.45 I
18, 4 -18.6 - location 2):
570
Fig. 2 Wind speed diagram at 50, 80 and 120 m on the most promising locations
The following figures represent the wind rose for central area of the location 1
and a wind speed diagram in the period January - December 2007 (S43 24'49''I18
28'51''). It is evident that the wind speed reached a value of over 20 m/s.
571
Ljubo Glamoi
Fig. 3 Wind rose for the location
Fig. 4 Wind speed diagram for location
Wind rose shows that the most intensive wind is from the southwest and
months of year when the wind is the strongest are January, February, November and
December, which can be seen from the monthly chart of wind speed. Maximum wind
speed mostly does not exceed 20 m/s.
The following photos give a geographical view of the location that have the
most significant wind potential in the Republic of Srpska. It is evident that this is a very
mountainous area. Photos are created using the program Google Earth.
572
Fig. 5 Geographical view of the locations in the southeastern part of the Republic of Srpska
Apart from these locations that are more explored, less promising locations are
considered for the possibility of their use for detailed analysis. Specifically considered
here are the locations in central and eastern part of the mountains or in certain
characteristic regions. Characteristic of the less promising areas is that wind speed at
50 meters rarely exceeds 5 m/s, which is, compared to the most promising locations
where the average wind speed ranged up to 7.7 m/s, significantly lower. Areas that
have been explored and which represents possible locations are the mountains
Manjaa Kozara Majevica, Romanija.
573
Ljubo Glamoi
Table 1 Overview of the most promising locations with basic characteristics

No.
1
2
3
4
5
6
7
8
9
Location
Manjaa
Kozara
Majevica
Romanija
Treskavica
Zelengora
Cvrnj
emerno
Bjelanica
Wind speed [m/s]

for H = 50 m
4,8
4,2
4,6
4,8
6.6
6,4
5.6
6,2
6,6
Wind power
W/m2
260
140
150
240
340
360
300
320
340
Wind
direction
South-West
East-West
South-West
South-West
South-West
East-West
South -West
South-East
South-West
Months when the

wind is strongest
November-February
November-March
December-February
November-March
November-March
November-March
November-January
October-March
October-February
INSTEAD OF CONCLUSION
From the above mentioned it is evident that the Republic of Srpska has some
potential for power generation from wind power. According to detailed analysis it is
possible to built wind turbine of about 600 MW. On prospective locations it is possible
to build wind turbines to 200 MW. Pursuant to the Directive 28/09, in Regulation on
renewable sources and cogeneration the indicative target by 2020 is established, that
is possible to install wind turbines of 200 MW, and that will encourage the electricity
produced from 100 MW. At this stage, the activities are carried out on the adoption of
bylaws which will determine feed-in tariffs to be applied in order to improve all forms of
renewable energy sources. At this point, the Republic of Srpska has no installed
capacity for electricity generation using wind power, which represents a significant
challenge for interested investors.
REFERENCES
[1]
[2]
[3]
574
Wind atlas Sander + Partner, 2010.

Mr. Grani, How to improve energy future?, 2010, Business Diary, Zagreb
www. izvorienergije.com / energija_vjetra
COVENANT OF MAYORS FORESEEN ACTIVITIES AT DISTRICT

HEATING COMPANY TOPLANA A.D. BANJA LUKA
Milovan Kotur1, Gostimir Radi2
Summary: During 2009, the City of Banja Luka joined the agreement Covenant of
Mayors. By signing this Covenant, the City has undertaken to decrease emissions of
CO2 in 2020 by 20% compared to 1990. One of the planned projects with the aim to
achieve the set target is modernization and reconstruction of the district heating
system in the City that will result in decrease of consumption of fuel (heavy oil) by 4500
tons, i.e., decrease of emissions of CO2 by 14,020.92 tons, which will in the year of
2020 amount to around 4.95% of the total foreseen decrease of CO2 emission.
Key words: Covenant of Mayors, District Heating Company Toplana a.d. Banja Luka,
modernization, CO2 emission reduction.
1. INTRODUCTION
Cities and urban areas home to 50% of the worlds population are
responsible for up to 75% of the global Greenhouse Gas (GHG) emissions. They cover
less than one percent of the earth's surface but 75% of all energy is consumed by the
worlds urban population [1].
The Covenant of Mayors is the mainstream European movement involving
local and regional authorities in the fight against climate change. It is based on a
voluntary commitment by signatories to meet and exceed the EU 20% CO2 reduction
objective through increased energy efficiency and development of renewable energy
sources [2].
During 2009, the City of Banja Luka joined the agreement Covenant of Mayors.
One of the planned projects with the aim to achieve the set target is modernization and
reconstruction of the district heating system in the City.
Toplana a.d. Banja Luka is the single provider of heating services in Banja
Luka, and, according to the available data, it provides services to around 30% of
households as well as commercial and administrative buildings. Toplana a.d. Banja
Luka supplies heating energy for heating of premises in Banja Luka only during the
heating season, which starts from 15th October and lasts until 15th April. Supply of
heating energy lasts, on average, around 188 days and the heating energy is delivered
daily from 6 a.m. to 22 p.m. During this time, Toplana a.d. Banja Luka tries to maintain
1
Master of Science in Mechanical Engineering, Milovan Kotur, Banja Luka, Faculty of Mechanical
Engineering, (koturm@gmail.com
2
Graduated engineer of Mechanical Engineering, Gostimir Radi, Banja Luka, Toplana A.D.
575
Milovan Kotur, Gostimir Radi
the indoor temperature at +19C (+/- 1C). Outline of heat supply area of Toplana a.d.
Banja Luka is shown by Figure 1.
According to the analysis of district heating at the City of Banja Luka, conducted
during the work on the study [3] 2006/07, the assessed annual efficiency of different
parts in the chain of production and distribution of heating energy amounted to:
Heating energy production efficiency 80%,
Heating energy loses from the network 11%,
Hot water loses from the network 13%.
Estimated efficacy (total produced energy in relation to the energy supplied to
consumers) of the system amounted to around 60%, while the efficacy of similar
systems in the Western Europe amounted to around 80%.
In addition to others, the study also contains the following conclusions:
There were loses in fuel in the work of Toplana a.d. Banja Luka because of
the lack of control systems as well as there were loses in energy due to
huge network leakages,
Complete modernization of the heating system has contributed to saving of
around 27% in consumption of fuel, 93% in consumption of water and 70%
in consumption of electrical energy.
Fig. 1 Area outline for delivery of heating energy and the main site of Toplana [3]
According to data delivered by Toplana a.d., in the heating season of 1989/90
heating of the City of Banja Luka was done by Toplana a.d. and partly though the Incel
-Energetika company. Total consumption of heavy oil in the stated heating season
amounted to 19,300 tons, while 209,100 MWh of heating energy was supplied to
consumers. Out of the supplied energy, 80% or 167,280 MWh was delivered to
residential customers, while the rest or 41,820 MWh was delivered to commercial
consumers (Municipal buildings and equipment/facilities - administrative and other
576
Covenant of Mayors Foreseen Activities at District Heating Company Toplana A.D. Banja Luka
facilities under the jurisdiction of the City of Banja Luka 10 %, and Tertiary /non
municipal/ buildings, equipment/facilities - buildings that are not under the jurisdiction
of the City of Banja 10 %).
The total connected heat load on district heating network in 1991/92 amounted
to 265 MW (all the consumers, including the industry.
According to collected data [4], district heating sector participated in 1990 in the
total emissions of CO2 by 8.8 %.
2
MODERNIZATION PLANS OF INIDIVIDUAL PARTS OF THE DISTRICT

HEATING SYSTEM
2.1 Boiler House

The total installed capacity of the boiler house is 246 MW (4 boiler of 58 MW
capacity and 2 of 7 MW boilers), of which 232 MW is currently operational. The boiler
house operates using heavy oil fuel. Due to the lack of proper control systems and, as
well as due to daily interruptions in the heat supply, the boilers operate at low average
efficiency.
Modernization plan foresees the following activities:
Purchase of new oil burners, pumps, engines and fans,
Installation variable speed drives to main engines,
Installation of local automation,
Installation of heat meters at the boiler house,
Installation of equipment for gas (oxygen) removal from the make-up water,
The objectives of implementing the modernization are:
Improved combustion of fuel and decrease of emissions of combustion
products into the air,
Fuel and energy savings,
Cost savings in maintenance and operation (staff),
Improved operation and safety.
Implementing the modernization of boiler houses envisages to achieve savings
in consumption of fuel until 2020 (compared to projected consumption without the
implementation of modernization) of 1,170 t heavy oil fuel, which is around 13,066
MWh of thermal energy, which would result in a reduction in CO2 emissions of around
3,645.42 tons.
2.2 Primary and secondary distribution network
The total length of district heating network is 155 km (the primary network is
around 45 km of pipelines, while the secondary network is around 110 km). The share
of modern pre insulated pipes is about 5% of the total network length.
The network is generally in bad condition and suffers from external and internal
corrosion. Due to network damages, significant part of water for additional filling is
constantly added to the network. To illustrate the poor condition of the network, it is
indicated in [3] that the annual consumption of water at Toplana a.d. Banja Luka is 45
times bigger than its typical rate in Western Europe.
Because of increasing the heating energy consumption in the City, due to
connecting of new consumers, it will be necessary, in the future, to increase the
577
capacity of the primary pipeline. This increase in capacity is best carried out together
with the reconstruction of the primary pipelines.
The plan of reconstruction foresees the following activities:
Replacement of the worst network sections,
Replacement of the valves in the manholes and other components.
The objectives of the implementation of reconstruction are:
Reduced heat and water losses
Improved reliability and quality of heat supply
Reduced maintenance costs
It is envisaged that, with reconstruction of network, the savings in 2020 are to
be realized in fuel consumption (compared to projected consumption without the
implementation of reconstruction) in the amount of 2565 t of heavy oil fuel (primary
network 1395 t of heavy oil fuel, secondary network 1170 t), which amounts to around
28,644.7 MWh of the heating energy (primary network 15,578.7 MWh, secondary
network 13,066 MWh), which would result in a reduction in CO2 emissions of around
7991.87 tons (primary network 4346.47 tons, secondary network 3645.4 tons).
2.3 Heat Substations
According to data given in the Study [3], the total number of heat substations is
209. Out of these, 188 heat substations are operational, while the remaining 21, which
served old industrial facilities, are now out of operation. About half of the operational
heat substations are owned by Toplana a.d. Banja Luka and the others are owned by
consumers. The existing heat substations are old - fashioned, using outdated
technology.
It is proposed to modernize all the heat substations that are connected to the
Toplana a.d. district heating network and that are operational. For this purpose it is
necessary to carry out:
Replacement of control valves and automatic regulators,
Replacement of heat exchangers, if necessary,
Installation of water flow meters between primary and secondary networks,
Installation of heat meters,
Automation.
The objectives of implementing the modernization are:
Cost savings,
Improved comfort and customer service,
Measurement of heat consumption of each group of buildings supplied by
each substation,
Monitoring of water consumption in secondary networks and buildings.
With modernization of heat substations, it is envisaged that in the 2020, the
savings realized in fuel economy (compared to projected consumption without the
implementation of modernization) of 720 t fuel oil, which is around 8,040.64 MWh of
thermal energy, which would result in a reduction in CO2 emissions of around 2,243.33
tons.
2.4 Installation of heat meters in buildings
According to data obtained in the Study [3], only about 5% of heat sales are
based on heat meter readings. It is proposed to install heat meters in all customer
buildings to facilitate monitoring of heat consumption in each building, and, in that way,
578
Covenant of Mayors Foreseen Activities at District Heating Company Toplana A.D. Banja Luka
creating the conditions for payment based on actual consumption of heat energy. In
addition, it will be possible to monitor the heat losses in the secondary district heating
network networks, helping to locate pipelines needing replacement.
As a result of implementing the mentioned activities, it is envisaged that in the
2020, the savings realized in fuel economy (compared to projected consumption
without the installing of heat meters) of 45 t fuel oil, which is around 502.54 MWh of
thermal energy, which would result in a reduction in CO2 emissions of around 140.3
tons.
2.5 Plans for development of the district heating system in the period of 2010 to
2020
In the draft of the Urban Plan of the City Banja Luka for the period from 2008
until 2020 [4], building of new plants in urban area and their connection to the district
heating system of Banja Luka are planned. In the following period, the plan is to
prepare district heating studies for certain city areas, which are not currently covered
by the district heating system, which should give responses about the size of the
heating consumption, the type of energy source that could be used in planned heating
plants, the possibilities using of renewable energy sources (geothermal, biomass, etc.)
in the City district heating system, etc.
3
CONCLUSION
Total investment funds needed to invest in the district heating system of Banja
Luka, to achieve the planned reduction in CO2 emissions in 2020, as stated in the
document Sustainable Energy Action Plan of the City of Banja Luka, amounts to 36.5
million Euro (Figure 2 and 3).
Fig. 2 Provided investments in the

district heating system of City Banja
Luka, in the Action Plan, until 2020 in
millions of Euros per sectors [4]
Fig. 3 Provided investments in the

district heating system of City Banja
Luka, in the Action Plan, until 2020 per
sectors in percentages [4]
As a result of modernization and reconstruction of the district heating system in

2020, a decrease of consumption of heavy oil fuel of 4,500 tons will be achieved in the
district heating system (Fig. 4 and 5), which corresponds to 50,253.88 MWh of heating
energy (Fig. 6), that is, this will result in a decrease of 14,020.92 tons of CO2 emissions
(Fig. 7), which will, in 2020, amount to around 4.95% of the total foreseen CO2
emission reduction.
579
Fig. 4 Foreseen decrease in consumption

of heavy oil fuel in the district heating
system in the City of Banja Luka, as a
result of implementation of the Action
Plan in 2020 per sectors in tons [4]
Fig. 5 Foreseen decrease of consumption of heavy oil fuel in the district heating
system of the City of Banja Luka as a
result of implementation of the Action
Plan in 2020 per sectors in percentages
[4]
Fig. 6 Foreseen saving of heating energy

in the district heating system of the City of
Banja Luka as a result of implementation
of the Action Plan in 2020 per sectors in
MWh [4]
Fig. 7 Foreseen decrease of CO2

emissions in the district heating system
of the City of Banja Luka, as a result of
implementation of the Action Plan in
2020 per sectors in tons [4]
Toplana a.d. Banja Luka is the largest District Heating Company witch use
heavy fuel oil as fuel in the Republic of Srpska and in B&H. Similar efficacy (total
produced energy in relation to the energy supplied to consumers) of the district heating
systems has most other District Heating Companies in the Republic of Srpska, B&H
and surrounding country which use fossil fuels, particularly heavy oil. Conducted
analyses showed great potential for savings in fuel consumption and CO2 emission
reduction by performing modernization of district heating system and that mean less
the environmental impact and less price of energy delivered to consumers.
LITERATURE
[1]
[2]
[3]
[4]
580
UNFCCC Enviromental Integrity Group (EIG), (2009). Cities and climate change.
Covenant of Mayors, from http://www.eumayors.eu, accessed on 2011-05-10
Makela, T., Sohail, H., Zachodni, K., Litterick, G., Hunt, J. (2007). Banja Luka Hot
Water Leakage Prevention and Environmental Enhancement Project.
UNDP BiH. (2010). Sustainable Energy Action Plan for Banja Luka.
PARTIAL REPLACEMENT OF HEAVY FUEL OIL WITH BIOMASS IN

THE DISTRICT HEATING COMPANY IN GRADIKA
Milovan Kotur 1, Branko Usorac2 Petar Gvero3, Gordana Tica4
Summary: Bosnia Herzegovina ratified the Kyoto Protocol in April 2007. BiH is a nonannex 1 country, which means that BiH can become eligible to use the Clean
Development Mechanism (CDM). The CDM helps industrialized countries to reduce
the costs of greenhouse gas abatement, but also helps developing countries in
achieving sustainable development. The proposed CDM Programme of Activities will
support fuel switch projects (fossil to bioenergy). The main activity of the proposed
small-scale CPA in Public Communal Company, Toplana A.D. Gradika is installation
of a new 6 MW biomass boiler in the boiler house. With implementation of this project,
the District Heating Company in Gradika could reduce the existing consumption of
heavy fuel oil during the heating season for about 1080 t, and reduce emission of CO2
for 3247 t/yr.
Key words: Kyoto Protocol, CDM Programme of Activities, small-scale CPA, district
heating, wood biomass boiler, emission of CO2.
1. INTRODUCTION
The Norwegian Ministry of Foreign Affairs financed a Capacity Building
Programme for developing greenhouse gas reduction projects in the district heating
sector in Bosnia and Herzegovina from autumn 2009 to December 2010. The final
result of this project is a development of PDD documentation for one potential small
scale CDM project and one Programme of Activities (PoA).
Today, most of the boilers in the district heating sector are fuelled with fossil
fuels in the Republic of Srpska and Bosnia and Herzegovina. On the other hand, the
Republic of Srpska has a large potential of biomass utilization. About 2371062 ha, or
40% of Bosnia and Herzegovina is covered with forest, and about 1250391 ha, or 53%
of this are is in the Republic of Srpska [4].
Since most of the fuel switching projects are not financially justifiable for the
project owners due to a long payback period, it would be an advantage to gain
1
Engineering, (koturm@gmail.com)
2
Bachelor of Engineering (Mechanical), Branko Usorac, Gradika, JKP Toplana a.d. Gradika,
(toplanagradiska@gmail.com)
3
Associate Professor, Petar Gvero, Banja Luka, Faculty of Mechanical Engineering, (petargvero@yahoo.com)
4
Assistant Professor, Gordana Tica, Banja Luka, Faculty of Mechanical Engineering, (ticagordana@yahoo.com)
581
Milovan Kotur, Branko Usorac, Petar Gvero, Gordana Tica
revenues from the Clean Development Mechanism. Due to this fact, the idea is that the
Fuel switch to bioenergy projects in Bosnia and Herzegovina will coordinate the SmallScale Programme of Activities (SSC-PoA) and will support the project implementers in
implementing the CDM Programme Activities (CPAs) in Bosnia and Herzegovina
through collaboration with district heating companies, industrial processing companies
and other possible stakeholders with fuel switch projects (fossil to bioenergy) in Bosnia
and Herzegovina [2].
The District Heating Company in Gradika was founded in 1980. The total
number of buildings supplied by the Public Communal Company, Toplana A.D.
Gradika, is about 1740 (residential buildings, public buildings such as kindergarten,
schools etc. and other facilities). The vast majority of these, about 50%, are residential
apartment buildings. Heated floor area in residential buildings is about 75 000 m2.
The Public Communal Company Toplana A.D. is the only provider and
distributor of district heating in Gradika. It produces heat in a central boiler house,
consisting of two 11.8 MW boilers with a combined capacity of 23.6 MW. The boilers
are fired by heavy fuel oil, and the total connected heat load in the town is about 16.8
MW. During the heating season, the Public Communal Company Toplana A.D.
Gradika, uses only one boiler because the existing disposition of pipelines in the
boiler house does not enable simultaneous operation of both heavy fuel oil boilers.
Average annual fuel consumption during the last three heating seasons (20082010) is about 1516 tonnes of heavy fuel oil, and heat supplied to the district heating
network is about 13,35 GWh/yr. Consumption of heavy fuel oil has been increasing
each year because of connection of new customers to the existing district heating
network. The proposed project activity includes installation of a new biomass boiler in
the boiler house. The biomass boiler should be used during heating seasons for
covering the base heat load, while one of the two present heavy fuel oil boilers will be
used for covering peak heat load, when the heat capacity of the biomass boiler is not
large enough to cover needed heat consumption.
2. TECHNICAL DESCRIPTION OF THE SMALL-SCALE CPA
The District Heating Company in Gradika intends to install a new 6 MW wood
biomass boiler for production of thermal energy for heating residential and commercial
facilities in Gradika. The new biomass boiler will be installed within the existing boiler
house of the company. During the heating season, the biomass boiler will provide the
base heat load. In that way, the Public Communal Company Toplana A.D. Gradika
has estimated less consumption of heavy fuel oil (which is currently the only fuel for
production of thermal energy) by approximately 1080 tonnes annually.
As part of the project, the wood biomass boiler will be connected with the
existing boilers in a parallel function enabling the use of both heavy fuel oil boilers for
covering peak heat load during the coldest winter days. As a result of implementation
of this project, the new installed heat capacity in production will be:
6 MW + 11,8 MW + 11,8 MW = 29,6 MW.
Biomass fuel should be transported by a truck from the local Forestry Company
or local biomass factory, about 30 km to a storage area, which will be built close to the
existing boiler house. The amount of transported biomass will be supported by
invoices.
582
Partial replacing heavy fuel oil with biomass in the District heating company in Gradika
The capacity of storage area will be big enough for two days consumption of
biomass in the biomass boiler. From the biomass storage, fuel will be transported to
the silo and then to the biomass boiler for combustion. Flue gases from biomass
combustion will be transported to the cyclone and then to the existing chimney. Hot
water from the biomass boiler will be sent to the accumulator tank and then delivered
to customers through the existing district heating network.
Heat energy delivered from the biomass boiler and heavy fuel oil boilers will be
measured by heat meters.
Disposition of equipment in the boiler house after the project implementation is
shown in Figure 1. The scheme of the boiler house after project implementation is
shown in Figure 2.
Fig. 1 Disposition of equipment in the boiler house after the project implementation [3]
583
Fig.2 Schema of boiler houses after project implementation [1]
584
Partial replacing heavy fuel oil with biomass in the District heating company in Gradika
3. BARRIERS FOR PROJECT IMPLEMENTATION

3.1 Investment barrier
As new buildings are constructed in Gradiska, they are required by the local
law to be connected to the district heating system if possible. Currently, the system is
already undersized for the peak demand in the network, and therefore would not have
the capacity for future demand. One of the possible solutions for this problem could be
modification of the system to enable both oil fired boilers to work simultaneously.
With the expansion of current production capacity on heavy fuel oil, the district
heating company will have:
Less investment in production equipment,
Experience and routines for operation and maintenance,
Lower operating and maintenance costs, excluding energy costs.
The main barriers for installing new biomass boilers are:
High investment costs,
Higher operating and maintenance costs.
District heating companies in Bosnia and Herzegovina are largely municipality
owned companies that provide heat to both commercial and residential sectors. District
heating boilers in the country are largely fuelled by fossil fuels, in this case heavy fuel
oil that incurs relatively high energy costs. The cost of heat to customers is regulated
by local municipal regulations. However, due to the high oil costs, heavy subsidy is
required to ensure sustainability of district heating companies. As the companies are
operating in a negative profit, investments to expand the system are not typically
viable. Initial economy calculation has showed that the proposed project activity could
not access appropriate capital without consideration of the CDM revenues.
Tabele 1 Initial economic evaluation in Euro [1]
Cost of investment
Annual operating costs incurred by the
project
Annual revenues incurred by the
project
Annual net savings
Economic lifetime
Payback time
IRR
Net Present Value
2,874,443.00
1,079,966.00
1,529,600.00
449,634.00
25
without CDM
with CDM
6 years and 11
months
6 years and 5 month
7.42%
9.08%
NPV = 29,618.00 NPV= 255,656.00
3.2 Technological barrier

Traditionally, boiler systems in Bosnia and Herzegovina have been either coal
or oil fired, and biomass fired boilers are relatively new in the market. Therefore, local
585
equipment and service providers of biomass boilers are limited, which in turn means
the investment and maintenance costs of such boilers are greater.
As the company currently operates oil boilers exclusively, there is an
associated business risk in the training of personnel on the operation and maintenance
of the boiler. There is also additional risk incurred by the potential overrun in time and
money associated with the installation and commissioning of a new biomass boiler due
to the lack of local expertise.
4. CONCLUSION
Installation of a new biomass boiler in the District Heating Company in
Gradika will directly contribute to the reduction of greenhouse gas emission produced
by the combustion of fossil fuels. With implementation proposed project activity, the
District Heating Company in Gradika could reduce the existing consumption of heavy
fuel oil during the heating season by 1080 t, or approximately 180 t of heavy fuel oil per
heating month. Increase in the installed capacity of boilers in the boiler house will
enable connection of new customers to the existing district heating network but also
the expansion of district heating network to other parts of the Gradika town which
have not yet been covered with the district heating network.
This project also aligns with the strategy of development Republic of Srpska
and Bosnia and Herzegovina, which promotes the use of renewable energy. Other
benefits that the project will contribute to are as follows:
Adjoining communities will directly benefit from the improvement of local air
quality by displacing the combustion of fossil fuel with biomass (that
produces less smoke and noxious gases),
Support of the development of biomass-based energy sources in the Host
County,
Increasing the competitiveness of the District Heating Company in
Gradika, consequently ensuring long term employment for the local
population, employed in the JKP Toplana A.D.,
Promotion of technology transfer to BiH, as the necessary equipment and
knowledge to operate it will be provided by foreign suppliers,
Reduction of fossil fuels dependency in BiH,
Reduction of CO2 emission for 3247 t/yr.
LITERATURE
[1]
[2]
[3]
[4]
586
JKP Toplana A.D. Gradiska, Norsk Energi Oslo, Department for Thermal
Engineering, Faculty of Mechanical Engineering University of Banja Luka. (2010).
Partial replacing heavy fuel oil with biomass in the District heating company in
Gradika, PoA CPA Gradika
JKP Toplana A.D. Gradika, Toplana A.D. Prijedor, Norsk Energi Oslo, CETEOR
Sarajevo. (2010). Boiler conversion to bioenergy in BiH, PoA BiH
RUDIS d.d. Trbovlje, Slovenia, (2008). Plan mainskih instalacija i mainske
opreme, Toplana Gradika.
Energetski institut Hrvoje Poar Zagreb, Ekonomski institut Banja Luka, (2010)
Plan razvoja energetike Republike Srpske do 2030. god.
THE BIOMASS PROJECT IN THE DISTRICT HEATING COMPANY

(DHC) IN PRIJEDOR, BIH
Milovan Kotur1, Zoran Kneevi2, Petar Gvero3, Gordana Tica4
Summary: Bosnia Herzegovina ratified the Kyoto Protocol in April 2007. BiH is a nonannex 1 country, which means that BiH can become eligible to use the Clean
Development Mechanism (CDM). The CDM helps industrialized countries to reduce
the costs of greenhouse gas abatement, but also helps developing countries in
achieving sustainable development. The proposed CDM Programme of Activities will
support fuel switch projects (fossil to bioenergy). The main activity of the small-scale
CPA is to replace heavy fuel oil with wood biomass. In the District Heating Company
Prijedor, one of the existing boilers of 30 MW will be reconstructed in order to use
wood pellets. With implementation of this project, the District Heating Company in
Prijedor could reduce the existing consumption of heavy fuel oil during the heating
season for about 4901 t, and reduce emission of CO2 for 14 381 t/yr.
Key words: Kyoto Protocol, CDM Programme of Activities, small-scale CPA, district
heating, wood biomass boiler, emission of CO2.
1.
INTRODUCTION
The Norwegian Ministry of Foreign Affairs financed a Capacity Building

Programme for developing greenhouse gas reduction projects in the district heating
sector in Bosnia and Herzegovina from autumn 2009 to December 2010. The final
result of this project is a development of PDD documentation for one potential small
scale CDM project and one Programme of Activities (PoA).
Today, most of the boilers in the district heating sector are fuelled with fossil
fuels in the Republic of Srpska and Bosnia and Herzegovina. On the other hand, the
Republic of Srpska has a large potential of biomass utilization. About 2371062 ha, or
40% of Bosnia and Herzegovina is covered with forest, and about 1250391 ha or 53%
of this area are in the Republic of Srpska [3].
Since most of the fuel switching projects are not financially justifiable for the
project owners due to a long payback period, it would be an advantage to gain
1
Engineering, (koturm@gmail.com)
2
Bachelor of Electronics and Tele-Communication Engineering, Zoran Kneevi, Prijedor, Toplana A.D.
Prijedor, (knezevic.zoki@gmail.com)
3
Associate Professor, Petar Gvero, Banja Luka, Faculty of Mechanical Engineering, (petargvero@yahoo.com)
4
Assistant Professor, Gordana Tica, Banja Luka, Faculty of Mechanical Engineering, (ticagordana@yahoo.com)
587
Milovan Kotur, Zoran Kneevi, Petar Gvero, Gordana Tica
revenues from the Clean Development Mechanism. Due to this fact, the idea is that the
Fuel switch to bioenergy projects in Bosnia and Herzegovina will coordinate the SmallScale Programme of Activities (SSC-PoA) and will support the project implementers in
implementing the CDM Programme Activities (CPAs) in Bosnia and Herzegovina
through collaboration with district heating companies, industrial processing companies
and other possible stakeholders with fuel switch projects (fossil to bioenergy) in Bosnia
and Herzegovina [2].
Toplana A.D. Prijedor, the district heating company (DHC) is a main producer
of heat for the town of Prijedor and it covers nearly 320 000 m2 of building surface for
heating. DHC is located in the outskirts of the town. Installed heat power is 2x30 MW
via two boilers. Total connected heat load in the town is about 30 MW (the second
boiler is technical reserve). Annual heat energy production is approximately 50 GWh.
DHC uses heavy fuel oil for combustion. The main activity of the small-scale
CPA is to replace heavy fuel oil with wood biomass. One of the existing boilers of 30
MW will be reconstructed in order to use wood pellets. This boiler will be the base load
boiler, while the other existing boiler will be reserve and peak load boiler. The needed
wood pellets will be produced by the DHC and it is a part of the project.
2.
TECHNICAL DESCRIPTION OF THE SMALL-SCALE CPA:
The entire project consists of two technical and technological units that are
closely related to each other:
Reconstruction of one of the existing heavy fuel oil boilers (with 30 MW
capacity) to biomass boiler and wood pellets as a primary fuel.
Production of wood pellets. The wood pellets will be used as fuel in the new
biomass boiler. Excess in production of pellets will be sold on the market
2.1 Production of pellets
Production of wooden pellets includes a complete introduction of the new
technological line for production of wood pellets (Figure 1). The wood pellets will have
the following parameters: 6 mm diameter, 10% of moisture, 1% of ash, and 5 kWh/kg
calorific value.
The capacity of the technological line for production of wood pellets will enable
production of 4 t of pellets per hour. This capacity will be sufficient for the continuous
production of heat during the heating season. In addition, pellets will be produced
outside the heating season and all production surpluses will be sold on the market.
Raw materials used for the production are wooden sawdust, waste wood and
wooden logs that are categorized as firewood. Warehouses for the reception of raw
3
materials are located near of the boiler house and have the capacity of 20 000 m .
The technological line will consist of several primary machines:
machine for grinding material (hammer type),
machine for drying material,
engine cooling,
interconnection conveyors and separators,
pellet machines,
silo for storing the finished pellets,
peripheral equipment.
588
The Biomass Project in The District Heating Company (DHC) in Prijedor, BiH
The complete technological line will be located in a separate building. The

warehouse can provide storage for 1000 tons of pellets.
Fig. 1 The new technological line for production of wooden pellets [1]
2.2 Reconstruction of 30MW boiler
Reconstruction of the heavy fuel oil boiler, type MKLV, 30 MW power with the
aim to use biomass requires finding a solution for various problems such as purchase
of the appropriate new burners with needed equipment, system for pneumatic transport
of wood dust, wood pellet mill / dust, dozer for pellets, silo to store pellets etc (Figure
2). Calculations show that the existing source of preheating air 40 000 m3/hr is
sufficient for combustion of wood dust. The existing system for removal of flue gases
will be used. Main problems are removing ash and length of flame. Heavy fuel oil
burners have flames with smaller lengths for the same power than wood dust burners
because of the difference in calorific fuel value. Calorific value of heavy oil is about 43
MJ/kg whereas it is about 19 MJ/kg for wood dust with 10% of moisture. Because of
these problems, significant changes will be necessary in the construction of the boiler,
and the existing boiler will have a role of heat exchanger.
589
Fig. 2 A block diagram of the new technological plant [1]

Energy from wood pellets will replace energy from 4901 t of heavy fuel oil. The
total amount of pellets needed per heating season is about 11272 t. To produce this
amount of pellets, DHC in Prijedor should provide at least 25362 m3 spatial raw wood
with 50% humidity.
3.
EMISSION REDUCTIONS
Emission reductions were calculated by using next equation [1]:

ERy = BEy PEy - LEy
Where:
ERy (tCO2) - Emission reductions in the year y,
BEy (tCO2) - Baseline emissions in the year y,
PEy (tCO2) - Project emissions in the year y,
LEy (tCO2) - Leakage emissions in the year y.
Estimated amount of emission reductions over the chosen crediting period (10
years) is shown in the Table 1.
Table 1 Estimated amount of emission reductions over the chosen crediting period [1]
Year
Year 2012
Year 2013
Year 2014
Year 2015
590
Estimation
of baseline
emissions
(tonnes of
CO2 e)
16 438
16 438
16 438
16 438
Estimation
of project
activity
emissions
(tonnes of
CO2 e)
1980
1980
1980
1980
Estimation
of leakage
(tonnes of
CO2 e)
77
77
77
77
Estimation of
overall emission
reductions
(tonnes of CO2 e)
14 381
14 381
14 381
14 381
The Biomass Project in The District Heating Company (DHC) in Prijedor, BiH
Year 2016
Year 2017
Year 2018
Year 2019
Year 2020
Year 2021
Total
4.
16 438
16 438
16 438
16 438
16 438
16 438
164 380
1980
1980
1980
1980
1980
1980
19 800
77
77
77
77
77
77
770
14 381
14 381
14 381
14 381
14 381
14 381
143 0
BARRIERS FOR PROJECT IMPLEMENTATION
4.1 Investmet barrier

Initial economy calculation (Figure 3) has showed that the proposed project
activity could not access the appropriate capital without consideration of the CDM
revenues. This is due to the high investment costs needed for reconstruction of the
heavy fuel oil boiler as well as investment in the pellets production facility. The largest
share of the investment is related to the pellets production. However, buying pellets is
not an alternative, due to small difference in price. According to the current price of
heavy fuel oil and pellets at the market, annual heavy fuel oil costs amount to about 3
430 700 KM and annual costs for the purchase of pellets are about 3 430 700 KM.
Fig. 3 Cash flow analyses without and with CDM [1]

4.2 Technological barrier
Traditionally, boiler systems in Bosnia and Herzegovina have been either coal or
591
oil fired, and biomass fired boilers are relatively new in the market. Therefore, local
equipment and service providers for biomass boilers are limited, which in turn means
the investment and maintenance costs of such boilers are greater.
As the company currently operates oil boilers exclusively, there is an associated
business risk in the training of personnel on the operation and maintenance of the
boiler as well as on the pellets production line. Also, a huge amount of wood will
needed to be provided for plant operation during the whole heating season so forestry
must improve their technology, road access to woods, transportation equipment etc.
5.
CONCLUSION
Reconstruction of one of the heavy fuel oil boilers in the District Heating
Company in Prijedor to a biomass boiler will directly contribute to the reduction of
greenhouse gas emission produced by the combustion of fossil fuels. With
implementation of the proposed project activity, the District Heating Company in
Prijedor could reduce the existing consumption of heavy fuel oil during the heating
season by 4901 t. Installation of a new technological line for production of wood pellets
will create new jobs in DCH Prijedor, but also in the local forestry company which
should be responsible for collection and ensure sufficient amount of biomass for
continuous production of wood pellets.
This project also aligns with the strategy of development Republic of Srpska
and Bosnia and Herzegovina, which promotes the use of renewable energy. Other
benefits that the project will contribute to are as follows:
Adjoining communities will directly benefit from the improvement of local air
quality by displacing the combustion of fossil fuel with biomass (that
produces less smoke and noxious gases),
Support development of biomass-based energy sources in the Host County,
Increasing the competitiveness of DHC, consequently ensuring long term
employment for the local population employed in the DHC and new
permanent employment for at least 15 people just in DHC and more people
in forestry,
Promotion of technology transfer to BiH, as the necessary equipment and
knowledge to operate it will be provided by foreign suppliers,
Reduction of fossil fuels dependency in BiH,
Reduction of CO2 emission for 14 381 t/yr.
LITERATURE
[1]
[2]
[3]
592
Toplana A.D. Prijedor, Norsk Energi Oslo, Department for Thermal Engineering,
Faculty of Mechanical Engineering University of Banja Luka. (2010). Biomass
project in District Heating Company (DHC) in Prijedor, BiH.
JKP Toplana A.D. Gradika, Toplana A.D. Prijedor, Norsk Energi Oslo, CETEOR
Sarajevo. (2010). Boiler conversion to bioenergy in BiH, PoA BiH.
Energetski institut Hrvoje Poar Zagreb, Ekonomski institut Banja Luka, (2010).
Plan razvoja energetike Republike Srpske do 2030. god.
D. TRAFFIC MEANS

MALICIOUS ECOLOGY ON VEHICLES AND TRAFFIC EXAMPLE
Dr in. Stojan Petrovi, Dr in. Boidar Nikoli, Dr in. Emil Hnatko, Dr in. Jovo
Mra, dr in. Stevan Veinovi
Summary: During the epoch of human progress and ecology there is a big interest
for malware ecology as well protection against them. Identifying the entire value stream
for each product is the first step in ecological thinking. By improvement a technical
facilities we degrease energetic and ecologic environment overload. For example,
vehicles are the cases where to positive effect of complex energy balance over the
propulsion system in them. Around 60 million vehicles are produced per year and 800
million vehicles are in traffic today in the world. The biggest power has been situated in
transport sector: up to 15 times bigger of all energy resources in the EU and up to 10
times bigger of all energy resources in Serbia.
Such huge number of vehicles has, of course, a certain negative influence on nature
and human environment. The overall pollution of environment air, water and soil
enforcement genetic bases uncontrolled changes wherever the entire world of our
Planet exists. The global conclusion is that there is no chance to change the road fleet
from fossil fuels to electric power.
Dr.in. S T O J A N P E T R O V I
Dr. in. B O I D A R N I K O L I
Dr in. E M I L H N A T K O
Dr.in. J O V O M R A
Dr. in. S T E V A N V E I N O V I
ZLONAMERNA EKOLOGIJA
U OBLASTI VOZILA I DRUMSKOG TRANSPORTA
MALICIOUS ECOLOGY
OF VEHICLE AND ROAD TRANSPORT
595
ENGINEERING
SCIENTISTS
Study the world:

as it is!
ENGINEERS
Create the world:
that has
never been!
S
C
I
E
N
C
E
1. Is the discipline and

profession of applying
technical and scientific
knowledge and
2. Utilizing natural laws
and physical resources
in order to design and
3. Implement materials,
structures, machines,
devices, systems, and
processes that realize
a desired objective and
meet specified criteria.

GLOBAL VEHICLE PRODUCTION SEGMENT OVERVIEW

AMERICANS DRIVE AN AVERAGE OF 13,500 miles PER YEAR,
WHILE THE FRENCH AVERAGE FEWER THAN 9,000 miles,
THE DUTCH 8,000 miles, AND
THE JAPANESE ONLY 6,000 miles ANNUALLY
Dramatically higher fuel prices in Europe and Japan cause

consumers to purchase smaller, more fuel-efficient vehicles
than in the United States.
Small cars make up 64 percent of all new car purchases,
compared to 29 percent in the U.S.
Gas prices are much higher in Europe and Japan: For
example, in Europe, where the price of fuel is as much as 2
times higher than in the U.S.

596
Malicious Ecology On Vehicles And Traffic Example
President Bush Directs

Administration To Take The First
Steps Toward Regulations Based
On
"Twenty In Ten" Goal.
Twenty Years from Now...

2025Dream
Sedan

37CIDFuelConverter?
MagicAireExhaust?
ContinuouslySmooth
Transmission?
AllWheelRegenerative
Braking?
xxxmpg
.00ygpmNOx
.00zPM
?
Goal is Zero Emissions

US California
FreedomCAR
January 2001.
the US Department of Energy and USCAR
announced a new partnership designed to initiate
research that will enable a transition to
a hydrogen transportation
economy while preserving transportation freedoms.

597
Last year, president George W. Bush proposed

$1.2 billion in research funding towards a bold
vision of a transportation system powered by
hydrogen.
"With a new national commitment, our
scientists and engineers will overcome
obstacles to taking these cars from laboratory
to showroom, so that the first car driven by a
child born today could be powered by
hydrogen and be pollution-free," he said.
We fully support the president's vision.
The learning we gain from this
program will provide direction for nextgeneration fuel cell programs. And our
Focus Fuel Cell vehicle will showcase
Ford's leadership in developing these
technologies."
Detroit According to the recently released 2007

Environmental Protection Agency (EPA) Fuel Economy
Guide, General Motors leads the industry with more
vehicles that achieve 30 mpg (9.4 lit/100 km), on the
highway, than any other manufacturer.
Mike Schwarz, director of Sustainable Mobility
Technologies, Ford Motor Company:
The president's (G.W. Bush) vision of a child born today
driving a hydrogen-powered vehicle is both compelling and
timely.
Delivering this vision is the responsibility of the

FreedomCAR program. FreedomCAR is the joint
government and auto industry research
partnership announced by Department of Energy
Secretary Spencer Abraham.
The "Freedom" in FreedomCAR represents:
Freedom from petroleum dependence
Freedom from pollutant emissions
Freedom to choose the vehicle you want
Freedom to drive where you want and

when you want
Freedom to obtain fuel

affordably and conveniently !!!

598
EPA recently revised methods for

estimating vehicle fuel economy to
better represent current driving styles
and conditions.
The new methodswhich apply to
model year 2008 and later vehicles
include the city and highway.
That program is overseen by the Department of Transportation, not by EPA,

and it involves different calculations.
Averaged over all vehicles of all types, city mileage ratings will drop
12% and highway numbers will fall 8%.
The new mileage tests don't affect calculations for what's known as CAFE,
Corporate Average Fuel Economy, or the average mileage of all vehicles
sold by an automaker. Averaged together, all cars an automaker sells
must hit 27.5 miles per gallon (10 lit/100 km) or the car company
has to pay a penalty to the government.
Trucks must hit 21.6 mpg (13 lit/100 km).
Ironically, EPA has for years been conducting tests along the lines of the
new fuel economy tests. But it was doing so only to gather emissions
information and wasn't sure it could simply use the data for fuel economy
estimates without Congressional approval !?
Get active!
We're the last generation that can stop it from reaching
a catastrophic tipping point.
Climate change kills 160,000 people every year.
Within 50 years, one-third of all species could face
extinction.
Why Esso is the world's number one climate criminal: Esso
actively undermined the Kyoto Protocol!

599
Charli E. Coon, J.D.,

is Senior Policy Analyst for Energy and
the Environment in the Thomas A. Roe
Institute for Economic Policy Studies
at The Heritage Foundation
USA FEATURES:
Why the Government's CAFE Standards for Fuel Efficiency
Should Be Repealed, not Increased
1. CAFE standards endanger human lives!

2. CAFE standards fail to reduce consumption!!
and
3. CAFE standards do not improve the
environment!!!
What is EUCAR?
The strategic and competent Association for
Automotive R & D in Europe
http://www.acea.be/eucarinternet/
European Approach to CO2 Reduction

CO2 g/km
CO2perate
CO2 R&D
190
140
120
1998
2004
Year
1995
2003
Review
2008
2012

600
REGULATIONS ON REDUCING TOXIC

UNREALIZEDEUROPEANFORECASTS
(Accordingtotheautooil"2program) EMISSIONS IS REGULARLY FOLLOWED
BY THE INCREASE IN FUEL
CONSUMPTION
Figuresshowsasimultaneousdecreaseinemissionoftoxiccomponents
withadvancedcatalytictechnologies.
Alsovisibleistheincreaseintheamountofcarbondioxideorinother
words,thefuelconsumption.Inpassengercarsthenewecological
regulationsresultinincreasedfuelconsumptionbyabout10%andinthe
trucksupto15%.
Justsuchaconclusionconfirmedbyfigures!
HYDROGEN: NOT ZERO POLLUTION

601
THE WELL-KNOWN CARBON DIOXIDE

CONCENTRATION CO2 = 0,038 %

Milutin MILANKOVI
cycles

CAFE:CORPORATE AVERAGE FUEL
ECONOMY
USA- CAF
The Corporate Average Fuel
Economy (CAFE) are
regulations in the United States,
first enacted by US Congress in
1975, and intended to improve
the average fuel economy of
cars. Canada have the weakest
standards in terms of fleetaverage fuel economy rating
e.g. 25 mpg in the US, versus
45 mpg in the European
Union!
EU- CAF
The Clean Air For Europe
initiative has led to a thematic
strategy setting out the
objectives and measures for the
next phase of European air
quality policy.

602
VEHICLES TREND
Ecological standards for vehicles demand that the
consumption of fuel should be minimal and that it
must change according to the laws of nature. It is
natural sesation.
That means:
the fuel consumption
has been
proportional
to the velocity.

603
DiesOtto: benefits of a diesel&otto engine

The Mercedes-Benz DiesOtto
a.- Germany: Der Diesotto-Motor ( 2011 Daimler AG);
b.- English: Combined Combustion System oder Combined Combustion
Engines
is an experimental automobile engine that is said to incorporate the benefits of
a diesel engine, but runs on regular old unleaded.
Its name is a portmanteau of Diesel and Otto, referring to the Diesel engine,
first introduced by German engineer Rudolf Diesel, and the Otto cycle, as a
reference to Nicolaus Otto (even though the invention of the gasoline engine is
variously attributed to both Nicolaus Otto and Alphonse Beau de Rochas).
The testing and demonstration vehicles include a long-wheel based MercedesBenz S-Class fitted with a 19 bhp (14 kW) electric motor to hide the transition
from spark to compression power, and the engine has been installed in the
Mercedes-Benz F700 concept car shown at the 2007 Frankfurt Motorshow.
Mercedes-Benz debuts DiesOtto

The concept engine has an in-line, four-cylinder configuration and
displaces 1.8 litres. Its peak power is no less than 175 kW
(235 hp), making its specific output 97 kW/L (130 hp/L), and its
peak torque is 400 Nm, making its specific torque 220 Nm/L.
Mercedes-Benz do not specify the RPM at which these specified
maxima are reached.
The DiesOtto features the following:
Variable compression ratio
Direct fuel injection
HCCI -Homogeneous Charge Compression Ignition
Variable valve timing
Twin variable geometry turbocharging.

ENGINE OF FUTURE- FUTURE OF ENGINES

604

FUTURE
vehicles,
engines,
emissions
and
LUBRICANTS
isues.
Always
one
step ahead!

605
FUTURE TRAFFIC WITHOUT CLASICAL FUEL

REFINERIES
PETROL= DIESEL = CRUDE OIL
ONE ENGINE FOR ALL FUELS - ONE FUEL FOR ANY ENGINES

606
A COMPARATIVE ANALYSIS OF CNG AND HYBRID BUSES VS

DIESEL BUSES
Zlatomir ivanovi1, Zoran Jovanovi2, eljko akota3
Summary: The paper presents a comparative analysis of CNG and DE (dieselelectric) hybrid buses vs conventional diesel buses on the basis of the collected and
analyzed data from different literature sources which are then presented as a survey.
The analysis includes capital and operation costs during life cycles of different bus
samples.
Key words: city buses, CNG, hybrid, comparative analysis, costs.
1
INTRODUCTION
Buses are responsible for 8% of carbon monoxide (CO), 6% of hydrocarbons

(HC), 9% of nitrogen oxide emissions (NOx), 7.7% of particulates (PM) and 5% of
carbon dioxide (CO2) emissions. The release of CO2 into the atmosphere occurs
whenever fossil fuel is burned and contributes to the greenhouse effect and global
warming. Thus thermal engine vehicles are responsible for 22% of all CO2 emissions in
the EU, with only 5% generated by buses and coaches [1].
The bus is a very efficient mode of public transport, which is cheap, flexible
and, in many cases, tailored to the needs of end-customers both in terms of capacity
and speed. Buses do not require any infrastructure except a depot and workshop, and
are easy to put in service. From an economic, environmental, and social point of view,
the bus still remains the most universal solution for a balanced and sustainable urban
development. Some 80% of all public transport passengers worldwide are carried by
buses.
Clean and energy efficient vehicles have an important role to play in achieving
the EU policy objectives of reducing energy consumption, CO2 emissions, and pollutant
emissions.

1
Dr Zlatomir ivanovi, Institute of Nuclear Sciences VINA, Department for I.C. Engines and Vehicles,
11001 Belgrade, E-mail: zzivanovic@vinca.rs
2
Dr Zoran Jovanovi, Institute of Nuclear Sciences VINA, Department for I.C. Engines and Vehicles,
11001 Belgrade, E-mail: zoranj@vinca.rs
3
Dr eljko akota, Institute of Nuclear Sciences VINA, Department for I.C. Engines and Vehicles, 11001
Belgrade, E-mail: sakota@vinca.rs
607
NEW TECHNOLOGY FOR PUBLIC TRANSPORT
The fundamental problem in limiting emissions of heavy-duty diesel engines is

to simultaneously decrease both NOx and particulates. Early injection timing reduces
the amount of particulates and fuel consumption, but increases NOx-emissions. The
NOx-PM dependency can be broken only by using exhaust after-treatment technology.
The key question is which technology to use for reducing NOx-emissions. The
formation of NOx can be limited either already in the combustion chamber
(recirculation of exhaust gas, EGR -technology) or by using exhaust after-treatment
technology (urea catalyst, SCR -technology). After-treatment will most likely be needed
for controlling particulate emissions when using EGR -technology (catalyst, particulate
catalyst or particulate filter).
A wide range of non oil-based options for road and air transport has been
developed in the last decade, and some technologies are already commercialized.
However, it is currently impossible to predict which technologies will emerge as the
front-runners for Europe. Five technological mainstreams are discussed today [1]:
Biofuels,
Natural Gas and LPG,
Hybrids,
Hydrogen and fuel cells, and
Battery Electric Vehicles.
The cleanest bus technologies available on the market today are CNG, diesel
Euro V, DE hybrid and electric vehicles. While CNG and diesel Euro V appear to be
the most ripe, and DE hybrid seem to be the most promising both for the emission
levels and the energy efficiency performance, electric vehicles have a more limited
range.
Finally, the very few hydrogen fuel cell powered vehicles circulate mainly,
when not exclusively, as prototypes. Under these premises, and also according to their
experience and expectations for the future implementations of urban sustainable
transport policies, CNG and DE hybrid are the two technologies to be addressed to for
a joined procurement of clean buses [2].
2.1 CNG bus technology
Up to the early 2000 years, compared to diesel buses the CNG technology
already offered better performance in relationship to particulate emissions. In the year
2003 natural gas buses managed to reach the EEV emission standard level, before
diesel V did so. On the contrary, in the near future diesel Euro V and CNG buses
performances will not differ much from each other.
In the recent past, at present and in the near future, the cost trend of the fossil
fuels - compared to the diesel cost - has favoured, is favouring and will favour the CNG
technology. In addition, public transport companies, whose natural gas filling station
dimensions have been optimally planned, will further benefit from the possibility of
saturating the filling capacity still at disposal.
As far as the vehicle cost is concerned, compared to diesel Euro V buses CNG
vehicles cost around 35,000 to 50,000 euros more, while maintenance costs are
getting year after year comparable. The above mentioned considerations match the
local public transport companies which have already invested in CNG bus fleets.
608
A Comparative Analysis of CNG And Hybrid Buses vs Diesel Buses
Another item to highlight refers to biogas as an alternative fuel to CNG

produced in landfill or water cleaning plants. Biogas represents the solution to energy
dependency, which is also economically convenient.
Natural gas is considered one of the most potential alternative fuels.
Worldwide there about 12,5 million NGVs supported by a network of 18 400 fuelling
stations. In Europe there are 1.305.579 NGVs (143.908 are buses or about 11%) [3].
2.2 Hybrid bus technology
Hybrid vehicles appear in the immediate future to represent a possible
alternative technology. The technology is indeed regarded by manufacturers as a way
to dramatically improve bus environmental performances.
Hybrid electric vehicles (HEV) have two sources; the addition being an electric
motor powered by an energy storage device, maintained at an optimal energy level by
the primary source, the engine. This configuration lowers the demand placed on the
primary power source, enabling both sources to operate at optimum efficiency.
While DE hybrid systems are gaining in popularity, vehicles using this
technology are still relatively expensive compared to most other alternatives, with a
total unit cost of over $500,000 for a full-size heavy-duty hybrid city bus, approximately
$200,000 more than a conventional diesel bus of similar design [4].
Operating costs of hybrid-electric buses also tend to be higher than other
alternatives currently, due to the added expense of battery maintenance and
replacement. Battery technology is advancing rapidly, however, which will likely drive
down both the capital and operating costs of hybrid-electric buses while increasing
their performance and durability.
3
COMPARATIVE ANALYSIS OF CNG AND DE HYBRID BUS VS DIESEL BUS
A general comparison of buses designed by applying different contemporary

technologies is based on several parameters, technological, financial, environmental
and planning-based, such as reliability, employment flexibility, fuel price, range,
exhaust gas emissions, noise, extra infrastructural needs. The considerations can be
summarized in table 1 [2]. Advantages of each compared technology are featured in
green, disadvantages in red.
Another comparison is based on a study of city bus life cycle cost (LCC),
analysis for 2007 model year buses [5]. It covers the folowing bus types: diesel buses
using ultra low sulfur diesel (ULSD), compressed natural gases (CNG) buses, and DE
hybrid buses. LCC factors included capital costs (bus procurement, infrastructure, and
emissions equipment) and operation costs (fuel, propulsion-related system
maintenance, facility maintenance, and battery replacement) available from the
literature.
A bus 12-year life cycle cost (LCC) analysis for a fleet of size of 100 buses was
performed based on information available in the literature, manufacturers
specifications, and fuel economy data gathered by WVU (West Virginia University).
Only technology-dependent factors relevant to bus propulsion were considered;
driver and management cost were excluded. Bus price, equipment and infrastructure
cost (to support novel technology), fuel cost, propulsion-related systems maintenance,
facility maintenance, and hybrid bus battery replacement were considered. Buses were
609
assumed to operate at a national average speed of 20.5 km/h, to travel for 60.000 km
per year, and to seat 40 passengers for the purposes of calculation.
Table 1 The comparison between the selected technologies
Diesel
Experienced
High range
Pollution (PM + NOx)
Noise
Increasing fuel costs
Dependence on mineral oil
CNG
Fuel filling station
infrastructure: new one
to be built
saturation of the
existing filling station
capacity
Reduced range
Low emission / no PM
Low fuel costs (74
ct/kg)
Higher vehicle costs
(+30.000 )
Dependence on
imported gas
Fig. 1 Capital cost per bus
610
DE Hybrid
High range +
Low emission
Reduced consumption
Reduced CO2
Reduced noise
Higher vehicle costs
(+150.000 )
Not yet experienced in
Europe
A Comparative Analysis of CNG And Hybrid Buses vs Diesel Buses
Fig. 2 Operational cost per bus
Fig. 3 Life cycle cost per bus per kilometer

In the case where only 20% of the bus procurement cost was considered, as a
result of 80% subsidies, all bus types had a sufficiently similar life cycle cost that
changes in fuel cost and battery technology could affect their relative positions in the
scale of cost, and all technologies were competitive on a cost basis.
4
CONCLUSION
Current developments in alternative bus technology are very promising. On the

basis of data taken from different literature sources it can be concluded that
development of alternative drive systems for buses is directed towards CNG and
hybrid technologies. At present, they satisfy current standards as regards harmful
gasses emissions and are commercially available.
611
While prototypes and small demonstration fleets may operate successfully,

further development toward technological maturity and operational efficiency requires
production of larger series of innovative buses with improved reliability and lower cost.
The overall impression was that diesel-electric and hybrid systems are a useful
and necessary stage in the development of electric bus and it is expected that these
technological developments will provide a platform for, and naturally integrate, other
future innovations such as fuel cell technology.
When considering or selecting bus technologies, it is important to recognize
that fuel economy and emissions depend strongly on bus route and bus operation
conditions.
LITERATURE
[1]
[2]
[3]
[4]
http://ec.europa.eu/transport/urban/studies/doc/2007_urban_transport_europe.pdf
http://www.compro-eu.org/doc/COMPRO_deliverable_D2.3_final.pdf
The GVR (Gas vehicles report), Volume 9 #10, Number 107, December 2010,
http://ntl.bts.gov/lib/32000/32700/32779/DOT_Climate_Change_Report__April_2010_-_Volume_1_and_2.pdf
[5] http://www.fta.dot.gov/documents/WVU_FTA_LCC_Final_Report_07-23-2007.pdf
612
ADJUSTMENTS OF VEHICLES WITH GAS DRIVE

Milan Milovanovi1, Dragoljub Radonji2, Saa Jovanovi3
Summary: Sequential systems for gas injection are applied on modern vehicles with
gas drive. Systems of different producers which have the possibility for auto-calibration
or manual adjustment are used. The paper presents the advantages of particular
systems at vehicle programming. The adjustment of vehicle is realised at standstill and
in the course of driving. The analysis of the adjustment effects was carried out by
means of computer and was conditioned by the possibilities of applied system for gas.
The results of analysis indicate the necessity for additional adjustment of system for
gas.
Key words: automobile, LPG, adjustment.
1. INTRODUCTION
All vehicles which use liquid fuels can be adjusted to application of LPG, in dual
fuel system. Application of bi-fuel systems on petrol engines is a long-term solution.
The main property of these systems is that the engine runs either on petrol or on gas.
Fig. 1 shows the scheme of one system for LPG injection. One more electronic unit for
gas is added to the existing electronic unit of the system for petrol injection, and that
unit makes the connection with the existing petrol system. System for gas makes
possible injection of gas into each intake manifold separately. According to defined
procedure, the engine starts running with fuel at start, and after certain period of time
or achievement of given temperature the control unit switches the engine running to
gas. The principle of functioning of electronic control unit for gas is calculating of gas
injection time, applied on gas injectors. Control of engine running is given to petrol
control unit, while electronic control unit for gas turns that information into suitable
control form for gas.
2. PROGRAMMING OF SEQUENTIAL SYSTEMS FOR GAS
When programming the electronic control unit of the sequential system for gas
injection, auto-calibration, which had been recommended by the system producer, is
usually used for initial adjustment of the system. The auto-calibration procedure, as
1
dr Milan Milovanovi, ZASTAVA VEHICLES GROUP, Kragujevac, SERBIA, siljakg@sbb.rs

prof. dr Dragoljub Radonji, Faculty of Machanical Engineering Kragujevac, SERBIA, drago@kg.ac.rs
3
Saa Jovanovi, ZASTAVA VEHICLES GROUP, Kragujevac, SERBIA, piter@ia.kg.ac.rs
2
613
well as the quantity of data used thereat, can vary a lot, which is conditioned by:
requests of the importer, i.e. service provider-installer, market demands i.e. attestissuing organization, system possibilities i.e. device quality, system price, regulations
of the country-installer etc.
Fig. 1 Scheme of system for gas
Fig. 2 Allowed system adjustment
For analysis of the influence of programming of sequential systems for gas

injection without auto-calibration, the system Sequent 24 of company BRC was used,
which does not provide the auto-calibration option; instead, adjustment of the system is
realised according to the procedure defined by the producer. The initial adjustment
includes adjustments at:
idle stroke
idle stroke with extra load (light, air-conditioning etc)
at 3000o/min, with load.
The initial adjustments are realized at a standstill, after installing the system in
the vehicle, i.e. after installing the system on engine in given case. The following
interventions on the investigated system are possible in the case of system
programming optimization:
at idle stroke
with load.
The system for gas Sequent 24 of company BRC does not allow other options
in the adjustment procedure, see Fig. 2. That fact can be facilitating for a less
demanding installer, but quite limiting in developmental system optimization. The
results of realized investigations showed that there are not many possibilities in the
optimization process of such a system. Consequently, the price of such a system is
considerably higher than the price of other systems.
In sequential systems for gas with auto-calibration the initial adjustments are
realized at a standstill, too, after installing the system for gas in vehicle. In general, the
adjustment of system configuration is realized when the engine is not running. After
that, the auto-calibration procedure is started, where adjustment is performed
according to the demands of the system producer. After the completion of autocalibration procedure, each system will define its system map, as shown in Fig. 3. The
map is constructed on the basis of two pieces of data: engine rpm and petrol injection
timing. In case that you are not satisfied with auto-calibration, possible modifications
can be realized by optimization of map value on gas, i.e. map coefficients loaded
614
Adjustments of Vehicles with Gas Drive
during automatic calibration process, both at minimum and beyond idle stroke range. In
order to modify the values, you simply mark one or several fields, press ENTER on the
keyboard after which three options for correction will appear: absolute, linear and in
percents (Fig. 4).
Fig. 3 System map
Fig. 4 System optimization
After selecting Switching from PC, the new icon - Switch - will appear on
the right side, which enables direct control of switch for gas, in the cabin, via computer.
When testing the vehicle on the road, simply click on Switch in case that you wish to
switch from petrol to gas or from gas to petrol. When you have completed the testing,
click on End switching from PC, see Fig. 5.
Fig. 5 Optimization at a standstill or on the

road
Fig. 6 Auto-calibration effects
3. OPTIMIZATION OF VEHICLE WITH GAS DRIVE

The optimization of running of engine or vehicle with installed sequential
system for gas injection can be realized by means of measuring on:
engine brake
dynamometer rolls and
on the road.
The paper will present the effects of vehicle optimization in the course of
driving, since it is the only feasible adjustment for most installers of system for gas.
615
Adjustment of vehicle running on the road is necessary as an additional adjustment.

For that purpose, test track was defined within the factory range, where the adjustment
optimization was performed in the following conditions:
Acceleration from as standstill (I-IV), vehicle runs on gas and petrol at the
same time
Test of vehicle elasticity, IV gear (0-120-40 km/h)
Drive on the track.
Figures 7 and 8 show the parameters measured in the first acceleration test.
The advantage of petrol as drive fuel, in considerably shorter acceleration time, can be
observed. Good congruence of acceleration time and good vehicle behaviour of the
vehicle were obtained in the test.
Fig. 7 Acceleration from a standstill (I-IV)

petrol
Fig. 8 Acceleration from a standstill (I-IV)

LPG
Fig. 9 and 10 show the results of vehicle elasticity test. Good vehicle behaviour
with gas drive was obtained, without jerks, with somewhat worse properties compared
with petrol drive.
Fig. 9 Vehicle elasticity test at running on

petrol
Fig. 10 Vehicle elasticity test at running

on LPG
After the completion of the first two tests, correction of programming was
performed. Optimization results are shown in Fig. 11. Adjustment optimization is
616
Adjustments of Vehicles with Gas Drive
completed with one such test for most installers of system for gas. After all specified
adjustment forms, data on adjustment effects regarding vehicle performance cannot be
obtained; instead, adjustment quality is conditioned by experience of system adjusters.
Fig. 11 Results of final vehicle test
Fig. 12 Comparative
presentation of speed properties
of petrol-LPG
In serial installing, effect of adjustment regarding vehicle performance and

especially regarding exhaust gases emission must be determined. With the purpose of
investigating the influence of system for LPG injection on engine with plastic intake
manifold TU5JP, installing of system Sequent 24 of company BRC, Italy, was
performed. Speed properties were recorded for petrol and LPG concurrently. Fig. 12
shows the obtained results. It can be observed that the change of speed properties
occurred with LPG drive, which had been expected, bearing in mind the programming
of system on gas. The programming itself had to be modified in relation to
recommended one, and had to be adjusted to test conditions. At recommended
programming of system, according to the producers instructions, speed properties
were slightly degraded.
The following step in optimization of the system for gas, in case of serial
installing, is optimization of vehicle running on dynamometer rolls. The methodology for
measuring power on vehicle wheel enables quick and accurate comparative measuring
with the aim of observing the influences of various modifications (which we made on
the vehicle) on force speed properties of the vehicle. The procedure makes it possible
to use the measured values of power on wheel in order to calculate the external speed
properties of the engine with sufficient accuracy (with information on transmission
usefulness degree and speed ratio in gearbox). The advantage is the fact that without
uninstalling the engine from the vehicle this important data is obtained, relevant for any
analysis related to the vehicle.
The measuring was realized on vehicle Zastava Florida with system OMEGAS
of company Landi Renzo. Fig. 13 presents the comparative results of measuring power
on wheel for petrol/LPG. Slight decrease of power can be observed in higher regimes,
since it is a well adjusted system.
617
Fig. 13 Comparative presentation of the

results of measuring on rolls for petrol/LPG
Fig. 14 Vehicle Citroen C1[3]
A vehicle Smart Fortwo 700cm3 45kW Turbo Euro 4, the system Sequent 24
LPG of company BRC, was installed. Fig. 14 shows the obtained results. Small
degradation of power was obtained due to application of LPG. The fact that it is a small
vehicle must be taken into consideration.
The presented investigation results indicate that degradation of vehicle
properties occurs when gas is applied. By careful adjustment, force properties of
vehicle almost identical to ones obtained with petrol application can be accomplished,
with considerable increase of consumption (up to 25% according to the
recommendations of the system producer). This must be taken into consideration when
searching for compromise in final vehicle adjustment.
4. CONCLUSION
The paper presents the investigation results for sequential systems for liquid
petroleum gas. System quality can influence the engine running considerably. In well
adjusted systems, there is little space for additional adjustments. Some devices allow
major additional adjustments. The optimization process is additionally influenced by
buyers requests, which complicates the process as well. Adjustment itself must be
adjusted to the possibilities of the system for gas.
LITERATURE
[1]
[2]
[3]
618
Milovanovi, M., Petrovi, S., Vitoevi, N., Spasojevi, S. (2006). Application of

gas as alternative fuel. ECO VARNA 2006 Conference Proceedings, p. 9-15.
Milovanovi, M., Dimitrov, A., Vitoevi, N., Uremovi, . (2007). Experience in
application of LPG on vehicles, ECO VARNA Conference Proceedings, p. 9-16.
BRC advertising material.
APPLICATION OF A CYCLIC BOUNDARY CONDITION FOR CFD

SIMULATIONS OF A 2-CYLINDER IC-ENGINE
Dalibor JAJCEVIC1 and Raimund ALMBAUER2
Summary: Computational Fluid Dynamics (CFD) has become an essential tool in the
design and development process of engineering devices. In the last decades, CFD has
also become a state of the art tool for the development of internal combustion engines
(IC-engines). It offers successful assessment of new technologies, e.g. new fuel
preparation methods, new combustion concepts and/or alternative fuels. The
simulation of an IC-engine is one of the most challenging applications of CFD
modelling and it consumes a lot of computer resources. Furthermore, for an efficient
simulation of more cylinders IC-engines periphery engine parts and all cylinders have
to be included in calculations. Growing calculation domains increase the calculation
time non-linearly and make them impracticable in the daily work of an engineer in the
engine development process. This paper covers the discussion of a methodology for
the efficient CFD simulation of a 2-cylinder 2-stroke high-performance engine using the
cyclic and time delayed boundary condition. This boundary condition allows a reduction
of the calculation domain and therewith of the calculation time. The boundary condition
data are calculated during the simulation and are set on the corresponding face in the
calculated domain. The essential advantage of the boundary condition is a reduced
simulation effort for cyclic and time delayed technical problems. For the IC-engine
simulation, this means that the complete simulation of the 2-cylinder engine can be
carried out without the geometry of the second cylinder.
Key words: 2-stroke, 2-cylinder, more cylinders, IC engine, CFD simulation
1. INTRODUCTION
The continuous improvement of numerical methods together with the increase
of computer power allows the simulation of ever more complex technical problems
even for bigger calculation domains. Growing calculation domains combined with the
simulation of physically complex phenomena increase the simulation effort non-linearly
and makes them impracticable in development process of IC-engines. The decrease of
simulation time can be realized by the reduction of the calculation domain and/or by
1
Dipl.-Ing. Dr.techn., Dalibor JAJCEVIC, Graz University of Technology, Christian Doppler Laboratory for
Thermodynamics of Reciprocating Engines, jajcevic@ivt.tugraz.at
2
Ao.Univ.-Prof. Dipl.-Ing. Dr.techn., Raimund ALMBAUER, Graz University of Technology, Christian Doppler
Laboratory for Thermodynamics of Reciprocating Engines, raimund.almbauer@tugraz.at
619
Jajcevic and Almbauer
the simplification of the technical problem. The physical phenomena in the simulation
have a big influence on the quality of results and simplifications may lead to a
decrease of simulation accurancy. Furthermore, the physically complex phenomena
are the main focus of almost all simulations in order to gain a better understanding of
the investigated technical problem. A non-calculated 3D domain can be replaced with
corresponding boundary conditions, which usually stems from 1D calculation or from
measurement data. These data are also incomplete and cause an inaccuracy in the
simulation and often lead to incorrect results. For instance, the constant boundary
conditions for the outlet face of the 2 stroke engine are usually defined with pressure
and temperature or with mass flow and temperature. These data are generally
approximated to be constant in space, which influences the results. Moreover, the
boundary conditions for the turbulence from a one dimensional (1D) model or from
measurements are not available. Hence, the boundary conditions are placed far from
the interesting domain, which directly increases the calculation time due to the
increase of considered domain. Here presented cyclic and time delayed boundary
condition allows a reduction of the calculation domain and therewith of calculation
duration. The boundary condition is calculated during the simulation and is set on the
corresponding face in the calculated domain. An important advantage of the proposed
method is the consistency of the fluxes of the individual conserved variables. It can be
used for technical problems, where a flow pattern repeats in successive cycles which
additionally can be time delayed. The data need not to be constant over the boundary
face, which is usually the case in the standard boundary conditions of all CFD
applications. In special applications it avoids data from a 1D code or measurements,
which additionally reduces the pre-processing time of the CFD simulation.
2.
TECHNICAL PROBLEM DESCRIPTION
The presented cyclic and time delayed boundary condition is a combination of

a cyclic and a symmetry boundary condition without pressure drop function, together
with a time delay option. There are technical problems, where a flow pattern repeats in
successive cycles and is additionally time delayed. A good example is a 2-stroke 2cylinder engine, where the second cylinder shows a time delay of 180 CA. Further,
cylinder 1 and 2 are coupled together via an exhaust system. The exhaust system has
an important influence on the scavenging process and for this reason on the mass and
the composition of the charge. The intake and the exhaust ports are open
simultaneously over a long period of the cycle and the gas dynamics inside the
exhaust system and the rest of the engine determine its performance concerning fuel
economy, emissions, and noise.
State of the art for the simulation of 2-stroke engines is the simulation of only
one or even a half of the cylinder, the exhaust, the intake ports and the crank case by
the use of the conventional symmetry boundary condition (e.g. simulation of 2-stroke
engines [2], [3], [4], and [5]). In case of an asymmetric design of the exhaust port a
conventional symmetry boundary condition cannot be used. Even if the exhaust port of
a 2-cylinder engine is symmetric, the simulation domain cannot be reduced to only one
cylinder, as the influence of the second cylinder is missing. The consideration of only
one cylinder causes a wrong gas dynamics inside the exhaust system and further
leads to unrealistic simulation results. Hence, in order to get effective and significant
620
Application of a cyclic boundary condition for CFD simulations of a 2-cylinder IC-engine
results for a 2-cylinder 2-stroke engine, the simulation of the complete geometry with
both cylinders and the unaltered exhaust pipe is required. E.g. [6] includes the
complete geometry of a 2-cylinder 2-stroke engine with both cylinders to simulate the
in-cylinder flow and the gas dynamics inside the exhaust system. Furthermore, the
simulation of a 2-stroke engine requires the calculation of several revolutions until the
gas dynamics inside the exhaust port achieves a cyclic steady condition. Hence, the
simulation consumes a lot of calculation time. E.g. [7] shows that a cyclic converged
solution is approached in about 10-12 engine cycles in the simulation of a single
cylinder 2-stroke engine.
3. CYCLIC AND TIME DELAYED BOUNDARY CONDITION
The basic idea of the cyclic and time delayed boundary condition introduced in
this work is a reduction of the calculation domain by applying the symmetry condition
with a minimal influence on the simulation results. An important advantage of the
proposed method is the consistency of the fluxes of the individually conserved
variables. The data do not need to be constant in space over the boundary face, which
is usually the case in almost all standard boundary conditions as well as in 1D-3D
coupling methodology, see [2]. A good example for the application of this boundary
condition is the simulation of a 2-cylinder 2-stroke engine. In this engine, the cylinders
are coupled together in a so-called two-into-one exhaust system, see Figure 1 left
side. Furthermore, the intake and the exhaust ports are open simultaneously over a
long period of the cycle and the gas dynamic behaviour inside the exhaust system has
a dominant influence on the scavenging process and on the mixture formation inside
the cylinder, so that both cylinders must be taken into consideration in order to get
accurate simulation results. Applying the new boundary condition, the domain of the
second cylinder (including cylinder, intake ports and crank case) can be replaced.
Fig. 1 Application of the cyclic and time delayed boundary condition

Figure 1 (right) illustrates the application in a 2-cylinder 2-stroke engine
simulation. In this case, cylinder 1 is completely calculated. For cylinder 2, the
621
symmetric condition is used, so that it is replaced by the time delayed data. The
boundary condition for the second cylinder is taken from the exhaust cylinder 1 domain
and for each cell respectively. This means, that for each cell in the exhaust of cylinder
1 a symmetrically positioned cell in the domain of exhaust cylinder 2 is found and
marked. Simultaneously, all cell properties are read and saved in an external output
data file. Therewith, the correct boundary condition for each cell can be defined and
the required flux can be calculated. For the flux calculation the time delayed data from
the external data file are read for each cell respectively applying ROE solver, see [1].
In this work, the flux calculation occurs after the defined time delayed period. Explicitly,
time delayed data are available and the flux calculation can be carried out. The
advantage of this boundary condition is that the requested data are only provided by
the simulation of cylinder 1. Neither additional information about the variable profile,
nor a 1D-model, nor measurement data for the replacement of the second cylinder are
requested. This type of boundary condition can be used for time delayed as well as
synchronous technical problems.
Fig. 2 Flow chart of cyclic and time delayed boundary condition

The boundary condition is written in the C programming language. The source
code containing user defined function (UDFs) is directly compiled and uses additional
macros and functions proposed by Fluent. Figure 2 presents a flowchart of the used
boundary condition. The code consists of four main blocks i.e., input data, read data,
flux calculation and source preparation. The code is executed every time step and
every iteration by the Fluent solver. The block INPUT DATA includes all required
geometries and condition data which have to be defined at the start of the simulation.
For each 3D cell in the selected boundary face, thermodynamic properties and
geometry data are read and saved (block READ DATA). These data cover static
622
pressure, velocity in x-, y, and z-direction, temperature, turbulent kinetic energy,

turbulent dissipation rate, species mass fraction, material data, flow area, and cell
volume. In this code part, the mirrored cells are found and marked with a coupling
number and an ID number. In dependence on the mirror plane position (symmetry
plane), the velocity components are reversed by one of the boundary local coordinate
directions. The required time delayed data are saved every time step in an output data
array and can be read at the start of the simulation as well. Therewith all required data
for the flux calculation are available and are calculated separately for two coupled cells
respectively (block FLUX CALCULATION). The fluxes are calculated by the
Approximate Riemann Solver of Roe according to [1]. Finally, the well-adjusted values
are submitted to the CFD-code Fluent as a source term (block SOURCE).
4.
ENGINE DESIGN
The experimental engine is a 2-cylinder 2-stroke engine from the BRPPowertrain Company, applying loop scavenging via five intake ports. The displacement
of each cylinder is approximately 300 cm. The engine can be also classified as a 2stroke engine with reed valve controlled sucked air mass, crank case charge, Schnurle
loop scavenging, variable exhaust port (RAVE Rotax Adjustable Variable Exhaust),
tuned exhaust pipe and high pressure gasoline direct injection (GDI). The intake ports
are located symmetrically around the cylinder with the same opening time. The
exhaust port consists of three ports, one main and two small lateral ports. In order to
reduce the scavenging losses and to widen the speed range for effective exhaust
tuning, the RAVE system controls exhaust port opening/closing time. Furthermore its
use and position depends on the engine speed (low engine speed = RAVE system
closed, high engine speed = RAVE system open). Due to a well-tuned exhaust system
the reflected pressure wave reduces the scavenging losses. The engine has two
separated intake ducts as well as two separate throttles. These intake ducts are further
connected to an air-box volume.
5.
SIMULATIONS RESULTS
In order to get the gas dynamics in a shorter calculation time, the simulation
was carried out without time consuming injection and combustion processes. At a
selected position of the piston (before the exhaust port opens) the volume of the
cylinder is initialized every cycle with the pressure, the temperature, and the accordant
composition of the exhaust gases. This data can be taken from the test bench or from
old simulation results. This simulation is run until a cyclic steady condition inside the
engine is achieved. This is the case when the pressure trends inside the exhaust
system and the crankcase volume and the sucked air mass do not differ between the
last two last cycles. In addition, the pressure trends must show a good agreement with
the test bench data. On the one hand, this strategy significantly accelerates the time of
the calculation, as injection and combustion process simulations are missing. On the
other hand, the gas dynamic behavior inside all parts of the engine achieves a steady
condition within three or four revolutions. The boundary conditions and the simulation
settings were specified using the data from the engine tests and from a preliminary
investigation [8], [9], and [10].
623
Fig. 3 Measurement point in the exhaust port

Figure 3 shows the measurement point for simulation validation. The indicated
pressure values inside the exhaust system (P-EXH-C1) are presented together with
the comparison to simulation results. The calculations were conducted for two different
engine speeds, namely 7900 rpm WOT (wide open throttle) and 6800 rpm LL (low
load).
Experiment
2,4
Simulation both cylinders

2,2
Simulation with one cylinder
Pressure [bar]
1,8
1,6
1,4
1,2
1
0,8
0,6
0,4
-180
-150
-120
-90
-60
-30
30
60
90
120
150
180
CA []
Fig. 3 Pressure trend inside exhaust system (P-EXH-C1),7900 rpm WOT

Figure 4 and Figure 5 show a comparison of the pressure trends of the test
bench measurements at 7900 rpm WOT and 6800 rpm LL and the simulation results
inside the exhaust system. The presented simulation results show a good prediction of
the pressure trend, which futher confirm the applicability of the presented cyclic and
time delayed boundary condition for the gas dynamic behavior simulation of a 2cylinder engine. The presented figures also confirm that this boundary condition is a
sensible simulation methodology which reduces the number of 3D cells and
simultaneously offers significant results in comparison to the simulation with both
cylinders. The number of 3D cells is reduced by approximately 40%. In this case the
simulation time in the solving phase of the simulation was reduced by approximately
50%. Furthermore, the simulation of only one cylinder reduces the requested time for
the moving of the second piston, the requested interface coupling between the 3D
domains, and further the requested time for the DPM (discrete phase model) solving as
well.
624
1,4
Experiment
Simulation with both cylinders
1,3
Simulation with one cylinder
Pressure [bar]
1,2
1,1
0,9
0,8
0,7
-180
-150
-120
-90
-60
-30
0
CA
30
60
90
120
150
180
Fig. 5 Pressure trend inside exhaust system (P-EXH-C1), 6800 rpm LL

6.
CONCLUSION
This paper covers the discussion of a methodology for the efficient simulation
of a 2-cylinder 2-stroke high-performance engine using a cyclic and time delayed
boundary condition. This boundary condition allows a reduction of the calculation
domain and therewith the time of the calculation. The boundary condition data are
calculated during the simulation and are set on the corresponding face in the
calculated domain. The essential advantage of the new boundary condition is the
solution for symmetric and time delayed periodic technical problems. This means that
the complete simulation of the 2-cylinder 2-stroke engine using the new boundary
condition can be carried out without the geometry of the second cylinder. In this case
the number of the 3D cell is reduced by approximately 40% (about 200000 3D cells)
and the simulation time by approximately 50%. Furthermore, due to the symmetric
situation inside the engine the simulation using the new concept leads to faster
convergence, and a smaller number of calculated engines cycles are needed
respectively. In comparison with the conventional boundary condition the data over the
boundary face need not to be constant in space and often additionally 1D calculation or
measurement data are obsolete. A requirement is only a symmetric and/or time
delayed flow situation. The presented results show a good agreement in comparison
with the test bench results and the complete simulation of the 2-cylinder 2-stroke
engine in both investigated engine operating points.
LITERATURE
Chapters in Books:
[1]
Toro E. F. (1997). Riemann Solvers and Numerical Methods for Fluid Dynamics.
Springer, Germany ISBN 3-540-61676-4, p. 313-341
625
Proceedings Papers:
[2]
Jajcevic D., Almbauer R.A., Schmidt S. P., Glinser K. (2008). CFD Simulation of a
Real World High-Performance Two Stroke Engine with Use of a Multidimensional
Coupling Methodology. SAE Paper 2004742(JSAE), 2008-32-0042(SAE)
[3] Schmidt S., Winkler F., Schoegl O., France M.M. (2004). Development of a
Combustion Process for a High Performance 2-Stroke Engine with High Pressure
Direct Injection. SAE Paper 2004-01-2942(SAE)
[4] Schmidt S., Schoegl O., Rothbauer R.J., Eichlseder H., Kirchberger R. (2007). An
Integrated 3D CFD Simulation Methodology for the Optimization of the Mixture
Preparation of 2-Stroke DI Engines. SAE Paper 2007-32-0029 (SAE)
[5] Chiatti G., Chiavola O. (2002). Scavenge Streams Analysis in High Speed 2T
Gasoline Engine, SAE Paper SAE 2002-01-2180
[6] Zeng. Y., Strauss S., Lucier P., Craft T. (2004). Predicting and Optimizing TwoStroke Engine Performance Using Multidimensional CFD. SAE Paper 2004-320039
[7] Bozza F., Gimelli A., Andreassi L., Rocco V., Scarcelli R. (2008). 1D-3D Analysis
of the Scavenging and Combustion Process in a Gasoline and Natural-Gas
Fuelled Two-Stroke Engine. SAE Paper 2008-01-1087(SAE)
[8] Jajcevic D., Almbauer R.A., Schmidt S. P., Glinser K. (2009). Simulation Strategy
and Analysis of a Two-Cylinder Two Stroke Engine Using CFD Code Fluent.
European Automotive Simulation Conference
[9] Jajcevic D., Almbauer R.A., Schmidt S. P., Glinser K. (2009) Simulation of
Scavenging Process, Internal Mixture Preparation, and Combustion of a Gasoline
Direct Injection Two-Cylinder Two-Stroke Engine. SAE Paper 2009-32-0046
[10] Jajcevic D., Almbauer R.A., Schmidt S. P., Glinser K., Fitl M. (2010). CFD study
of spray de-sign for a GDI high performance 2-stroke engine. SAE Paper 201032-0014
626
CALCULATING THERMODYNAMIC PROPERTIES BY CREATING

AND USING MODEL OF A DIESEL ENGINE WITH SIX CYLINDERS
Melisa Velic1, Semir Mulalic2, Adnan Pecar3
Summary: In this note we have made a model of a normally aspirated diesel engine
with six cylinders, by using AVL-BOOST software produced by the AFL firm from Graz.
We determined the performances of the mentioned engine by using a number of input
data regarding engine dimensions, initial conditions of pressure and temperature, on
the boundaries of the system. We have analyzed the influence of the heat release
characteristic, based on the single Vibe function. As a result, we got the indicated
pressure diagram, which we compared to an experimentally measured pressure within
the cylinder. We also analyzed cylinder temperature and influence of single Vibe
function on it.
Key words: thermodynamic properties, model, engine
1. INTRODUCTION
Process of engine simulation with mathematical model represents a very
important and almost mandatory operation in the technological process of engine
design. Improving the functionality of highly developed technical products, such as
internal combustion engines, is only possible if the physical processes that occur at the
surface and in the inside of the engine can be thoroughly explained and quantified.
For purposes of this paper, a model was designed and a simulation of a sixcylinder normally aspirated engine was carried out, using the software tool AVLBOOST. For the approximation of the rate heat release a single Vibe function was
used, as well as for evaluating the effects of shape parameter m of single Vibe function
on effective pressure, temperature and engine power.
2. MODEL
A model was made for a naturally aspirated six cylinder engine. Constructive
characteristics for this engine are shown in the following table.
BSc., Melisa Velic, Sarajevo, MFS, (melisa_velic@hotmail.com)

BSc., Semir Mulalic, Sarajevo, MFS, (semir_sa_9@hotmail.com)
3
BSc., Adnan Pecar, Sarajevo, MFS, (pecar_5721@hotmail.com)
2
627
Table 1 The basic structural information required to create engine model

Experimentally measured power at 2200 rpm
Cylinders
Cylinder diameter
Cylinder volume
The degree of compression
Rod length
Engine speed
Firing order
Length of intake pipe
Diameter of intake pipe
Length of the exhaust pipe
Diameter of exhaust pipe
Intake valve seat diameter
Exhaust valve seat diameter
Static angle of pre-injection
132 kW
6
123 mm
9,9760 dm3
18:1
260 mm
2200 rpm
1-5-3-6-2-4
700 mm
60 mm
700 mm
60 mm
53,8 mm
44,1 mm
23 deg
Engine model, made by using software tool AVL BOOST, is shown in the
following figure.
Fig.1 Four-stroke diesel naturally aspirated engine model. Labels on the engine
model: SB- systems border, MP-measurement point, C-cylinder, J-junction, numbers
are intake and exhaust pipes
Fig. 1 shows one-dimensional model of in-line engine, which was used to carry
out the thermodynamic calculation. It can be concluded that the engine model is very
simple.
Global parameters are defined in the model framework:
The choice of the working properties of matter
The number of revolutions of crankshaft
Fuel Type
Cylinder properties
Environmental conditions
628
Calculating Thermodynamic Properties by Creating and Using Model of A Diesel Engine With
Six Cylinders
The order of ignition in the cylinders

Friction in the engine
After the global parameters, the parameters for each of the elements that make
the model were defined. The largest number of defined parameters is related to the
cylinder. Rate heat release, heat transfer model and movement of intake and exhaust
valves are defined in the cylinder.
When defining the characteristics of heat release, there are a number of
available approximate functions. As noted above, for the purposes of this study
Woschni model of heat transfer and single Vibe functions were used.
A number of short junction pipe were used for the modeling of the intake and
exhaust manifold. At the same time the restrictions in form of given lengths and
diameters of intake and exhaust manifold, which leads from the system boundary to
the cylinder, were respected.
The model was designed with the assumption that the parameters in all the
cylinders are equal, because the values entered for the first cylinder were used for the
calculation of parameters in all six cylinders.
2.1 BASICS OF MATHEMATICAL MODEL
The calculation of the thermodynamic state of the cylinder is based on the first
law of thermodynamics:

The variation of the mass in the cylinder can be calculated from the sum of the
in-flowing and out-flowing masses:
Where:

change of the internal

energy in the cylinder
piston work
fuel heat input
wall heat losses
crank angle
enthalpy of blow-by
blow-by mass flow
enthalpy flow due to blowby

mass in the cylinder
specific internal energy
cylinder pressure
mass element flowing into the

cylinder
mass element flowing out of
the cylinder
enthalpy of the in-flowing mass
enthalpy of the mass leaving
the cylinder
evaporation heat of the fuel
629
cylinder volume
fuel energy
wall heat loss
fraction of evaporation heat

from the cylinder charge
evaporating fuel
Together with the gas equation

Where:
gas constant
cylinder temperature
establishing the relation between pressure, temperature and density, equation for the
in-cylinder temperature can be solved using a Runge-Kutta method. Once the cylinder
temperature is known, the cylinder pressure can be obtained from the gas equation.
The Vibe function is often used to approximate the actual heat release
characteristics of an engine:
Where:

total fuel heat input

start of combustion
combustion duration
m
a
cylinder temperature
shape parameter
Vibe parameter a = 6.9 for
complete combustion
3. ANALYSIS OF THE RESULTS

Results shown in graphs are obtained after simulating six cycles of engine
operation. Because of the equal conditions prevail in all engine cylinders, the results
for the other cylinders are identical, yet phase-shifted.
In this paper, the diagrams of rate heat release, cylinder pressure and cylinder
temperature are presented. Results are shown for three values of the shape parameter
of single Vibe function, parameter m. All results are shown in function of crankshaft
angle.
Figure 2 represents heat rate release functions approximated by the single Vibe
function, with different shape parameters m. Crank angle of combustion duration =
74 deg and Vibe parameter a = 6.9 for complete combustion, are held constant.
630
RateHeatRelease[J/deg]
Calculating Thermodynamic Properties by Creating and Using Model of A Diesel Engine With
Six Cylinders
200
m=0.15
150
m=0.4
100
m=0.25
50
0
220
240
260
280
300
Crankangle[deg]
Fig. 2 Cylinder Rate Heat Release
Pressure[Pa]
By analyzing diagram in Figure 2 it can be clearly seen that the rate heat release
curve is approximated with function which generally takes the form of a triangle. By
increasing the parameter m the maximum of rate heat release function increases. For
small changes in parameter m function gradient does not change much.
10000000
9000000
8000000
7000000
6000000
5000000
4000000
3000000
2000000
1000000
0
m=0.4
m=0.15
m=0.25
experiment
200
400
600
800
Crankangle[deg]
Fig. 3 Cylinder pressure change
With pressures in Fig. 3, a very sharp gradient in increase and pressure drop
can be seen. Maximum pressure of 97 00000 Pa or 97 bar for the value of the
parameter m = 0.15, can also be seen. One can also see an excellent superposition of
the experimentally obtained pressure and the pressure obtained for the parameter
m=0.25. These functions have their maximum pressures in the value of 93 bar. The
lowest pressure value in the value of 88 bar was obtained for m = 0.4.
With no pressure oscillations it can be concluded that the burning is normal and
there is no knock combustion.
631
Temperature[K]
2500
m=0.15
2000
m=0.4
m=0.25
1500
1000
500
0
0
200
400
Crankangle[deg]
600
800
Fig. 4 Cylinder temperature change

With increasing pressure temperature rises in the cylinder and reaches its
maximum of 2400K, a few degrees of crankshaft after reaching the maximum
pressure. It may be noted that different values of the parameter m, that is, the different
Rate Heat Release of single Vibe approximations have negligible temperature impact.
4. CONCLUSION
Approach to modeling of thermodynamic processes in the IC engine covers
processes that have a direct impact on engine energy indicators. Aspects of the
influence of hydrodynamic and dynamic characteristics of the fuel supply here are
simplified as a constant input data required for calculation.
Comparing the cylinder pressure diagram under the value of m = 0.25 and a
diagram of pressure resulting from the experiment, it can be concluded that the model
is quite good because the pressure in both cases is approximately 93 bar.
The pressure diagram clearly shows that the shape parameter m of the single
Vibe has a large impact on the cylinder pressure, while its influence on the cylinder
temperature is rather small.
The modeled engine is compression ignition engine, so the question can be
raised in aspect of choosing the right Vibe function. Specifically, for the purpose of this
paper a single Vibe function was used, which realistically describes the process of
combustion in spark ignition engine. It would be necessary to make the same
calculation with double Vibe function.
LITERATURE
[1] Filipovi, I. (2009). Modeliranje procesa u motorima, Mainski fakultet Sarajevo
[2] Hamzabegovi, M. (2010). Izbor modela procesa sagorijevanja kod dizel motora sa
direktnim ubrizgavanjem, Mainski fakultet Sarajevo, diplomski rad
[3] Doboviek, ., ernej, A., Mirkovi, M., Filipovi, I. (1980). Simuliranje procesa
klipnih motora SUS, Mainski fakultet Sarajevo
[4] Stoji, T., (2001). Motori SUS-Ciklusi, teorija, razvoj, Mainski fakultet Sarajevo
[5] AVL Boost: User`s Guide
632
DETERMINING THE DURATION OF VIBRATION MEASUREMENTS

OF HUMAN BODY IN LABORATORY CONDITIONS BY SUBJECTIVE
METHODS
Mile Raievi1, Miroslav Demi2, Neboja Rako3, Predrag Milenkovi4
Summary: Human response to WBV is a complex phenomenon. During exposure to
WBV many different psychological, psychophysical and physical factors, such as
individual susceptibility, body characteristics and posture together with the frequency,
direction, magnitude and duration of vibration are relevant in investigation of unwanted
effects.
Standards for measuring and evaluating human exposures to vibration use uncertain
frequency weightings and time dependencies but allow legislation for the protection of
those exposed. The standards may appear useful, but it is necessary to distinguish
between standards and knowledge between what is accepted to reach a consensus
and what can be accepted as proven. Paper will give review of scientific papers and
show significant differences in conclusions about vibration duration and implemented
method influence on whole body comfort during laboratory investigations.
Key words: human, vibration, duration,
1.
INTRODUCTION
Apart from the intensity of vibration amplitudes and frequency range at which
they occur, influence on the comfort and duration of the vibration is obvious. Although
such generalization is logical, it is very difficult to establish quantitative relation
between the discomfort and exposure time. There are a number of researches of time
dependences in literature which are related to subjective human responses to
vibration, where many of whom are completely opposite conclusions.
M. J. Clarke's paper [9] gives an overview of works from the point of impact of
duration of vibration on comfort and dexterity during vibration activity. In the available
literature he classified three types of experiments:
1. Capturing the behaviour of passengers in real conditions where the test can
last more than 4 hours
1
mr, Mile Raievi, Kragujevac, Fiat automobile Srbija, mile.raicevic@fiat.com

Prof.dr., Miroslav Demi, Kragujevac, Mainski fakultet Kragujevac, demic@kg.ac.rs
3
Neboja Rako, Kragujevac, Fiat automobile Srbija, rako.nebojsa@fiat.com
4
mr, Predrag Milenkovi, Fiat automobile Srbija, milenkovic.predrag@fiat.com
2
633
2. experiments on pulsators where examinees give subjective ratings of

vibration effect
3. experiments on pulsators where the subjects are asked to perform simple
actions during the vibration exposure.
Analyzing the works from the period 1930 1979 he did not fint any strong
indications which would be the thesis that there is a trend in reduction of comfort or
ability due to prolonged duration of vibration. Since the data on reduction of comfort
due to prolonged duration of vibration is maintained during testings on pulsators he
warns that these data should be interpered very cautiously given the fact that there is
not enough comparative data and that during the experiment vibrations were
constantly present (since no non-vibration control was used). All the data that he
could find indicate the reduction of comfort which could have arisen due to the causes
which were present even when there are no vibrations. He concludes that the
reduction of comfort may have happened due to the inadequately designed seat or
effect of fatigue which was present regardless of the presence of vibration.
In the work [10] similar research analysis can be found as in work [9] for the
period until 1985. In this work, authors also point to the conflicting views of different
authors, as well as the conflicting results of certain researches regarding the impact of
duration of vibration on the human body. The authors also point that practically in all
the works vertical vibrations were studied, and that the effect of horizontal ones could
be made in such a way to completely change the results.
From the physiological point of view, they point out, that there are few studies
which show certain results of influence of exposure time on physiological parametres.
However there are no studies which have completely covered this issue, and
especially in terms of mental workload during the experiment that can shed entirely
new light to the problem.
After th work [9] there were two studies (Griffin and Witham 1980, Kjellberg and
Wikstrom 1985) which indicated that the discomfort increases with the time of
exposure to the vibrations. During these studies the longest period of exposure to
vibrations was 2 minutes. On the other hand, a number of studies which were longer
exposed to vibrations claim that such an influence does not exist.
The authors indicate the fact that with the subjective method the authors
necessarily give the examinees a clear picure of what is expected of them. This way
the examinees' awareness is directed to responding in accordance with the
recommendations.
With the method of comparison of pairs of stimuli there is also a suspicion in
validity of results. Especially when the examinee is required to compare a signal which
lasts for 1 minute with a similar one which would have the same level of comfort for a
longer period of time.
2.
THE REQUIREMENTS OD STANDARD ISO 2631 AND THE INFLUENCE OF

DISEASE OF MOVEMENT
Further on will be stated 5.5. a point of general requirements of standard ISO

2361/1 from 1997:
The length of measurements should be such as to provide with a reasonable
statistic accuracy and ensure that evaluated vibrations are of typical exposures.
634
Determining the duration of vibration measurements of WB in laboratory by subjective methods
Overall measurement should be presented. If the complete presentation consists of

different periods with different characteristics, separate analyses of different periods
may be required.
Note: For stationary random signals, measuring accuracy depends on the filter
bandpass and duration of measurements. For instance, to obtain a measurement error
less then 3 dB with a confidence level of 90% it is required a minimum duration of 108
s for the lower limit frequency 1 Hz i 227 s for the lower limit frequency (LLF) of 0.5 Hz
when analysed with the tertiary octave width. Measuring period is, normally, much
longer, so it represents well exposure to vibrations.
From the above text we can conclude that the duration should be determined
based on requirements for measurement accuracy. Yet the determination of
measurement duration is not that simple and it requires taking into consideration other
parametres which will be discussed in subsequent paragraphs.
If a man is considered as a viodynamic model in Figure 1. then as an
attenuation factor values we can [6] adopt [1 =0,679 [ 2 = 0,034 [ 3 = 0,11 [ 4 =
0,499
[5
=0,1.
Fig. 1 [6]
In professional literature as the most critical zone in terms of oscillatory comfort
is the area of 4-8 Hz for vertical vibrations and about 2 Hz for longitudinal and lateral
vibrations. As a boundary between low and high frequences in the Z direction, usually
2 Hz is taken, because the studies have shown [19] that at lower frequences froces
acting on human's bodyproportional to input acceleration. In the work [20] it is stated
that the biggest convergence to resonant frequences of the head, while biiodynamic
modeling, occurs when the body is exposed to vertical vibrations of 4 Hz, but is almost
equally important so-called resonant frequency of the torso at 2Hz. Due to the stated,
f r min 2 Hz will be adopted as a value of minimal resonant frequency.
Frequency bandwidth (half-power point bandwith) for the usual cases where
the attenuation is low, it can be shown with a formula [21]:
635
2 [ f r min
Br
2 0,11 2
0,44
Br | 0,44 Hz
Bias error [22] can be shown with a formula:
Hb
1 Be
3 Br
Hb
0,05 Be
0,107
Be | 0,1Hz T
10 s
Where Be is a minimum frequency resolution band-width:

Accidental error [21] can be shown with a formula:
Hr
1
n d J xy
J xy 0,7 correlation coefficient adopted.

H r 0,1 based on standard requirements for 90% reliability of results.
The number of repetitions or recordings during the experiment based on the
formula is n d
143 and from there it follows that the total time for conducting the
experiment is Ttotal
1430s .
In the opening elaboration at an international conference dedicated to the

inflience of vibrations on a human in 2008 [1], where what we know is highlighted, and
what we do not know on this issue, in the conclusion Griffin's words are stated in 2006:
Measuring standards and assessing human exposure to vibrations use vague
frequency bands and time dependents, but they are used in the legislation for
exposure protection. The standard is useful, but a clear difference should be made
between a standard and understanding between what a consensus was reached and
knowledge.
However, it is clear from the previous text that in terms of the length of
laboratory testings of human comfort, on pulsators, a consensus has not even close
been reached, and that there have been different conclusions during the research.
3.
THE DATA FROM THE PREVIOUS EXPERIMENTAL STUDIES WITH

SUBJECTIVE METHODS AVAILABLE IN LITERATURE
In the table are shown data relevant for experimental investigation in laboratory
condition by subjective method.
636
3M
airplane
standing
Sin
Z
A.G.Woods
(1967) [2]
4M
airplane
standing
Sin
Y
A.G.Woods
(1967) [2]
10M
70.2
airplane
standing
ran
zy
P.Donati
A.Grosjean
P.Mistrot
L.Roure[3] 1983
Shoenberger
R.W. [4] 1976
15-34M
7-8F
66
Two types Sitting,

of seat with upright
backrest
Sin
Ran
10M
12M
Sitting,
Hard-toback (with upright
& without a
seatbelt)
Sin
Ran
Corbridge C.,
Griffin M.J. [5]
1986
Leatherwood
J.D.
Dempsey T.K.
Clevenson A.C.
[7] 1980
10-20M
0-20F
Tvrdo,
drveno sa
naslonom
Sitting,
upright
sin
Airplane,
two types
Sitting
Sin
Ran
Z
Y
Roll
Z
X
Pitch
Output
function
Rms
0.071g
0.141g
0.212g
Rms
0. 1g
1-10
Hz
120s
Subjective
scale 1-6
0.8-7
Hz
180s
Subjective
scale 1-6
rms
0. 05g
0.15
0.25
0.3
0.1
0.2
rms
1.2
1.6
2-7
Hz
Subjective
180s
scale 1-6
po
signalu.
2-10
Hz
100 s
comparin
g pairs
stimuli
11-63
Hz
16-40
Hz
30 s
comparin
g pairs
stimuli
0.5-5
Hz
10
curves of
equal
comfort
scale 1-4
rms
0.251.76
0.380.9
rms
0.250.75
126
54
Length
A.G.Woods
(1967) [2]
Frequency band
(Hz)
(
Amplitude
Seat
Type
Number of subjects
and gender
Author(s)
Characteristics of the vibrations

Posture, position
Characteristics of the
seat
Weight
(kg)
Subjects
characteristics
0.0250.075 g
0.0250.075 g
0.020.5
rad/s2
0.0250.075 g
0.0250.075 g
0.020.5
2
rad/s
3.5 ,7
3.5 ,7
3
5
5
3
637
Kjellberg A.
Wikstrom B.
Dimberg U. [8]
1985
15
74,1
Ordinary
drivers
seat
Sitting
0.95,
3.1
1.1,
1.4, 2
6.7
1.3
,
1.6,
2.4, 3.5
2
m/s
Meister A. and
others: [11]
1984
71,1
Hard seat Sitting

anatomical
shape
Yoshihiko K.,
Sugimoto G.,
Suzuki Y. [12]
1986
Mansfield N.J.,
Griffin M.J. [13]
2000
5M
2F
Car seat
Sitting
Z
Y
1.41
2
2.82
562
*
2
4
8
16
*
10M
10F
M 73.2 Car seat

F 57.8
Sitting
0.2
0.4
0.8
Kaneko C.,
Hagiwara T.,
Maeda S. 2005.
[14]
11M
2F
61.3
flat Sitting
Dickey J.P. and

others [15]
2005
Bhalchandra
J.A. [16] 2008
10
Car seat
Sitting
0.2
0.4
0.8
1.2
1.8
0.55
Do 30 10 s
Hz
2
s
pauza
10 s
1
5s
100
x 45 =
Hz
225 s
10M
3F
Hard seat, Sitting

backrest
o
90
Mansfield N.J.
and others [17]
2007
36
Car seat
Sitting
Marjanen Y.,
Rane S.,
Mansfield N.
[18] 2008.
Mili . [19]
2001.
12M
10F
74.8
59.6
Car seat
Sitting
30
85.9
Lorry seat
Sitting
Hard
seat
X
Y
Z
ran
X
Y
Z
ran
X
Y
Z
ran
Z
X
ran
64 min
Crossmodality
matching
14 x 3 comparin
min
g pairs
stimuli
50 s do Subjective
4 min
scale
Comparin
g pairs
Category
judgment
method
6.3
Subjective
scale
0.323
0.539
0.756
180 s
Subjective
scale
0.3
0.6
1-20
Hz
80 min.
Subjective
scale
10 s
x 30
Subjective
scale
Total
90 min
Comparin
g pairs of
stimuli
* - explained in the text

Woods A.G. in the work [2], examines the impact of turbulence during a flight to
the comfort of passengers and pilots reaction. An airplane seat was used, an
examinee was in a position typical for the plane. Comfort was rated on a scale from 16. A series of experiments was conducted as it follows:
1. impulses are vertical sinus vibrations, frequency band is from 0.8-7 Hz with
rms values 0.071, 0.141, 0.212 g, there were three subjects involved in the
experiment, and the question was to estimate the comfort level for 10 minutes
based on the 3 minute exposure to vibrations
2. impulses are lateral sinus vibration, frequency band is from 1-10 Hz with rms
values 0.75, 1, 2 g, there were four subjects involved in the experiment, and
638
the question was to estimate the comfort level for 10 minutes based on the 3
minute exposure to vibrations
3. impulses are random vertical vibrations, frequency band is from 2-7 Hz
whereby certain conditions were repeated only once, with rms values 0.05,
0.1,0.15, 0.2, 0.25, 0.3 g, there were four subjects involved in the experiment,
and the question was to estimate the comfort level for 10 minutes based on 3
minute exposure to virations for each of the levels, whereby one minute breaks
were made for subjects to rest and to change parametres
4. impulses are random lateral vibrations, frequency band is from 2-7 Hz whereby
certain conditions were repeated only once, with rms values 0.075, 0.1, 0.2,
0.25 g, there were three subjects involved in the experiment, and the question
was to estimate the comfort level for 10 minutes based on 3 minute exposure
to vibrations for each of the leveles, where one mnute breaks were made for
subjects to rest and to change parametres.
Donati P., Grosjean A., Mistrot P., Roure L. In the work The subjective
equivalence of sinusoidal and random whole body vibration in the sitting position (an
experimental study using the floating reference vibration method) [3] tried in a series of
experiments by the method of comparing pairs of stimuli to determine subjective
equality between sinus and narrowband random vibrations. There were 34 men in the
experiments for vibrations in direction of z axis, 15 men and 8 women for vibrations in
the direction of u axis and 17 men and 7 women for vibrations in direction of h axis.
Examinees were played a signal which lasted 100 s, and then by a switch placed on a
steering wheel random signals were adjusted in order to achieve the same overall
feeling of comfort. In one part of the experiment adjustment of pairs of stimuli with the
method of limit was used (floating reference method) where certain pairs of stimuli
were offered to the subject to compare. For example three pairs werer offererd: sinus
signal 4 and 3.25 Hz, sinus signal 4 and 5 Hz, and then sinus signal 4 Hz with random
central frequency 4 Hz.
Shoenberger R.W. in the work Comparison of the subjective intensity of
sinusoidal, multifrequency and random whole body vibration [4] showed two
experiments. In the first experiment there were 10 men involved. Impulsive sinus
vibrations were with one or more components like in the table. Each of the subjects
was asked to adjust the intensity of vibrations at the adjusted frequency of 25 Hz in
order to achieve the same comfort level as with the signal to which it was exposed just
before. The exposure to impulsive signal is 30 seconds, and for adjusting the same
comfort level the subject had 15-30 seconds. Between the two exposures to vibrations
there were 5 minute breaks. The second experiment is very similar to the first one but
sinus impulsive vibrations were replaced with random tertiary octave. Random
vibrations had central frequencies at 16,20,25,31.5 i 40 Hz. Target frequency is also 25
Hz, and acceleration is slightly lower values ranging from 0.38-0.9 g. Reduction of rms
is a consequence of existence of peaks at random impulse. There were 12 examinees
involved in the second experiment.
Corbridge C., Griffin M.J. in the work Vibration and comfort: vertical and lateral
motion in the range 0.5 to 5.0 Hz [5] showed three experiments. Curves of equal
comfort were created based on the experiments. In the first experiment it was
completed for 20 male and 20 female examinees with vertical sinus vibrations ranging
from 0.5 to 5 Hz. Method of constant stimuli was used. The examinees were submitted
to recommended (referent) impulse with the frequency of 2 Hz, which lasted 10
639
seconds and amplitudes (rms 0.75 or 0.25 m/s2 ). Referent impulse was followed by a
test impulses randomly chosen from the line of frequences and amplitudes which
lasted 10 seconds. After referent and test impulses the subjest was supposed to
choose which out of two impulses he would want to be reduced using a button if he
was supposed to be resubmitted to their influence. Each of the two rreferent impulses
was compared to 11 test tertiary octave centred sinus impulses with the frequency
from0.5 to 5 Hz. Each examination period lasted between 1 and 1.5 hours including a
short period for a break after 45 minutes. The second experiment is similar to the first
one provided that as an impulse random vertical vibrations of 1 octave width were
chosen. Random vertical octave vibrations are of 0.5, 1, 2 and 4 Hz central frequency.
Test impulsive signal was with r.m.s. from 0.1 to 1 m/s2 and compared to 2 Hz by a
sinusn referent signal. Test signal lasted for 41 seconds like referent one did. Referent
signal was 2 Hz with rms 0. 5 m/s2. Examination period was the same as in the first
experiment. In the third experiment influence of lateral sinus vibrations with the
frequency of 0.5 to 5 Hz was examined. Completely the same methods and techniques
were applied as in the first experiment. Referent signal was a sinusn one of 2 Hz which
lasted 10 seconds and fixed r.m.s 0.75 m/s2. Referent imoulse was compared to 11
tertiary octave centered sinus impulses with the frequency of 0.5 to 5 Hz, with rms 0. 4
and 3.15 m/s2.
Leatherwood J.D.,Dempsey T.K., Clevenson A.C. in the work A design tool for
estimating passenger ride discomfort within complex ride environments [7] use
simulator with 3 degrees of freedom with internal arrangement like on an airplane. The
seat is an airplane seat of the tourist and first class, so that a survey of the influence of
kind of a seat on oscillatory comfort was conducted. In different variant of the
experiment there were 2200 examinees involved who gave their subjective grades of
comfort on the scale from 1-4. What matters from the point of this works topic is that
210 examinees were used to study the influence of duration of vibrations. These data
were used to define corrective factor of the overall discomfort due to the duration of
vibrations. The authors do not state the data of duration of signal. The interesting data
is that the authors state that there is acclimation of examinees to vibrations after one
hour duration of vibrations and that corrective discomfort factor changes after an hour
due to the duration of vibrations. As a relation which describes the dependence of
discomfort on the duration for the exposure greater than 120 minutes the next
formulas are stated:
'DISCdur= 0,0031 0.012 t ; za 1 t 60 minutes
'DISCdur= 0.072 ; za 60 t 120 minutes
where:
'DISCdur correction of discomfort for the duration t,
t the duration of vibrations in minutes.
Kjellberg A. Wikstrom B. Dimberg U. In the work Whole body vibration:
exposure time and acute effects experimental assessment of discomfort [8], study
the influence of vibrations of the whole body to the discomfort during the exposure to
64 minute vibrations. In the work cross-matching method is applied. Matching metod
(comparing pairs of stimuli) is used in order to compare the discomfort eg. between two
signals where one of them lasts for 1 minute, and the other for 60 minutes. Thereby a
subject is asked to adjust the new signal, which compared to the remembered one
during the 1 minute exposure, causes the same discomfort as is if it was exposed for
640
60 minutes. In the work cross-matcing method is applied where the vibrations level of a
human body is transformed into the sound scale. Due to the vibrations the examinees
are given the opportunity to transform certain discomfort level to the similar one or the
same discomfort level which produces sound. Examinees had done the training in
comparing vibrations in Z and sound vibrations direction before the experiment. The
examinees compared signals in 16 points during the 64 minute exposure to vibrations.
The conclusion of the study is that the discomfort shows the growth with the increase
of the exposure time.
In the work [11] the authors start from the fact that the employees who are
exposed to vibrations which are stochastic and therefore they must be considered as
broadband ie. represented with the width over 1/3 octave. In order to detrmine the
interaction of different frequencies frequently descrete vibration question (sin) and the
question of one octave broadband vibration (OWV) with the identical RMS values are
compared. Centered frequency (central) is the same. The study is multidisciplinary
because the data are collected are elelctromyografic.
The study was conducted on electrohydraulic pulsator with the maximal
amplitude of 100 mm. To achieve constant parametre values the pulsator takes 1
minute.
The experiment was done so that complete comparison between 14 exposures
characteristics which are shown in the table 1 was carried out.
1/3 octave SIN
or 1 oct.- OWV
SIN
OWV
SIN
OWV
SIN
OWV
SIN
OWV
SIN
OWV
SIN
OWV
SIN
OWV
Intensity
I1 light
I2 great
I1
I1
I1
I1
I2
I2
I1
I1
I2
I2
I1
I1
I2
I2
Frequency
Acceleration RMS
2
(m/s )
2
2
4
4
4
4
8
8
8
8
16
16
16
16
2
2
1.41
1.41
2.82
2.82
1.41
1.41
2.82
2.82
2.82
2.82
5.62
5.62
Mode
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
Sinuosoid
Octave width p.
The procedure related to the duration of exposure to vibrations is given in the

following table.
Tests and examinations during the period
Preparations which include the check of electric
connections, safety, measuring the temperature of
milieu, determining health condition of an
examinee, setting EMG and ECG electrodes
Duration
(minutes)
Present WBV or not
20
No
0.5
No
0.5
No
Detrmining EMG with the lowered rest

Determning EMG with the raised rest
641
Taking into
complaints
consideration
symptoms
and
The first pair of stimuli to compare
The first exposure to vibrations
Adjusting the second exposure
The second exposure to vibrations
Comparing the pair of stimuli from 4

aspects of subjective evaluation
The fourteenth pair of stimuli to compare
No
2
0.15
2
0.5
Yes
No
Yes
No
As a conclusion of this research the fact that the dependency between the
effects of vibrations at different frequencies exists is stated. In terms of comparison
tertiary octave band and broadband vibrations the research dos not provide with the
clear conclusion. Differences at multiple frequencies were noted, but causes and
assessment of the effects are not clearly indicated.
In the work [12] authors present the results of a research by which an index is
formed (vibration meter VM) which should objectively assess vehicles comfort during
the driving. The examination is conducted on a pulsator, with a car seat and vibrations
in two directions, on Z and Y axis.
Goals of the research were:
where and which direction to measure vibrations in order to assess the
comfort
how to deal with the changes due to impact vibrations which occur over time
how to present the overall influence of vibrations when certain components
are measured in different places.
Random vibrations similar to those measured on the car floor and with
centered frequencies of 1.2 Hz , 5.5 Hz and 11.5 Hz were used like motive vibrations.
The duration of vibrations is 50 seconds for each of the experimental conditions. There
were five men and two women involved in the experiment, whereby the examinees
were well informed about the goals and problems of the examinations. Seven levels of
subjective assessment of vibrations were determined, whereby referent signal was on
the level four. The most uncomfortable level is level 1, and the most comfortable is
level seven. Frequent signals are randomly released and repeated twice during the day
in order to verify assessment repeatability. As a final result middle value of the two
assessments was used.
Coefficients obtained by the formulas were connected to subjective evaluations
of examinees whereby multiple regression analysis was used. The influence of certain
vibration components and capturing points was monitored through correlation
coefficients.
The authors gave the following conclusions based on this research:
capturing the acceleration of vertical vibrations between the seating part
and seat cushions, lateral accelerations on the rest and vertical feet
acceleration it is possible to evaluate the comfort
there is the value of vertical and lateral vibrations value to which is possible
to observe the evaluation of comfort only by vertical oscillations
index by which is possible to present the comfort level is provided based on
captured accelerations of vertical vibrations between the seating part and
642
seat cushions, lateral accelerations on the rest and vertical feet

accelerationa.
In the work [13] the authors start with the fact that drivers due to the comlexity
of car vibration often cannot notice slight improvements of comfort. This fact is very
significant when taken into consideration that certain car components can significantly
influence comfort evaluation, therefore it is very important to properly evaluate the
differences in points of perception.
The experiment was conducted with 10 male and 10 female examinees. Four
impulses were used on the Z axis with captured accelerations on a car seat of a small
vehicle, for two types of bases. An examinee was submitted to the referent signal for
10 seconds, then a break of 2 seconds, and then a recorded signal was released for
2
10 seconds. The size of the referent signal was adjusted to 0.2, 0.4 or 0.8 m/s RMS
values depending on the examined conditions. The shape of the referent signal and
the movement test were always the same, but the size of test movement was always
slightly bigger that the referent movements. The order of comparing signals was
random. After each comparison of pairs the examinee was asked the following:
Did you feel greater discomfort during the first or the second impulse?
When the examinee answers that the impuls was more uncomfortable the
answer correct is written, and when he/she answers that it was more comfortable the
answer incorrect is written. An amplitude was adjusted by UDTR method (up and
down transformed response). When the answer was incorrect the amplitude was
increased by 1.04. When the answer to the three successive repetitions was the same
the amplitude was reduced by 1.04. If there were no three same successive answers,
the amplitude was not changed.
After processing the data the authors came to the conclusion that with the
increase of amount of vibrations the difference in points of perception of vibration
grows. It is also concluded that there is no significant difference in responses of male
and female examinees.
In the work [14] the authors establish the legitimacy between the values of
vibrations in Z direction and 5 criteria of subjective comfort evaluation. The examinees
give assessments in accordance with subjective scale from ISO 2631-1: not
uncomfortable, slightly uncomfortable, averagely uncomfortable, uncomfortable and
2
very uncomfortable. For RMS values 0.2, 04., 0.8, 1.2 and 1.8 m/s three captures
were conducted for each value with the length of 5 seconds with 2 second breaks in
between. There are 5 signals, each with three realisations, for five conditions, therefore
each of the examinees was 225 s totally exposed. During each capturing the
examinees assessed the comfort. Before the experiment the authors had a doubt
whether the examinees would be able to assess the comfort in 5 seconds, but, as
stated in the text, during the experiment realisation it became clear that it was possible.
The signal had a random frequency of 1-100 Hz centered according to ISO 2631-1.
In the conclusion the authors state that it was shown that it is possible to obtain
that kind of data that there is no overlap in subjective assessment defined by standard
and frequently filtered values of RMS acceleration. It has been also concluded that
there was certain discordance in the data from the standrad and those obtained from
the experiment in the part which is related to the categories: averagely uncomfortable,
uncomfortable and very uncomfortable. These categories are at lower values in the
standard of frequently evaluated RMS. It has been noted that also the discomfort level
increases when the vibration contains components with low frequent frequency.
643
In the work [15] Dickey J.P. and others try to determine whether there is dependency
of subjective feeling of comfort depending on the duration of vibrations and time for
rest between two impulsive signals. The authors regard that in that way they will help
to properly determine the duration of impulsive vibrations for laboratory studies of
vibrations of a human body.
The experiment was conducte with 10 examinees. A car seat was used, and
the examinees were sitting in a relaxed manner. Vibrations were divided into 4 blocks
and: with 1 level of freeway along Z axis, with 3 levels of freeway in XY plane, with 3
levels of freeway in YZ plane and with 6 levels of freeway. Each of the blocks had 37
individual sinus vibrations with identical characteristics: 6.3 Hz in vertcal direction and
value 0.55 m/s2 of rms. Inside the overall duration of the experiment of 62 minutes per
examinee, presentation of vibrations lasted 43 minutes. Subjective scalehas nine
comfort levels from 0 which is marked as 0 discomforts, up to 8 maximal
discomfort. During the breaks the examinees verbally stated subjective evaluation of
discomfort.
In the conclusion the authors state that they did not find the statistic
dependency in comfort evaluation for the presentations of 15 opposite 20 seconds.
Thay have not also discovered the difference between subjective comfort evaluations
for the break duration of 5 and 10 seconds between two impulsive signals.
In the work [16] the beginning is the fact that most of the studies uses different
subjective scales which provide with discrete data. In this research the authors tried to
obtain continuous data about subjective feeling of comfort. Apart from these the work
especially takes care of the influene of vibrations transmitted over feet, as well as of
the influence of vibrations on heartbeat.
The research is conducted on a pulsator which is capable of conducting multiaxis movement with the impulse and desired bandwidth. The impulsive signal was
obtained from real measurements of vibrations during a drive by car, train and by
vehicle for timber transport. Impulsive 4 minutes long signals were obtained by
processing these captured signals. For the needs of the experiment three rms levels
2
were chosen 0.323, 0.539, 0.756 m/s . Three rms levels and the same conditions for
each were supposed to provide with better evaluation of the influence of vibrations on
an examinee.
During the experiments were used three different ways of sitting in order to
monitor vibration influence transmited over legs:
1. backrest at 90 degrees and knees angle 90 degrees
2. backrest at 90 degrees, feet on an insulating pad
3. backrest at 90 degrees, legs raised in order to unable the transmition of
vibrations via feet.
For each of the conditions the examinee was submitte to 3 minute vibrations. In
order to enable continuous monitoring of a sense of comfort in hands of the examinees
was a potentiometre with rotational command by which he/she gives assessment. The
examinee assesses comfort during 3 minute exposure to vibrations by subjective scale
of 1-5 according to ISO 2631-1.
In the conclusion is stated that the new method of capturing vibrations of 12
axes was successfully applied in accordance with ISO standard. It is also stated that
during the research no differences were noted in comfort when operated with different
vibrations on feet.
644
The work [17] presents experimental research by which a car seat model was
assessed when the examinee was exposed to vibrations for a longer period of time. It
begins from the fact that discomfort can be divided into static, dynamic and temporal.
Fig. 3 Model which shows seat discomfort made of static, dynamic and temporal part
(Mansfield 2005)
In the experiment was used a pulsator with 6 degrees of freeway able to
reproduce the incoherent random 1-20 Hz frequency signals of a random signal. An
European middle-class car seat filled with foamy material was used in the experiment.
For assessing the discomfort a subjective scale of 7 levels of assessments was used.
At the beginning of the experiment the examinees adjust the seat for the most
comfortable position provided that the knee angle is at 120 degrees with the feet
supported on the platform. 36 examinees are divided into three groups consisting of 12
examinees. After ending the experiment in one group, they moved to another one.
The first group was sitting for 80 minutes in a car seat without vibrations. The
second one was submitted to vibrations for 80 minutes in the direction of all three axes
with the vibration strength of 0.3 m/s2 rms. The same conditions as in the second group
were true as in the third one except that the vibration strength was 0.6 m/s2 rms.At the
beginning of examination the examinees assess the discomfort without vibrations
static discomfort. Then the discomfort is assessed at the end of the first minute, and
then after each 10 minutes until the end of 80th minute when the vibration effect stops.
From 80-90th minute the examinees were sitting in the seat without vibrations and in
the 90th minute gave their last discomfort assessment.
With the group sitting in a seat without vibrations there was an increase of
discomfort over time. Average discomfort value shows monotonous growth for this
group of examinees. With two other groups which were exposed to vibrations there
was greater discomfort increase in regard to the group which was not exposed to
vibrations. Ovo je manje izraeno u prvih 10 minuta dejstva vibracija.
The conclusion of the work is that there is an increase of discomfort with the
increase of sitting time, as well as with the increase of vibration strength.
In the work [18] Marjanen Y., Rane S., Mansfield N. conducted a research on a
pulsator with 6 levels of freeway. In order to decrease the number of variables they
reduced the standards requests for measuring vibrations from 12 axes to the system
of 9 axes. Namely, during the experiment the examinees legs were supported on a
mobile platform so that there was not greater vibration influence via their feet.
This is also significant because the authors start with the fact that the devices
which enable measurements according to ISO 2631-1 are unavailable to many who
wanted to examine the influence of vibrations on human body. The standard allows
645
several possible scenarios to simplify this procedure, but it does not explicitly show
which axes to use or which scenarios. It is stated in the standard that it is necessary to
use translatory axes on the examinees seat, provided that it is possible to use one
axis if the vibrations on the other axis are at least 25% of the dominant axis. Also
rotations around the axes are present, so it is very important to examine which of these
freeway degrees we can exclude from the measuring in order to make the
examination simpler.
There is a note in the standard where it is said that it is not necessary to
measure vibrations at three axes on the backrest of a seat using magnification factor
1.4. Based on this there can be a confusion whether it is necessary to use this factor or
only when vibrations in this part are excluded for some reason. The authors did not
obtain the data of the way how this factor was derived so this can lead to different
interpretation of the standard and bad interpretation of research results.
The goal of the experiment is to establish a connection between the value of
subjective discomfort scale and value of vibrations calculated based on the standard.
Signals from the realistic captures of vibrations shown in the table were used for the
impulse. Before each of the sequences for assessing oscillatory comfort the
examinees were submitted to the effect of five vibration levels from the comfortable, to
the most uncomfortable one in order to memorize the impulses and be able to assess
the discomfort level. Duration of the signal was 15 seconds where the first 5 seconds
do not count during data processing beause that time is necessary for the examineesto
adapt to the signal and to adjust the discomfort evaluation (because the scale on the
measuring device was restored). After each sequence, which included 30 captures,
there were 5 minute breaks. During those breaks the examinees came down the
pulsators and walked in order to reduce the tiredness influence during examination.
The examination procedure lasted for about 40 minutes per examinee.
A subjective scale with continuous measuring discomfort was used to evaluate

the discomfort. Discomfort was evaluated through the line length on a display for
discomfort evaluation. The line length was maneuvered via a command which was in
the examinees hands. At the end of each session an examinee restores the scale to a
position wirhout discomfort.
According to the number of axes combinations which are allowed by the
standard, the following practical scenarios are formed for correlation analysis between
discomfort degrees and vibrations (table 3). The scenarios are chosen according to
practical reasons which are useful for researches and include possible interpretations
of the standard.
646
The standard allows the use of different axes combination and vibrations
measuring locations, but it does not provide with clear instructions when to use which
combination. The examination shows the increase of the correlation of subjective
discomfort evaluation and physical magnitudes which are characterized by vibrations,
but in most of the cases the measurement in the direction of translator axis seats is
enough, if magnification factor is 1.4 for horizontal axes. Also rotational movements
around the axes have low influence regarding the efforts which are supposed to be
made for measuring these movements and in most of the cases it is not necessary to
measure them.
In the work [19] a series of examinations of vibrational loads while driving was
conducted with Zastava-Iveco lorries. The measurements were conducted in the X,Y
and Z axes directions on the floor in front of the co-drivers seat, on the co-drivers seat
and on the steering wheel. Captured like this, and then processed signals were used
for examination of the influence of multicomponent random vibrations on humans
fatigue.
There were 30 examiness involved in the experiment. It is a lorry seat, and the
examineess position is in accordance with ergonomic recommendations for the lorry
drivers position. Generating of random impulses in the form of narrowband tertiary
octave spectra movements is realised by a PC. Central frequencies of vibration spectra
are: 0.63, 0.8, 1.00, 1.25, 1.60, 2.00, 2.5, 3.15, 4.00, 5.00, 6.30, 8.00, 10.00, 12.50,
16.00.
The examinations were conducted as it follows. For the determined vibration
spectrum with corresponding central frequency of tertiary octave band and initial
amplitude, the examinees adjusted the vibration amplitudes to the level which
according to subjective evaluation was supposed to provide with comfortable drive in
certain temporal duration by potentiometre. Adjusting test stimuli, depending on the
examinees, lasted from 10 to 30 seconds, and writing down temporal signal of
acceleration started when the subject gives a sign that he/she has determined the level
of random vibrations, which according to his evaluation provides with comfort in
required temporal duration. The examination of each subject lasted foro 90 minutes
with the note that the breaks were made in order not to get tired.
647
4.
CONCLUSION
An overview of experimental researches of vibration influences on human body

by subjective methods is given in the work. The purpose of using subjective methods is
to: understand the subjective comprehension of physical vibration characteristics; to
determine the relations and establish a good scale between subjective observations
with physical characteristics of vibrations; to establish target values to form vibrational
environments taking into consideration the limits obtained by subjective methods.
It is seen that the authors used different durations of experimental researches
over time. Lately, the division of discomfort to static, dynamic and temporal is used the
most. By that the discomfort also includes certain phenomena which are not directly
connected to vibrations.
Based on overview requests of standard and experimental researches it can be
concluded that the problem of determining duration of experimental researches is
current and still solved.
BIBLIOGRAPHY
[1]
Proceedings of the Second American Conference on Human Vibration, Chicago,

IL June 46, 2008.
[2] Woods A.G.: Human response to low frequency sinusoidal and random vibration,
journal Aircraft engineering, july 1967.
[3] Donati P., Grosjean A., Mistrot P., Roure L.: The subjective equivalence of
sinusoidal and random whole body vibration in the sitting position (an
experimental study using the floating reference vibratiom method), journal
Ergonomics, vol. 26, number 3, 1983.
[4] Shoenberger R.W.: Comparison of the subjective intensity of sinusoidal,
multifrequency and random whole body vibration, journal Aviation, space and
environmental medicine, August 1976.
[5] Corbridge C., Griffin M.J.: Vibration and comfort: vertical and lateral motion in the
range 0.5 to 5.0 Hz, journal Ergonomics, vol.29, 1986.
[6] Rodean S.: Study about damping ratio of the human body-seat system, Fascicle
of Management and Technological Engineering, Volume VII (XVII), Oradea
University, Romania, 2008
[7] Leatherwood J.D.,Dempsey T.K., Clevenson A.C.: A design tool for estimating
passenger ride discomfort within complex ride environments, Human factors,
vol.22, pages 291-312, 1980.
[8] Kjellberg A. Wikstrom B. Dimberg U.: Whole body vibration: exposure time and
acute effects experimental assessment of discomfort, Ergonomics, vol.28,
pages 545-554, 1985.
[9] Clarke M.J.: A study of the avalaible evidence on duration effects on comfort and
task proficiency under vibration, Journal of sound and vibration, vol.65, str. 107123, 1979.
[10] Kjellberg A., Wikstrom B.: Whole-body vibration: exposure time and acute effectsreview, Ergonomics, vol.28, pag. 535-544, 1985.
[11] Meister A. and others: Evaluation of responses to broad-band whole-body
vibration, Ergonomics, vol.27, pag. 959-980, 1984.
648
[12] Yoshihiko K., Sugimoto G., Suzuki Y.: A new ride comfort meter, SAE technical
aper 860430, 1986.
[13] Mansfield N.J., Griffin M.J.: Difference tresholds for automobile seat vibration,
Applied ergonomics, vol.31, pages 255-261, 2000.
[14] Kaneko C., Hagiwara T., Maeda S.: Evaluation of wholebody vibration by the
category judgment method, journal Industraial health, vol. 43, pages 221-232,
2005.
[15] Dickey J.P. and others: Vibration time and rest time during sinusoidal vibration
experiments: do these factors affect comfort ratings, Proceedings of the First
American Conference on Human Vibration, 5-7 june 2006.
[16] Bhalchandra J.A.: Development of continouse subjective and physiological
assessment techniques for the determination of discomfort and activity
interference in transportation systems, Indian Institute of technology, Roorkee,
India, 2008.
[17] Mansfield N.J. and others: Relative influence of sitting duration and vibration
magnitude on sitting discomfort in a car seat, Presented at the 42nd United
Kingdom Conference on Human Responses to Vibration, held at ISVR, University
of Southampton, Southampton, England, 10 - 12 September 2007
[18] Marjanen Y., Rane S., Mansfield N.: Optimising the standardised method to
evaluate discomfort from multi-axis whole bodz vibration, Presented at the 43rd
United Kingdom Conference on Human Responses to Vibration, Leicester,
England, 15 - 17 September 2008
[19] Mili .: Identifikacija elemenata istovremenog uticaja viekomonentnih vibracija
na zamor korisnika teretnih motornih vozila, doktorski rad, Mainski fakultet
Kragujevac,2001.
[20] Seagull J.F., Wickness C.D.: Vibration in command and control vehicles: visual
performance, manual performance and motion sickness: a review of literature,
Technical report HFD-06-07/FEDLAB-06-01, University of Illinois, 2006.
[21] Bendat S., Piersol A.: Engineering application of correlation and spectral analysis,
John Wiley and sones, 1980
649
DUAL PORT INDUCTION SYSTEM FOR DMB 1.4 MPI ENGINE

Miroljub Tomi1, Stojan Petrovi2, Slobodan Popovi3, Nenad Milji4
Summary: The paper presents the research results of DMB 1.4 l spark ignition engine
equipped with prototype variable induction system. This variable geometry induction
system was developed in order to investigate the possibilities of improving induction
generated swirl and turbulence in combustion chamber and in this way to accelerate
combustion, especially at low load engine condition. Induction channel in engine head
was separated longitudinally into two sections using thin tin longitudinal wall and
deflector valve at channel entrance was used to direct air flow into half channel cross
section (at low engine load) or to enable full channel cross section (at full load). The
results of combustion analysis show that combustion is accelerated and cycle
variations are reduced at certain working conditions. In addition, specific fuel
consumption is reduced, but very small, smaller than it was expected. Further
geometry optimization and appropriate process managing is required.
Key words: Spark ignition engine, variable induction geometry, cylinder pressure
recording , combustion analysis.
1
INTRODUCTION
During engine induction system design, researchers have always been faced
with the problem of optimal flow characteristics realization at engine different seeds
and loads. Induction system optimal geometry can not be the same for all engine
operating conditions. For example, inlet channels optimal length, enabling the
utilization of pressure waves for resonant filling is longer for low engine speed and
shorter for maximal speed. Therefore, either the compromise channel length must be
accepted or a solution enabling variable pipe length should be applied, what
significantly increase the engine cost. However, this problem, although very important
is not topic of this paper.
Another problem, also very important, is the realization of optimal flow (air flow
velocity, swirl and turbulence) in inlet channels and combustion chamber, what has
crucial influence on mixture formation and homogenization and flame propagation
through combustion chamber, i.e. heat release rate. At engine partial load and low
1
Prof. dr Miroljub Tomi, University of Belgrade, Faculty of Mech. Eng. (mtomic@mas.bg.ac.rs)

Prof. dr Stojan Petrovi, University of Belgrade, Faculty of Mech. Eng. (spetrovic@mas.bg.ac.rs)
Mr Slobodan Popovi, University of Belgrade, Faculty of Mech. Eng. (spopovic@mas.bg.ac.rs)
4
Mr Nenad Milji, University of Belgrade, Faculty of Mech. Eng. (nmiljic@mas.bg.ac.rs)
2
3
651
engine speeds, small channel flow area is favorable in order to unable sufficient flow
velocity. But such reduction of channel cross section is unfavorable at engine high load
conditions since it has negative influence of volumetric efficiency. This problem can be
successfully solved in the case of four valve engines, i.e. engines with two inlet valves
and two separate inlet channel per cylinder, where a secondary channel can be
throttled at low engine load (example Opel twin port system [1]). However, in the case
of two valve engines, i.e. one inlet valve per cylinder, the problem is more complex for
construction, and there are no similar solutions in engine commercial production. The
aim of this work is the investigation of possibilities to realize the variability of inlet cross
section in the case of two valve engine and to study the benefits of such solution.
An original idea is the principle shown in fig. 1. In order to accelerate inlet air
flow in the entrance to the cylinder at engine low loads, it is necessary to reduce its
cross section at least in final part of the channel. Also, air flow tangential entry into
cylinder is favorable to enable swirl motion in combustion chamber. This can be
achieved by separation of inlet channel in engine head in two sections using
longitudinal tin wall and deflector plate located at the entrance of cylinder head. Such
design enables practically two inlet channels, so one of them can be partially blocked
and air flow directed to the other one. This is schematically shown in fig. 1, where
longitudinal tin wall and deflector are shown only for first cylinder to simplifies the
drawing and in practice other cylinders have the same system.
a)
b)
Fig. 1 Schematic of Dual port variable induction system; a) low engine load; b) full
engine load. (1 tin wall, 2 deflector, 3 throttle valve)
At engine partial loads and low speeds, the deflector closes one section of inlet
port and the air flow is directed into another channel in which flow velocity is increased
and the tangential entry into cylinder is achieved (fig. 1-a). At engine full load, when
inlet air flow is high enough, it is necessary to enable low flow resistance in order to
achieve high volumetric efficiency and consequently high power output. Then,
deflectors are placed in the open position, so that both inlet channel in engine head are
active and flow area is increased (fig. 1-b).
652
Dual port induction system for DMB 1.4 MPI
DESIGN OF DUAL PORT VARIABLE INDUCTION SYSTEM
Laboratory prototype of variable induction system called Dual port, has been
developed for spark ignition engine DMB 1.4 MPI, produced by engine factory 21 may
Belgrade [2]. It is four cylinder engine with swept volume of 1372 ccm and bore/
stroke ratio 80.5/67.4. The engine has 9.2 compression ratio and water cooling. The
prototype was designed using commercial software for 3-D modeling Autodesk
Inventor and made using an CNC machine. Prototype design and experimental testing
have been carried out at the Faculty of Mech. Eng., University of Belgrade.
In order to simplify prototype design, the part of existing induction system is
used: intake air plenum and initial part of inlet pipes. Other, newly designed part of inlet
pipes has complex geometrical shape because in its initial part cross section must be
circular to match the existing pipes, while at outlet cross section must be tetragonal to
enable deflector rotating motion. The transition between the circular and rectangular
cross-section is continual to enable good flow characteristics. Since such a complex
geometry is hard to be properly realized in closed space, even using CNC machine,
the assembly is made of two halves upper and lower that are connected by screws.
The connection between the pipes and new part is also realized by screws. Simpler
connection would be by welding but in this case simple disassembling would not be
possible and eventual corrections of geometrical shape would be very complicated.
Screw connection between the pipes and new part also enable the insertion of different
pipe in order to investigate the influence of pipe length.
Fig. 2 and 3 show prototype of Dual port variable induction system. In fig. 2deflectors are in their closed positions directing air flow into primary inlet channels in
engine head and enabling flow acceleration and tangential entry into cylinder. This
corresponds to engine low load. In fig. 2-b deflectors are in open positions and both
parts of inlet channel are active (full engine load).
a)
b)
Fig. 2 Design of induction system Dual port with directing deflectors in a) closed
position and b) open position.
Certain modifications also had to be done in cylinder head. Fig. 3-a shows
inserted tin wall plates which separate intake channel in the head. Fig. 3-b shows
complete assembly of cylinder head with Dual port variable induction system.
653
a)
b)
Fig. 3 Detail of engine head with longitudinal tin plates inserted in head induction
channels (a) and assembly of engine head and the prototype of variable induction
system Dual port (b).
In laboratory conditions the deflector positions were adjusted manually. After
system experimental testing and optimal deflector position mapping, deflector actuating
can be controlled by the engine Electronic Control Unit using appropriate actuator.
Since only two deflector positions are required, a relative simple and cheap solution
with pneumatic membrane device can be used. Membrane moving which is transmitted
via lever system to the deflectors is activated by ECU using pneumatic valve and
pressure from engine induction system.
3
TESTING PROCEDURE AND EXPERIMENTAL RESULTS
Fig. 4 Detail of combustion chamber with pressure transducer location (engine 3.

cylinder)
654
370
1.00
360
0.98
380
1.05
370
1.00
360
0.95
geo open port
etaeo open port
340
etaeo closed port

lambda open port
330
350
0.96
geo closed port
0.94
340
0.92
330
0.90
geo closed port

etaeo open port
0.85
etaeo closed port

lambda open port
lambda closed port
0.34
lambda closed port

0.34
320
0.32
310
0.32
310
0.30
300
0.30
300
0.28
290
0.28
290
0.26
280
0.26
280
0.24
270
0.24
270
0.22
0.22
260
260
2
10
pefo [bar]
etaeo [-]
320
etaeo [-]
geo open port

350
lambda [-]
1.02
geo [g/kWh]
380
lambda [-]
geo [g/kWh]
The expected effect of increased turbulence and swirl motion (deflector closed)
is the acceleration of combustion process at engine partial load, what should increase
fuel economy. To study these effects, a number of engine load characteristics at
different speeds were recorded. In addition, in-cylinder pressure in 50 consecutive
cycles was recorded at each operating point. Individual cycles as well as average cycle
are later analyzed and heat release and heat release rate were determined using the
methods of pressure record thermodynamic analysis [3,4,5,6].
Fig. 5 shows specific effective fuel consumption (ge0) and thermal efficiency
(Keo) vs. engine load at 2800 and 3300 rpm at the same ignition timing. All values are
plotted comparatively for two deflector position closed port (flow is directed into one
half of inlet channel) and open port (full inlet channel cross section is active). Since
mixture strength has significant impact on fuel economy, the air access ratio (O) is also
shown in order to separate the different influences. Comparative are only the points
where air access ratio is the same and the difference in fuel economy is the
consequence of the rate of combustion.
0.20
2
10
pefo [bar]
Fig. 5 Engine characteristics vs. load at n=2800 and 3300 rpm; geo -specific fuel
consumption, etaeo- effective thermal efficiency, lambda-air access ratio.
As can be seen, specific fuel consumption is lower and thermal efficiency is
greater in the condition of increased flow velocity in inlet channel (closed port), except
at full engine load where mixture strength is richer and consequently fuel economy is
worse. Nevertheless, the differences in fuel economy are relatively small, about couple
of percent, except at n=2800 rpm where the differences achieve a few per cent.
Fig. 6 shows the universal diagram of specific fuel consumption relative
difference recorded in engine whole operating range. This relative difference is defined
as (geclose-geopen)/geopen, so that negative difference means that the consumption is
lower in the case of increased flow velocity (closed port). Positive difference (lower
consumption with open port condition) is mainly the consequence of the difference in
mixture strength.
655
10.0
9.0
0.01
-0.01
-0
.01
6.0
-0.04
2000
2250
2500
2750
0.
03
02
-0.01
-0.02
-0.03
-0.04
-0.05
0.
0.0
0.01
-0.01
3.0
1.0
1750
4.0
2.0
0.
09
-0.
01
01
0
0 0.0 .06 0
.0
0.002.03.04 5
7
0.
01
0.08
2
-0.0
7.0
-0.
0.02
0.01
8.0
5.0
0.07
0.01
0.01
pefo [bar]
3000
3250
3500
3750
4000
n [min-1]
Fig. 6 Universal diagram of specific fuel consumption relative difference

The typical results of cylinder pressure recording and thermodynamic analysis
for a working regime, where the effects of dual port are the most expressed and an
improvement of engine performances is the biggest, are shown at figures 7. At other
engine working points, these effects are smaller. Besides indicated cylinder pressure p,
figure 7 show normalized integral heat release X (ratio of released heat vs. total heat
corresponding to the mass of fuel per cycle) and the rate of heat release dX/dD (heat
released per crank angle). It can be seen that peak pressure is 10% higher in the case
of closed port, as well as heat release rate is higher (a 20%) and combustion duration
is shorter (a 25%).
At every engine operating point a series of 50 consecutive cycles were
recorded in order to analyze cycle variations of working process. Thus, thermodynamic
analysis was carried out on single recorded cycles (diagrams c and d) and on
average cycle (diagram f).
For analyzing the speed and stability of the combustion process the following
parameters has been considered:
Variations of cycle maximum pressure pmax;
Variations of combustion process defined as angular duration of combustion
10-95% of burnt fuel mass (X=0.1-0.95)
Comparison of heat release X and rate of heat release dX/dD obtained by
average cycle analyzing.
As can be seen, combustion process is faster in the case of increased air
velocity in inlet channel (closed port) what indicate heat release and the rate of heat
release of average cycle. Besides, the cyclic variations of combustion are considerably
lower what can be concluded from both cycle maximum pressure variations and heat
release cycle variations.
656
OPEN PORT
35
30
30
25
25
20
20
15
15
10
10
5
0
100
200
a)
1.1
1.0
0.9 OPEN PORT
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
320
340
360
300
c)
CLOSE PORT
40
35
OPEN PORT
25
20
15
10
5
0
90 180 270 360 450 540 630 720

alpha [CA deg]
e)
CLOSE PORT
100
200
b)
380
400
420
alpha [CA deg]
45
30
400
V [cm3]
X [-]
X [-]
40
35
p [bar]
45
1.1
1.0
0.9 CLOSE PORT
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
320
340
360
d)
1.1
1.0
0.9
CLOSE PORT
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
320
340
360
f)
300
400
V [cm3]
380
400
420
alpha [CA deg]
5.5
5.0
4.5
4.0
3.5
3.0
OPEN PORT 2.5
2.0
1.5
1.0
0.5
0.0
-0.5
380
400
420
alpha [CA deg]
dX/dalpha [rad-1]
40
p [bar]
45
X [-]
p [bar]
Fig. 8 Cylinder pressure (a,b) and heat release X (c,d) for 50 consecutive cycles;
cylinder pressure (e) and heat release X and heat release rate dX/dalfa (f) for average
cycle, in function of crank angle (alfa). Engine load pe=6.2 bar and speed n=2300 rpm.
657
11.0
-5
-4
-5
-3
-4
-4
-6
3.0
1750
2000
-7
-8 -9
2250
-9
-5
-10
-5
-5
-9 -6
-8
-2
-4
-6
-5
-1
-2
-3
-5
-12
13
-6
-4
-3
-9
-10
1
-1
2500
-6
-7
4.0
-3-4
-6
-4
-6
-6
-5
-3
0
1
5.0
-7
-3
-6
-5
-2
-11-9 -7
-12
-8
-4
-4
-4
-8
-5
-7 -4
-7
-5
-3
-10
7.0
6.0
-5
-5
8.0
-6
9.0
-4
-8
-9
-7
10.0
-6
imep [bar]
Figure 8 shows the universal diagram of the difference of angular combustion

duration variations for the case of closed port and open port condition. Lower cyclic
variation of combustion results in lower variation range of combustion angular duration,
what means higher combustion process stability. Therefore negative difference of
variation range on the diagram means lower variation in the case of increased air
velocity in inlet channel (closed port). As can be seen negative difference is evident in
all engine load and speed and ranges from a few crank angle degrees to even -10
crank angle degrees. As a result of increase intake port velocity and generated swirl
with closed secondary port, a higher stability of combustion process leads to lower
cycle variation of maximum pressure and mean effective pressure .
-6
-7
2750
-5
3000
-7-6
3250
3500
3750
4000
n [min-1]
Fig. 9 Universal diagram of the difference of angular combustion duration

variation range.
4.
CONCLUSIONS
Based on the comparison of fuel consumption and thermal efficiency, as well

as on combustion analyses, the following conclusions can be formulated:
The acceleration of air flow in inlet channel by the reduction of flow area
using designed Dual port variable induction system evidently accelerate
combustion process at all engine loads and speeds. Besides, the cyclic
variations of combustion process duration are significantly lower what
results in lower cyclic variations in maximum pressure and mean indicating
658
pressure. All this shows that the stability of engine working process is
increased.
Combustion acceleration and stability enable increased fuel economy of the
engine at low engine loads and speeds where engine volumetric efficiency
has not been decreased by increased intake air velocity. Nevertheless,
measured fuel consumption reduction is at the level of couple per cent,
lower than expected.
Further investigation and optimization of variable induction system is
required in order to consider the possibilities of better utilization of increased
rate of combustion. There is the indication that the optimization and
adaptation of some control parameters (for example ignition timing) would
enable better effect of faster combustion on engine fuel economy.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
Peter Gebhard; Uwe D. Grebe; Gerold Reinheimer; Gerd Liebler; Rainer Prfer;
Jan Zimmermann; Frank Dickgreber - The New Generation of Small Gasoline
Engines for Opel Corsa, Agila and Astra, MTZ worldwide Edition: 2004-03
Jurkovi, T., Petrovi, S., Tomi, M., upurdija, A., Uremovi, ., Revitalization of
1.4 lit. DMB Engine, JU-03050, International scientific conference Science and
Motor Vehicles 03 , Belgrade, 2003.
D. M. Brunt and C. Pond, Evaluation of techniques for absolute cylinder pressure
correction, in SAE Technical Paper Series, no. 970036, Feb. 1997.
D. M. Brunt and D. A. L. Emtage, Evaluation of burn rate routines and analysis
errors, in Sae Technical Paper Series, no. 970037, Feb. 1997.
M. Tazerout, O. L. Corre, and P. Stouffs, Compression ratio and tdc calibrations
using temperature- entropy diagram, in SAE Technical Paper Series, no. 199901-3509, 1999.
M. Tomic, S. Popovic, N. Miljic, S. Petrovic, M. Cvetic, D. Knezevic, and
Z. Jovanovic, A quick, simplified approach to the evaluation of combustion rate
from an internal combustion engine indicator diagram, Thermal Science, vol. 12,
no. 1, pp. 85102, 2008.
659
FUNCTION AND ADOPTION OF IC ENGINES DUAL MASS

FLYWHEEL
Blaevi A.1, Bibi D.2, Filipovi I.3
Summary: In order to increase energy efficiency of the vehicle powertrain, the
innovative solutions in the construction of internal combustion engines, its aggregates,
systems and equipment must be obtained. A trend that is prevalent today is widely
known as downsizing. Namely, downsizing involves reducing of engine volume
(consequently, its mass) and keeping or even increasing the power output compared to
a reference IC engines in the same time. However, structural changes coincide with
changes of the IC engines dynamic load output and its oscillatory behavior. Certain
modes where the resonance and the excessive amplitude of oscillation of IC engine
crankshaft appears, can lead to its breakage. For this reason, besides classical
damping system of the oscillation amplitude, in form of elastic dampers at the one of
the crankshafts ends can be introduced, as well as the dual mass flywheel so enabling
dynamic relief of IC engines while simultaneously balancing the angular velocity of the
IC engine.
In this paper, analysis of the dual mass flywheel application will be presented
compared to the oscillatory behavior of IC engine with or without other damping
systems used to decrease torsional oscillations. Besides the analysis, an optimization
of the parameters, based on dual mass flywheel, will be performed (mass distribution
and stiffness of the springs) in order to achieve a better output of the IC engines
oscillatory behavior.
Key words: torsional vibration, dual mass flywheel, IC engine
1. INTRODUCTION
The IC engine is a cyclic machine which transformes internal chemical fuel
energy into power, through the combustion process. Due to the cyclic repetition of the
process, unbalanced fuel quality, as well as the randomness of the combustion
process, the engine itself is exposed to the various alternating mechanichal loads. In
order to achieve a more balanced rotating speed, the engine is fitted with a flywheel.
Despite that, an imbalance of the current effective torque exists, which is basically the
result of the excitation for the torsion vibration of the crankshaft. The vibration of the
1
Almir Blaevi, Sarajevo, Faculty of Mechanical Engineering Sarajevo, blazevic@mef.unsa.ba

V. Prof. Dr. sc. Devad Bibi, Sarajevo, Faculty of Mechanical Engineering Sarajevo, bibic@mef.unsa.ba
3
Prof. Dr. sc. Ivan Filipovi, Sarajevo, Faculty of Mechanical Engineering Sarajevo, filipovic@mef.unsa.ba
2
661
crankshaft is undesirable, and may lead to the crankshaft failure, as a result of extreme
dynamic loads, especially in the case of the high output IC engines. The modern trend
of decreasing engine volume, known as downsizing additionaly increases the
dynamic load of the crankshaft, and in the terms of torsion vibration extremely affects
the crankshaft lifetime. Due to lack of possibility of changing the structural integrity, or
even the design of the crankshaft, the use of vibration dampers is inevitable. The
vibration dampers are used to decrease the cryticall vibration amplitudes, thus
preventing crankshaft failure. Dampers are usually mounted on the opposite side of the
flywheel. Dampers can be, considering their design, divided into several groups, of
which the elastic dampers are used the most. Today, the use of dual-mass flywheels is
becoming a general practice for most engine manufacturers. A dual-mass flywheel
enables the solution of two crutial problems (rotating speed balance, and vibration
damping) of an operating engine, as a system integrated into a whole. Next to these
crutial problems, the dual-mass flywheel decreases the engine noise and the work of
the transmission system makes easier, thus increasing the driving comfort. The dualmass flywheel constists of the two masses attached by a spring of an appropriate
stiffness. The total mass of dual-mass flywheel is equal to the mass of a classic
flywheel. Figure 1 illustrates the dual-mass flywheel manufactured by LUK.
Fig. 1 Dual-mass flywheel (LUK)

Due to the complexity of the dual-mass flywheel construction, it is extremely
hard to define all of its parametrs nessecery for the torsion vibration calculations. This
paper shows one possible way for optimising the stiffness and the inertia of the dualmass flywheel in order to achieve the best damping effect in a resonant area of the IC
engine. The following text describes a possibility use of the dual-mass flywheel on a
specified turbocharged diesel engine, originally fitted with a classic torsion vibration
damper, installed on the free end side of the crankshaft.
2. THE EQUIVALENT VIBRATION SYSTEM
A frequently used equivalent system for the torsion vibration analysis is a linear
system, in which all the elements are reduced to a mutual axis of rotation, regarding
the equality of the potential and kinetic energy of the matching elements. Figure 2
illustrates an equivalent torsion vibration system of an six-cylinder, four-stroke, turbo662
Function and adoption of IC engines dual mass flywheel
charged, water cooled diesel engine fitted with a dual-mass flywheel. Experimental
results have been made for the observed engine, and the results will be used for the
verification of the model used for the calculations. The inertia values (T) of certain
cylinders are different between each other, due to different counterweights fitted on the
cranks of the crankshaft.
M1
q2
q1
c1
M3
M2
e1
q4
q3
c2
c3
n2
q5
c4
e3
e2
n1
M4
e4
n3
c5
n4
q9
q7
q6
c6
c7
e6
e5
q8
M6
M5
n5
e7
c8
e8
n6
Fig. 2 Equivalent vibration system

The table 1. shows the basic construction data of the engine, as well as the
parametrs of the equivalent system.
Table 1 Basic engine data and parameters of the torsional vibration system
Engine Volume
Bore
Stroke
Engine Power
Engine Torque
Firing order
Moment of inertia
Stiffness
Index
Vh
Dk
h
Pe
Me
--
T1; T2,4,5,7; T3,6; T8; T9

c1, c2,3,4,5,6; c7, c8
Internal damping
H1,2,3,4,5,6,7;H
External damping
Q1,2,3,4,5,6
Numerous value
11045
125
150
185
890
1-5-3-6-2-4
0.102; 0.147;
0.083; 1; 0.9
4.32; 2.80; 4.11;
optimization
calculated;
optimization
10.8
Unit
cm3
mm
mm
kW
Nm
-kgm2
x106 Nm/rad
Nms/rad
Nms/rad
3. VERIFICATION AND MODEL OPTIMIZATION

The form of the system of the differential equations which is going to be used in
this paper in order to analyse the torsional vibration system shown in figure 2. is given
below:
663
T i - i v i - i H i 1 - i - i 1 H i - i - i 1 c i 1 - i - i 1 c i - i - i 1 M i
$$
(1)
$$
Where - is the angle of torsion due to vibration, - i - are the matching

derivatives of the angle, whilst the Mi represents the excite momentum of the disk i.
Considering the real shape of the excite momentum, it must be transformed into the
Fourriers order:
f
j 0 ,5
j 0 ,5
a j cos( jZ t ) b j sin( jZ t )
Mi
(2)
For solving the system of equations (1) a self-developed program code was
built, in the Matlab commercial software. Figure 3 illustrates the comparison of the
calculation results and the experiment results for an engine without a torsion vibration
damper.
0.7
Calculation
Vibration amplitude []
0.6
6 order/I
Experiment
0.5
0.4
0.3
9 order/I
7.5 order/I
0.2
0.1
0
1000
12 order/I
1200
1400
1600
1800
2000
2200
Engine speed [min-1]
Fig. 3 Verification of the calculation results for an engine without a torsion vibration
damper
Figure 3 shows a fairly good match of the results obtained by the calculation to
those which are measured by the experiment. After such verification of the model, a
calculation of the angular vibration amplitudes is performed, with the accent made on
the belt pulley, and regarding the model shown in figure 2. Since the charachteristics of
the dual-mass flywheel are relatively unknown, an inertia ratio of T8/T9 = 1/0.9, has
been adopted, while the stiffness c8 has been altered in an interval of 0.3x106 to
1.2x106 Nm/rad. Internal damping has been kept constant, with a value of H8 = 30
Nms/rad. Although, in order to obtain more detalied optimisation of the dual-mass
flywheel, these parameters would require a lot more attention. However, in this paper,
it is chosen to observe the influence of the spring stiffness. Figure 4 shows the angular
amplitude change as a function of the dual-mass flywheel spring stiffness.
664
Function and adoption of IC engines dual mass flywheel
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.2
0.4
0.6
0.8
1.2
Spring stiffness x 106 [Nm/rad]
Fig. 4 Changes of the amplitude on the belt pulley for a 6th excite order depending of
the spring stiffness
Figure 4 shows that the spring stiffness of c8=0.3x106 Nm/rad generates an
exceptable angular amplitude, for the 6th excite order. After such optimisation, a
complete calculation for the mentioned stiffness is performed, using an inertia of the
supplementary mass in a value of a T9 = 0.9 kgm2. Figure 5 shows the calculation
results, compared to an engine equipped with a classic elastic torsion vibration damper
(experimental results).
0.70
Calculation with dual-mass flywheel
0.60
Experiment with elastic damper
0.50
0.40
0.30
6 order/I
4.5 order/I
6 order/I
0.20
3 order/I
0.10
9 order/I
7.5 order/I
4.5 order/I
0.00
1000
1200
1400
1600
1800
-1
Engine speed [min ]
2000
2200
Fig. 5 Results of the amplitudes on the belt pulley with dual-mass flywheel
665
As figure 5 shows, the dual-mass flywheel with the following parameters

c8=0,3x106 Nm/rad i T9 = 0.9 kgm2 can succesfully reduce critical angular amplitudes
on the belt pulley, and keep them in the boundaries achieved by a classic torsion
vibration damper. The excite orders 7.5 and 9 are moved out of the working engine
speed range, albeit it was not in the case of using a classic damper. Further
optimisation could generate an even greater reduction of the amplitudes of the 6th
order, or even completely moving it out from the working speed range.
4. CONCLUSION
In this paper an analysis of the influence of the spring stiffness of the dual
mass flywheel on the angular vibration amplitudes during a resonance vibration period
has been made, using the self-developed calculation program that calculates the
angular vibration amplitudes and the experimental results as well.
The given results indicate that the dual mass flywheel has the same efficiency as a
classic elastic damper in damping torsional vibrations of the crankshaft.
With a more detail analysis of the dual mass flywheel parameters (T8, T9, c8, H8), the
same satisfactory efficiency can be achieved as when using a conventional damper.
LITERATURE
[1] Filipovic. I. (2007). Motori s unutarnjim izgaranjem, dinamika i oscilacije. Mainski
fakultet Sarajevo, Sarajevo.
[2] Bibi. D. (2004). Uticaj viskoelastinog priguivaa na torziono-oscilatorno
ponaanje motornog mehanizma. Magistarski rad. Mainski fakultet Sarajevo,
Sarajevo.
[3] B.I.C.E.R.A. (1958). A Handbook of Torsional Vibration. Cambridge. University
Press.
[4] Wilson K. (1963). Practical Solution of Torsional Vibration Problems. Chapman &
Hall Ltd., London.
[5] Bibi D., Filipovi I., Blaevi A. (2010). Contribution in defining parameters of the
equivalent system of torsional vibrations with the significant internal damping
elements. Mobility & Vehicle Mechanics, International Journal for Vehicle
Mechanics, Engines and Transportation Systems, volume 37, Number 2, 2011,
pgs. 69 78, University of Kragujevac Faculty of Mechanical Engineering,
Kragujevac.
666
HYDRAULIC PUMP IN THE OF VEHICLE STEERING SYSTEM

Jelena Eric Obucina1, Jovanka Lukic2
Summary: The paper presents a review of the hydraulic systems with power steering
pump that is used in the control system for motor vehicles. Power steering system of
vehicle reduces the steering effort on cars by using an external power source to assist
in turning the wheels. Most power steering systems work by using a belt-driven pump
to provide hydraulic pressure to the system. This hydraulic pressure is generated by a
rotary vane pump which is driven by the vehicle's engine. As the speed of the engine
increases, the pressure in the hydraulic fluid also increases, hence a relief valve is
incorporated into the system to allow excess pressure to be bled away. While the
power steering is not being used, i.e. driving in a straight line, twin hydraulic lines
provide equal pressure to both sides of the steering wheel gear. When torque is
applied to the steering wheel, the hydraulic lines provide unequal pressures and hence
assist in turning the wheels in the intended direction. Vehicle producers and
constructors are dealing with noise and vibration problems of hydraulics vehicle
steering system. In this paper, possibilities for reducing noise and vibration of vehicle
will be presented.
Key words: hydraulic system, power steering system, pump, vehicle.
1. INTRODUCTION
If you've ever driven an older build the vehicle that does not have power
steering, then you know how hard the vehicle can be to turn at lower speeds. However,
modern vehicles almost always come equipped with power steering pumps that allow
you to easily turn your vehicle, and even turn the steering wheel when the vehicle is at
a complete stop.
The power steering pump in your vehicle is also a hydraulic fluid pump. In fact,
power steering fluid is nothing more than hydraulic fluid. Whenever you turn your
steering wheel, the hydraulic pump in your car's steering system pushes power
steering fluid to the steering gear via a rotary vane hydraulic pump. In the majority of
vehicles, the motor in the pump is powered by a belt and pulley.
As the belt powers the motor vanes and rotor in the power steering system,
hydraulic fluid creates high-pressure that is used to help you turn your vehicle. The
1
Master of Technical Science, Jelena Eric Obucina, Trstenik, High technical mechanical school of applied
studies, (jecaeric79@yahoo.com)
2
Associate Professor, Jovanka Lukic, Kragujevac, Faculty of Mechanical Engineering, (lukicj@kg.ac.rs)
667
pressure created is applied to the torsion bar which is attached to the steering wheel
and pinion or worm gear (which turns the wheels). The more force or torque that you
apply to the steering wheel, the more that the torsion bar twists and turns the wheels.
2. HYDRALIC SERVO PUMP
The hydraulic power for the steering is provided by a rotary-vane pump. This
pump is driven by the car's engine via a belt and pulley. It contains a set of retractable
vanes that spin inside an oval chamber.
As the vanes spin, they pull hydraulic fluid from the return line at low pressure
and force it into the outlet at high pressure. The amount of flow provided by the pump
depends on the car's engine speed. The pump must be designed to provide adequate
flow when the engine is idling. As a result, the pump moves much more fluid than
necessary when the engine is running at faster speeds.
The pump contains a pressure-relief valve to make sure that the pressure does
not get too high, especially at high engine speeds when so much fluid is being
pumped.
2.1 Pump maintenance
Hydraulic oil is also lubrication pumps, so that does not need any lubrication.
However, it is important that at regular intervals controlled by the oil level in the tank. It
is recommended that controls be at least every 5,000 km traveled road, and according
to the manufacturer of the vehicle.
Since there is a possibility that at various points within the circular flow of oil
intake air, and the fact that the oil and becomes air emissions, remain the tiniest air
bubbles, it must control the oil level when the pump is running.
Forced air is under pressure back flow, and after retiring pump expands due to
normal atmospheric pressure. Therefore, after the exclusion of the pump, the oil level
in the reservoir rises to about 1-2 cm.
Before starting the pump, it is recommended to control the oil level in the tank,
and that the necessary amendments so that the oil level is about 2 inches above the
top mark on the meter when the pump is not working. Only after this should include a
pump (engine), to observe the oil level and if necessary supplement. If the oil level
drops significantly (more than 2-3 cm), then the air must be discharged into a complete
hydraulic unit.
In order to avoid suction of air service in search of the vehicle should be
checked for leaks, and if necessary further tighten all threaded fittings and pipes, high
pressure pump. When repairing the pump or control, to replace the cartridge filter in
the tank of oil.
For the flawless operation of the pump, it is necessary to use suitable hydraulic
oil. Hydraulic oil and lubrication also takes the pump, and the entire unit needs only
one oil.
2.2 Testing vane pumps
In situations where a certain type of car (engine) selected and installed vane
pump or when the manufacturer is replacing them, the final evaluation of the
668
Hydraulic pump in the of vehicle steering system
performance decision is made on the basis of the conducted investigations. Testing

methodology provides that measures the following characteristics:
speed pumps at the min and max engine revs;
flow pump;
pressure in the pressure vessel;
underpressure in the suction line;
outside temperature pumps;
oil temperature in the tank.
2.2.1
Test conditions
Measurement of size is carried out under the following conditions:
rotate the wheel-load vehicles in place;
rotate the wheel nominally loaded cars in place;
nominally loaded car movement in a circle diameter of 12 m (outer front
wheel). Speed of movement 8 y 10 km / h, and base asphalt;
climbing vehicle on the curb height of 15 cm, angled.
2.2.2
Test results
Test results are presented in the form of tables and diagrams which are the
basis for analysis. The obtained values for measuring characteristics are evaluated
whether they are below the "critical" value, or whether they are in the allowed area.
Fig. 1 Curve to pressure smokers depending on the time t
Fig. 2 Curve changes in pressure in the pressure water pump when the car was
moving the left side
669
Fig. 3 Curve changes in pressure in the pressure water when the car was moving the
right side
Fig. 4 A sudden increase in pressure in the pressure water pump

2.3 Causes of noise vane pumps, primary and secondary measures
The cause of noise in pump, there is change in pressure between the suction
and pressure parts. So they bit steep gradients of pressure in certain areas of thrust.
This kind of excitement and noise created by the forces pushing further into the current
course remain dominant, so that the noise due to flow can be neglected.
When we talk about the vane pump causes noise frequency spectrum consists
of the pulsations of pressure fluid in the chambers caused by the equation of motion of
working bodies, as well as the characteristics of the workplace, especially during
reverse fluid working chamber during the transition from the suction pump to push the
zone.
After crossing the working chamber with a closed volume of working fluid
through the wall the board, it comes in contact with the respective discharge and
suction opening in the switchboard. If at that moment the pressure in the chamber is
different from the pressure in the opening with which it is connected, there is a jump in
pressure, followed by the hydraulic shock and the appropriate sound effect. At the time
the merger was the thrust chamber with an opening formed opposite fluid flow from the
opening in the chamber, followed by hydraulic shock with a significant loss of pressure
and with pulsing pressure in the chamber and potisniom water, conditioned inertial
forces and fluid compressibility. Education of the pressure wave is repeated with each
subsequent cycle.
The oscillations of pressure in the pump rostiru through a working fluid in the
hydraulic pressure line installations rise to it because of inertia and compressibility of
fluids, oscillatory processes and the corresponding vibration of the mechanical
elements of the hydraulic installation (piping, etc.).. In some conditions it is possible
resonance with rapidly increasing amplitude of oscillation.
670
Hydraulic pump in the of vehicle steering system
The noise level generated by the pump during operation influence the shaft
speed and pressure at the outlet. With increasing speed shaft and fluid pressure at the
output level of noise increases.
As part of measures to reduce noise from the main interest is the excitation
force, because the sound at the source of combat gives less problems. Unfortunately,
the primary measures are often used in narrow limits and high costs of implementation.
There is also reducing the degree of efficiency and reduced service life. Optimum
balance between part kotradiktorno goals are determined by market conditions or the
requirements of the end user.
One power station is a strong generator of sound, in addition to it because of its
compact design all produced sound itself vent. The absolute size of the force excitation
despite leading to a certain level of sound. Besides mounting the outer part, as a
mechanical suspension, the pump is increased the excitation source.
To release rigid connection (motor-pump unit) of a mechanical arm (steel
frames) is often applied elastic suspension (spring steel). Favorable conditions arise if
the net mechanical oscillation frequency is lower than the rotating drive shaft.
Irradiated volume in pump and powertrain encapsulation often dim. This measure
leads to increasing need for space and increased costs.
In many cases, the desired reduction in noise of a power device is not achieved
by coating aggregates. The cause lies in the sound that creates a fluid pressure in the
roundabout. Bulk current pulsation pump operates across the roundabout as a sound
source connected to each customer depending on the impedance of consumers. So
that the irradiated volume of the total hydro can be larger than from the pump, if it
tends to be noise from pulsations of current density and pressure at the pump keep as
little as possible.
The area of secondary measures is widespread, if they are outside the pump.
As secondary measures inside the pump there are damped mechanical properties and
structure changes. Extreme measures secondary cause additional costs, which are
larger if the application site remains within the transmission chain from the noise
source.
3.
CONCLUSION (10 PT BOLD)
The term power steering is usually used to describe a system that provides
mechanical steering assistance to the driver of a land vehicle, for example, a car or
truck. The power steering system in a vehicle is a type of servomechanism. For many
drivers, turning the steering wheel in a vehicle that doesn't have power steering
requires more force (torque) than the driver finds comfortable, especially when the
vehicle is moving at a very slow speed. Steering force is very sensitive to the weight of
the vehicle, and nearly so much to its length, so this is most important for large
vehicles. In a vehicle equipped with power steering, when the driver turns the steering
wheel, she or he feels only a slight retarding force, so a vehicle equipped with power
steering can be driven by any healthy driver, even when the vehicle is being parked.
This is because the power steering system furnishes most of the energy required to
turn the steered wheels of the car. Most power steering systems in cars and light
trucks today are hydraulic. This paper describes the hydraulic pump for power steering,
671
both in terms of maintenance and testing of hydraulic requirements, as well as from the
point in terms of noise reduction.
LITERATURE (BOLD)
Journal Papers:
[1] R. Rajamani (2005.). Vehicle Dynamics and Control, University of Minnesota, USA
[2] J. Eric (2007). Master thesis, Faculty of Mechanical Engineering Kraljevo.
[3] M. Mini (1992). Control systems of vehicles, ABC Glas, Beograd.
[4] ZF Friedrichshafen AG http://www.zf.com
[5] Z. Adamovic (1998). Maintenance technologies, University of Novi Sad, Technical
Faculty "M. Pupin" Zrenjanin.
672
IGNITION SYSTEM OF MULTIPROCESSING OTTO/DIESEL ENGINE

Aleksandar Davini1, Radivoje Pei2, Dragan Taranovi3, Miroslav Ravli4
Summary: Essence of multiprocessing operation principle of an engine is to achieve
Otto and Diesel working process in the same cylinder, depending on operating regime
and using two different types of bio-fuels. Thus, this engine has two fuel supply
systems and an ignition system. The main problem of the ignition system is reliability of
a spark plug during longer operation of the engine in Diesel mode. Due to potential
forming of electrically conductive deposit on a ceramic spark plug isolator, instead of
using classical ignition system (one spark per cycle), multi-sparks producer type of the
ignition system is used that is able to obtain continuous, arbitrary long electric
discharge between the spark plug electrodes. Possible occurrence of electrically
conductive deposit leads to establishing of surface electric arc and momentary burning,
which enables self-cleaning of the spark plug. Description of proposed ignition system
and comparison of its electric characteristics with standard BOSCH MP 3.1 ignition
system in laboratory conditions are presented in the paper. Comparative results of
engine tests of these two ignition systems also are given in the paper.
Key words: ignition system, multiprocessing engine, spark plug, self-cleaning
1. INTRODUCTION
One possibility to increase the economy of the IC engine and to influence row
emissions is to introduce the multiprocessing operating principle. The essence of this
principle is the ability to lead the operating process by OTTO, DIESEL, HCCI cycle
during engine operation, whereat the most favorable cycle is selected depending on
the operating regime.
At the Faculty of Mechanical Engineering from Kragujevac, there is an ongoing
effort to develop a multiprocessing Otto-Diesel engine that operates by Diesel cycle
(mode) at lower loads and by Otto cycle during start and at larger loads [1,2,3]. The
basic engine (DMB 3DA450) is an air cooled diesel engine with single cylinder and
direct injection. In order to achieve multiprocessing features, the following was done:
1
MSc Aleksandar Davini, assistant, Kragujevac, Faculty of Mechanical Engineering in Kragujevac, Serbia,
E-mail: davinic@kg.ac.rs
PhD Radivoje Pei, professor, Kragujevac, Faculty of Mechanical Engineering in Kragujevac, Serbia, Email: pesicr@kg.ac.rs
3
MSc Dragan Taranovi, assistant, Kragujevac, Faculty of Mechanical Engineering in Kragujevac, Serbia, Email: tara@kg.ac.rs
4
MSc Miroslav Ravli, Kragujevac, PRIZMA, Serbia, E-mail: miroslav.ravlic@prizma.co.rs
2
673
Davini A., Pei R., Taranovi D, Ravli M.
geometric compression ratio is reduced from 17.5 to 12.5 by increase of

piston chamber volume,
electrically controlled throttle valve and fuel injector are added on intake
system as elements of the system for external air-fuel mixture forming for
operation at Otto mode,
engine is equipped with ignition system,
the existing diesel fuel injection system is added with actuator for switchingoff,
laboratory microcontroller system for engine operation control is developed.
In diesel mode, due to relatively low compression ratio, a highly inflammable
biodiesel fuel (methyl ester of soybean oil) is used, while in Otto mode, due to relatively
high compression ratio, bio ethanol is used as high-octane fuel [4, 5, 6].
The main problem in the ignition system is reliability of ignition in Otto mode
after prolonged period of engine operation in diesel mode. Due to diffuse combustion
of non-homogeneous diesel fuel mixture and lower spark plug temperatures, there is a
danger of forming of electrically conductive deposit on ceramic insulator. In coil ignition
system (one spark per cycle), regular arc discharge between electrodes (volumetric
discharge) stops and surface arc through deposit at the spark insulator is set up, which
leads to irregular operation of the ignition system.
In order to prevent this problem, a multi-sparks producer ignition system is
applied that is able to provide continuous, arbitrary long electric discharge between
spark plug electrodes. In case of eventual appearance of electrically conductive
deposit on ceramic insulator, a surface electric arch is established with enough energy
to burn the formed deposit. In this way, despite the spark plug with low heat range, its
self cleaning is obtained.
2. CONCEPT AND REALISATION OF IGNITION SYSTEM
Ignition system is integrated in multiprocessing engine control system,
presented in Fig.1. The basic task of the control system is to engage corresponding
mode of operation (Otto or diesel) based on information on engine operating regime
and to set optimal parameters of fuel ignition and injection [7]. A microcontroller
developmental system, 9, performs processing of data regarding operating regime and
open loop control algorithm. Programming, monitoring, input of constant values and
possible setting of ignition and injection parameters are done with computer, 10.
Operation of the ignition system is based on digital and analogue signals. Digital
signals are: top dead centre signal (TRIG), crank shaft position signal (CDM) and
intake valve position signal (SINC). Analogue signal is the signal of intake manifold
depression signal (pus).
TRIG and CDM signals are given by angle encoder, 1, (mechanically coupled
and phase synchronised with engine crank shaft), via optical transmitter, 3, and pulses
multiplier, 8. The encoder sync signal in regard to engine operating cycle, SINC, is
given by inductive switch, 5, that registers opening of the intake valve. Calculation and
setting of ignition timing angle in Otto mode is done based on TRIG signal frequency
(engine speed) and pus signal (engine load) and according to experimentally defined
dependences (ignition map). Microcontroller capabilities allowed performing control
and possible correction of TRIG signal on a hardware level for each cycle, by which the
674
Ignition system of multiprocessing Otto/Diesel engine
influence of disturbances is avoided. Ignition advance angle may be set with 1 CA

resolution, while period of engagement is fixed (typically at 10o CA). In this manner, a
command ignition signal, IGN, is formed. Executive part of the ignition system is
ignition module, 13, with spark plug, 14.
1 - Angle encoder, 2 - Throttle actuator, 3 - Optical transmitter, 4 - Throttle potentiometer,

5 - Inductive switch, 6 - Differential pressure gauge, 7 - Depression meter, 8 - Pulses multiplier,
9 - Microcontroller Development Board, 10 - Computer, 11 - Injectors driver, 12 - Ethanol
injectors, 13 - Ignition module, 14 - Spark plug, 15 - AUX drivers, 16 - Diesel injection STOP
actuator.
Fig. 1 Block chart of multiprocessing engine control system

Fig. 2 presents a concept of a new multi-sparks producer ignition system
module. The trigger circuit galvanically separates a command signal, IGN, and turns
on/off astable multivibrator which generates excitation signal with fixed frequency of
10.5 kHz. By alteration of ton/toff ratio within a single period of excitation signal,
characteristics of the multi-sparks producer may be optimised.
Fig. 2 Block chart of Ignition module
675
3. TESTING OF MULTI-SPARKS PRODUCERS ELECTRIC CHARACTERISTICS

Electric characteristics of multi-sparks producer are first determined in
laboratory conditions. Voltages and currents of primary and secondary winding of high
voltage transformer (HV) are simultaneously measured in continual sparks-producing
regime with analogue-digital oscilloscope HAMEG HM1508. Measurement results for
one of the series of impulses in primary and secondary circuit of high-voltage
transformer are presented in Figs. 3 and 4, respectively.
Fig. 3 Voltage (Up) and current (Ip) in primary winding of high-voltage transformer
Fig. 4 Voltage (Us) and current (Is) in secondary winding of high-voltage transformer
The following may be concluded based on analysis of process in primary
winding (Fig.3):
after switching on, primary current increases monotonously until the
moment of switching off, which means that the ferritic core is not in
magnetic saturation,
from the moment of switching off, and during the following 6 s, there is a
period of complex transitional phenomena caused by self-induction in
primary circuit, induction from secondary circuit and effects of supply filter.
Then, primary current abruptly decreases with the peak in voltage
(dissipation of energy at transistor). Negative primary current is a
consequence of accumulated energy from magnetic circuit of the
transformer. Security diode integrated into transistor rectifies the selfinduction current towards a voltage source,
there is no fluctuation of primary voltage until the end of the cycle, while
current decreases.
676
By analysis of voltage and current flows in secondary winding of the

transformer (Fig. 4), the following is concluded:
immediately before the start of a new cycle, there is electric discharge on a
spark plug from the previous cycle (voltage and current of the secondary
winding are negative).
increasing primary current changes polarity of secondary voltage and
current by induction, thus, during 12 s there are some transient
phenomena when electric arc shortly exhibits properties of negative
resistance. Namely, due to stopping of spatial ionic load and change of its
travel direction, the arc voltage rapidly raises.
further on, arc voltage stabilises, while discharge current monotonously
raises,
from the moment of switching off of primary circuit, arc voltage and current
abruptly change polarity and stay negative until the end of the cycle.
4. COMPARATIVE ENGINE TESTS OF THE IGNITION SYSTEM
Comparative engine tests of the ignition system are conducted with a goal to
verify performance of multi-sparks producer in real operating conditions. In this regard,
beside cold start, engine idle speed is the most demanding regime (in this case
n=1000 rpm, Ps75 W). Ignition advance angle (app) and air excess ratio () are varied.
Comparison criteria are derived from analysis of combustion process course for 50
consecutive cycles, by using AVL Indimer 619 measuring system. Combustion
dynamics is followed by means of average values of crank shaft angle Q from the
moment of ignition to the moments of releasing 5%, 10%, 50% and 90% of cycle
quantity of heat (Q). Stability of the combustion is followed by means of standard
deviation of values of these angles, Q. Duration of sparking was 20 CA during
operation with multi-sparks producer.
Fig. 5 Ignition angle and dynamics of

combustion (KV = CA)
Fig. 6 Ignition angle and stability of

Fig. 5 shows effects of ignition advance angle on dynamics of combustion

process during engine operation with =1.07 mixture, for both ignition systems. It may
be observed that dynamics of initial combustion period, Q5%, as well as centre of
combustion, Q50%, is almost identical for both ignition systems. By application of multisparks producer, the combustion duration decreases by 27 CA. It is considered to be
677
a consequence of generally better combustion stability between cycles. From Fig. 6, it

may be seen that small differences in equability at the beginning of combustion
progressively increase as process advances. This phenomenon is pronounced in the
case of early ignition, when conditions for ignition are less favourable, so it seems that,
in this case, multi-sparks producer shows some advantages.
Varying of air excess ratio (Fig. 7) also shows, in regard to combustion process
dynamics, that there is no significant difference between tested ignition systems. With
respect to combustion stability and based on results in Fig. 8, it may be concluded that
the multi-sparks producer is in certain advantage.
Fig. 7 Air excess ratio and dynamics of

Fig. 8 Air excess ratio and stability of

Influence of the operating regime (load, Wi, and engine speed, n) and ignition
angle on ignition delay (zu CA and tzu ms) during operation with multi-sparks producer
is presented in Table 1. It is indicative that conditions for air excess mixture ignition
become more favourable with the increase of the load and with the decrease of ignition
angle, which is recognised by shorter duration of ignition delay.
Table 1 Influence of operating regime (Wi, n), ignition angle (pp) and
air excess ratio () on delay of ignition (zu CA and tzu ms)
Operating regime
Idle
0.96
1.13
pp CA
25
20
25
30
35
25
tzu ms
0.71.4
0.50.8
0.61.2
1.12.3
1.62.3
0.61.0
zu CA
4.28.3
3.04.8
3.67.1
6.613.7
9.713.9
3.65.9
1.03
25
0.20.3
2.23.3
1.07
n=1000 rpm
Wi=0.1 kJ/dm3
n=1830 rpm
Wi=0.34 kJ/dm3
Time period between the start of sparks produce and the forming of flames
core (a) may be determined by oscilloscopic monitoring of electric parameters of multisparks producers sparks, as illustrated in Figure 9. After ignition and forming of flame
in the cylinder, due to significantly larger amounts of electrified ions in space between
spark plug electrodes, shape and amplitudes of currents through secondary circuit
change (b), Fig. 9. These signals may be used in OBD for detection of ignition
absence. Similar results are gained by authors in [8].
678
Fig. 9 Electric current in primary (Ip) and secondary (Is) winding of high voltage
transformer
5. CONCLUSIONS
During longer operation of multiprocessing engine in diesel mode, possible
appearance of electrically conductive deposit on spark plug electrodes leads to forming
of surface electrical arc and its immediate burning, in the case of working with multispark producer type of ignition system, which enables self-cleaning of the spark plug.
By application of multi-sparks producer, combustion duration is decreased by 27 CA,
which is due to generally better combustion process stability between cycles.
Tests with different values of air excess ratio have shown that, with respect to
combustion process dynamics, there is no significant difference between the multisparks producer and coil ignition system. With respect to combustion stability, it may
be concluded that the multi-sparks producer is in certain advantage, especially in the
case of early ignition when conditions for ignition are less favourable.
By monitoring of shapes and amplitudes of current and voltage in primary
and/or secondary circuit of the ignition system with multi-sparks producer, it may be
concluded when ignition in the cylinder occurred or has there been ignition absence,
which may be used in OBD systems.
ACKNOWLEDGMENTS
The paper is the result of the research within the project TR 35041 financed by
LITERATURE
[1]
[2]
Davini A., Pei R., Veinovi S. (2004). The new concept of the IC engine for
realization the Otto and the diesel process in the same cylinder, XIII International
Scientific Symposium MVM2004, Kragujevac, p. 1-6.
Pei R., Petkovi S., Hnatko E., Veinovi S. (2008). Anthropogenic global
warming, the Kyoto protocol and transport ecology, (in Serbian), Tractors and
power machines, vol.13, no. 3, p. 43-50.
679
[3]
[4]
[5]
[6]
[7]
[8]
680
Pei R., Davini A., Veinovi S. (2008). New engine method for biodiesel cetane
number testing, Thermal Science, vol. 12, no. 1, p. 125-138.
Pei R., urevi K., Vemi M., Veinovi S. (2004). The bio-fuels and the biolubricants as future of the agricultural power trains, (in Serbian), Tractors and
power machines, vol. 9, no. 3, p.19-25.
Furman T., Nikoli R., Tomi M., Savin L., Simiki M., (2007). Characteristics of
fuels for diesel engines, (in Serbian), Tractors and power machines, vol. 12, no. 5,
p. 21-26.
Pei R., Nestorovi D., Jei D., Veinovi S. (2003). Global trends in the fuels
and lubricants future of the agricultural power trains, (in Serbian), Tractors and
power machines, vol. 8, no. 3, p.11-17.
Taranovi D., Pei R. (2009). Effects of applications of biofuels on control
algorithms of IC engines, (in Serbian), IX International Conference on
accomplishments in electrical, mechanical and informatics engineering DEMI
2009, Banja Luka, p. 703-706.
Weyand P., Piock W.F., Weiten C., Schilling S., (2008). Adaptive Multi Charge
Ignition for Critical Combustion Conditions, MTZ Motortechnische Zeitschrift vol.
69, no. 07/08, p. 16-23.
IMPACT OF THE SPECIFIC ABSORBERS ON THE DYNAMIC LOAD

OF THE IC ENGINES CRANKSHAFT
Filipovi I.1, Milainovi A.2, Blaevi A.3, Pecar A.4
Summary: New development of the IC engine is followed by the increase of the
3
specific parameters: specific engine power per volume (kW/dm ), specific engine
3
2
torque per volume (Nm/dm ), specific engine load per piston crown area (kW/dm ), etc.
All of the parameters mentioned above increase dynamic load of the engines crank
mechanism, particularly the engines crankshaft. Common methods reducing dynamic
loads of the engines crankshaft, mainly as a result of the torsion vibration, are the
torsion vibration dampers mounted on. The most widely used the torsion vibration
dampers in the vehicles engines are viscoelastic and viscous dampers. However, there
are also its drawbacks such as reliability, increased weight of the crankshaft, etc. In
order to obtain the highest possible reliability of the dampers by diminishing peaks of
the extreme torsion loads of the crankshaft and, at the same time, reducing the weight,
face a new solution in the form of the vibration dampers built in the existing
components of the engine crankshaft. So dampers (absorbers) developing today are:
dampers in the form of the dual mass flywheels and
dampers built in the counterweights of the engines crankshaft.
This paper gives comparative analysis of the damping efficiency of the dynamic
viscoelastic dampers and absorbers built in the counterweights on the dynamic load of
the crankshaft. At the end, the adoption of the dynamic absorber with the optimal
characteristics will be shown.
Key words: torsion vibration, crankshaft, absorber, IC engine
1. INTRODUCTION
The variable load transferred by the crankshaft, due to the basic work process
of an IC engine, the randomness of the combustion process, as well as the inconstant
stiffness of the crankshaft on different sections of it, cause a vibration motion of the
entire crank mechanism.
Considering the construction of the crank mechanism, and the way in which the
torques are transmitted, the crankshaft vibrates in three ways: torsional, bending, and
1
Doc. Dr. sc. Aleksandar Milainovi, Faculty of Mechanical Engineering Banja Luka, acom @blic.net
Almir Blaevi, Faculty of Mechanical Engineering Sarajevo, blazevic@mef.unsa.ba
4
Adnan Pecar, Faculty of Mechanical Engineering Sarajevo, pecar_5721@hotmail.com
2
3
681
axial. Due to the intensity of the crankshaft forces and torques, as well as the bearing
methods, torsion vibrations are of the highest intensity. They cause high frequency
dynamic loads, which may lead to material fatigue, and eventually, cranskhaft failure.
Avoiding IC engine crankshaft fatigue is extremelly difficult, and can rarely be achieved
using constructional adjustments, due to the crankshaft working enviroment. Thus, the
most common way of avoiding the crankshaft fatigue is reducing the vibration
amplitudes, using a torsion vibration dampers (TVD). Conventional damper types often
installed on ICE are:
- mechanichal spring or friction dampers,
- elastic dampers,
- viscous dampers,
- dynamic absorbers.
whose appliance is based on the engine type. Elastic dampers have been used the
most, so far. Today, however, aiming to reduce the crankshaft aditional masses, as
well as reducing damper cost, and increasing its reliability, has lead to the
developmnet of dampers integrated into some fundamental parts of the crankshaft
(dualmass flywheel, counterwight dampers).
In this paper, a comparisson analysiss and an adoption of the optimal charachteristhics
crankshaft counterweight damper, shown in figure 1. The damper is mounted on the
side of the crankshaft which is opposite to the flywheel.
Fig. 1 Design of a crankshaft counterweight torsion vibration damper
2. RESEARCH OBJECT
The results were analysed on a four-cylinder, four-stroke, naturally aspirated,
air cooled vehicle diesel engine, as described in lit [1]. Engine power is 150 kW at 2150
rpm, with a maximum torque of 787 Nm at 1500 rpm. The physical model used for the
analysis is a linear torsion-equivalent system, as show in figure 2.
682
Impact of the specific absorbers on the dynamic load of the IC engines crankshaft
Fig. 2 The equivalent torsion-vibration sistem of the analysed engine

The figure illustrates the concentrated masses of the flywheel, on the right side,
along with its inertia member (T7), belt pulley on the left (inertia member T2) and a
counterweight damper (inertia member T1), which can be shown in such a way, due to
its proximity to the belt pulley. The basic data for the shown equivalent torsion vibration
system (inertia momentum (T, stiffness (c), excitation (M ) ) is defined in [1],[2], and
shown in table 1.
Table 1 The basic elements of the model shown in figure 2.
Inertia
Stiffness
c1 variable
T2=0,096 kgm2
T3=...=T6=0,1144 kgm2
T7=1,77kgm2
c2 = 2,18 106 Nm/rad
T1variable
c2 = c3= c5= 3,56 106 Nm/rad
Damping (external)
Q1 = Q2 = Q3 = Q4 = 10
Nms/rad
c6 = 5,47 Nm/rad
The internal damping of the crankshaft H2, ... H6 is calculated by a model

described in [3]. The inertia of the damper, its stiffness and damping are an object of
research, and shall be described as a part of the analysiss of their influence on the
torsion vibrations.
3. THE RESULTS OF THE ANALYSIS
The analysis required a calculation program to be developed. The developed
program calculates the torsion angles caused by torsion vibration, and the
experimental results are used from the lit. [2], [4]. The developed program is based on
the vibration equations of concentrated masses, such as:
Q - H '- c 'Ti i
i i
i
i
i
i
Mi
(1)
which are, in fact, linear differential in homogeneous equations of the second degree.
The method used for the solving of the equations is described in detail in lit. [3].
For the verification of the mathematichal model, as well as the developed calculation
683
program, an engine described in table 1. has been used. The engine was unequipped
with a torsion vibration damper. The comparisson of the results of the experiment and
the calculation results are shown in figure 3, regarding theIvibration mode, and the 9.5,
10th and 12th excitation order.
Fig. 3 Diagram of the amplitude of vibration obtained computationally and

experimentally for three excitation orders
The above figure shows a satisfying match of the calculated and the
experimental results, which indicate that the chosen physical model and the numerical
solution used in the calculation program have been properly selected. Thus, all further
calculations of the torsion vibration can be considered highly accurate.
For the analysis of the torsion vibration in this paper, the following parameters
have been modified: inertia (T1), spring stiffness (c1), while the internal damping (H1)
was held unchanged. The impact of the stiffnes of the springs which are a part of a
counterweight damper is shown on figure 4. The figure shows vibration angles of 9.5,
10 and 12th excitation order for the I vibration shape, for three differrent stiffnes values:
c1 = 0,3 106 Nm/rad, 0,2 106 Nm/rad and 0,1 106 Nm/rad. The figure shows the
influence the stiffness makes to the belt pulley vibration angle amplitudes. The same
figure shows the appropriate vibration angle amplitudes for the belt pulley of the same
engine when equipped with a classic elastic damper.
684
Impact of the specific absorbers on the dynamic load of the IC engines crankshaft
Fig. 4 Diagram comparing the vibration amplitude of the motor pulley for different
values of the damper spring stiffness
By comparing the angular amplitudes of the belt pulley of an engine equipped
with a classic damper to the angular amplitudes of an engine equipped with a
counterweight damper, it can be stated that the same effect can be achieved when
using a counterweight damper as when using a classic elastic damper, by choosing an
optimal damper spring stiffness. The system oscillation frequency is significantly
moved to the right side of the working range, but this has no significant effect to the
dynamic durability of the crankshaft. Stiffness of c1=0,1106 Nm/rad, has given
appropriate angular vibration amplitudes, and therefore it has been held constant,
while optimising the inertia of the counterweight supplementary mass. The inertia has
been altered in a range moving from T1 = 0,02 kgm2 to T1 = 0,08 kgm2. The analysis
was made for theIvibration mode, and the 10th excitation order. The results are shown
in figure 5. The same figure shows the angular vibration amplitudes of the same engine
when equipped with a classic elastic damper.
Fig. 5 Diagram comparing the vibration amplitude of the motor pulley for different
values of the damper inertia moment
685
The given results indicate that the highest damping quality is achieved when
using an inertia of T1 = 0,03 kgm2. In this way, the angular vibration amplitudes of the
belt pulley are even lower than when using a classic elastic damper, with a slightly
increased system vibration frequency.
A more detalied optimisation should take into consideration an alteration of all
relevant damping charachteristics: inertia (T), stiffness (c), and internal damping
coefficient, which exceeds the boundaries of this paper.
4. CONCLUSION
In this paper, based on a self-developed calculation program for calculation of
angular vibration amplitudes, and using an experimental verification, an analysis of the
following elements has been made:
- influence of the inertia of the crankshaft counterweight damper on the angular
vibration amplitudes during a resonance vibration period,
- influence of the counterweight spring stiffness on the angular vibration
amplitudes during a resonance vibration period.
The given results indicate that a crankshaft counterweight damper has the same
efficiency when damping torsion vibration of a crankshaft, as a classic elastic damper.
With a more detailed analysis of the damper parameters, a satisfactory efficiency can
be achieved as when using a conventional damper.
LITERATURE
[1]
[2]
[3]
[4]
[5]
686
Milainovi A., Filipovi I., Hribernik A. (2009). Contribution to Definition of

Torsional Stiffness of Crankshaft of Diesel Engine Used in Heavy Duty Vehicles,
Journal of Automobile Engineering, Part D., Volume 223, Number 7.
Milainovi A. (2007). Mathematical Modeling and Experimental Investigation of
Nonlinear Torsional Vibration of Crankshaft of IC Engine, PhD Thesis, University
of Banja Luka.
Filipovi I. (2007). IC Engine Dynamic and Vibration, Faculty of Mechanical
Enfineering, Sarajevo.
Bombek G., Milainovi A., Filipovi I. and Hribernik A. (2005). Determination of
Torsional Vibrations of a Diesel Engine Crankshaft, In Proceedings of thE
Innovative Automotive Technology, University Ljubljana.
Milainovi A., Filipovi I. and Hribernik A. (2007). Mathematical Model for
Choosing the Flywheel of an IC Engine, In Proceedings of the 8th International
Conference on Accomplishments in electrical and mechanical engineering (DEMI
2007), Banja Luka.
IMPROVEMENTS OF GROUND VEHICLES FUEL ECONOMY USING

REGENERATIVE BRAKING
Jasna Gliovi 1, Jovanka Luki2, Danijela Miloradovi3
Summary: The electrification of conventional vehicles, ranging from passenger
vehicles and trucks to ground vehicles such as agricultural tractors, construction
equipment and military vehicles, can potentially offer improvements in fuel economy
and emissions. Nearly all conventional vehicles are powered by a gasoline or diesel
fueled internal combustion engine that converts the energy stored in the fuel to rotating
power out of the engine. Internal combustion engines are relatively inefficient, typically
converting only 25 to 35 percent of the input energy in the fuel to the output shaft to do
useful work. With heavily fluctuating fuel prices, the total cost of ownership of loaders,
excavators, and other classes of ground vehicles is nowadays strongly influenced by
the fuel costs. Moreover, there is growing concern about CO2- emissions caused by
the burning of fossil fuels as well as about the long-term availability of these fuels. The
military is particularly interested in the electrification of vehicles because the attributes
that these vehicles possess offer exceptional potential for the military to achieve its
long-term objectives of developing a highly mobile, lethal fighting force.
Key words: ground vehicles, regenerative braking, fuel economy
1. INTRODUCTION
The increasing demand for oil is a global issue, and solutions to this problem
are driven by the sudden increase in demand from developing industrial nations, such
as China, sustained demand from developed nations, such as the USA, and the
knowledge that soon the demand for oil will exceed the production capability (Fig. 1).
Hybrid vehicles are just one of a number of solutions suggested by the transport
industry to reduce dependency on oil and harmful GHG emissions. Hybrid has its
origin in the Greek language and means: Mixture or combination of two things. Hybrid
vehicles use two types of energy to attain optimum propulsion. Combustion engine
demand is reduced by hybrid technology, giving the user two big advantages: lower
fuel consumption and a significant reduction in emissions. The most common hybrid
technologies are electric and hydraulic hybrids [1].
MSc Jasna Glisovic, assistant, Kragujevac, Serbia, Faculty of Mechanical Engineering, jaca@kg.ac.rs
PhD Jovanka Lukic, associate professor, Kragujevac, Faculty of Mechanical Engineering, lukicj@kg.ac.rs
3
MSc Danijela Miloradovic, assistant, Kragujevac, Serbia, Faculty of Mechanical Engineering, neja@kg.ac.rs
2
687
Improvements of ground vehicles fuel economy using regenerative braking
Fig. 1 Actual and projected worldwide discovery, extraction and demand for
conventional oil (From 1920 to 2040 in billions of barrels)[1]
Hybrid vehicles are called hybrids because they use both a internal combustion
engine (ICE) and an electric motor to obtain maximum power and fuel economy with
minimum emissions. What all hybrids have in common is the ability to generate electric
current, store it in a large battery, and use that current to help drive the car. Hybrids
have regenerative braking systems that generate electric power to help keep the
batteries charged. When the driver applies the brakes, the electric motor turns into a
generator, and the magnetic drag slows the vehicle down. For safety, however, there is
also a normal hydraulic braking system that can stop the car when regenerative
braking isnt sufficient. Theres no difference in maintenance or repair except that the
brake pads tend to last much longer because they dont get used as much. The big
difference is that regenerative brakes capture energy and turn it into electricity to
charge the battery that provides power to an electric motor. Hybrids can also conserve
energy by shutting down the ICE when the vehicle is in Park, idling at a light, or
stopped in traffic, or when the electric motors energy is sufficient to drive the vehicle
without assistance from the ICE [2].
With heavily fluctuating fuel prices, the total cost of ownership of loaders,
excavators, and other classes of ground vehicles is nowadays strongly influenced by
the fuel costs. Moreover, there is growing concern about CO2-emissions caused by the
burning of fossil fuels as well as about the long-term availability of these fuels. The fuel
economy and efficiencies of the drive train and the hydraulic implements have
therefore become extremely important parameters in the design of future ground
vehicles. Hybrid transmissions are now considered to be a solution for ground vehicles.
Furthermore, hybrid electric vehicles need sophisticated electric transmissions with
delicate and expensive inverters, converters, and batteries. Taking the extreme power
transients in mobile machinery and the rough operational conditions of off-road drive
trains into account, it is questionable whether the delicate hybrid electric drive trains
can be considered a viable, inexpensive, and robust option for ground applications.
Yet, there is still the alternative of the hydraulic hybrid drive train-a full series hybrid
system with an in-wheel motor in each wheel and hydraulic transformers for efficient
power control. For military vehicles, the drivers for hybrid electric vehicles are quite
different; although the fuel economy remains a desirable feature for both the
commercial and military sectors. The military needs for hybrid electric reside in their
ability to generate significant level of electric power onboard the vehicles to meet the
demand of the warfighter thus expanding his mission capabilities in terms of mobility,
survivability and lethality. However, the fielding of hybrid electric military vehicles has
688
been much slower than the commercial vehicles due to the technical challenges that
must be overcome before the hybrid technology can be considered viable for military
applications [3].
2. FUEL SAVINGS FOR GROUND VEHICLES
2.1 Hybrid-electric ground vehicles
The idea of hybrid-electric drive is based on maximizing the energy efficiency
of fuel. In the classical drive the energy disappears by transforming into the heat in all
drive components from the fuel tank to the drive wheels of vehicles, and most in the
internal combustion engine that converts chemical energy of fuel into useful work. The
causes of this are numerous, but one of the biggest is that the engine is running in a
very wide range of number of revolutions and loads while working with varying degrees
of efficiency ration.
Fig. 2 Diagram of the relative fuel consumption of a typical internal combustion engine
and the paths of energy through a typical gas-powered vehicle in city driving
Figure 2 shows the relative fuel consumption of a typical ICE and the paths of
energy through a typical gas-powered vehicle in city driving. If, for example, the engine
delivers 100 kWh of energy from areas of least fuel consumption, it will be spent 22
liters of fuel, and if that same energy draws out from the area of the engine works at
low load, it will be spent up to two or even three times more fuel. If the engine could
keep the regimes of minimum consumption, which means a load of more than 50%
and the number of rotation around the maximum torque, then you could save a
considerable amount of fuel for the same usable energy. This is not possible in a
classical drive chain (motor-transmission-wheels). It is necessary, therefore, to
separate the engine physically from the drive wheels and to establish a completely
different structure of the propulsion system, in order to achieve full control of engine
operating modes according to the number of rotation and load [4].
There are tree basic types of hybrid-electric drive: serial, parallel and plug-in
(Fig. 3). A series hybrid uses a gasoline or diesel ICE, coupled with a generator, to
generate electricity but not to drive the car. The engine can send the electric current
directly to the electric motor or charge a large battery that stores the electricity and
delivers it to an electric motor on-demand. The electric motor propels the vehicle, using
its power to rotate a driveshaft or a set of drive axles that turn the wheels. A parallel
hybrid uses both an electric motor and an ICE for propulsion. They can run in tandem,
or one can be used as the primary power source with the other kicking in to assist
689
when extra power is needed for starting off, climbing hills, and accelerating to pass
other vehicles. Because both are connected to the drive train, theyre said to run in
parallel. Because plug-in hybrids feature larger batteries that can be charged at any
ordinary electrical socket, they have the capacity to extend the ability of the electric
motor to drive the car farther without the need for starting the ICE and therefore
substantially increase the vehicles fuel efficiency. Estimates have ranged as high as
100 mpg!
Fig. 3 The basic structure of serial, parallel and plug-in hybrid-electric drive
2.2 Hidraulic hybrid ground vehicles
The HRB Hydrostatic Regenerative Braking System helps meet increasingly
strict environmental regulations and reduces operating costs by saving fuel. The HRB
system will provide superior return on investment. The HRB stores a vehicles kinetic
energy, which would otherwise be lost during mechanical braking operation. This
energy is then available for powering the vehicle and reducing primary energy use.
Fig. 4 Characteristics of a electric hybrid and hydraulic hybrid

Characteristics of electric hybrid: Excess motor power is continuously
accumulated in a battery over a longer time period (blue) and accessed as needed
(light blue); High energy density and low power density: The battery can absorb a great
deal of energy, but the charge time is relatively long, so it is not possible to fully
recapture the braking energy; Energy is stored in batteries; Typically found in
passenger cars.
Characteristics of hydraulic hybrid: The kinetic energy from braking is fed to a
hydraulic accumulator (blue) and immediately reused for starting (light blue). Hydraulic
hybrids are ideal for vehicles with frequent, short start-stop cycles, such as public
transit buses, refuse trucks, forklifts, pneumatic tire rollers, telehandlers, swap body
movers and much more; High power density and low energy density: there are limits to
the amount of energy the system can accumulate. However, it takes less time to collect
690
and store this energy, which can be called upon as needed. The full braking energy is
then fed to a hydraulic accumulator and stored.
There is a common consensus that hybrid drive trains can strongly increase the
efficiency of a drive train. However, the increased efficiency does not by itself result in
a reduction of the fuel consumption or the cost of ownership. The electric drive train
components result in an increased vehicle weight, which increases the fuel
consumption of the vehicles. Moreover, the added cost for the electric components
gives doubt about the cost-benefit-relationship, especially of the full hybrid drive train
concepts. Nevertheless, the trend of hybrid electric transmissions has also come to the
ground vehicles market. Kobelco and New Holland have presented an excavator, in
which the swing movement of the upper carriage is realized by means of a hybrid
electric drive. Because of the hybrid system, a smaller internal combustion engine
could be applied and the fuel consumption is reduced by 40%. In Europe, Deutz and
Atlas Weyhausen have shown a hybrid electric wheel loader, in which the flywheel of
the internal combustion engine is replaced by an electric motor/generator. According to
the developers, the fuel consumption is reduced by 20%. Finally, also Volvo
Construction Equipment (Volvo CE) has announced a concept of a hybrid electric
loader called the Gryphin. The mechanical drive train in the Gryphin is completely
replaced by an electric drive, having in-wheel electric motors in all four wheels. It has
an in-wheel hydrostatic machine in each wheel, hydraulic transformers instead of
electric power controllers, and hydraulic accumulators instead of batteries. Added to
this, the hydraulic hybrid system also includes recuperation modes for the hydraulic
cylinders [3].
2.3 Military Applications of Hybrid Cars and Trucks
Combat vehicles can also have benefit from the hybrid-electric drives. Some of
the most important are the following: Fuel savings-As with civilian vehicles and in
combat and other vehicles for military purposes, narrowing the area of combustion
engines can contribute to significant fuel savings (in some driving conditions even
more than 50%). Fuel savings in military vehicles is important not only because of the
fuel but also because there are other repercussions: simplifying and reducing costs of
logistics, increasing the autonomy of the vehicles and so on. The issue of
environmental protection is also not unimportant-lower fuel consumption means lower
emissions and other toxic substances. Low acoustic signature of the vehicle-Hybridelectric drive reduces the possibility of observing the vehicle through a lot of elements.
One of the most important is that within the limits of available energy in batteries, the
vehicle can be powered purely electrically powered, which means no noise, heat
radiation and no emission gases. This ride can be of crucial importance in military
operations. But even when the vehicle is driven in a combined operation, observing of
the vehicles has been reduced because:
IC engine runs at lower numbers of rotation, and there is no sudden
changes, which means less noise,
hydraulics for propulsion of the auxiliary systems to power the vehicle can
be completely eliminated, thereby eliminating the noise of hydraulic pumps
and motors, which is usually considerable,
reduced noise from cooling fans and others.
691
Fig. 5 Hybrid Electric Combat Systems

With hybrid-electric drive, vehicle has electricity available in unlimited quantities
that can be used to drive auxiliary devices, and other needs of the dome cars, and
even for purposes outside the vehicle. New technologies of electric weapons and
electromagnetic protection (EM shielding) are useless without a strong source of power
supply. There are many other benefits of hybrid-electric drive as: modularity of the
system easier to maintain, complete automation and control reduce the human factor
and provide a better diagnosis, flexibility in the design of the propulsion system
enables better use of space, and others.
3. CONCLUSION
Alternative drive is the ultimate solution for propulsion of any kind of vehicle.
Hybrid-electric drives as bridge towards a more radical future will certainly spread
rapidly on vehicles like cars, buses, light utility vehicles and ground vehicles, also. Fuel
economy however has become equally important. The proposed hybrid drive and
control system enables the design of a new generation of ground vehicles with a
strongly reduced fuel consumption, while maintaining the productivity. Military,
especially combat vehicles will probably be the last to replace fossil fuels by an
alternative, given its energy density. Heavy hybrids have still too much drawbacks and
it is hard to believe their mass application on combat, especially tracked vehicles. Light
and mild hybrids seem a more realistic model for the near future which may marry
conventional with new technologies to get substantial improvements not only in fuel
economy and performances but also in many other aspects concerning combat
vehicles.
This paper was realized within the researching project ref. no. 35041, funded
by Ministry of Science of the Republic of Serbia.
LITERATURE
[1]
[2]
[3]
[4]
692
Gilbert, R., Background paper for a post Kyoto transport strategy (06/07/02).
Pistoria, G. (2010). Electric and hybrid vehicles. Elsevier, Netherlands
Achten, P., Innas B.V. (2009). A Series Hydraulic Hybrid Drive Train For Off-Road
Vehicles. Fluid Power Journal, Off-Highway, p. 8-12.
Ilijevski, Z., Hibridno-elektrini pogon vozila za vojne potrebe, from http://www.
hrvatski-vojnik.hr/hrvatski-vojnik/0212005/hibrid.asp, accessed on 2010-09-07.
INFLUENCE OF TRAFFIC ON AIR QUALITY IN NI

Predrag ivkovi1, Mladen Tomi2, Gradimir Ili3, Mirko Dobrnjac4, Vladimir
Lazovi5,
Summary: In this paper traffic induced pollutant emission estimation on the City of Ni
territory was done. It is well known that traffic is the largest single air pollution source in
Europe, as well as in Serbia and in Ni. Emission range estimation was performed
using COPERT software. For the purpose of emission data validation, traffic intensity
measurements were done on the main crossroads in the City. Simultaneously, CO2
concentration, as the representative imission component, was measured on 1 m
distance from the road. Fleet composition was obtained from the Ministry of Internal
Affairs of Republic of Serbia. On the basis of the District heating plant in Ni and
available statistical data, obtained on the Republic Bureau of Statistics web-site,
estimation of the traffic induced pollution, as overall pollution component in the City of
Ni was done.
Key words: COPERT, traffic, emission, CO2, CO
1. INTRODUCTION
Traffic induced pollutants (CO, NOx, SO2, SOX, VOC) are well documented in
a number of investigations [1, 2]. Recently, with the increasing global warming
problem, the CO2 emission becomes more important. It is known that the traffic os one
of the largest CO2 emission sources. It is estimated that traffic represents about 10% of
global, and about 20% of european anthropogenic CO2 emission[1]. As the need for
mobility rises, especially in the countries with rapid urbanization as India i China [3,4],
traffic becomes the largest air pollution source in urban areas [3,5]. Although CO2
measurement is not very complicated, the definition of traffic as a CO2 emission source
is highly complicated, considering its stohastic nature [4].
Lately, many different methods for traffic induced emission were developed.
Such estimations are very important for the more efficient air quality management.
Traffic induced emission depends on a number of parameters. Vehicle type, engine
size, age, fuel used, driving style, are only some of the parameters influencing the CO2
emission. In this paper, so-called macro approach was adopted. On the basis of
1
PhD, Predrag ivkovi, Ni, Mainski fakultet Univerziteta u Niu, pzivkovic@masfak.ni.ac.rs

BSc, Mladen Tomi, Ni, Mainski fakultet Univerziteta u Niu, mladentomic@yahoo.com
3
PhD, Gradimir Ili, Ni, Mainski fakultet Univerziteta u Niu, gradei@masfak.ni.ac.rs
4
PhD, Mirko Dobrnjac, Banjaluka, Mainski fakultet Univerziteta u Banjaluci, dobrnjac.mirko@gmail.com
5
BSc, Vladimir Lazovi, Ni, Mainski fakultet Univerziteta u Niu, lazovicvladimir@ymail.com
2
693
Predrag ivkovi, Mladen Tomi, Gradimir Ili, Mirko Dobrnjac, Vladimir Lazovi
information available, the overall emission is being estimated. Such data can be used
for creating the national inventory of gasses [2,6,7]. Newest investigations are making
effort to estimate emission as the function of the engine size, type and the average
speed of a vehicle [8, 9,10]. On the basis of this research, different models were
developed, and are being used for emission prediction [1,2,11, 12].
In this paper, COPERT [13] software was used. From the traffic intensity
measurements on the main crossroads in the City, and from the statistical data
obtained from the Ministry of Internal Affairs of Republic of Serbia, CO i CO2 emission
levels were estimated, and afterward validated.
2. CALCULATING EMISSION
City of Ni fleet composition data used in this paper were obtained from the
Department of IT - PU Ni. Based on these data, one can notice that passenger cars
occupy by far the largest share of the fleet. Vehicles with petrol engines are the most
common, making almost 2/3 of the fleet. The average age of vehicles is about 14
years. This data were used for the rate of CO2 emission estimation. The values are
calculated using the methodology COPERT. Basically, COPERT was used as a tool for
assessing emissions from traffic at the national level. COPERT operation is based on
the analysis of large amounts of data from several European vehicles testing facilities.
This is very reliable methodology, since the actual high-level agreement of data
obtained [15]. Contribution to the total emission (EC) in gkm-1s-1 for each sub-category
of vehicles is calculated from the formula:
n
EC
i 1
numi EFi
't
(1)
If we adopt a particular vehicle speed u as a constant, then emission factor can

be calculated as:
EFi
a u2 b u c
(2)
Former expression, from which we calculate the emission factor, is generally

approximated by a second degree polynomial function of (average) vehicle speed
[12,9]. Although EFi is not constant, it is assumed that, for the adopted average vehicle
speed in the City of 23km/h, can be treated as constant.
2.1 Fleet in the City of Ni
Personal vehicles
By its number, passenger cars, by far, represent most of the fleet, with number
of 60720 a 91.12% of the fleet. Before inputting the data into the COPERT, statistical
averaging was carried out while processing the data obtained from the Department of
IT - PU Ni, in order to adjust them for use in COPERT. For the average mileage per
year in the City, the data of 3500 km per year was adopted, with an average speed of
23 km/h. The speed was adopted by monitoring the driving habits. In average, driver
crosses 10 kilometers per day in the City.
694
Influence of Traffic on Air Quality in Ni
Buses
In the City of Ni total of 477 buses has been registered. However, when
calculating the emissions, in doing so, the public transport company, which, according
to data from 2009. [15], possesses 124 buses and traveled 8 609 250 km, have far
more influence than other transport companies.
Trucks
During traffic flow counting, light trucks, which are mostly up to 2000cm3 engine
size, were counted among passenger cars. For heavy trucks, it has been adopted that
all vehicles over 2000cm3 from IT Sector category data, represent this category.
Motorcycles
During the time period when the monitoring was done, the number of
motorcycles on the streets was negligible. Also, motorcycle number itself is negligible
in relation to motor vehicles.
Fig. 1 The Composition of the Fleet in Ni, and Fleet structure according to
Fuel Type, Engine Size and Age
Table 1 Number of vehicles and average annual mileage on the City of Ni territory
Vehicle Type
Personal Vehicles
Trucks
Buses
Vehicle No.
58049
2286
124
Annual Mileage in the City

3500
2000
69500
695
2.2 Emissions
Calculating emissions was performed in the software COPERT on the basis of
the data obtained from Department of IT - PU Ni. Vehicles were divided as:
Personal Vehicles
Buses
Trucks.
Results are shown in the tables 2 andi 3.
Table 2 Overall annual CO2 i CO emission per vehicle category
Vehicle type
Annual CO2 emission (t)
Annual CO emission (t)
Personal Vehicles
67034,28
2054,03
Trucks
2996,1
9,84
Buses
17669
40,82
Table 3 Specific CO2 and CO emission per kilometer and per per vehicle category
Vehicle type
Spec. CO2 emission (kg/km)
Spec. CO emission (kg/km)
Personal Vehicles
0,330
0,010
Trucks
0,660
0,02
Buses
2,050
0,0047
3. RESULT DISCUSSION
3.1 CO Emission
For a theoretical consideration of CO2 emissions in gasoline engines, octane
combustion model can be used:
2C8H18 25O 2 o 16CO 2 18H2O
(3)
From the above equation it can be concluded that the complete combustion of
228 kg of octane (petrol), produces 704kg of CO2. If we assume that during a city
cruising, a passenger car consumes 1l of gasoline for a 8km trip, we get the CO2
emissions of around 0.290 kg per km. One can notice that exhaust gasses consists of
52,941% superheated steam and 47,059% CO2 (for incomplete combustion sum of CO
and CO2). While analysing a group of over a 100 vehicles, the average volumetric ratio
of CO in the exhaust gasses was 1,93%. Results obtained by COPERT on the level of
molar masses shows that emissions are 0,0075 kmol/km for CO2 and 0,00036 kmol/km
for CO, which is equal to the volumetric ratio of CO in the exhaust gasses of 2,15%.
Results obtained with analysis in COPERT, assumes that the passenger vehicle emits
per kilometer on average 0.330 kg of CO2, which is the error of 12.1% in agreement
with theoretical results, as well as [8].
3.2 CO2 Emission
For CO2 emission analysing, measurements on the location Jagodina Mala will
be used. During the experiment, simultaneous vehicle counting and CO2 concentration
measurement on the distance of 1m from the road for neutral atmospheric conditions.
Number of vehicles was written down every 5 min, so the measured CO2 concentration
696
Influence of Traffic on Air Quality in Ni
was averaged over the same period. Applying values from Table 3 in the formula (1),
the diagram shown in Figure 2 was obtained. Good aggreement between measured
CO2 concentrations and predicted CO2 emission from COPERT can be noticed.
700
600
y= 5E 07x 2 0.005x + 613.94
R 2 = 0.5127
500
400
K onc entrac ijaC O 2(ppm)
300
E mis ijaC O 2(g/km)
200
y= 6E 07x 2 0.0007x + 223.43
R 2 = 0.4098
100
0
0
5000
10000
15000
Fig. 2 Comparison of specific traffic CO2 emission nd measured CO2 concentration on

the location in the City of Ni
J ag odinMala29.10.09.
350
300
automobili
kamioni
B rojVoz ila
250
200
150
100
50
0
5
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 0
Fig. 3 Measured daily traffic intensity distribution for one location and distribution
aproximation for all measuring locations
4. CONCLUSION
Finally, one can conclude that software package COPERT is efficient tool for
traffic induced pollution estimation. As can be seen in the Table 4, traffic is the most
influential emission factor in the City of Ni. With almost 88000 tCO2 per annum, it
covers over 38% of all emissions. This confirms that the traffic is the largest single air
pollution source.
697
Table 4 Category ratio in the CO2 balance in the City of Ni [16]

Heat source
District heating plant
Heating with wood
Heating with coal
Traffic
Sum
Power [TJ] IPCC emiss.fac.[kgCO2/TJ]CO2 emision [t]

1284,96
56100
72086.26
315,44
112000
35329.28
356,73
198500
35156.55
87699.38
230271.47
Ratio
31.30%
15.34%
15.26%
38.09%
100.00%
LITERATURE
[1] Nejadkoorki F., Nicholson K., Lake I., Davies T., An approach for modeling CO2
emission from road traffic in urban areas, Sc. of the total env., 406 (208) 269-278.
[2] Anali Machado1 et all,Air emissions inventory of the public transport in maracaibo
municipality. Part I: Passenger car, Revista Tcnica de la Facultad de Ingeniera
Universidad del Zulia, ISSN 0254-0770 versinimpresa
[3] Nesamani KS, Estimation of automobile emissions and control strategies in India,
Science of the Total Environment, (2010), doi:10.1016/j.scitotenv.2010.01.026
[4] Yuchuan DU, Yuanjing GENG, Lijun SUN, Simulation model based on Monte
Carlo method for traffic assignment in local area road network, Front. Archit. Civ.
Eng. China 2009, 3(2): 195203, DOI 10.1007/s11709-009-0032-3
[5] Sturm P, Almbauer R, Sudy C, Pucher K. Application of computational methods
for the determination of traffic emissions, Air Waste Mng. Assc. 1997;47:120410.
[6] Baldasano J.M., Guidelines and formulation of an upgrade source emission model
for atmospheric pollutants In: Power, H., 1998.
[7] Vehicle speed emission factor database for all pollutants (version 02/3). UK:
National Atmospheric Emissions Inventory; 2003.
[8] Andre, U. Hammarstrom, Driving speeds in Europe for pollutant emissions
estimation, Transportation Research Part D 5 (2000) 321-335;
[9] Washburn S., Seet J., Mannering F., Statistical modeling of vehicle emissions
from inspection/maintenance testing data: an exploratory analysis, Transportation
Research Part D 6 (2001) 21-36;
[10] S. Nesamania, Lianyu Chub, McNallyc M.G., Jayakrishnan R., Estimation of
vehicular emissions by capturing traffic variations, Atmospheric Environment 41
(2007) 29963008;
[11] de Haan P., Keller M., An Emission factors for passenger cars: application of
instantaneous emission modelling, Atmospheric Environment 34 (2000) 46294638;
[12] Kassomenos P., Spyros Karakitsios, Costas Papaloukas, Estimation of daily traffic
emissions in a South-European urban agglomeration during a workday. Evaluation
of several what if scenarios, Science of Total Enviroment, 370 (2006) 480490;
[13] Ntziachristos L., Samaras Z., COPERT 3Computer Programme to calculate
Emissions from Road Transport, Methodology and emission factor (version 2.1).
Copenhagen: European Environment Agency; 2000a.
[14] Marmur A, Mamane Y. Comparison and evaluation of several mobilesource and
line-source models in Israel. Transp Res Part D Transp Environ 2003;8:24965.
[15] http://www.jgpnis.rs/index.php/ukdes.html
[16] R. Srbija,Republiki zavod za statistiku, Bilans uglja i Bilans za ogrevno drvo 2009.
698
INVESTIGATION OF DYNAMICS CHARACTERISTICS OF HYBRID

VEHICLES
Boran Pikula1, Ivan Filipovi2, Mirsad Trobradovi3
Summary: The hybrid vehicles become more and more popular on the automotive
market. Almost all automobile fairs, in recent history, were marked by car
manufacturers with presentation of the new hybrid vehicle, even plug-in electrical
vehicles. In case of the hybrid vehicles, the combinations of IC Engine and
Electromotor can be found in different form of parallel and serial connections. Using
one of the mentioned forms, the best use of total energy (power and torque) for vehicle
propulsion with lower fuel consumption can be achieved. In order to achieve lower fuel
consumption by use of hybrid vehicles in urban areas with limited speeds, only the
electromotor is used. However, what can we expect regarding the vehicle dynamics
characteristics from these vehicles? The answer to this question is given in this paper
with the basic objective to present dynamics characteristics of hybrid vehicles during
different driving cycles.
In order to investigate the dynamics characteristics of hybrid vehicles, the
measurements of velocities, accelerations, and braking or accelerating distance were
done by modern devices with working principle based on GPS and accelerometers.
Taking in to account the importance of other engine parameters of IC engine,
electromotor and vehicles, by use of special device for collecting data from the CAN
BUS, the following parameters are recorded simultaneously: engine load, gas pedal
position, engine speed, percents of batteries charged, braking by engine, etc. On this
way all necessary data for the analysis were collected.
Key words: hybrid vehicle, vehicle dynamics characteristics
1. INTRODUCTION
Increasing climate change and increasing oil prices on the world market for
years are imposing general trends in the development of modern vehicles lowering
harmful exhaust emissions and fuel consumption reduction. Increasing energy
efficiency and environmental acceptability is undoubtedly an imperative in the
development of new vehicles and new technologies for vehicles. In this sense, on the
present-day level of technology development, hybrid vehicles represent the optimal
1
Prof. Dr. sc. Boran Pikula, Sarajevo, Faculty of Mechanical Engineering Sarajevo, pikula@mef.unsa.ba
3
Mr. sc. Mirsad Trobradovic, Sarajevo, Faculty of Mechanical Engineering Sarajevo, trobradovic@mef.unsa.ba
2
699
compromise of required performances, and every day are becoming more pervasive in
the automotive market.
By definition, any vehicle that has two or more sources of energy and power
converter is called a hybrid vehicle. For the purpose of regeneration of the part of the
energy generated during braking, hybrid cars typically have at least one two-way
source of energy and a power converter. Today's hybrid vehicles typically have two
sources of energy, i.e. power converters - internal combustion engine and electric
motor. For regeneration of the energy generated during braking the motor/generator
with the battery is used. IC engine and electric motor, as well as any other energy
sources in hybrid vehicles may be functionally linked in several ways, and therefore
hybrid vehicles are divided into four groups: series hybrid, parallel hybrid, seriesparallel hybrid and complex hybrid (Figure 1).
a) series hybrid, b) parallel hybrid, c) series-parallel hybrid, d) complex hybrid
Fig. 1 Classification of hybrid vehicles schematic view of component connection

Modern hybrid vehicles provide an optimal compromise between the
characteristics of classic vehicles and electric vehicle. These vehicles retain good
dynamic performances and long operating range, which modern internal combustion
engine provides, while ensuring energy conversion efficiency and low exhaust
emissions.
To evaluate dynamic characteristics of a modern hybrid vehicle, testing of the
basic dynamic parameters during the motion of vehicles are carried out, for different
driving regimes and road conditions. The results are presented in this paper.
2. INVESTIGATION OF DYNAMICS CHARACTERISTIC OF VEHICLE
In order to study the dynamic characteristics of the vehicles the measuring of
different dynamic parameters during acceleration and braking was done, in different
driving regimes of the vehicle. Testing was performed on a modern hybrid vehicle,
Toyota Prius III. Schematic representation of the drivetrain, while acceleration and
braking, is shown in Figure 2
Its important to mention that the tested hybrid vehicle has several possibilities
700
Investigation of dynamics characteristics of hybrid vehicle
of drivetrain control, which can be selected as desired. In addition to "standard" mode,

the PWR mode (power) can be selected, in which the control unit optimizes the
drivetrain in order to achieve better acceleration performance of the vehicles, and ECO
mode (ecology), where the optimization is done by fuel consumption.
a)
b)
Fig. 2 Operating principles of hybrid vehicle powertrain, during accelerating (a) and
braking (b) - schematic representation
Recording of the velocity and acceleration, or deceleration of the vehicle was
carried out using a measuring device Racelogic DriftBox, based on GPS technology.
The device uses a 10 Hz GPS module connected to accelerometer to measure the
acceleration in the longitudinal and lateral direction. Since the other parameters of the
engine, electric motor and vehicle are of great importance too, with the help of a
special device which collects data from the CAN BUS (DashDaq II), engine load,
throttle pedal position, engine speed, battery charge level and state of fuel injection
system (FIS) are registered at the same time. This way the relevant data used in the
analysis is collected.
2.1 Testing of vehicle acceleration
For analysis of the vehicle characteristics during acceleration, the testing of the
vehicle acceleration was carried out with maximum accelerator pedal position (max.
engine load) on straight road, and in urban driving conditions.
The testing of the vehicle acceleration was carried out on straight road with
negligible tilt. Testing was done during the winter, on dry road, with winter tires. The
study was conducted in PRW mode. Test results are shown in Figure 3.
Fig. 3 Vehicle velocity and acceleration, during vehicle acceleration test

701
To accelerate the vehicle from standstill to the speed of 100 km/h was required
11.5 s, which is due to vehicle payload (gross vehicle weight 1380 kg + 3 persons) and
winter tires, very satisfying result. Factory data accounted for 10.4 s.
It is interesting to notice the character of the vehicle acceleration curve, which
is the result of application of CVT transmission. Changing gear ratios in transmission is
continuous, with uninterrupted power flow, which increases the efficiency of power
transmited from engine to wheels. At the same time the driving comfort is increased.
Testing of the vehicles acceleration performance in urban driving conditions,
was performed in two different modes of the drivetrain control, PRW and ECO.
Characteristic parameters such as vehicle velocity, engine load, engine speed, battery
charge level were analyzed. The results are shown in Figure 4.
a)
b)
Fig. 4 Vehicle acceleration a) ECO mode, b) PWR mode

Analyzing the obtained results the differences in the vehicle drivetrain mode
can be seen, for two different control modes. In ECO mode (optimum fuel
consumption) vehicle is first powered by an electric motor, so that after 1 s IC engine
would start too. In PWR mode (speed performance) IC engine is started from the
beginning of the vehicle drive.
Maximum gear ratio in the ECO mode is 4.2 and is achieved 2 s after the start
of the acceleration. In PWR mode, the maximum gear ratio is 6.3 and is achieved 0.8 s
after the start of the acceleration. This way greater traction force and better vehicle
acceleration are enabled.
The drop in the engine load and engine speed at time t = 2 s for testing in PWR
mode is the result of short-term release of the gas pedal the vehicle was tested in
urban conditions, where other participants in traffic are present too.
Based on the vehicle velocity the change of the gear ratio is calculated, which
is shown in Figure 5. Figure 5 also shows the trend of the vehicle velocity and the
702
Investigation of dynamics characteristics of hybrid vehicle
engine speed for a vehicle with similar characteristics with conventional 6-speed
gearbox. In comparison to conventional transmission, CVT transmission provides
continuously gear change, with uninterrupted power flow, while retaining the IC engine
in the optimal regime of engine speed with lower fuel consumption.
Fig. 5. Vehicle and engine speed vehicle with manual and CVT transmission
2.2 Testing of vehicle braking characteristics
To analyze the characteristics of the vehicle during of braking the testing of the
deceleration of the vehicle for two different driving regimes is carried out: alternating
accelerating and full braking of the vehicle, as well as maintaining vehicle velocity in
downhill movement. Vehicle velocity, engine speed, battery charge level and state of
the fuel supplying system (FIS: 1 - engine off, 2 - engine on, 4 - engine braking: ignition
and fuel supply off). The results are shown in Figure 6
a)
b)
Fig. 6 Characteristic parameters of vehicle braking a) intensive braking, b)

maintain desired vehicle speed
703
In the first case (Fig. 6a) testing was conducted on a test section of 2 km in
length, with several cycles of vehicle accelerating and full braking. Braking is done with
the transmission on (gear lever in position D - drive). As can be seen from the diagram,
during each braking the IC engine turns off (FIS state 1). Part of braking energy returns
to the system via the electric motor/generator, as evidenced by increasing the battery
power state (an increase of 8%).
From the diagrams can also be noticed that every stopping and non-moving of
the vehicle is followed by turning off of the IC engine. Tested hybrid vehicle is equipped
with start/stop system that switches off the engine whenever the vehicle is stationary.
This way the fuel consumption and engine noise are reduced. After re-pressing the gas
pedal start/stop system starts the IC engine. According to the manufacturer, start/stop
system reduces fuel consumption and CO2 emissions by as much as 8%, depending
on the vehicle.
In the second case (Figure 6.b), testing was conducted on a road section of 1
km in length, with a fall of 3.1%. Braking is performed on the downhill section, in order
to maintain the desired vehicle speed (lever in position B - engine braking). Braking is
done by engine (FIS state 4), where the ignition and fuel supply are turned off. At the
same time this process is accompanied by a significant increase in the battery power
state (an increase of 5.5% on a section 1 km).
3. CONCLUSION
Hybrid vehicles, designed as a functional combination of IC engines and
electric motors, are becoming evidently present in the automotive market. Modern
hybrid vehicles offer the optimum compromise between the characteristics of classic
vehicles and electric vehicle they maintain good dynamic performance and long
operating range, while reducing fuel consumption, exhaust emission and noise.
Confirmation of this hypothesis was given by conducted tests of dynamic
characteristics of a hybrid vehicle, shown in this paper.
LITERATURE
[1] Ehsani, M., Gao, Y., Emadi, A. (2010). Modern Electric, Hybrid Electric and Fuel
Cell Vehicles, 2nd Edition. CRC Press, Taylor & Francis Group, Boca Raton, USA.
[2] Larminie, J., Lowry, J. (2003). Electric Vehicle Technology Explained. John Wiley
& Sons Ltd, Southern Gate, Chichester, UK
[3] Miller, J. M. (2004). Propussion Systems for Hybrid Vehicles. The Institution of
Engineering and Technology, Stevenage, Herts, UK
[4] Gillespie, T. D. (1992). Fundamentals of Vehicle Dynamics. Society of Automotive
Engineers, Warrendale, PA, USA
[5] Daci, S., Trobradovi M., Pikula B. (2008). Osnove dinamike vozila zbirka
zadataka. Mainski Fakultet Sarajevo.
[6] http://techdoc.toyota-europe.com
704
LANDFILL GAS AS A FUEL FOR A VEHICLE FLEET FOR THE CITY

LANDFILL
Vladan Ivanovic1, Decan Ivanovic2, Vladimir Pajkovic3
Summary: By building sanitary landfill Livade in Podgorica, the conditions were
made for building system for landfill gas capturing, which is the first phase in control of
emission from landfill and its efficient usage.
Extraction and combustion of methane, the main component of landfill gas represents
a significant contribution to reducing emissions of greenhouse gases effect.
This paper describes one possible use of landfill gas as a fuel for vehicle fleet of
landfill. This way of using landfill gas has significant ecological and economical effects
on the operation of the landfill.
Key words: landfill gas, sanitary landfill, municipal solid waste
1. INTRODUCTION
Landfill of municipal solid waste is located south-east from Podgorica, 5 km
from city centre, on the location Livade, just next to local road Podgorica Dinosa,
near the Cijevna river, which flows to the south. The landfill area, until now, has been
the place for unselective and uncontrolled waste dump for more than 50 years, on
classical way and without appropriate measures in human environment protection. On
the landfill primarily is dumped waste from communities Podgorica and Danilovgrad,
and since the beginning of 2008. waste from some seashore communities.
With main project of rehabilitation and organization of city landfill, it is expected
that the waste be dumped in organized and arranged way, according to the regulations
that apply in countries of European Union, where priority is given to human
environment protection.
Because of that, as a part of solution to the waste disposal and storage
problem, considering that release of landfill gas represents physical (explosion),
chemical (harmful substance in surrounding air) and physiological (odour) risk for
people who work and live in the vicinity of landfill, system for landfill gas capturing is
predicted, its transport in controlled way and in the first phase its combustion
(treatment) [1,2].
The sanitary landfill (in its final form) should consist of six cells dimensions 100
1
Prof. Vladan Ivanovic Ph.D, Montenegro, Faculty of Mechanical Engineering, vladaniv@ac.me

Prof. Decan Ivanovic Ph.D, Montenegro, Faculty of Mechanical Engineering, decan@ac.me
3
Doc. Vladimir Pajkovic Ph.D, Montenegro, Faculty of Mechanical Engineering, pajkovic@ac.me
2
705
V. Ivanovic, D. Ivanovic, V. Pajkovic
200 m, which are located in the central part of complex. Every cell has implemented
system for collecting and dispositioning leachate waters, system for capturing and
dispositioning of biogas, and system for collecting and dispositioning of pure
atmosphere water. In first period construction of 1.500.000 m3 of landfill is predicted,
with three cells of 500.000 m3 each and it is planned that construction of every cell
should be completed in year that preceded its usage.
Waste layers are stamped down with 6-7 crossings by 29 t compactor. In that
way we reach the density of 750-850 kg/m3. For calculation the average density of 800
kg/m3 is used, which means that total mass of waste of one cell is 400.000 t. Three
cells will be built during start year, 4th year and 12th year; that is, filling of second cell
begins in 2009, and of third in 2017.
2. LANDFILL GAS CAPTURING WELLS
The biogas capturing system predicted by the project is based primarily on
network made of vertical extraction wells, which are during the filling of landfill made by
principle of elevation (fig.1).
Fig. 1 Phase of extraction wells elevation

Determine factor while calculating capturing network, which is organised on
vertical elements, is called radius of influence. Radius of influence is distance in
which well, with applied sub pressure, preserves capability to develop its own action
efficiently. The project of capturing network predicts complete coverage of landfill area
with wells with radius of influence of approximately 25 metres.
For each parcel is predicted 21 extraction wells divided into three routes (fig.
2). In addition, in order to include complete landfill, after the parcel merging, along the
route that separates the parcels 1 and 2, and parcels 2 and 3, another line of extraction
wells will most likely be added.
706
Landfill gas as a fuel for a vehicle fleet for the city landfill
Fig. 2 A cross section through the sanitary cells

3. POSSIBILITY OF USING LANDFILL GAS
Landfill gas, produced by anaerobic digestion, is made as a mixture of
hundreds of different gases. By volume, can be said that it consists primarily of two
gases: 40-60% methane (CH4) and 60-40% carbon dioxide (CO2). In combination with
these gases, oxygen (O2) and nitrogen (N2) are often present, because as the air are
presents in the free interstices in the trash or are drawn from the atmosphere by
dynamic aspiration. In small quantities there are ammonia, sulfur, hydrogen, carbon
monoxide and non-methane organic compounds (NMOCs).
Experiences in many countries indicate that the landfill gas can be utilized to
replace other energy sources. As the energy value of landfill gas containing 50% of
methane is around 17500 kJ/m3, or about 5 kWh/m3 at your disposal is a large energy
potential, since its production carried out continuously for 24 hours.
The best type of project for a particular landfill will depend upon a number of
factors, including existence of an available energy market, project costs, potential
revenue sources, and many technical considerations [9].
LFG can be used directly as a medium quality fuel, primary or supplementary,
burned in boilers or industrial furnaces. In this way, gas is with the pipeline leads to a
nearby customer where it is used with appropriate modifications in existing or new
combustion devices. This manner of using does not require special treatment of landfill
gas. Moisture removal and basic filtration of gas are considered.
707
The usual way of using of LFG is as fuel in gas turbines or reciprocating

engines for power generation or more efficiently in cogeneration or trigeneration plants
with gas machines. These processes required equipment for purification and cooling of
landfill gas because moisture, hydrogen sulfide, siloxans and dust that landfill gas
contains adversely affect on the work of gas plants.
LFG can be used to produce the equivalent of pipeline-quality gas (natural
gas), compressed natural gas (CNG), or liquefied natural gas (LNG). Pipeline-quality
gas can be sold into a natural gas pipeline or used for an industrial purpose. CNG and
LNG can be used to fuel vehicles at the landfill. For these processes it is necessary to
remove carbon dioxide and almost all of the harmful components from the LFG.
Technology for this type of using exist and it is not a technical but economical aspects
of the wider usage of these processes.
4. PURIFICATION METHODS
Purification of the landfill gas is a necessity before it can be effectively utilized
as a fuel or an energy source. A comparison between typical landfill gas properties and
standard utility pipeline quality gas specifications illustrates why this is necessary.
Landfill gas typically has about half the calorific value of an equivalent amount of
pipeline gas. Its hydrogen sulfide content can be as much as 250 times the maximum
allowed under pipeline specifications, with 500 times the allowed amount of water
vapor. The landfill gas can have carbon dioxide, oxygen, and nitrogen quantities 10 to
12 times greater than those allowed by the gas utilities.
The goal of a landfill gas purification process is to produce a stream of
methane with levels of purity as high as is physically and chemically possible. The
purification process usually involves several steps.
First, intrusion of atmospheric nitrogen and oxygen into the well field must be
limited as much as possible. Second, the moisture content of the landfill gas stream
must be removed. Next to be eliminated is the hydrogen sulfide and the other trace
contaminants. The last, and most significant, component to be removed before the
conversion to useable methane is complete is the carbon dioxide, which can constitute
the bulk of the landfill gas stream. Various methods are available for removal of each
type of unwanted compound from the gas stream.
Solvent absorption is a preferred method of extracting carbon dioxide from
industrial gases in general and for purifying natural gas in particular. In this process,
liquid chemicals are used to absorb carbon dioxide, which is then released when the
absorbing liquids are subject to higher temperatures. The first step occurs in an
absorption column filled with the extracting liquid. As the landfill gas is bubbled through
this column from below, the carbon dioxide is stripped from the gas stream, the rest of
which continues to the top of the column in purified form. The carbon dioxide laden
liquid is pumped to a desorption column where it is heated to 120C, releasing the
carbon dioxide. The stripped liquid is then recycled back into the absorption column.
Pressure swing adsorption (PSA) is a non-cryogenic separation process that
functions under near ambient temperatures. It is based on the fact that different gases
have different propensities to be attracted to different materials and surfaces. A
nitrogen removal process utilizing PSA would involve the use of an adsorption bed
consisting of porous materials with a natural inclination to attract nitrogen. Most of the
nitrogen will remain in the bed, allowing the passage of a purified gas stream. Later,
708
Landfill gas as a fuel for a vehicle fleet for the city landfill
the nitrogen can be removed for the adsorption bed by means of reducing its pressure.
A variation of this process is temperature swing adsorption where gas release is
triggered by changes in temperature, not pressure.
Membrane separation is often used in conjunction with other gas separation
techniques. This technique is an absorption process that utilizes hollow fiber
membranes as contact media for gas flows. Porous polymer membranes are the
preferred contact media. The membranes can operate at 140 bar pressure differential
and at temperatures as high as 90C to 180C. Operationally, membrane separation is
based on the physical principle that certain gases permeate across a barrier faster
than others. The hollow fibers utilized by these membranes are no bigger than twice
the diameter of a human hair. By applying gas flows at given operational pressures
and temperatures the desired impurities can be forced through the membrane (or
conversely retained by the geomembrane) for later sequestration.
Fig. 3 The CO2 Wash Process

CO2 washing is an innovative new approach that takes advantage of the
different reaction of the landfill gas components to lowered temperatures. At the same
pressures, carbon dioxide gas tends to liquefy at a higher temperature than methane.
So when the landfill gas stream is subjected to refrigeration, carbon dioxide will
transition to liquid while methane remains gaseous. In a vent stack, the now liquid
carbon dioxide falls backwards, counter to the upward direction of the gas stream, as a
mist of small droplets. As they fall, the droplets act as a carbon filter, picking up the
trace impurities as they drop. Meanwhile, the now almost pure stream of methane
continues upward through the vent stack to its final destination. The chilled liquid
carbon dioxide can be utilized as a commercial refrigerant (dry ice).
5. CONCLUSION
Landfill gas (LFG) generated at landfills can serve as a nonconventional source
of energy if cleaned of certain impurities and additionally as a transportation fuel if it is
adequately cleaned. This fuel can then be used as a fuel source for the refuse trucks
operating at landfills. The refuse trucks primarily operate in the residential and
commercial areas and any fuel consumption and emissions improvements will have
709
considerable positive environmental benefits for the community.

Using the LFG for vehicular fuel is a much more recent application than others
such as electricity and heating. There are, therefore, very limited studies [4-6] on the
economic feasibility of this option and the possible environmental benefits.
LITERATURE
[1]
Ivanovic V., Ivanovic D. Dragovic B. (2009). Capture and usage of landfill gas,
Energy-Economy-Ecology, No 3-4, p. 251-256. (in Serbian)
[2] Ivanovic V., Ivanovic D. (2010). Instability in production of landfill gas during
landfill building, M&T&M -Machines, Technologies, Materials, Issue 4-5/2010, p.
52-56
[3] Houdkova L. at all (2008). Biogas a renewable source of energy, Thermal
science, Vol. 12, No. 4, pp. 27-33.
[4] Zletsman J. at all (2008). Feasibility of Landfill Gas as a Liquefied Natural Gas
Fuel Source for Refuse Trucks, Journal of the Air & Waste Management
Association, Vol. 58, p. 613-619.
[5] NSCA Report (2006). Biogas as a road transport fuel, An assessment of the
potential role of biogas as a renewable transport fuel, Research undertaken for
the National Society for Clean Air and Environmental Protection
[6] Persson M., Jonsson O., Wellinger A. (2006). Biogas Upgrading to Vehicle Fuel
Standards and Grid Injection, IEA Bioenergy
[7] UK Environment Agency (2004). Guidance on the Management of Landfill Gas
[8] UK Environment Agency (2002) Guidance on Landfill Gas Flaring
[9] US EPA (1996) Turning a Liability into an Asset: A Landfill Gas-to-Energy Project
Development Handbook
[10] US EPA (2005). Landfill Gas Emissions Model (LandGEM) Version 3.02 Users
Guide
710
NVH INVESTIGATION OF POWER STEERING SYSTEM HYDRAULIC

PUMP
Jovanka Luki1, Radivoje Pei2, Dragan Taranovi3
Summary: Paper presents a method for analyzing the noise and vibration problem
caused by hydraulic pump of a vehicle power steering system. The method is based on
experimental research and coherent analysis of observed system. Measurements are
performed order to determine dominant excitation of hydraulic system in different
operating conditions. Pump number of revolution was constant and was varie.
Pressure of oil in hydraulic system is varied too. A method of assessment and
determination of dominant NVH pump excitation is presented. Experimental results are
input and output data in coherent analysis. Results showed that number of measuring
channels can be reduced, especially in high frequency domain. Obtained results can
be used in experiment design.
Key words: vehicle, steering system, hydraulic pump, NVH, noise, vibration, coherent
analysis
1. INTRODUCTION
Contemporary vehicles are faced with undesirable fluid borne, structure borne
and airborne noise. Fluid borne noise is typically generated in the fluid reservoir or in
the hydraulic lines and could propagate through the mounting brackets. Structure
borne noise transmitted to the driver via body structure through the pump mount,
engine mounts, lines and system mounting brackets. Moan is the structure borne
noise. Moan frequency is driven by natural frequency and harmonics of the pump
rotational Vane Passing Frequencies or Blade Pass Frequency, depending on type of
the pump [2, 4, 9,10].
The noise generated by pump of power steering systems is an example of this
kind of problem. Vane pumps are used in the majority of steering systems in
automotive applications. These pumps generate noise due the vane passing
frequency. Basically, the hydraulic pump noise can be classified as moan or whine,
regarding the operating condition [2, 4, 5, 6].
1
PhD Jovanka Luki, Assoc. Professor, Kragujevac, Faculty of Mechanical Engineering Kragujevac,
(lukicj@kg.ac.rs)
2
PhD Radivoje Pei, Full Professor, Kragujevac, Faculty of Mechanical Engineering Kragujevac,
(pesicr@kg.ac.rs)
3
MSc Dragan Taranovi, Assistant, Kragujevac, Faculty of Mechanical Engineering Kragujevac, (tara@kg.ac.rs)
711
The noise and vibration problems in hydraulic pumps of hydraulic power

systems, as well as steering system, mainly moan and whine noise, have similar
generation mechanisms. For a diagnostic procedure, it is necessary to find the
frequencies and amplitudes for which these noises become annoying for the
passengers. In most cases, the moan and whine noise are related with different
running conditions of the power steering system and the diagnostics must be done for
each one apart. However, the moan and whine problems are directly related with the
engine rotation and the frequencies of noise and vibration can be pointed knowing the
construction and operation characteristics of the hydraulic pumps, [15].
The moan problem can be noted as a noise and/or a vibration. This kind of
problem occurs when the vehicle engine is in idle condition and when the wheels are
being steered. In this case, the system needs hydraulic assistance and the pump is
charged.
The whine noise has the same generation characteristics as well moan noise.
However, the whine noise arises for engine rotations out of the idle condition. In most
cases, this category of noise becomes annoying for the vehicle passengers when the
-1
engine is accelerated around 2000 min . Above this frequency, the engine noise
masks the whine noise and the analysis become more difficult. In the same way as
well as moan noise, the whine noise is also noted at the harmonics of the vane passing
frequency. An important characteristic of these two types of noise is that the harmonics
does not appear to have the same contribution to the annoyance sensation inside the
vehicle cabin. NVH problems of hydraulic pump in vehicle determine experimental
conditions, [2-7,10].
Aim of this work is to determine dominant hydraulics system parameter on
pump noise. Experimental investigation is conducted in order to get a input data for
suitable NVH modeling of hydraulic system.
Experimental research was conducted in laboratory condition. Performed
research was conducted in order to get input data for coherent analysis and to
determine dominant excitation of hydraulics systems. Measurement set up is designed
to investigate the level of influence of hydraulic system parts.
Fig. 1 Mesurement rig

712
NVH investigation of power steering system hydraulic pump
Measurement rig used in this research is given in Figure 1 and is consisted of:
electric engine which drives pump by belt transmission, thermal control unit, system for
control pressure and oil reservoir. Hydraulic pumps (P), is driven by electric engine
SEW Eurodrive, type DV132S4TF/IG1. At engine and pump accelerations are
measured by three axial transducers PCB 356A16, (A). Outlet pump pressure is
measured by pressure sensor Shaevitz, Type P1221-0002-03 M0, (PS). Pump volume
flow is measured by Kracht flow meter, type VC3F1DS, (F). Sound Pressure Level
(SPL) is measured by microphone Microtech Gefel,l type MK 250, (F). Pump velocity in
radial direction is measured by digital vibrometer Polytec, type PDV100, (V). Pump
number of revolution is measured by optical sensor Baumer, type: FZAM 181155
V152, (OS). NVH Measuring signals are acquired and stored by Pak Mueller system
for data acquisition and HVH analysis, [8]. Measurements are performed in order to get
proper data for further NVH investigation. Measuring protocol is defined in order to
obtain necessary data for analysis of dominant factors which have influence on SPL of
pump. Characteristic of acceleration transducers could not give completely analysis in
human hearing frequency domain. Acceleration transducers have flat characteristics in
frequency domain till 4.5 kHz. The human ear can nominally hear sounds in the range
20 Hz to 20 kHz. The upper range tends to decrease with age, most adults being
unable to hear above 16 kHz. Region 2 kHz to 5 kHz is very important for
understanding of speech, [7,9]. Digital vibrometer is used to get data about pump
behavior in high frequency domain, as well as pressure transducer and volume flow
transducer.
Measurements are preformed under different pressure condition: 30, 50, 60, 70
and 80 bar. Pump number of revolution was constant: 1000, 2000, 3000 and 4000 min1 and variable from 500 to 4000 min-1.
Sampling frequency was 48000 Hz, maximal frequency was 18750 Hz, number
of sampling was N=65536 -, frequency resolution was f=0.732 Hz, duration of signal
was T=1.38 s, and Hanning filter was applied. Sampling frequency was 2.56 times
maximal frequency, [8]. Overlapping of 50% of measured signal, averaging procedure
was applied.
Experimental results are partially given in Figure 2. In Figure 2, the auto power
spectrum of SPL is given. Under constant pressure (50 bar), overall SPL increases
with respect to increase of pump number of revolution. At higher pump number of
revolution, the auto power spectrum of SPL has more signal variations. At constant
pump number of revolution, the auto power spectrum of SPL decreases with increase
of pump pressure, Figure 2. Signal oscillations are firmly marked in frequency region till
10 kHz and are less as well as pressure increase. Low pump pressure indices more
pulsation in the Auto power spectrum of SPL. The resonant frequency is around 6.5
kHz. This peak is more pronounces under high pressure, Figure 2. More results can be
seen in [6].
3. DATA ANALYSIS
Aim of the work was to investigate the influence of different hydraulics system
parameters on SPL of pump. Analysis of coherent functions is conducted according to
computational algorithms given in chapter 10. 3, pages 249-263, [1].
713
Fig. 2 Auto power spectrum of sound pressure level at fixed pump pressure, pump
number of revolution was np=4000 min-1: a) p=50 bar, b) p=60 bar, c) p=70 bar, d)
p=80 bar
In order to perform coherent analysis, software for calculation of partial and
multiple coherence functions was developed according to [1]. Measuring system, given
in Figure 1, can be presented as Multi Input One Output system (MIMO), Figure 3.
According to Figure 3, input data are:
- x1 pump acceleration in axial direction,
- x2 pump acceleration in tangential direction,
- x3 pump acceleration in radial direction,
- x4 pump pressure,
- x5 engine acceleration in axial direction,
- x6 engine acceleration in tangential direction,
- x7 engine acceleration in radial direction and
- x8 pump velocity in radial direction.
Output data, y, is
- x9 sound pressure level.
System with two input one output, Figure 4, could be used in high frequency
domain with respect to characteristics of applied transducers. Input data are: x1 pump
velocity in radial direction and x2 pump pressure. Output data is y=x4 pump SPL.
Partially, results given here, present one way of contribution to NVH pump
modeling in order to get validated data for further model development, Figure 5 714
NVH investigation of power steering system hydraulic pump
Figure 6. More results can be found in [6].
Fig. MIMO System in frequency domain

till 4.5 kHz
Fig.4 MIMO System in high frequency

domain
Performed coherent analysis showed that model given in Figure 6 can be

substitute by model 2x1 given in figure 4. Accelerations of engine and pump can not be
considered as well input data, because they have no influence on SPL. It is valid in
laboratory conditions.
Figure 5. Coherence functions (Pump

velocity in radial direction / pump SPL),
np=3000 min-1, J289.5!, J289.5! and J289.5! are
partial coherence functions when engine
accelerations in axial, tangential and
radial directions are removed respectively
Fig. 6 Coherence functions: 223 ordinary coherence function between

pump pressure and pump SPL, and
223.1! partial coherence function
between pump pressure and pump SPL
when pump velocity is removed, np=4000
min-1, p=80 bar
Results showed that existing measuring rig could be updated by digital pump
pressure control unit implementation. Used measurement rig also can be used in
developing phase of active noise control.
For example, investigation of new duct application for pump supplying,
investigation at operating conditions in vehicle and developing NVH model of pump
and ducts mounting.
715
4. CONCLUSION
Obtained results showed that performed coherent analysis can be applied in
order to get sufficient input data for NVH modeling of vehicle hydraulic system.
Results of coherent analysis showed that:
Partial coherence functions are low when linear effects of pump
accelerations are removed which means that vibration of pump can not be
neglected and are dominant influence factor on pump SPL.
Ordinary coherence function between pump velocity and sound pressure
level is high, and partial coherence functions are lower than ordinary
coherence function.
When oil pump pressure is increased all values of coherence functions
increase.
The same analysis method can be applied both on vane and radial piston
pump.
Coherent NVH analysis showed that reduced measuring rig can be applied.
Depending on volume flow characteristics in high frequency domain, system
with two input and one output signals can be used if the volume flow is
constant.
Applied method can be very useful in initial experiment phase.
LITERATURE
[1]
Bendat, J. S., Piersol, A. G., (1980). Engineering Applications of Correlation and

Spectral analysis, John Wiley & Sons, New York.
[2] Carbary, K., Ulep D., et all, (2003). Power Steering Pump Sound Quality and
Vibration Test Stand Development, Noise & Vibration Conference and
Exhibition, Traverse City, Michigan, May 5-8, SAE Paper No. 2003-01-1662
[3] Goenechea, E., Gels, S., (2008). Laermbekaempfung in Der
Hydraulik,
Geeignete Massnahmen, um die Schallpegelwerte zu senken, O+P Konstructions
- Jahrebuch 2008/2009, p. 22-31.
[4] Harrison, M., (2004): Vehicle Refinement Controlling Noise and Vibration in Road
Vehicles, SAE International Commonwealth Drive, Warrendale, PA 15096-0001
[5] Junior, M. T., et all, (2003). Analysis of moan and whine noise generated by
th
hydraulic pumps of power steering systems, 12 NVH Congersso Brasil, SAE
Paper No. 2003-01-3581.
[6] Lukic, J. (2009). NVH Investigation of vehicle hydraulics pump, University of
Pitesti, Scientific Bulltein, Faculty of Mechanics and technology, Automotive
series, Vol.15, No.19, p. 102-108.
[7] Mser M., (2007). Technische Akustik, Springer-Verlag Berlin Heidelberg.
[8] Mueller BBM: PAK- Software for measure, analyze and interpret sound and
vibration signals
[9] Pflger, M., Brandl, F., Bernhard U., Feitzelmayer K. (2010). Fahrzeugakustik,
Springer Wien New York
[10] Rsth. M., (2007). Hydraulic Power Steering System Design in Road Vehicles
Analysis, Testing and Enhanced Functionality, Dissertation, Linkping University,
Sweden.
716
POSIBILITY OF DETERMINATION THE ROUTES FOR

TRANSPORTATION OF HAZARDOUS GOODS ON THE BASIS OF
THE RISK LEVEL
Vojislav B. Krsti1, Boidar V. Krsti 2, Vuki N. Lazi3
Summary: This paper contains presentation of methodology fur establish the routes
for transportation of hazardous goods. The selection of routes for transportation of
hazardous goods is based on determination of the extent of risk.
Key words: transport, hazardous goods, risk
1.
INTRODUCTION
Transport of hazardous goods is a kind of transport related to the highest risks

and possible dangers for population and environment. In order to reduce possible
consequences for population and environment from mobile sources of hazardous
materials (transportation modes), it is necessary to establish the routes for their moving
and that should be done on the basis of the level of risk acceptable by social
community of a certain area.
In order to establish the routes for transport of hazardous goods within a
certain area, it is necessary to conduct wide-ranging researches of features of
transport demands for each class of hazardous goods and separately for each source.
Also, it is necessary to determine the features of traffic flow (flow volume, structure), to
explore demographic features (population and density) and to define sensitive areas.
In addition to conducting wide-ranging researches, it is also necessary to define
methodology for route selection based on determination of risk levels for each section
within the considered route and to apply it to the local conditions, which is the actual
aim of this paper.
2.
PRESENTATION OF METHODOLOGY FOR ESTABLISHING THE ROUTES

FOR TRANSPORT OF HAZARDOUS GOODS
The methodology developed by the authors of the paper unites all possible
elements which influence or might influence the selection of optimal route for road
transport of hazardous goods. It consists of 11 steps (phases) and it is based on
1
University in Belgrade Faculty of Traffic and Transport Engineering, Beograd,

University in Kragujevac Faculty of Mechanical Engineering, bkrstic@ kg.ac.rs
3
University in Kragujevac Faculty of Mechanical Engineering, vlazic@ kg.ac.rs
2
717
establishment of routes for transport of hazardous goods on the basis of absolute risk,
that is, the risk which can be quantified.
According to the international classification [1], hazardous goods are divided
into 9 different classes, and each class of hazardous goods has its own features and
degrees of danger. Within each separate class there are many different substances
that are called hazardous because they by their nature represent hazard for the
environment and, depending on their features, may cause certain consequences.
It is very important to mention the fact that, in accordance with the selection of
type of hazardous goods and the features of hazard, this methodology can be applied
to the transport of all classes of hazardous goods, except to the class 7, that is,
radioactive substances. The reason for this limitation lies in the fact that radioactive
substances, as a difference from other hazardous goods have certain specifics, that is,
only these substances emit the radiation which is very dangerous because human
senses are not able to identify them. In case of transport of all other types, i.e. classes
of hazardous goods, it is possible to apply this methodology for the purpose of
establishing the routes for vehicles movement.
Another very significant characteristic of hazardous goods, beside their
features and hazard degree, is their influential zone. Size of influential zone of
hazardous goods depends on their type and quantity, weather conditions (rain, snow,
wind, fog, or similar), as well as of the features of terrain; and it goes from 25 to 1600
meters.
In order to obtain a detailed insight into the flows of hazardous goods, it is
necessary to identify the sources, that is, places where the hazardous goods is stored,
produced and similar, i.e. so called fixed sources of hazardous goods. This represents
the second step within the methodology. For each source of hazardous goods, it is
needed to determine the total quantities of hazardous goods according to their types at
daily, weekly, monthly and yearly basis. When the total quantities of hazardous goods
are defined depending on their type, it is necessary to explore by which modes of
transportation the hazardous goods is transported from those places to their
destinations.
Therefore, for the specific serviced area, the following is necessary:
- To define the total quantities of hazardous goods by classes which are transported,
stored and similar within the plants of factories,
- To define the percent of hazardous goods transported from all fixed sources by
certain means of transportation,
- To sum up the quantities of goods for each class of hazardous goods, especially
those that are transported from each fixed source by different modes of transportation
with the aim to obtain the total quantity of hazardous goods transported within that
serviced area for different periods of time (daily, weekly, monthly and yearly
quantities),
- To establish the time irregularities of hazardous goods transport, i.e. to define hourly
irregularities during the day, daily irregularities during the week, weekly irregularities
during the month and monthly irregularities during the year for the whole system, for
subsystems and for each fixed source,
- To produce the origin destination matrix movements of vehicles for each class of
hazardous goods by modes of transportation at the daily level for each day during the
week for the chosen period (representative period),
- To load the road network with goods flows at the daily level for each day of the week
718
Posibility of Determination the Routes for Transportation of Hazardous Goods on the

Basis of the Risk Level
within the representative period, and
-To determine the features of goods flows relative to the defined serviced area (source
within the territory, destination out of it; both source and destination within the defined
area, etc.)
After establishing the features of transport demands in space and time, it is
necessary to define the restrictions, which represents the fifth step of the methodology.
The restrictions that occur in establishing the routes for movement of vehicles which
are used for transportation of hazardous goods are reflected in different types of
restrictions defined within the framework of legal regulations, in physical types of
restrictions and those restrictions that are related to the sensitive areas of nature
(ecological zones).
As the restrictions which may exist in the legal regulations are those kinds of
rules by which the transport of hazardous goods at the certain roads and transportation
routes in the certain periods of time is forbidden, or regulations (at the local level)
which precisely determine which roads may be used for hazardous goods
transportation and in which period of time. When we consider the spatial restrictions in
the legal regulations, they are related to the restriction of transport of hazardous goods
by transport routes or roads with certain physical limitations reflected in the height of
overpass culvert, allowed axle load of road surface, as well as in ban of transport by
those transport routes that go through or are next to the main city parts. Time
restrictions which occur in the legal regulations are represented by time periods during
which the transport of hazardous goods on certain transportation routes or roads is
forbidden. These periods are usually defined during traffic peak hours, that is, when
the traffic flows are greatest. This type of restriction is introduced for the reason of
minimizing possible consequences in case of incident situation.
Defining of sensitive environmental areas (so-called ecological zones) by state
or city authorities can have strong influence on the selection of route for transport of
hazardous goods, because possible influence of hazardous goods on these areas
could have immeasurable consequences. Types of sensitive areas of environment
which are located along the route of the road used for transportation of hazardous
goods can be defined from different sources and on the basis of level of measures
applied for locating sensitive areas of environment where the state or state authorities
can prescribe even the ban on passing by these areas. Very important fact that may
influence the feasibility of solution is to prevent the route used for transportation of
hazardous goods within its influential area (by making intervals of hazardous goods
influence) to cross or touch those sensitive areas of environment. If it is not the case,
then the solution regarding the route is doomed because of this kind of failure,
regardless how good it is from the technical and technological aspect.
The sixth step of the methodology, defining the alternatives for analysis, as well
as the seventh step, defining of the criteria for the route selection, represents two steps
on the basis of which it is possible to eliminate certain routes that any case cannot
fulfill the conditions for usage for transport of hazardous goods.
There are several common clusters when establishing the routes, and they
comprise of the following decisions [2]:
Tunnels and long bridges may be banned for transportation of hazardous
goods. When establishing the routes, these decisions are common for all long tunnels
(with ventilation systems) and they are generally applied to the ban on transportation of
inflammable, poisonous gases, and/or explosives through tunnels, or time restriction,
719
that is, periods of time when their transport is allowed.

Directing of flows of hazardous goods to the bypass directions results in
transportation of hazardous goods around rarely populated, and not through densely
populated areas.
Sources and/or destination places are the same when the routes are
established and they are related to places such as ports, terminals, as well as chemical
and industrial plants.
Setting the routes for movement of vehicles which transport hazardous goods
without interruptions from source to destination, allow the deliveries of hazardous
goods to transit through city areas, state or certain regions.
Remaining decisions are rarely used when establishing the routes for
transportation of hazardous goods, and they are developed for particular, that is,
unique situations.
After determining the common clusters when establishing the routes, the first to
be considered are those routes:
- That fulfill all the aims defined by the authorities from the aspect of competence;
- That are in accordance with the existing models for establishing the routes for
hazardous goods and which allow access to the terminals and other facilities (factories
and similar);
- That have no obvious physical or legal restrictions that may seriously hinder their
usage or even lead to the ban on their usage;
- That have continuous route (without interruptions) and that can be connected with the
roads in other regions (when the roads within one region and city are observed) or with
the roads in other countries.
In formation of concepts of the establishing the routes, it is also necessary to
define the criteria for their selection. When defining criteria, it is necessary to make
sure that they are not too numerous and that they are sufficient for defining the basic
postulates when making the decision on the selection of optimal route for
transportation of hazardous goods.
The basic criteria used as relevant when making the selection of routes in the
framework of the methodology are the following:
- Consideration of existing distribution of source destination localities;
- Partial usage of roads which are most often used for movement of vehicles for
transportation of hazardous goods;
- Usage of roads reserved for freight transport to the higher extent;
- To minimize to the lowest possible extent the distance passed between source and
destination spots, and
- To decrease the size of potentially affected area by narrowing the selection of roads
in the existing conditions.
After determining the alternatives for analysis and criteria for selection of
routes, the remaining possible routes for transportation of hazardous goods which
fulfilled the previously defined criteria have to be inspected from the aspect of risk. In
order to establish the level of risk, it is necessary to define the parameters that
influence the extent of risk level, that is, to define the model on the basis of which the
selection of routes will be made.
The model with defined parameters is presented in the following item of the
paper.
720
Posibility of Determination the Routes for Transportation of Hazardous Goods on the

Basis of the Risk Level
3.
PRESENTATION OF METHOD FOR DEFINING THE RISK LEVEL
When speaking of risk, two basic components for defining the risk are
probability of occurrence of incident situation and consequences, or negative influence
on the environment in case of occurrence of incident situation (accident). This helps us
a lot when defining not only the parameters as an input in the model, but also the entire
model for selection of routes based on the risk analysis. However, parameters which
define risk do not have the same level of significance, so it is necessary to define their
weightings (impact factors). In order to define the weightings, the authors of this paper
conducted the experts survey on the basis of which the weightings for each of the
parameters influencing the risk extent are determined.
For each parameter it is necessary to define the level of risk, that is, to quantify
the risk. In order to achieve that, it is needed to define the interval of values that the
risk may take up, depending on its level. In order to make the model as
understandable as possible, the risk values that certain parameter may take up amount
from 0 to 100, where 0 represents the value where the risk does not exist, while 100
represent the highest possible level of risk.
On the basis of the defined criteria and the alternatives for each section of
route separately, it is necessary to make the calculation of risk. On the basis of
establishing the level of risk for each parameter, probability and extent of the
consequences, the decision is made whether that section is acceptable for transport or
not.
Decision whether certain section satisfies the criteria from the aspect of risk is
made on the basis of comparisons of obtained values with the allowed level of risk.
That comparison is conducted by means of the risk matrix, where on the basis of
probability value (obtained as an arithmetical mean of values of all parameters
influencing the probability) and the values of level of risk from the possible
consequences the comparison is made with the allowed level of risk. The easiest way
to show this is graphically, where on the basis of probability value (which is placed on
Y-axis) and the value of consequences (placed on the X-axis) the meeting point is
obtained. If that point is above the allowed level of risk (shown with dashed red line in
the picture), the section is unacceptable from the aspect of risk; therefore it is excluded
from further analysis.
After conducted analysis of risk, it is necessary to make the final step of the
defined methodology, which are identification, revision and approval of the routes. A
great advantage of this methodology is its flexibility which is reflected in the fact that it
is possible to repeat the inspection of the level of risk for certain routes, if one or more
sections of a route do not satisfy the criteria from the aspect of risk, that is, it is
possible to influence the reduction of probability of occurrence of incident situation by
improving the elements of road construction (improving the state of road surfacing,
changing the elements of regulation, etc.), as well as by more efficient traffic
management (reduction of traffic flow volume and change of its structure).
3. CONCLUSIONS
The methodology presented in this paper is based on the selection of routes for
transport of hazardous goods on the basis of the risk extent. The presented
methodology consists of 11, mutually dependant steps. In the framework of the
721
methodology there is the model on the basis of which the extent of risk is established
for the observed section according to the defined parameters and their weightings.
The only limitation of the defined methodology is that it cannot be applied for
establishing the routes for transportation of hazardous goods of class 7 radioactive
substances, due to the specifics of this type of hazardous goods, which are strength
and extent of radiation.
In accordance with the defined methodology, the last item of the paper
presents its implementation to the City of Belgrades road network. As the output of
methodology, we obtained the routes for transport of hazardous goods which are
acceptable from the aspect of allowed extent of risk. Beside the routes, the last item of
the paper also presents the results of research of oil and oil derivates transport
demands at the territory of Belgrade (modal split, spatial flows features) which are
necessary for quantification of risk for each section.
BIBLIOGRAPHY
[1]
[2]
[3]
[4]
[5]
722
United Nations, European Agreement Concerning the International Carriage of

Dangerous Goods by Road, New York and Geneva, 2009.
Nelson, C., Cataford, A., Transportation of dangerous goods policy and
evaluation framework, Transportation Association of Canada, Charlottetown,
Canada, 2006.
National Highway Institute (NHI) and Federal Highway Administration (FHWA),
Highway Routing of Hazardous Materials: Guidelines for Applying Criteria.
Arlington, VA, November 1996.
Krsti, B.: Exploitation of motor vehicles and motors, Faculty of Mechanical
Engineering from Kragujevac, Kragujevac, 2009.
Erkut E., Verter V.: Modeling of transport risk for hazardous materials, Operation
Research, Vol. 46, No.5,, pp. 625-642., 1998.
POSSIBILITIES DETERMINATION OF THE OPTIMAL STRATEGY

FOR PREVENTIVE MAINTENANCE OF THE CARDAN SHIFT
VEHICLE USING POLYCRITERION OPTIMIZATION
Boidar V. Krsti1, Vojislav B. Krsti2, Ivan B. Krsti3
Summary: This work presents a possibility to find the optimum solution in the
maintenance cardan shift when the criteria functions are maximal avalability and
minimal costs maintenance. These two criterions lead to several solutions of the
maintenance of cardan shift therefore it was necessary to seek for a trade off solution.
Key words: Motor vehicle, Maintenance, Optimization, Reliability, Availability, Costs
1.
INTRODUCTION
In this paper are presented results of determination the reliability parameters of

the cardan shift vehicles, based on which the optimal value of the cardan shift vehicles
maintenance periodicity was determined, for the criteria of the maximal availability and
the minimal maintenance costs. By correct forming of the maintenance model it is
possible to perform the optimization, namely to select the most favorable maintenance
system. Such a problem can be solved if all the important requirements and restrictions
are precisely determined. As the optimal periodicity of the preventive maintenance
procedure of the analyzed technical system, determined according to the criterion of
the maximal availability differs from the optimal periodicity determined according to the
criterion of the minimal costs, it is necessary to apply the multicriteria analysis methods
and to determine the value of the required optimal periodicity, taking into account both
mentioned optimization criteria. This actually is the basic goal of this work.
By the method of the compromise selection, the limits are determined of the
optimal periodicity for the preventive maintenance procedure conducting, which
correspond to extreme values of the adopted criterion functions. By applying the
method of the multicriteria analysis, one determines the discrete value of this
periodicity, whose exactness depends on selection of the discretization step of the
analyzed time interval. Generally considering, the presented methodology of the
multicriteria decision-making can be applied for obtaining the reliable value of the
preventive maintenance procedure conducting periodicity of the technical systems.
There one needs to know the availability of data, which are acquired by analysis of the
1
Faculty of Mechanical Engineering, Serbija, bkrstic@ kg.ac.rs

Faculty of transport and traffic engineering, University of Belgrade, Beograd, Serbia
3
Faculty of Tehnical Sciences, K. Mitrovica, Serbia
2
723
technical systems during their operation and maintenance, based on which one can
determine the indicators of their reliability, as well as the characteristics of their
maintenance.
2.
DETERMINATION OF PARAMETERS OF THE VEHICLES CARDAN SHIFT

RELIABILITY
The reliability parameters of the analyzed cardan shift vehicles need to be

determined for the planning of measures for its maintenance, prediction of the working
life and estimates of its quality.
Table 1 Possed road before appearance of middle bearing of the cardan shaft of vehicle
Working
Working Working Working Working Working Working Working Working
time till
time till time till
time till
time till
time till
time till
time till
time till
failure
failure (h) failure (h) failure (h) failure (h) failure (h) failure (h) failure (h) failure
(h)
(h)
16,60
19,65
31,45
49,50
51,00
53,90
80,60
80,70
92,00
93,00
94,00
96,10
96,50
97,50
98,50
98,70
99,00
99,50
99,70
102,0
104,0
106,0
106,5
107,00
107,50
100,10
110,00
119,00
119,50
120,00
125,00
126,00
129,00
130,00
134,00
140,00
145,00
150,00
161,00
161,50
161,60
162,00
162,50
163,00
164,00
164,50
165,00
165,10
165,70
165,90
166,00
168,0
169,50
169,90
170,00
171,00
172,40
172,80
173,20
173,50
174,00
174,50
175,00
175,50
176,09
176,10
176,70
177,00
177,10
178,00
178,90
179,00
180,00
181,00
182,00
182,10
182,30
182,40
182,70
183,00
185,00
186,00
188,00
189,00
189,30
189,90
189,95
191,00
191,50
192,00
193,00
194,00
198,00
198,90
199,00
200,00
201,00
202,00
203,00
204,00
204,50
205,00
206,00
206,50
207,90
208,00
209,00
209,90
210,00
211,50
212,00
212,30
213,00
220,00
230,00
231,00
231,50
232,00
245,00
250,00
260,00
265,00
267,00
270,00
280,00
290,00
299,00
301,00
302,00
303,00
304,00
305,00
306,00
306,50
307,00
308,00
309,00
309,90
310,00
320,00
329,00
330,00
340,00
350,00
360,00
370,00
380,00
386,00
389,00
389,90
390,00
390,80
391,00
392,00
393,50
394,00
395,00
396,00
397,00
398,00
398,80
401,0
401,5
402,0
410,0
415,0
416,0
418,0
420,0
425,0
426,0
428,0
429,0
431,0
431,5
432,4
432,5
433,0
434,0
434,5
435,0
438,0
439,6
439,9
431,0
431,5
432,0
433,0
435,0
438,0
438,5
439,0
439,8
439,9
440,0
440,5
441,0
441,5
442,0
442,5
442,7
442,8
442,9
443,0
443,5
443,6
443,9
Determination of the cardan shift vehicles reliability distribution law represents

the Basis for evaluation of its state, as well as for decision-making on when, i.e., after
how many working hours, one should conduct procedures of the preventive
maintenance. Determination of the most acceptable model of the reliability distribution,
based on data on its behavior, from the aspect of irregularities appearance, is a
complex task and it is solved by application of the corresponding algorithm1, with
724
Posibilitu of Determination the Routes for Transportation of Hazardous Goods on the Basis Of
the Risik Level
application of the probability theory, mathematical statistics and the reliability theory2.
Based on voluminous research of the cardan shift vehicles behavior, in the real
exploitation conditions, for a longer period of operation, in Table 1 are presented the
working times until occurrence of failure and between the two consecutive failures.
Estimated values of the cardan shift vehicles reliability indicators, based on
data from Table 1, are determined using the known methodology1 and presented in the
table 2.
Based values of deviation of the reliability theoretical values, obtained by
testing the corresponding hypotheses by application of the known methodology2, from
results of the estimated values, obtained from the exploitation data (Table 1), one
came up to the conclusion that the Weibulls two-parameter distribution, with the shape
parameter 1,7 and scale parameter 144, was the most acceptable for the analyzed
vehicle sample.
Adopting this reliability distribution law for the cardan shift vehicles of the
analyzed technical system, expressions for determination of the reliability function R(t),
the density function f(t) and failure intensity (t), can be written in the following forms:
R (t ) e
f (t )
O (t )
t )
( 144
1,7
1,7
144
1,7
144
(1)
t
( 144
) e
0,7
t )0,7
( 144
t 0,7
144
(2)
(3)
Based on the previous expressions, the optimal periodicity of the working time
can be determined, after which either of the following should be performed: preventive
inspections, preventive substitutions, repairs or general revisions, as well as providing
the optimal values of spare parts stocks3.
3.
DETERMINATION OF THE MOTOR VEHICLE CARDAN SHIFT PREVENTIVE

MAINTENANCE PERIODICITY BASED ON CRITERION OF THE MINIMAL
MAINTENANCE COSTS
Providing for the required availability and reliability of the cardan shift vehicles,
with minimal maintenance costs, is possible if one correctly determines the periodicity
interval of that maintenance3.
The total cardan shift vehicles maintenance costs can be expressed in the form
C (t )
C k (C k C p ) R(t )
T
(4)
R(t )dt
0
where: C(t) are the total specific maintenance costs; Ck are the corrective maintenance
costs and Cp are the preventive maintenance costs. By application of expression (4)
for various periodicities of the starter vehicles maintenance, the values for the
maintenance costs are obtained shown in Table 2 and in Figure 1.
Based on results shown in Table 2 and Figure 1, one can conclude that the lowest
maintenance costs of the analyzed technical system chardan shift vehicles are
obtained for the maintenance periodicity of 350 working hours. Table 2. Maintenance
expenses of chardan shift for different of the preventive maintenance
725
C(t)
350
t (h)
Fig. 1 Graphical representation of the dependence of the costs on periodicity of
chardan shift vehicles maintenance
Table 2 Maintenance expenses of chardan shift for different of the preventive
maintenance
Maintenance
frequency (h)
Costs of
corrective
maintenance
Ck
Costs of
preventive
maintenance
Cp
Reliability
R(t)
100
150
200
250
300
350
400
450
500
400
400
400
400
400
400
400
400
400
400
100
100
100
100
100
100
100
100
100
100
0,95
00,88
00,79
0,7010
0,6054
0,5138
0,4293
0,3534
0,2870
00,2300
6,36
38,4
75,82
08,43
36,34
59,84
79,30
95,27
08,12
0,97
0,93
0,91
0,85
0,88
0,80
0,79
0,85
0,91
0,97
0,97
0,93
0,91
0,85
0,88
0,80
0,79
0,85
0,91
0,97
5
50
R(t )dt
Total specific
costs C(t)
fi,,2
4.
3
3
DETERMINATION OF THE MOTOR VEHICLES CHARDAN SHIFT

PREVENTIVE MAINTENANCE PERIODICITY BASED ON CRITERION OF
MAXIMAL AVAILABILITY
Since the most acceptable model of reliability, distribution is determined and

since the time picture of the chardan shift of the analyzed technical system is
completely known (time in operation, time spent on waiting to operate while in order,
time spent while out of order), it is possible to apply the maintenance model based on
Table 3 Motor vehicle availability as a function of its chardan shift vehicles preventive
maintenance periodicity
Maintenance periodicity (h)
Preventive maintenance time tp (h)
Unreliability F(t)
Reliability R(t)
Number of corrective maintenances
between the two preventive ones
Time of corrective maintenance tk (h)
726
50
5
0.04
0.95
100
5
0.132
0.868
150
5
0.15
0.84
200
250
5
5
0.208 0.280
0.792 0.720
300
5
0.29
0.71
350
400
5
5
0.310 0.316
0.690 0.684
450
5
0.31
0.68
0.01
0.11
0.019
0.71
0.38
2.18
0.824 0.165
4.89 9.35
0.28
16.1
0.450 0.698
26.45 41.63
1.07
64.3
the Risik Level
Time spent on waiting while in order
tr (h)
Availability G(t)
fI,,2
10
0.96
0.96
20
0.988
0.988
25
0.99
0.99
59
74
89
103
121
133
0.981 0.978 0.973 0.962 0.958 0.953
0.981 0.978 0.97 0.962 0.958 0.953
G(t)
150
t(h)
Figure 2. Graphical representation of the motor vehicle availability dependence on

starter vehicles maintenance periodicity
availability3. By application of this model, the exploitational reliability of the technical
system, from the aspect of the chardan shift vehicles, can be determined by using the
expression:
G( t )
t p tcr
F( t )
t p tcr to
tk
R( t )
(5)
where: t - is the periodicity of maintenance; tcr - is the time spent to operate while in order;
tp - is the time of preventive maintenance; tk - is the time of corrective maintenance.
By varying the periodicity of preventive maintenance, one obtains the functional
dependence based on which the optimal value of the preventive maintenance
periodicity can be determined, based on the maximal availability criterion. Results of
determination of availability, for various maintenance periodicities of the chardan shift
vehicles of the considered technical system, are shown in Table 3 and in Figure 2.
Based on results shown in Table 3 and in Figure 2 it can be concluded that the
highest availability of the analyzed technical system, from the aspect of its chardan
shift vehicles, is obtained for the maintenance periodicity of 150 working hours.
5.
DETERMINATION OF THE OPTIMAL PERIODICITY OF THE PREVENTIVE

MAINTENANCE OF THE MOTOR VEHICLE'S CHARDAN SHIFT BY
APPLICATION OF THE MULTICRITERIA OPTIMIZATION
The value of the preventive maintenance periodicity of the analyzed technical

system chardan shift vehicles lies between the times that correspond to maximal
availability and to minimal costs. This period can be discretized. Each discrete value
can be associated with considered concept of the preventive maintenance. In that way,
one obtains the corresponding number of preventive maintenance variations, which
differ from each other only in working time lengths after which the procedures of
preventive maintenance are being conducted. Since the values of optimal periodicities
of the considered chardan shift vehicles preventive maintenance, obtained by criterion
of maximal availability and criterion of minimal costs differ from each other (parts 3 and
4 of this paper), in this part are presented results of determination of the periodicity by
application of the multicriteria optimization method, which is known in literature as the
5
MCDM (Multi Criteria Decision making) problem . The basic characteristic of the
727
MCDM problem, thus accordingly of the problem considered in this work, is that the
best alternative is found in the sense of several attributes, simultaneously, or in the
limited set of available alternatives.
Table 4 Availability of the motor vehicle from the aspect of its chardan shift vehicles
Alternative
number
(i)
Chardan shift
maintenance
periodicity
(h)
1
2
3
4
5
150
200
250
300
350
Vehicle's
availability (G)
from the chardan
shift vehicles
aspect
0,997
0,981
0,978
0,973
0,962
fi,1
(fi,,1)n
Total specific
maintenance
costs
(C)
fi,,2
(fi,,2)n
ai
0,997
0,981
0,978
0,973
0,962
0,447
0,437
0,446
0,435
0,426
39,4
34,22
23,63
22,30
19,43
39,4
34,22
23,63
22,30
19,43
39,4
36,22
35,63
22,30
19,43
0,4432
0,4372
0,4465
0,4331
0,4311
In literature can be found a large number of multicriteria optimization methods5.

One of the most frequently used methods is the Analytical Hierarchy Process (AHP)4.
The AHP method was developed based on the principle of decision making, human
knowledge, as well as on data that are available to experts in the process of decision
making. That process is a creative one, which is based on three main concepts:
analytics, hierarchy and process3.
The nature of the optimality criterion can be benefit wise and coastwise5. When
the benefits optimality criterion is used, the higher its value is the better and vice versa.
When the costs optimality criterion is used, the less their values are the better and vice
versa.
The set of alternatives i is being represented by the set of alternative indices i =
(1, i, I) where I is the total number of the considered alternatives. The problem is
represented by matrix F = [fik],1uK. Here fik denotes the optimality criterion k for
alternative i. In the general case, the optimality criteria are of various natures; they
have different values and different units. This means that the optimality criteria values,
for alternative are not comparable. From that reason it is necessary to perform the
normalization procedure by which all the values of fik are being mapped within interval
[0, 1]. At present, a large number of the normalization types are being used7: simple,
linear, vectorial, etc. regardless of which type of normalization is being used, different
expressions are used for benefit wise and coastwise optimality criteria. When the
vectorial normalization is applied, the decision making process can be represented by
matrix F = [fik],1uK, where (fik),n is the normalized value of the optimality criterion k for
alternative i. To each considered alternative, certain value is being associated4.
Normalization of values fi,1 is being done by application of the expression for
vectorial normalization and by application of the benefit wise optimality criterion. For
solving the concrete task the following expressions can be used:
7
( f i ,1 ) n
f i ,1 ( ( f i ,1 ) 2 )1 / 2
(6)
i 1
( f i,2 ) n
(1 / f i , 2 ) ( (1 / f i , 2 ) 2 )1 / 2
(7)
i 1
The value of the factor based on which the best alternative of the maintenance
728
the Risik Level
periodicity ai is being determined by application of the assumption that validities of the

adopted optimization criteria (maximal availability and minimal maintenance costs) are
equal and that they are set as normalized, what is the case in this concrete task, by
application of the following expression7:
ai
1 K
( f ik ) n
kk1
1
>( f i ,1 ) n ( f i , 2 ) n @
2
(8)
The optimal value of the periodicity of conducting the preventive maintenance

of the motor vehicle's chardan shift is within interval of 150 to 350 working hours,
because the limits of that interval were obtained based on criterion of the chardan shift
maintenance minimal costs and criterion of the maximal availability in that interval. The
interval is being divided into 5 equal parts (Table 4).
Elements of matrix F are being obtained in such a way that they are being
made equal to values of the analyzed technical system availability, from the aspect of
its chardan shift vehicles, for various periods of the preventive maintenance that
correspond to individual alternatives (fi,1) and by making them equal to values of the
total clutch maintenance costs for different periods of preventive maintenance that
correspond to individual alternatives (fi,2).
Based on data obtained by monitoring the analyzed technical system, from the
aspect of its chardan shift vehicles, in real exploitation conditions, by application of
expression (5) for determination of availability, obtained were the values of elements fi,1
(namely the availabilities) of matrix F, while by application of expression (4) for
determination of the maintenance costs obtained were the values of elements fi,2
(namely the maintenance costs) of matrix F (Table 4).
The best alternative is one for which the value of factor ai has the highest
value. The values of this factor, calculated by application of expression (8) are given in
table 4. based on those values and by recognizing the aforementioned, it can be
concluded that the optimal value of periodicity of conducting the preventive
maintenance procedures of the analyzed technical system, from the aspect of its
chardan shift vehicles, is after every 250 working hours.
6.
CONCLUSION
Results of voluminous investigations of the motor vehicles chardan shift

reliability parameters, which were obtained by monitoring the behavior of the analyzed
motor vehicle in the real exploitation conditions, from the aspect of failure occurrence
of its chardan shift, and with application of the corresponding scientific knowledge from
the area of probability., mathematical statistics, systems theory and reliability theory,
have served as a basis for finding the optimal periodicity of the chardan shift vehicles
maintenance, taking into account the criteria of maximal availability and minimal costs
of its maintenance.
Since the optimal periodicities of conducting the chardan shift vehicles
preventive maintenance, determined by criteria of maximal availability and minimal
maintenance costs differ from each other, it was necessary to apply one of the
multicriteria analysis methods and to determine the value of the required optimal
periodicity of conducting the preventive maintenance procedures, taking into account
both optimization criteria.
The value of optimal periodicity of conducting the preventive maintenance
729
procedures of a chardan shift vehicles was determined according to maximal

availability criterion to be 150 working hours, while according to criterion of minimal
maintenance costs that value was 350 working hours.
By application of the multicriteria analysis the value of the required optimal
periodicity of conducting the preventive maintenance procedures of a chardan shift
vehicles, with taking into account both optimization criteria, was 250 working hours.
The presented methodology of the multicriteria decision-making can be applied
for obtaining the reliable value of optimal periodicity of conducting the preventive
maintenance procedures also of other parts of the analyzed technical system. There
one needs available data, which can be obtained by analysis and monitoring of the
considered technical system, thus the reliability indicators of the system can be
determined, as well as the characteristics of its maintenance.
BIBLIOGRAPHY
[1]
[2]
[3]
[4]
[5]
730
British Standard, BS5760, Part 2, Guide to the assessment of reliability, Reliability

of systems, equipments and components, BSI, London, 1981.
S. Vukadinovi: Elements of the Probability Theory and Mathematical Statistics,
Economy Review, Belgrade, 1981. (In Serbian)
T. L. Saty: How to Make a Decision: The Analytic Hierarchy Process, EJOR 48, 9,
1990.
D. Milanovi et al.: Management Information Systems, Megatrend University,
Belgrade, 2003.
S. Opricovi: Multicriteria Optimization, Faculty of Civil Engineering, Belgrade,
1986. (In Serbian)
RESEARCH OF THE EXTERNAL AERODYNAMICS OF THE

VEHICLE MODEL
Pikula Boran1 Filipovic Ivan2, Kepnik Goran3
Summary: As the manufacturers of the motor vehicles generally use their own wind
tunnels to test the real vehicle's performances, especially in the final stage of vehicle
design, though the development of prototype starts from defining external
aerodynamics of the vehicle model using experimental research and numerical
simulation. Today, there are a number of different commercial software designed for
the numerical simulation of the external aerodynamics, that show the different air flow
images around and especially behind the vehicle depending on the different turbulence
models. Therefore, the improvement of prototype implies the testing of the vehicle
model in the wind tunnel in order to reach the optimal numerical model for the air flow
simulation around the vehicle.
This paper presents the results of the numerical simulations of the air flow around the
vehicle model compared to the experimental results. The results obtained by numerical
simulations and experimental research, enable an analysis appointing the existing
problems and appropriate numerical methods considering improvement of the vehicle
aerodynamics based on the development of the vehicle model.
Key words: Vehicle aerodynamics, CFD, wind tunnel.
1.
INTRODUCTION
During the vehicle motion the road resistances can appear as follows: rolling
resistance, aerodynamic drag, accelerating resistance, road grade and eventually
trailer resistance. The consequence followed by increase of these resistances is the
fuel consumption and the pollutants emission raise. As the accelerating resistance
dominates in the overall road resistances during the urban driving due to the frequent
changes of the vehicle speed meanwhile air drag dominates during the drive along the
highways, where the vehicle speed achieves 80 km/h or more [1]. Therefore,
considering the motor vehicle development, a maximum attention is dedicated in the
field of the air drag reducing. The parameters that directly affect to the air resistance
1
B. Pikula, Ph. D, Associate Professor, Mech. Eng. Faculty, Sarajevo, Bosnia & Herzegovina,
(pikula@mef.unsa.ba)
2
I. Filipovic, Ph. D, Full Professor, Mech. Eng. Faculty, Sarajevo, Bosnia & Herzegovina,
(filipovic@mef.unsa.ba)
3
G. Kepnik, B. Sc., Assistant, Mech. Eng. Faculty Sarajevo, Bosnia & Herzegovina, (kepnik@mef.unsa.ba)
731
Pikula B., Filipovic I., Kepnik G.
are: A - frontal area of the vehicle, v speed of the vehicle and cx aerodynamic drag
coefficient.
Considering the wishes of the users of the transportation means related to
achieve higher speed, options like reducing the frontal area of the vehicle and the drag
coefficient cx to reduce aerodynamic drag must be achieved. As reducing the frontal
area of the vehicle have an impact to the comfort reduction, the most influential
parameter for the aerodynamic drag optimization becomes the aerodynamic drag
coefficient cx.
Determination of the drag coefficient requires very complex experiments in the
wind tunnels [2]. The experimental method requires existing of very expensive
equipment, like the wind tunnel at the first place. Only very successful car
manufacturers and some test centers are able to have expensive wind tunnels
enabling the real motor vehicles research. External aerodynamics research also can
be performed by the vehicle model, scaled in specified ratio as it is presented in [3]. In
this way, using the less complex laboratory equipment, the good results can be
obtained which could be useful in the verification of the many numerical software
simulation methods related to the air stream around the vehicle body.
In this paper, the possibilities of the external aerodynamics research around
the vehicle body, experimental results gained and their comparison with the numerical
methods are presented. Based on the experimental results and using the numerical
methods, appropriate recommendations are given for the further research work.
2.
EXPERIMENTAL RESULTS OF THE AERODYNAMICS OF THE VEHICLE

MODEL
The first steps in the field of the vehicle model external aerodynamics research
by these authors had started at Mechanical Engineering Faculty, University of Sarajevo
in year 2006. In these days, the tests had been based on determination of the
aerodynamic drag force when the balance (equality) between the weights of the known
mass and the air drag force have been established. Continued research that followed
during 2008 has been based on the implementation of the modern measurement
methods where the use of the strain gauges could accurately determine the
aerodynamic drag force and determine the aerodynamic drag coefficient as well.
Obtained results of this experiment are shown in [3].
However, basic drawback of the previously performed measurements laying on
the fact that the vehicle model is 35 times of its length away from the outlet nozzle of
the wind tunnel and didnt give the possibility to determine the pressure field around
the model of the vehicle what is very important in order to define the external
aerodynamics. Considering those drawbacks, the idea consists of forming the
extension nozzle came up and so, in fact, placing the vehicle model in the nozzle
superstructure, as shown in Fig. 1. Thus, the uniform velocity of the air flow in front of
the vehicle model have been achieved, and in addition to that, the use of the nozzle
superstructure made the installation of the devices measuring the pressure around the
vehicles model possible to perform.
732
Research of the external aerodynamics of the vehicle model
Description:
1 Measuring probe carrier for
determining the pressure around
the vehicle model
2 Nozzle superstructure
(extension) of the wind tunnel
3 Vehicle model
Fig. 1 Superstructure of the nozzle extension of the wind tunnel

As the vehicle testing model of the external aerodynamics, the model of
Peugeot 407 Coupe in the scale of 1:18 have been used. At that time, the values of the
air velocity, pressure around the vehicle model and the air resistance were measured.
For this purpose, the measuring equipment have been used was next: pressure gauge
FLUKE FL 922 with the Pitot Prandtl probe and with the characteristics as follows:
4000 Pa / 1 Pa / 1% + 1 Pa, and dynamometer LUTRON FG 5005 with the
characteristics as follows: 5000 g / 1 g / 0,4 %. Using the pressure gauge FLUKE
FL 922 data containing the values of the air stream velocity in front and behind of the
vehicle model have been obtained, as shown in Fig. 2, as well as the values of the
static pressure, for a different flow velocities, around the vehicle model, as shown in
Fig. 3.
Fig. 2 Velocity profile of the air stream in front and behind of the vehicle model
Based on the results shown in Fig. 2, it is obvious that the superstructure
attached to the wind tunnel nozzle achieves that the air stream in front of the vehicle
model be kept uniform what justifies the initial assumptions. However, the velocity of
the air flow have been measured behind the vehicle model as well, where the
733
Pressure, bar
disruption of the air flow and possible turbulence was noticed behind the vehicle model
and the platform keeping the vehicle model in place in order to measurement makes
possible. Since this paper only analyzes the flow around the vehicle model in terms of
the speed and the pressure of the air stream, on this occasion, the value of the
aerodynamic drag coefficient of the vehicle model is not going to be determined.
The values of the static pressure around the vehicle model, for the velocities of
air flow given in Fig. 2, are given in Fig.3.
High of vehicle model, mm
21.8 m/s
25.5 m/s
28.8 m/s
32.5 m/s
35.8 m/s
Vehicle model length, mm
Fig. 3 Distribution of the static pressure values along the vehicle model for different
velocities of the air stream
Considering the results shown in Fig. 3 it can be concluded that the increased
air flow rate leads to the increased pressure in the front of the surface of the vehicle
model below the hood and on the windshield as well. Following the shape of the roof
towards the rear glass and the hatchback, the pressure decreases. Significant
734
fluctuations in the pressure caused by the formation of the boundary layer and
turbulence along the end side of the vehicle model can be seen as well. Especially this
formation of the turbulence after the rear glass and hatchback of the vehicle model has
a dominant influence in defining the turbulence, where many numerical models are
trying to give the answer what is going to be presented in the next chapter.
3.
NUMERICAL SIMULATIONS OF THE VEHICLE MODEL EXTERNAL

AERODYNAMICS
The fact is in order to do the research of the external aerodynamics, the

substantial funds for the experimental setup are necessary even for testing vehicle
model analyzed in this example, but applying the numerical methods, the spending can
be significantly reduced. Although, the use of the numerical methods makes sense in
order to obtain the first results, which are very helpful in the phase of the prototype
development. Therefore, almost all manufacturers of motor vehicles practicing to do
numerical simulation at first, then design a vehicle model, perform the experimental
tests, complete the verification of the experimental results with the results of the
numerical simulations and continue with further development in order to achieve the
optimal solution. In this way the significant savings in time and money can be
accomplished.
Today, various software implementing the numerical simulation in the field of
fluid mechanics can be found on the market. The software chosen for the purposes of
this work was Ansys Fluent, because it has the different methods of the turbulence
calculation built in [4].
In order to reduce the hardware resources, the longitudinal cross-section of the
adopted vehicle model have been chosen. The numerical mesh is generated
automatically with some additionally refined characteristic area. The problem of the
external aerodynamics of the selected vehicle model has been analyzed as a quasisteady problem where k-H model of the turbulence calculation was used. It is important
to note that in the case studies of the external aerodynamics as an unsteady problem
i.e. transient problem, the use of DES and LES model of the turbulence calculation in
lieu of k-H is recommended. The same parameters of the air flow around the vehicle
model obtained by the experiment, have been used in the numerical simulation as well,
thus the initial conditions applied for this case are the next:
- air flow velocities: v0 = 21.8; 25.5; 28.8 and 32.5 m/s,
- static pressure: p0 = 101325 Pa,
- density of air: U = 1.21 kg/m3.
Due to the limited space provided for the text of this paper, the comment and
the analyze of the results given below refer only to the calculations of the pressure and
the velocity of the air stream flowing along at the speed of 32.5 m/s (Fig. 4 and Fig. 5).
The results of the numerical simulations at the other velocities above (21.8; 25.5 and
28.8) will be shown in order only to give a picture of that how the air streamlines
around the vehicle model are changing by changing the initial speed, as it is shown in
Fig. 6.
735
Fig. 4 Air pressure around the vehicle model
Fig. 5 Velocity of the air stream around the vehicle model

In Fig. 6, the results of the air velocity for a given speeds of the vehicle (the air
stream velocities) of 21,8; 25,5 and 28,8 m/s are presented.
736
Fig. 6 Representation of the streamlines around the vehicle model for a different
velocities (21.8; 25.5 and 28.8 m/s) of the air.
Regarding the results of the numerical calculation, it might be concluded that its
similarity with the experimental results is more than satisfactorily considering the
values of the static pressure (Fig. 4). This is obvious in terms of the static pressure
values profile in the frontal area of the vehicle below the hood and on the windshield.
737
Emergence of the turbulence behind the hatchback for different cases of the initial
velocities of the air stream, presented in Fig. 5 and Fig. 6, is corresponding with the
experimental results shown in Fig. 2. Also, it is obvious that increased initial velocity
leads to an increase of the turbulence behind the vehicle model, what is clearly visible
in Fig. 6.
4.
CONCLUSION
This paper shows that the numerical simulation of the external aerodynamics of
the vehicle model as well as of the real vehicles now may be implemented by the
average hardware resources. The development and improve of the external
aerodynamics of the vehicles using a computer with a various CFD (Computer Fluid
Dynamics) software installed (ANSYS FLUENT, AVL Fire etc.), not only makes
possible the solving of the engineering problems quick and efficient but it has become
the standard in the companies engaged in the production of the vehicles, planes and
vessels. These are all reasons why CFD is becoming the standard for any
development of the vehicles aerodynamics in the future especially considering the time
and costs savings avoiding the modeling of the external aerodynamics of the vehicles
by the experiment.
Although, in order to make a serious approach to the above problems, in the
case when precise results are required, the experiment, that most often consists of the
wind tunnels tests must be performed. However, there are some aspects of the
calculation of the specific components of the air stream that still require the
professional skills and experience in the research work (air stream around the engine
and the equipment of the motor vehicles) owing to the complex analysis in the form of
3D simulations for what the high-capacity computer systems are required.
REFERENCES
[1] Giilespie T.: Fundamentals of Vehicle Dynamics, SAE R-114, Warrendale, Pa.,
USA, 1992
[2] Hucho W.H.: Aerodynamics of Road Vehicles, Forth Edition, SAE R-117,
Warrendale, Pa., USA, 1997
[3] Pikula B., Mei E., Hodi M.: Determination of Air Drag Coefficient of Vehicle
Models, International congress Motor Vehicles and Motors 2008, Kragujevac,
October 8-10, 2008
[4] www.ANSYS.tutorials.com
738
ROAD TRAFFIC SAFETY PERFORMANCE IN MONTENEGRO

Vladimir R. Pajkovi1
Summary: The road safety situation in Montenegro is very poor, compared to both the
EU average and neighbours in the region. The fatality rate on Montenegros roads
measured as deaths per capita is 50% higher than that of EU average. Even within a
regional context, Montenegro has a very high fatality rate. Taking into account that car
ownership in Montenegro is still considerably lower than the EU average, it is expected
that this situation will deteriorate further if no actions are taken. Efforts have been made
in recent years to address this problem, but reforms were insufficient and inadequate
partly because addressing the situation properly requires considerable investments.
Key words: road traffic, traffic safety, road safety performance
1. INTRODUCTION
The safety performance of the road traffic system is the result of the (right)
combination of the functionality, homogeneity, and predictability of the road network,
the road environment, and the traffic involved. Assessment of road safety performance
is based on statistical data. Statistical indicators are proposed as supportive tools in
comparing the level of road safety between various countries and are useful in terms of
monitoring changes over time. Also, in order to make progress in road safety,
comprehensive data collection and in-depth data analysis are essential in terms of
designing effective safety strategies, setting challenging targets, determining
intervention priorities and monitoring programme effectiveness.
In relevant statistics (ITF, CARE-EU road accident database) there is no data
about road safety situation in Montenegro. Regardless that, some indicators as road
infrastructure condition, average age of the vehicle fleet and commonly ignoring traffic
regulations (speed limits, drink-driving laws, etc.) suggested that this situation cant be
satisfactory. Provisional accident data collected for this paper reveal that road safety
has become a serious and increasing economic and social problem in Montenegro.
2. ROAD INFRASTRUCTURE
Montenegro has a road network totalling approximately 7,000 km with approx.
900 km of main and primary roads, 950 km of regional and secondary roads, and
around 5,000 km of local roads. This is equivalent to a road density of 500 km per
1
Doc. Vladimir Pajkovic, PhD, Montenegro, University of Montenegro, Fac. of Mech. Engng, pajkovic@ac.me
739
Vladimir R. Pajkovi
1,000 km2. This figure is broadly consistent with the density of regional comparators in
South-eastern Europe (with an average of about 555 km per 1,000 km2 of territory), but
below the levels in the new EU member states (Table 1). Other factors, such as the
distribution and density of population, and the country's geography play a considerable
role in determining country's road network. These effects become evident when
making the comparison on a different measure of road density viz. road kilometres
per 1,000 people. With this measure, with more than 11 km of road per 1,000
inhabitants, Montenegro is ahead of most of its regional comparators and comparable
to those of the new EU countries, [1].
Table 1 Road network density
Road density
km of roads per
2
1000 km
km of roads per
1000 people
Montenegro
500
11.1
South-eastern Europe, average

Albania
Bosnia and Herzegovina
Croatia
Macedonia, FYR
Serbia
555
657
427
506
513
500
5.9
3.5
5.6
6.4
6.4
5.2
New EU member states, select

Czech Republic
Estonia
Hungary
Slovenia
1427
1646
1320
1733
1007
19.9
12.5
41.2
15.7
10.2
Situation is different regarding road infrastructure quality. In the recent Global

Competitiveness Report 20092010, [2], Montenegro ranked only 119th (of 133
countries surveyed) for the quality of its road infrastructure (only Bosnia and
Herzegovina scored lower among the countries in the region). Most of Montenegros
road network is high risk, including those segments with a high traffic volume. Also,
nearly one-half (47%) of the entire road network is in poor or very poor condition,
reflecting inadequate maintenance (the full breakdown of the main road network is
15% good, 39% fair, 37% poor and 10% very poor, [3]). Inadequate maintenance
increases the rate of network deterioration, creating considerable contingent liabilities
(in terms of future rehabilitation costs) and economic costs. To improve the road safety
situation on the main road network, costs an estimated 105138 million, [1].
3. LEVEL OF MOTORISATION
Montenegros motorisation is growing rapidly since 2004. During 20042008
the number of registered vehicles has grown by more than 70% (fig. 1), from around
123,000 to more than 210,000. In nationals vehicle fleet prevailed a passenger cars
(around 80%). The average passenger car in use is greater than 14.5 years old, [4].
Older cars were imported in a considerable number between 2004 and 2007, when
Montenegro had no adequate regulations to prevent vehicles with older technologies
740
Road traffic safety performance in Montenegro
and lower safety performances from being imported.

Vehicle registration in Montenegro
(1999=100)
175
150
125
100
75
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
50
Fig. 1 Level of motorisation in Montenegro

Compared to both the EU average and neighbours in the region, Montenegros
level of motorisation is moderate. The rate is 280 passenger cars per 1,000 population
(based on Eurostat 2009 data), whereas the EU-27 average, for example, is 460 per
1,000 population (fig. 2).
Albania
Turkey
Macedonia FYR
Bosnia & Herz.
Serbia
MONTENEGRO
Croatia
EU-27
Iceland
0
100
200
300
400
500
600
700
Fig. 2 Number of vehicles per 1,000 population (2009)

Therefore, this rate of motorisation is not considerable affecting road safety by
itself. Only combined with both inadequate road infrastructure and low vehicle safety
performance (on average) this rate of motorisation become a limiting factor of traffic
safety situation in Montenegro. Add to that the fact that number of vehicles on
Montenegros roads multiplies by as many as 20 times in the tourist season.
741
Vladimir R. Pajkovi
4. ROAD SAFETY PERFORMANSE

In 2010 in Montenegro there were 8,103 road accidents in which 95 people
died and 2,120 were severely or slightly injured. Compared to 2000, the number of
road accidents went up by 44.7%, the number of killed by 17.3% and injured by 9.7%.
Over the last decade more than 1,000 people died on Montenegrin roads and more
than 23,300 were injured (Table 2).
Table 2 Traffic accidents and casualties on Montenegrin roads
Year
Accidents
Total
2000=100
Killed
Total
Injured
2000=100
Total
2000=100
2000
5597
100
81
100
1933
100
2001
5275
94.2
105
129.6
1957
101.2
2002
5503
98.3
81
100
1834
94.8
2003
5094
91.0
84
103.7
1702
88.0
2004
5377
96.0
91
112.3
1750
90.5
2005
6192
110.6
95
117.3
1942
100.4
2006
7185
128.4
85
104.9
2257
116.7
2007
8882
158.7
122
150.6
2796
144.6
2008
10170
181.7
111
137.0
2473
127.9
2009
9043
161.5
100
123.4
2542
131.5
2010
8103
144.7
95
117.3
2120
109.7
In the same period nearly all EU and high income countries record a decrease
in number of traffic casualties year after year. Some of these countries, such as
France, Spain and Portugal, had an average annual reduction of more than 7%, [57].
Montenegros figures generally point in opposite direction until 20072008 trend is
towards an incline (with small exceptions) and declines only in last 23 years (fig. 3).
Years 2007 and 2008 were particularly evil, with more than 5,400 casualties
(deaths and injuries) on Montenegrin roads. This was the price of uncontrolled growth
of motorisation (in precedent years) and inadequate road infrastructure that couldnt
bare an increasing traffic volume. The highest density of fatalities is found on main
north-south route Border with Serbia Bijelo Polje Podgorica Cetinje Budva and
Podgorica city zone. Similarly to other European countries the victims of road
accidents are first of all passenger car users (drivers and occupants) amongst
fatalities there were mostly car drivers (49% in 2007 and 42% in 2008, of all fatalities
on Montenegrin roads) and car passengers (27% in 2007 and 38% in 2008), fig. 4.
Also, passenger car drivers have been the main cause of accidents in Montenegro, in
742
Road traffic safety performance in Montenegro
Evolution of fatalities, accidents and injured

Montenegro
(2000=100)
200
180
160
140
120
100
80
Accidents
Fatalities
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
60
Injured
Fig. 3 Evolution of traffic accidents and casualties

particular ignoring road traffic regulations. The most frequent causes of road accidents
caused by drivers include high speed, failure to give way, illegal overtaking and illegally
driving through pedestrian crossing. Drinking and driving is a major problem in
Montenegro as well.
Motorcyclist,
bicyclist
5%
Pedestrian
20%
Motorcyclist,
bicyclist
4%
Pedestrian
15%
Car driver
49%
Car
passenger
27%
Car driver
42%
Car
passenger
38%
Fig. 4 Fatalities user type in 2007 (left) and 2008 (right)

High fatality rate Montenegro shows compared to both EU countries and
neighbours in the region. According to data from 2008, with over 170 fatalities per one
million population, Montenegro is ranking not only significantly higher than EU counties
(averaging under 80 fatalities per million population), but has also a higher fatality rate
than the neighbouring countries (fig. 5).
Having traffic safety figures so bad, efforts have been made in recent years to
address this issue. In 2010 National Coordination Board was established in Montenegro
to monitor road safety parameters. In addition, government adopted Strategy for
improvement of road traffic safety. Main goal of this strategy is to reduce number of
fatalities by 30% and injuries by 20% before 2014 (compared to 2007). Long-term goal
is to reduce these figures by 50% and 30% respectively by 2019, compared to the
same year. Sadly, the growing interest in reducing road traffic risk is not accompanied
743
Vladimir R. Pajkovi
with active road safety policies. Proposed measures are inadequate and insufficient
partly because addressing the situation properly requires considerable investments.
Netherlands
Sweden
United Kingdom
Switzerland
Germany
Italy
Macedonia FYR
Albania
Slovenia
Slovakia
Serbia
Greece
Bulgaria
Romania
Croatia
MONTENEGRO
0
30
60
90
120
150
180
Fig. 5 Fatalities per million population (2008)

5. CONCLUSION
Road traffic safety performance in Montenegro is very poor. Fast and dynamic
growth of motorisation over the last 5-6 years took the government by surprise. With
the road infrastructure unsuitable for the growing needs, a largely differentiated fleet of
cars, difficulties with effective traffic enforcement and too few effective preventative
schemes, the effects of growing motorisation have turned into a painful experience.
Over the last decade (20012010), compared to 2000, the number of road accidents
went up by 44.7%, the number of killed by 17.3% and injured by 9.7% (opposite
direction than the trend in EU which is towards a decline in number of fatalities and
injuries!). The fatality rate on Montenegros roads measured as deaths per capita is
also 50% higher then that of EU average. Even within a regional context, Montenegro
has a very high fatality rate. The economic cost of this road safety situation is
estimated to exceed 2% of GDP. In recent years efforts have been made to address
this problem, but reforms were insufficient and inadequate partly because addressing
the situation properly requires considerable investments.
LITERATURE
[1] World Bank Report No. 46660 ME (2008). Montenegro Beyond the Peak:
Growth Policies and Fiscal Constraints, Vol. II
[2] World Economic Forum (2009). Global Competitiveness Report 20092010
[3] COWIS A/S (2008). World Bank: Road Safety Survey Montenegro, Final Report
[4] Pajkovi, V., Simovi, S., Damjanovi, M. (2010). Influence of the vehicle technical
failures on road safety in Montenegro. Road accidents 2010 Conference
Proceedings, p. 5864. (in Serbian)
[5] CARE-EU road accidents database (2007). Road accident statistics in Europe
[6] OECD/ITF (2009). ITF Trends in the Transport Sector statistics 2008
[7] OECD/ITF (2011). IRTAD Road Safety 2010
744
SURVEY OF METHODS FOR CALCULATING THE WAVE ACTION IN

THE MANIFOLDS OF INTERNAL COMBUSTION ENGINES
Dobrivoje Ninkovic1
Summary: In the thermodynamical design process of IC engines one can identify
three primary areas of concern as regards the quest towards maximising the engine
performance: the cylinder process, the gas exchange process, and the unsteady flow
in the gas conduits, frequently referred to as wave action. Whereas the cylinder
process can be analysed in isolation from the other two when the valves are closed,
the wave action substantially influences the performance of the gas exchange. This is
even more important in the case of turbocharged engines, for the knowledge of the
interaction of flow pulsations with the turbine is instrumental in arriving at an optimal
performance of the entire turbocharged system.
As one of the pioneers in the area of turbocharging of large Diesel and gas engines,
ABB Turbo Systems Ltd. (formerly Brown Boveri & Co.) have also been at the forefront
of the development of the calculation methods for performance prediction and
optimisation. By staying abreast with the current research efforts, and through original
contributions to the numerical methods for predicting one-dimensional wave action in
the engine pipework and the associated turbocharging system components, it is
ensured that the companys in-house simulation software delivers the accuracy
demanded by the ever-increasing performance and emission requirements.
Surveyed in the paper are the methods for calculating the wave action in the manifolds
of IC engines, starting with the first developments (which nevertheless are still at the
base of the most modern calculation methods), up to and including the most recent
developments. Current research activities are also indicated. Although it has been
claimed that the one-dimensional methods were to be replaced by the two- or threedimensional schemes, it is shown in the paper that they still constitute the backbone of
the contemporary simulation practice and are capable of being further refined.
Key words: Wave Action, IC Engines, Simulation, Unsteady Flow, Turbocharging,
Engine Manifolds
1.
INTRODUCTION
The steadily increasing regulatory and other demands on the energy efficiency,
fuel economy, and minimization of the exhaust emissions of modern internal combustion engines call for using ever more sophisticated methods and tools in the research
1
Dr. Dobrivoje Ninkovic, ABB Turbo Systems Ltd, 5401 Baden, Switzerland, dobrivoje.ninkovic@ch.abb.com
745
Dobrivoje Ninkovic
and development processes. Beside measurements, which always deliver the final
facts as to the performance of a device or a system, the simulation and optimisation
methods have always been indispensable parts of the engine development process [1].
From the beginning of the proliferation of digital computers, ABB Turbo
Systems Ltd., world's largest independent manufacturer of turbochargers for large
Diesel and gas engines, has been actively developing and using computer programs
for the performance prediction of turbocharged engines, which unite complex
mathematical models of matter and energy exchange in all components of the
turbocharging system with a substantial experimental data base. The system can be
modelled at various levels of detail, starting with simple sizing procedures, continuing
with lumped parameter models, commonly referred to as the filling and emptying
method, and at the highest resolution level, the piping in the system is discretised in
the space domain, in the vast majority of cases by assuming one-dimensional flow.
The pipes are then connected by means of generic flow resistances (if there are only
two pipes to be joined with each other), or by employing special elements, called pipe
junctions (three or more pipes at the same connection point) [2]. This approach is a
prerequisite for calculating the propagation of large-amplitude waves in the engine
manifolds, an effect of utmost importance given the periodic operation mode of IC
engines. From the theoretical point of view, one is dealing with a distributed-parameter,
unsteady flow model, in the engine literature customarily referred to as wave action.
The term wave action was probably used for the first time in the context of IC
engines in a seminal paper by Bannister and Mucklow of 1948 [3] to describe the
unsteady flow phenomena occurring in a pipe after a sudden release of gas under
pressure from an attached cylinder, a situation closely related to the propagation of a
gas pulse discharged from an engine cylinder into the exhaust collector. Although
detailed measurements of similar phenomena were performed a decade earlier in
Germany [5][6], Bannister and Mucklow were the first to present a theory explaining
the phenomena measured, leading to a request from the audience to extend their
theory so as to be used as a prediction, i.e. a simulation tool in engine design.
The theoretical approach of Bannister and Mucklow is founded on the equations relating the pressure and velocity in an isentropic, large-amplitude wave, derived
by Earnshaw in 1860 [11]. The same equations can be used to calculate the superposition of simple waves and decomposition of complex waves into their left and right
running components. The great advantage of the method is the insight it affords into an
essentially complex phenomenon. Although several other researchers subsequently
employed this theory in their studies (e.g. [12]), it had to wait for more than forty years
to be implemented as a useful simulation method [4], mostly because suitable
computing devices were not available at the time of its inception.
Approximately at the same time, a grapho-analytical procedure appeared for
the prediction of wave action in fluids, known as the method of characteristics [7][8]. As
the previously mentioned method, it was also based on a theory published at approxith
mately the same time in 19 century, this time by Riemann [13]. In combination with a
method for connecting the pipes with the boundaries such as valves, throttles etc.
developed by Jenny [9], it quickly established itself as the standard technique for
studying the propagation of large-amplitude waves in the manifolds of IC engines. Its
attractiveness was due to the incremental manner in which the so-called position and
state diagrams depicting the wave action were calculated and plotted. After it was
implemented as a computer program under the name mesh method of characteristics
746
Survey of Methods for Calculating the Wave Action in the Manifolds of Internal
Combustion Engines
by Benson et al. [16], it remained in use well into the computer era of engine
simulation, to be definitively displaced by the numerical methods in the eighties.
However, the metod of characteristics remains the essential part of almost all
models of pipe boundary conditions (valves, throttles, turbocharging turbines, etc.) [17],
and is therefore to be found in all commercial and proprietary codes for unsteady flow
simulation in IC engines.
The term numerical methods refers to a collection of algorithms for obtaining
numerical solutions to the one-dimensional, hyperbolic, partial differential equations of
mass, momentum, and energy conservation in internal flow systems of IC engines,
referred to as the Euler equations. Their development coincided with, and was indeed
made possible by, the introduction of useful digital computers in the sixties, following a
publication by Lax [10]. Together with the models of boundaries, numerical methods
constitute the backbone of the contemporary performance prediction software suites
for the design and optimisation of IC engines. Although it is now a mature field, new
papers and reports addressing individual issues apperar steadily; and there is no
shortage of comprehensive surveys. To name but a few, the two volumes of
Winterbone and Pearson [18][19] represent the reference work for this discipline; and
Depcik et al. cite more than 100 references in their paper [20]. Serrano et al. [21] and
Torregrosa et al. [22] recently surveyed the field from the standpoints of thermal
contact discontinuities and tapered pipes, respectively.
It is not the purpose of this paper to attempt yet another detailed survey of this
important area of engine simulation, but rather to point out some of the issues from the
standpoint of the requirements arising from the use of these algorithms in the in-house
software of a turbocharger manufacturer. This software is an essential part of the large
engine turbocharging process R&D, and plays a vital role in supporting engine
manufacturers in their product development activities.
2.
THEORETICAL BACKGROUND
Since the diameter to length ratio of engine manifolds is typically a small

number, it is customary to treat the gas flow therein as essentially one-dimensional.
Furthermore, based on the order of magnitude analysis of the governing equation
terms [23], viscosity effects in the momentum equation are negligible in comparison
with the pipe wall friction; and the convective radial heat transfer through the pipe wall
dominates over both the longitudinal heat conduction and the viscous losses in the
energy equation. With this in mind, the equations describing the conservation of mass,
momentum and energy in an one-dimensional channel can be compactly written as:

(1)
where

(2)
where and stand for the pressure, density, velocity, internal energy and
747
Dobrivoje Ninkovic
enthalpy, resp., and the subscript refers to the stagnation conditions. Pipe area is
designated by is the friction force term and represents the amount of heat
exchanged per unit mass of the flowing gas. Note that constant gas composition is
implicit in the above formulation. This, so-called conservation form of the equation, was
widely used as a basis for developing various numerical algorithms for use in engine
simulation packages, albeit with conservation errors when applied to pipes with
variable cross-sectional area. The latter was significantly improved by inserting the
area into the state vector, as proposed by Corbern and Gscon [24], i.e.

(3)
This form is recommended by most authors as a much better basis for

developing solution algorithms for one dimensional pipe flow. One notable exception is
Depcik et al. [20], who still use the form of Eq. (2) in their chosen solver, proposing
simultaneously to address the variable area issue in the future work.
The closure of the above equation systems used to be effected by means of
the perfect gas equation of state, characterized by constant specific heats. This
question shall be dealt with in a subsequent section.
2.1 Numerical solution of the governing equations
As mentioned above, the first algorithm employed for solving the Euler
equations was based on the method of characteristics. Being of the first order, it was
not accurate enough, and the increased accuracy was sought in higher order schemes.
The paper by Lax and Wendroff [25] provided the basis for quite a number of secondorder schemes. In its original version, the scheme called for calculating a Jacobian
matrix at each mesh and time instant, and was referred to as the one-step method.
This was deemed a disadvantage, and Richtmyer and Morton [26] proposed a two-step
version of the scheme, which disposes of the Jacobian calculation, evaluating the
derivatives at a suitable midpoint in the computational stencil instead. The two-step
version of the LW (Lax-Wendroff) method served as a basis for several well-known
schemes, including the widely used MacCormack algorithm [27]. Bulaty and Niessner
[28] compared the one- and two-step versions of the algorithm from the point of view of
the computational speed, and found out that they practically do not differ. The present
author recently repeated the test in the context of variable gas properties in pipe flow
calculation, and observed the same conservation accuracy, but the current one-step
scheme in the ABB simulation software, by virtue of using an efficient mechanism for
keeping track of the fluid properties, was faster by 40% than its two-step counterpart.
Since the numerical schemes are obtained by discretizing the above differential
equations, truncation errors are inevitable. They give rise to nonphysical dissipation
and dispersion effects in the numerical solutions. The former tend to impair the resolution of steep fronts, and the latter give rise to large oscillation at the discontinuities.
Quite a lot of research effort has been invested into developing methods for reducing
these errors to an acceptable level. In the context of the LW schemes, two methods
have withstood the test of time, representing the state of the art even today. Based on
the development of Boris and Boook [29], BBC (now ABB) engineers Niessner and
748
Combustion Engines
Bulaty [30] established several flux-corrected transport schemes (FCT), and Davis [31]
devised a version of the LW algorithm that possesses the TVD (Total Value Diminishing) property. Both belong to the so-called post processing algorithms in that the
solution obtained by means of a native scheme is subsequently corrected in order to
remove the nonphysical deviations.
Another class of solution algorithms makes use of the high-resolution
Godunov-based schemes, whose concept involves gradient limiting, i.e. they attempt
to deal with the spurious numerical effects before they occurred, which places them
into the pre-processing category of algorithms [18]. The generic gradient-limiting
mechanism employed in this class of schemes is known as MUSCL (Monotonic
Upstream-Centred Scheme for Conservation Laws). Although they seem to be widely
used in the field of CFD, and Winterbone & Pearson [18] in 2000 expected their
increased use in the field of IC engine simulation in the future, only one major effort
materialized in the intervening time [32], albeit with somewhat mixed results.
Although there were efforts to use the Finite Element Method for predicting 1D
wave action, and good accuracy was demonstrated [33][34], the computation time
penalties (almost two orders of magnitude slower than e.g. LW+TVD) precluded their
wider use in the practice [18].
Finite volume methods are not widely used in the IC Engine simulation practice. However, one well-known commercial code employs an algorithm of this kind [35].
As already mentioned above, Prof. Blair developed a method [4] that does not
start with the Euler equations, using instead the wave propagation formulae of
Earnshaw [3]. He pursued this idea actively for decades and employed it as the basis
for his engine simulation code [14]. The method was validated by comparing its results
with those obtained by using other simulation algorithms, and against measurements
performed on a specially designed unsteady flow test rig, e.g. [36][37]. Recently, a
second-order version of the algorithm appeared [15], improving on the treatment of
heat transfer and including a method for mass conservation.
Another nontraditional method appeared in 1991 [38], unique in its treatment of
time and space in an equal manner, and capable of high simulation accuracy and
robustness. Known as the CE/SE method (Conservation Element / Solution Element),
it does not need any of the techniques mentioned above to combat nonphysical
deviations in the solution, using instead weighted averaging of the neighbouring cells.
Briz & Giantanasio [39] added the source terms (the original derivation contained only
the homogenous part of Eq. (1)) to the model and used it to simulate unsteady flow in
IC engines. The method has been thoroughly analyzed both from the theoretical and
experimental standpoints in a large number of papers [40], and is being increasingly
applied to a wide range of flow problems and further developed. This also holds for the
IC engine applications, as evidenced by several recent papers, e.g. [21][22].
3.
SIMULATION ACCURACY
Only the basic set of model components has been included in the above
exposition of wave action simulation modelling. Nevertheless, this basic set has been
successful in covering a good part of of the requirements arising in the practice. For
example, heat transfer through exhaust piping walls can still be neglected when
simulating wave action in large engine turbocharging; and the perfect gas hypothesis is
still a part of the boundary element modelling in a number of cases. Constant gas
749
Dobrivoje Ninkovic
composition in all pipe meshes has also produced ample accuracy in most unsteady
flow simulations for quite a long time. However, simulation accuracy requirements have
been increasing with time, dictating thus improvements in the modelling of the wave
action simulation.
3.1 General accuracy issues
Insufficient conservation accuracy in the unsteady flow simulation was noticed
relatively early in the development of finite difference schemes [41], and has been a
research subject ever since [24][22][21]. It can be caused by a number of factors,
starting from the discretization issues in arriving at a numerical scheme for solving the
Euler equations, scheme properties, treatment of the gas properties, boundary
conditions modelling, treatment of discontinuities, etc.
As reported by Bulaty et al. [42], conservation accuracy in the ABB simulation
software, i.e. in the own version of the LW-FCT algorithm, was brought to a satisfactory
level by formulating the problem in terms of Eq. (3) above, and using half-meshes at
the boundaries, as suggested in [24]. Gas composition in the pipes was kept constant
within an engine period, and constant values of the specific heats were therefore used
in all meshes. This produced very good agreement between the simulation results for a
six-cylinder engine and the measurements, both at the system level and in the time
domain [42].
Fig. 1 Comparison of data ([36], Figs. 20 & 21) with simulations (Depcik [32], Fig. 2.28)
Crucial part of an algortihms performance test is the fidelity of measured waveform prediction. The de-facto necessary condition for an algorithm to be used for wave
action simulation is a good prediction accuracy with the shock tube problem [18]. However, this is not the sufficient condition, as the experience shows: Bulaty & Niessner
[23] found out that the so-called artificial compression method of Harten [43] performed
well with the shock tube problem, but was not usable in engine work, especially in
situations involving temperature transients. This seems also to be the case with the
MUSCL algorithm chosen by Depcik [32], for in spite of very convincing shock tube
results, it produced a rather large pressure error (Fig. 1) when reproducing the wave
action measured in a constant pipe diameter case on a test rig of Kirkpatrick et al. [36].
750
Combustion Engines
With reference to Fig. 2, simulating the the same case with the ABB version of the LW
scheme produces a more satisfactory agreement.
Kirkpatrick et al., SAE 941685, Fig. 20
Kirkpatrick et al., SAE 941685, Fig. 21
1.35
1.3
Blair data
ABB LW +FCT
Blair data
ABB LW +FCT
1.3
1.25
1.25
1.2
1.2
P
ressure[bar]
P
ressure[bar]
1.15
1.15
1.1
1.1
1.05
1.05
0.95
0.001
0.002
0.003
0.004 0.005
0.006
Time [sec]
0.007
0.008
0.009
0.01
0.95
0.01
0.012
0.014
0.016
Time [sec]
0.018
0.02
Fig. 2 Comparison of data ([36], Figs. 20 & 21) with simulations (ABB LW+FCT)
3.2 Variable gas properties
Accounting for variable gas properties must be considered from two points of
view, both of which have the conservation aspect to them, but also differ in their main
intents. At the points in the system that are prone to recirculation typically, in the
neighbourhood of the valves conservation is the main aspect. At large valve overlaps, as is often the case with large engines, exhaust gas pushed through the inlet
valve into the attached channel is later readmitted into the cylinder; and if the
recirculation is not accounted for, mass and species conservations are in error.
Similarly, at the exhaust valve, scavenging air ejected into the valve channel and later
readmitted must be properly accounted for. In both cases, wave action is not the prime
factor; it is rather the conservation. As a matter of fact, the main point here is the
engine mass conservation.
Farther away form the recirculation zones, accuracy of the wave action calculation is of the same importance as the conservation issues. The behaviour of the
caclulation scheme at the discontinuities (geometrical, thermal, and compositional) are
then the factors to consider when evaluating the results.
Various levels of accuracy are possible when considering variable gas properties. As already mentioned [42], the composition in a pipe or a group of pipes can be
assumed constant within an engine period, and the calculations made with constant or
temperature-dependent properties. However, the species balance can only be calculated if the gas composition is accounted for in the governing equations. If chemical
transformations are to be considered, as is the case in modelling e.g. a catalytic
converter, the model must be augmented by all pertinent chemical reactions.
In order to enable considering the species balance, the state vector must be
augmented by a term expressing the working gas composition. In the Eq. (4) below it is
the vector of mass fractions ( assuming no chemical reactions):
751
Dobrivoje Ninkovic

(4)

For a gas mixture with Nspecies, only the first ones have to be considered, i.e.
(5)
since the sum of the fractions must be unity, i.e.

.
As regards the equations of state to use, constant specific heats are nowadays
employed only within the individual integration steps. Globally, the ideal gas assumption ( ) prevails, with a choice of accurate correlations [44][45] for
calculating the specific heat (ABB software uses the NASA correlation [44]). A real gas
version of the LW algorithm was developed in a different application area [46], but at
the pressure levels currently encountered in the IC engine manifolds, there is still no
need for considering the pressure effects when calculating the wave action in pipes.
Blair et al. [37] established a basis for studying the effects of varying gas
properties upon the wave action by performing a series of measurements on their
single-pulse test rig. Referring to Fig. 3 below, valve 'S' separates the parts of the rig
that can be filled with different gases. The authors used air and CO2, creating thus a
composition discontinuity between a predominantly two-atomic and a three-atomic gas.
Fig. 3 Sketch of the test rig of Blair et al. [36]

From the four tests reported in [37], we are presenting here the one marked as
CCA in the paper. It denotes the case where the part to the left of valve S was filled
with CO2 at 1 bar and 293 K, and the rest communicated with the atmosphere through
the open end at the right. The vessel at the far left side also contained CO2, but at a
pressure of 2.4 bar and the same temperature as in the rest of the rig. Referring to Fig.
4 below, pressure traces calculated with the current version of the ABB unsteady pipe
flow algorithm (LW + FCT) at the three measurement stations are compared to the
measured ones. The agreement is obviously very good, both from the temporal side
(the arrival times of both the compression and expansion waves are almost exact) and
from the standpoint of amplitude accuracy. This can not be said of the simulation
results obtained by the original authors. With reference to the bottom plot of Fig. 4,
while the arrival time of the incident compression wave is well predicted, the front rise
752
Combustion Engines
time is not (the simulation suggests a shock there), and the pulse amplitude is
generally higher. Furthermore, neither the temporal nor the amplitude aspects of the
reflected expansion wave (right hand part of the plot) are predicted well.
While the variable gas property predictions of the LW+FCT algorithms presented here are accurate enough for practical use, there are occasions where this algorithm creates nonphysical values of the component fractions [48]. This does not seem
to be the case if the TVD scheme is used [47]. The latter has also advantages in the
case of tapered pipes.
Blair gas discontinuity test, case "CCA", MS1
1.8
Blair data
ABB LW+FCT
Pressure [bar]
1.6
1.4
1.2
1
0.8
0.6
0.01
0.02
0.03
Time [sec]
0.04
0.05
0.06

1.8
Blair data
ABB LW+FCT
Pressure [bar]
1.6
1.4
1.2
1
0.8
0.6
0.01
0.015
0.02
0.025
Time [sec]
0.03
0.035
0.04

1.8
Blair data
Blair sim.
ABB LW+FCT
Pressure [bar]
1.6
1.4
1.2
1
0.8
0.6
0.01
0.015
0.02
0.025
Time [sec]
0.03
0.035
0.04
Fig. 4 Simulation results vs. data of [37]

3.3 Tapered pipes
The assumption of one-dimensional flow in pipes reaches its limit when applied
to tapered (conical) manifolds. Depending upon the angle of the taper, it produces
conservation errors, but can also lead to a total breakdown of the calculation. Reducing
the mesh size helps in any case, but does not cure the problem.
Blair et al. [49] studied the problem experimentally and compared four prediction methods with the data obtained. While almost all predictions were satisfactory at
lower velocities and/or small taper angles, this was no more the case when the latter
conditions were less well met. However, Blair et al. have probably not used the
modifications to the LW algorithm suggested by Corbern and Gscon [24], which
improve the properties of the LW algorithm (both papers appeared at approximately
the same time).
There has been a revived interest in the unsteady flow calculation in tapered
ducts, especially in conjunction with the CE/SE method. The main reason is apparently
753
Dobrivoje Ninkovic
not an increased calculation accuracy of this method, but to improve it in order to profit
from the higher computational efficiency of the CE/SE method in comparison with the
traditional schemes [22][50].
3.4 Boundary conditions
The subject of boundary element modelling in unsteady flow simulation is a
vast one, and thus lies outside of the scope of the present paper. However, since it
influences markedly the accuracy of the former, it shall be here briefly commented
upon, especially as regards the points where there are adverse effects upon the accuracy of the wave action prediction.
Generally, one can not be satisfied with the situation regarding the modelling of
boundary conditions in unsteady flow simulation. The subject was briefly discussed in
[1], and in much more detail in the references quoted therein. It is a point of fact that
major proportion of wave action calculations are still carried out by using the method of
characteristics, as developed by Jenny [9], and subsequently expanded by Benson
[16]. In doing so, the boundary is modelled as an adiabatic flow element, and the mass
flow rate is calculated by means of the discharge coefficient model. The latter is,
however, not a physical model, but merely a calculation device, which can not be
brought in connection with the contemporary reasoning about flow losses, such as e.g.
the concept of lost work [51]. The boundary condition complex has been extensively
treated by Blair both theoretically and experimentally, especially in connection with
calculating the engine valve and port flow [52][14].
While some elements encountered in the turbocharging simulation practice can
be assumed to behave adiabatically (e.g. throttle and bypass valves), there are others
that are definitely diabatic. Engine valves are a case in point; and the turbocharging
turbine, while being almost adiabatic, exchanges work across its boundary. Apart from
the work of Benson [17], who implemented a particular turbine model within the
framework of the method of characteristics, the present author is not aware of publicly
available boundary condition models that implement the steady flow energy equation in
full or in part. Furthermore, the pipe inflow model of Blair [14] seems to be the only one
that inherently considers the pressure increase downstream of the vena contracta in a
flow restriction. The present framework also precludes the use of other loss
formulations, such as Fanno or Rayleigh flow, and thus prevents using the large body
of empirical knowledge on flow losses acquired in the field of hydraulics, e.g. [53].
4.
CONCLUSIONS
The field of one-dimensional wave action simulation in the manifolds of IC

engines is a mature one, offerring the practitioner a choice of well tested and reliable
methods. The accuracy of predicting the propagation of large-amplitude waves is of
special importance in the design and optimization of turbocharged systems.
The algorithms encountered in the contemporary commercial and proprietary
simulation packages are mostly the descendants of the Lax-Wendroff method in
various forms. The deficiencies inherent in the basic scheme (numerical diffusion and
dispersion) are dealt with by using either the flux-corrected transport (FCT), or one of
the TVD formulations. The latter method is superior to FCT as regards variable gas
properties and tapered pipes.
754
Combustion Engines
Newer schemes, such as e.g. the CE/SE, are implemented in a couple of
simulation suites used by research groups at Universities. Although the CE/SE method
was very promising at the beginning, there was the need to improve its performance
with tapered pipes.
Although it has been claimed that the one-dimensional methods were to be
replaced by the two- or three-dimensional schemes, it is shown in the paper that they
still constitute the backbone of the contemporary simulation practice and are capable
of being further improved.
LITERATURE
[1] Ninkovic, D. (2008). Large Engine Turbocharging Simulation: The Role, State of
the Art, and Development Trends. MVM2008, Paper MVM2008IL03.
[2] Ninkovic, D. et al. (2006). The ABB Turbo Systems Program Suite for the Design
and Optimisation of Turbocharging Systems. MVM2006, Paper MVM20060088.
[3] Bannister, F.K., Mucklow, G.F. (1948). Wave Action Following Sudden Release of
Compressed Gas From a Cylinder. Proc. I. Mech. E., Vol. 159, p. 269-287.
[4] Blair, G.P. (1991). An Alternative Method for the Prediction of Unsteady Gas Flow
Through the Internal Combustion Engines. SAE Paper 911850.
[5] Mayer-Schuchard, C. (1936). Schwingungen von Luftsulen mit groer Amplitude.
Forschungsheft 376, Beilage zu Forschung auf dem Gebiete des
Ingenieurwesens. Ausg. B, Bd. 7.
[6] Lettau, E. (1939). Messungen an Gasschwingungen grosser Amplitude. Deutsche
Kraftfahrtforschung, 39, p. 1ff.
[7] Sauer, R. (1942). Charakteristikenverfahren fr die eindimensionale instationre
Gasstrmung. Ingenieur Archiv, Vol. 13, No. 2, pp. 79-89.
[8] De Haller, P. (1945). On a Graphical Method in the Gas Dynamics. Sulzer
Technical Review, No. 1, Winterthur, pp. 6-24.
[9] Jenny, E. (1949). Berechnungen und Modellversuche ber Druckwellen grosser
Amplituden in Auspuff-Leitungen. Dissertation, ETH Zrich.
[10] Lax, P.D. (1954). Weak Solutions of Nonlinear Hyperbolic Equations and Their
Numerical Computation. Communications on Pure and Applied Mathematics, Vol.
7, pp. 159-193.
[11] Earnshaw, S. (1860). On the Mathematical Theory of Sound. Phil. Trans. Roy.
Soc., London, Vol. 150, pp. 133-148.
[12] Wallace, F.J., Mitchell, R.W.S. (1953). Wave Action Following the Sudden Release
of Air Through an Engine Port System. Proc. I. Mech. E., Vol. 1B, pp. 343-356.
[13] Riemann, B. (1858). ber die Fortpflanzung ebener Luftwellen von endlicher
Schwingungsweite. Gott. Abh. (Math.), 8, pp. 43-65.
[14] Blair, G. P. (1996). Design and Simulation of Two-Stroke Engines. SAE
International, Warrendale, PA, U.S.A.
[15] Gibbes, G., Hong, G. (2010). Improving the Accuracy of a 1D Gas Dynamics
th
Model. Proceedings 17 Australasian Fluid Mechanics Conference.
[16] Benson, R.S., Garg, R.D., Woolatt, D. (1964). A Numerical Solution of Unsteady
Flow Problems. Int. J. of Mech. Sciences, 6, pp. 117-144.
[17] Benson, R.S. (1982).The Thermodynamics and Gas Dynamics of InternalCombustion Engines. Edited by J.H. Horlock and D.E. Winterbone, Clarendon
Press, Oxford.
755
Dobrivoje Ninkovic
[18] Winterbone D.E., Pearson R.J. (2000). Theory of Engine Manifold Design: Wave
Action Methods for IC Engines. Professional Engineering Publishing, London.
[19] Winterbone D.E., Pearson R.J. (1999). Design Techniques for Engine Manifolds:
Wave Action Methods for IC Engines. Professional Engineering Publishing,
London.
[20] Depcik, C., van Leer, B., Assanis, D. (2005). The Numerical Simulation of VariableProperty Reacting-Gas Dynamics: New Insights and Validation. Num. Heat
Transfer, Part A, Vol. 47, pp. 27 56.
[21] Serrano, J.R., Arnau, F.J., Piqueras, P., Onorati, A., Montenegro, G. (2009). 1-D
Gas Dynamic Modelling of Mass Conservation in Engine Duct Systems With
Thermal Contact Discontinuities. Mathematical and Computer Modelling, Vol. 49,
pp. 1078-1088.
[22] Torregrosa, A.J., Serrano, J.R., Arnau, F.J., Romero, J.V. (2008). Experimental
Validation of a New Semi-Implicit CE-SE Scheme for the Calculation of Unsteady
One-Dimensional Flow in Tapered Ducts. Int. J. Num. Meth. Engng, Vol. 74, pp.
1473-1494.
[23] Bulaty, T., Niessner, H. (1985). Calculation of 1-D Unsteady Flow in Pipe Systems
of IC Engines. Journal of Fluids Engineering, Vol. 107, pp. 407-412.
[24] Corbern, J.M., Gscon, LL. (1995). A New Method to Calculate Unsteady 1-D
Compressible Flow in Pipes With Variable Cross Section. Application to the
Calculation of the Flow in Intake and Exhaust Pipes of IC Engines. Transactions
ASME ICE, Vol. 23, pp. 77-87.
[25] Lax, P., Wendroff, B. (1960). Systems of Conservation Laws. Comm. Pure &
Applied Math., Vol. 13, pp. 217-237.
[26] Richtmyer, R.D., Morton, K.W. (1967). Difference Methods for Initial Value
Problems. Wiley-Interscience, New York.
[27] MacCormack, R.W. (1969). The Effect of Viscosity in Hypervelocity and Impact
Cratering. AIAA Paper 69-354.
[28] Bulaty, T., Niessner, H. (1998). Internal documentation. ABB Turbo Systems.
[29] Boris, J.P., Book, D.L. (1973). Flux-Corrected Transport, I. SHASTA A Fluid
Transport Algorith That Works. J. Comp. Phys., Vol. 11, pp. 38-69.
[30] Niessner, H., Bulaty, T. (1981). A Family of Flux-Correction Methods to Avoid
Overshoot Occurring With Solutions of Unsteady Flow Problems. Proc. of the
Fourth GAMM Conf. on Numerical Methods and Fluid Mechanics, pp. 241-250.
[31] Davis, S.F. (1987). A Simplified TVD Finite-Difference Scheme via Artificial
Viscosity. SIAM J. Sci. Stat. Comput., Vol. 8, pp. 1-18.
[32] Depcik, C.D. (2003). Modeling Reacting Gases and Aftertreatment Devices for
Internal Combustion Engines. Ph.D. Dissertation, Univ. of Michigan.
[33] Onorati, A., Perotti, M., Rebay, S. (1997). Modelling One-Dimensional Unsteady
Flow in Ducts: Symmetric Finite Difference Schemes Versus Gelerkin Discontinuous Finite Element Methods. Int. J. Mech. Sci., Vol.39, pp. 1213-1236.
[34] Onorati, A., Ferrari, G. (1998). Modeling of 1-D Unstready Flow in IC Engine Pipe
Systems: Numerical Methods and Transport of Chemical Species. SAE 980782.
[35] Morel, T., Keribar, R., Blumberg, P.N. (1988). A New Approach to Integrating
Engine Performance and Component Design Analysis Through Simulation. SAE
880131.
756
Combustion Engines
[36] Kirkpatrick, S.J., Blair, G.P., Fleck, R., McMullan, R.K. (1994). Experimental
Evaluation of 1-D Computer Codes for the Simulation of Unsteady Gas Flow
Through Engines A First Phase. SAE 941685.
[37] Blair, G.P., Kirkpatrick, S.J., Fleck, R. (1995). Experimental Validation of a 1D
Modelling Code for a Pipe Containing Gas of Varying Properties. SAE 950275.
[38] Chang, S.C., To, W.M. (1991). A New Numerical Framework for Solving Conservation Laws - The Method of Space-Time Conservation Element and Solution
Element. NASA TM 104495.
[39] Briz, G., Giannattasio, P. (1993). Applicazione dello schema numeric Conservation
Element - Solution Element al calcolo del flusso instazionario nei condotti dei
th
motori a c.i.. Proc. 48 ATI National Congress, pp. 233-247.
[40] http://www.grc.nasa.gov/WWW/microbus/, accessed on 20.04.2011.
[41] Niessner, H., Bulaty, T. (1986. Improving Conservation of Mass, Momentum, and
Energy by Appropriate Flux Correction at the Boundaries. ZAMM, 66, pp. 250-274.
[42] Bulaty, T., Codan, E., Skopil, M. (1996). A Flexible System for the Simulation of
Diesel Turbocharged Engines and Turbocharging Systems. Proc. ASME ICE, Vol.
26-3, pp. 57-63.
[43] Harten, A. (1977). The Artificial Compression Method for Computation of Shocks
and Contact Discontinuities: I. Single Conservation Laws. Comm. Pure Appl.
Math., Vol. 30, pp. 611-638.
[44] McBride, B.J., Zehe, M.J., Gordon, S. (2002). NASA Glenn Coefficients for Calculating Thermodynamic Properties of Individual Species. NASA/TP-2002-211556.
[45] Bcker, D., Span, R., Wagner, W. (2003). Thermodynamic Property Models for
Moist Air and Combustion Gases. Trans. ASME, Journal of Engineering for Gas
Turbines and Power, Vol. 125, pp. 374-384.
[46] Ninkovic, D. (2006). Improving the Accuracy of the Reciprocating Compressor
Performance Prediction by Considering Real Gas Flow in the Valve Chambers and
Associated Piping. Proc. IMechE Conf. on Fluid Machinery for the Oil, Petrochemical and Related Industries, pp. 177-186.
[47] Pearson, R.J., Winterbone, D.E. (1997). The simulation of Gas Dynamics in
Engine Manifolds Using Non-Linear Symmetric Difference Schemes. Proc. Instn.
Mech. Eng., Vol. 211, Part C, pp. 601-616
[48] Pearson, R.J., Winterbone, D.E. (1993). Calculating the Effects of Variations in
Composition on Wave Propagation in Gases. Int. J. Mech. Sci., 35, pp. 517-537.
[49] Blair, G.P., Kirkpatrick, S.J., Mackey, D.O., Fleck, R. (1995). Experimental Validation of a 1-D Modelling Codes for a Pipe System Containing Area Discontinuities.
SAE 950276.
[50] Payri, F., Galindo, J., Serrano, J.R., Arnau, F.J. (2004). Analysis of Numerical
Methods to Solve One-Dimensional Fluid-Dynamic Governing Equations Under
Impulsive Flow in Tapered Ducts. Int. J. Mech. Sci., Vol. 46, pp. 981-1004.
[51] Greitzer, E.M., Tan, C.S., Graf, M.B. (2004). Internal Flow: Concepts and Applications. Cambridge University Press.
[52] Blair, G.P. et al. (1998). Some Fundamental Aspects of the Discharge Coefficients
of Cylinder Porting and Ducting Restrictions. SAE 980764.
[53] Idelchik, L.E. (2005). Handbook of Hydraulic Resistance. Jaico Publishing House.
757
TEST BENCH FOR NON-STANDARD MEASUREMENT

CHARACTERISTICS OF RECIPROCATING COMPRESSOR
Dragan Taranovic1, Radivoje Pesic2, Jovanka Lukic3, Aleksandar Davinic4
Summary: A dynamic test of the reciprocating compressor requires non-standard
measurement of input parameters (pressure, temperature, flow, and humidity), output
parameters (pressure, mass of compressed gas) and process parameters (pressure in
the cylinder, piston position) of reciprocating compressors. For non-standard testing is
necessary to establish a special test bench. Non-standard measurements required
automatic adjustment of power unit and actuators parameters in the test bench to
enable the change of compressor working point and to control input and output
compressor parameters in order to be constant during a measurement time. The paper
describes the test bench and some results of non-standard measurement
characteristics of reciprocating compressor. Presented solution enables the automation
of the test process through setting of the operating regimes, control of operation
conditions, measurements, and data processing.
Key words: Test bench, Reciprocating compressor, Measurement
1. INTRODUCTION
Testing of all produced parts is essential to achieve the optimal product with
respect to ecology, economy and their utility and exploitation characteristics. In order to
perform the necessary tests, test bench is necessary to conduct measure
characteristics of these parts. For each part testing, it is necessary to develop a
specific measurement installation. Depending on the testing type, test benches can be
classified into: standardized test benches where typical, standard parameters of
produced parts are tested and non-standard test benches where testing parameters
and characteristics are not covered by standard tests. Non standard tests are
necessary in the product development phase and in post production phase when it is
necessary to improve parts characteristics and to monitor parameter changes in use.
1
Dragan Taranovi, M.Sc., teaching assistant, Faculty of Mechanical Engineering from

Kragujevac, Serbia, e-mail: tara@kg.ac.rs
2
Radivoje Pei, Ph.D., full professor, Faculty of Mechanical Engineering from Kragujevac,
Serbia, e-mail: pesicr@kg.ac.rs
3
Jovanka Luki, Ph.D., assoc. professor, Faculty of Mechanical Engineering from Kragujevac,
Serbia, e-mail: lukicj@kg.ac.rs
4
Aleksandar Davini, M.Sc. teaching assistant, Faculty of Mechanical Engineering from
Kragujevac, Serbia, e-mail: davinic@kg.ac.rs
Kragujevac,
Kragujevac,
Kragujevac,
Kragujevac,
759
D. Taranovic, R. Pesic, J. Lukic, A. Davinic
Standard test of reciprocating compressors are used mainly for inspection of

nominal parameters in stationary conditions: the maximum pressure of compressed air,
the maximum amount of compressed air and energy efficiency. Standardized test
bench is defined for measuring these characteristics.
In order to determine the characteristics of reciprocating compressors in the
development stage or in order to improve characteristics of the existing compressors it
is necessary to measure characteristics that are not included in standard tests.
Non-standard testing is performed at a specific test bench which enables
measurement of stationary and dynamic parameters of compressors. Test bench
contains control unit which maintains constant test conditions and controls dynamic
changes of the test conditions.
At Faculty of Mechanical Engineering in Kragujevac was a need to develop
non-standard test bench for specific testing of reciprocating compressors of vehicle air
brake systems. Development of the test bench for non-standard parameters
measurements of reciprocating compressors and other testing machines, devices and
circuits with similar principles of working (other types of compressors and IC engines...)
is conducted.
Test bench is designed for:
standard, control and research tests of air compressors,
research of compressor's working process,
tribological tests of compressors and IC engines,
determination of mechanical losses in compressors and IC engines,
comparative tests of lubricants for compressors and IC engines,
determination of fluid characteristics in the air-conditioning systems...
2. OVERVIEW OF REALIZED TEST BENCHES
Testing of reciprocating compressors in development phase requires a specific,
non-standard test bench which is realized variously in different research institutions.
Diversity is reflected in the type of power unit and measuring and control elements of
test bench. Measuring elements are chosen depending on needs for stationary or
dynamic parameters of compressors. Control elements in the measuring installations
are rare, although they are necessary to maintain constant test parameter or to control
dynamic changes of test conditions.
Test benches for compressors testing are consist of standard measuring
components. Analysis of individual components choice and their influence on the
measurement results are not performed. Inappropriate element of the test bench
installation can corrupt the measuring characteristics of the installation while the
element with oversized characteristics can raise the test bench costs.
Standardized measurement tests are performed on the test benches which are
described in the literature [1,2]. Typical standards for compressors applied in motor
vehicles are ECE 13 and ISO 1217. For standard tests there are commercially realized
test benches (at example, American company Climatic Testing Systems [3]). Different
test benches for non-standard compressor tests are described in literature. They have
different characteristics: different power units with different power and various
transducers in the measuring installation. The power unit can be a diesel engine [4],
electric motor [5,6], gas turbine [7] ...
760
Test bench for non-standard measurement characteristics of reciprocating compressor
There are small number of automated measuring installation for measuring

compressors characteristics. Automated measurements installations for measuring
compressors with a PID controller, which was conducted by Y. He in the China, [7],
enabled more precise measurements by reduction of the time and number of involved
researchers in the measurement realization. Test bench for non-standard
measurement characteristics of compressors used in refrigeration systems was
realized by B. Wang and associates at Tsinghua University in the Beijing, based on
their own mathematical model [8]. Measuring installation for the testing of compressors
with data acquisition system, but without automatic control of test bench, was realized
by G. Raush with colleagues at the Technical University of Catalonia [9].
3. TEST BENCH REALIZATION
Installation is primarily aimed for testing of performances of vehicle braking
system's compressors and of small, general purpose compressors. Typical
characteristics of vehicle braking system's compressor are presented in Fig. 1 and 2.
These data are reference for dimensioning of compressors test bench.
3
Power, kW
8 bar
7 bar
2.5
5 bar
350
Flow of air, dm3/min

5 bar
300
7 bar
8 bar
250
200
1.5
150
100
0.5
0
400
50
800
1200 1600 2000 2400 2800 3200

speed, rpm
Fig. 1 Compressor driving characteristics
0
400
800
1200 1600 2000 2400 2800 3200

speed, rpm
Fig. 2 Compressor production

characteristics
Test bench provides: measuring of the compressor parameters, control of

measurement installation and processing of measurement data. The test bench
consists of units given in the block diagram in Fig. 3. Measuring signals are marked by
red line and control signals are marked with blue line. Schematic diagram of the test
bench is shown on the Fig. 4
The power unit provides controlling of propulsion power of measuring
installations by application of three phase electric motor (power 4 kW). Engine speed is
regulated by frequency converters. Speed and torque submitted through output shaft of
motor to compressor is continuously measured. Pulse load of electric motor caused by
process in the compressor, is reduced by application of flywheel on the output shaft.
Load of compressor is regulated by servo valve in load compressor unit, which
controls the compressed air flow. Flow, temperature and humidity of compressed air
are measured by appropriate transducers. Pulsation of compressed air on the
compressor outlet is compensated by using the appropriate tank.
761
CONTROL & MESAUREMENT & DATA PROCESSING
COMPRESSOR
LOAD
POWER UNIT
WORKING
CONDITIONS
COMPRESSOR
Fig. 3 Test bench block diagram

2
4
10
11
1
3
12
1. Stand; 2. Electric motor; 3. Electric cabinet; 4. RPM transducer;

5. Torque transducer; 6. Elastic clutch; 7. Flywheel; 8. Flange; 9. Elastic clutch;
10. Compressor; 11. Crankshaft angle indicator; 12. Adjustable stand
Fig. 4 Test bench with power unit/compressor assemble

Working condition unit provides compressors working conditions similar to
operational conditions. In this unit there are three sections: section for circulation of
lubricating oil through the compressor, section for temperature control of cooling fluids
and section for air conditioning compressors. Depending on, the type of compressors
or testing device, there is no need to use all three sections of this unit. For example,
the section for circulation of lubricating oil used in compressors which have a common
lubricating system with internal combustion engine and engine is not subject of
research.
The work of the reciprocating compressors is not controlled. Measurement of
the pressure in the cylinder depends on the angular position of the crankshaft which is
measured with an accuracy of 1.
Control & Measurement & Data Processing Unit is based on the USB data
acquisition system from National Instruments and LabVIEW software. This unit fully
controls the entire of test bench. At the same time is managing power unit in order to
obtain preferred speed and compressor load unit for obtaining preferred air flow and to
maintain preferred operating conditions.
Electric power part is in electric cabinet and can be use for restricted control in
manual operation of test bench.
762
Test bench for non-standard measurement characteristics of reciprocating compressor
Measuring signals from all subsystems are acquired and saved in the computer
memory in order to be processed and analyzed by appropriate tables and graphics.
The same measuring signals are used in the control area of measuring installation.
Photo of the test bench part is given in the Fig. 5.
Fig. 5 Photo of test bench

4. EXAMPLE OF TEST BENCH APPLICATION
Realized test bench allows research of the new constructions and
technological solutions to the certain parts of engines. Its basic advantage is in the
realistic conditions of research constructions, materials, technologies and lubricants
which give special value in the R&D testing.
The main instruction with tribological optimization orders the transfer of all
bordering cases in hydrodynamic lubrication. In accordance with this request the new
solution piston with tribo-pads is shown in Fig.6.
Fig. 6 Piston with tribo-pads
Fig. 7 Effect of a tribological pads on mechanical

friction
Results of tested solutions with tribo-pads is presented in Fig. 7. The increase

of normal forces (from compression) vs. the number of rpm shows that mechanical
losses are lower 5% then with the classical piston made by aluminum and cylinder
made by grey cast.
763
5. CONCLUSION
Standard and research tests of reciprocating compressor demand
corresponding test bench. Basic criteria for its design are contained in technical
conditions of standards or are defined based on non-standard programs of research.
Thanks to the principle, construction and technological similarity between
reciprocating compressors and IC engines, it is possible to apply unique test methods
and equipment.
Contemporary test bench has automated management of the test process,
data acquisition and data processing.
Presented solution of test bench for testing of reciprocating compressors of air
brake systems satisfies all conditions listed in standards and enables numerous nonstandard and research tests.
ACKNOWLEDGMENTS
The paper is the result of the research within the project TR34051 financed by
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
764
. ., . ., . . (1964).
, , .
Davini A., Taranovi D., Pei R. (2004) Installation for air compressors testing
and tribological tests of IC engines, Mobility & Vehicle Mechanics - International
Journal for Vehicle Mechanics, Engines and Transportation Systems, vol. 30,
Special Edition, p. 151-165.
Climatic
Testing
Systems.
Compressor
Test
bench.
from
http://www.climatictesting.com/PDF Files/CTB_DS_3.pdf, accessed on 2011.,
Galindo J., Serrano J.R., Guardiola C., Cervell C. (2006), Surge limit definition in
a specific test bench for the characterization of automotive turbochargers,
Experimental Thermal and Fluid Science, vol. 30, no. 5, p. 449-462,
Cuevas C., Winandy E., Lebrun J. (2008), Testing and modeling of an
automotive wobble plate compressor, International Journal of Refrigeration, vol.
31, no. 3, p. 423-431,
Lebrun J., Cuevas C. (2009), Testing and modeling of a variable speed scroll
compressor, Applied Thermal Engineering, vol. 29, no. 2-3, Pages 469-478,
He Y. (2010), Proportional-Integral-Differential (PLC) Control System Design of
Compressor Performance Test Bench, 2010 International Conference on
Electrical and Control Engineering, p. 1511-1514,
Wang B., Li X., Shi W., Yan Q. (2007), Design of experimental bench and
internal pressure measurement of scroll compressor with refrigerant injection,
International Journal of Refrigeration, vol. 30, no. 1, p. 179-186.
Raush G., Rigola J., Prez-Segarra D., Oliva A. (2004), A novel Sigma-Delta
ADC application oriented to test hermetic reciprocating compressors,
Measurement Science and Technology, vol. 15, no. 11, p. 2207-2214
THE RESEARCH OF CORRELATIONS BETWEEN MOTORCYCLE

OSCILLATORY PROCESSES DURING THE NONSTEADY MODES
OF MOTION
Branislav Aleksandrovi1 Rajko Radonji2, Marko apan3, Aleksandra Jankovi4
Summary: The issues of motorcycle motion safety in highway traffic have been
presented in the paper itself. The influence of oscillations during the nonsteady modes
of motion upon the system behaviour has been emphasized. The appropriate models
have been formed, in a sense of supporting the projection of experiments, conducting
the research and the analysis of obtained results. Correlative dependencies have been
defined, the ones between the level indicators, time history and the power of relevant
signals. The illustrative results have been presented.
Key words: motorcycle, motion safety, oscillations, experiment
1. INTRODUCTION
The specific constructive and dynamic characteristics of a motorcycle, as a
single track vehicle, has a considerable influence upon its behavior as regards traffic
along with the risk of occurrence of a traffic accident. Statistic data referring to traffic
accidents confirm the fact that a motorcycle is a very often cause of traffic accidents.
The results of statistical analysis of traffic accidents, which are obtained in one Serbian
city, and displayed in figure 1a, [1], refer to an increase of the number of traffic
accidents caused by a motorcycle, during the observed period lasting from 2005 to
2008. This trend is followed by an increase of material costs, figure 1b, and the traffic
accidents are divided according to the degree of an injury inflicted to the human factor:
those who are seriously injured, those whose injuries are not severe, those who get
killed. What should be taken into consideration when it comes to these analyses is the
increased number of motorcycles in traffic, the ones belonging to various categories,
with different technical conditions, connected to mostly younger drivers, and what
should also be kept in mind is inadequate training and disobeying regulations. The new
legal highway traffic regulation, which has already become effective in Serbia, is
supposed to contribute, among other things, to the increase of safety of a motorcycle
rider behavior as a traffic participant, [2].
1
MSc Branislav Aleksandrovi, Kragujevac, Faculty of Mechanical Engineering (bane1510@yahoo.com)

Full prof. Rajko Radonji PhD, Kragujevac, Faculty of Mechanical Engineering (rradonjic@kg.ac.rs)
Marko apan, PhD student, Kragujevac, Faculty of Mechanical Engineering (djapan@kg.ac.rs)
4
Full prof. Aleksandra Jankovi PhD, Kragujevac, Faculty of Mechanical Engineering (alex@kg.ac.rs)
2
3
765
Branislav Aleksandrovi, Rajko Radonji, Marko apan, Aleksandra Jankovi
2. PAPERS GOALS
In our previous research, [3], [4], [5], [6], the following problems have been
dealt with: modeling and motorcycle dynamics simulation, experimental research of
dynamic stress of its frame, identification of maneuverability characteristics,
experimental research frame vibrations during the process of starting and braking,
respectively. Considering the topic of this paper, what should be emphasized here is
some specific qualities of a motorcycle as a single track vehiclle compared to double
track vehicles and the ones with more than two tracks, in accordance with
presentations and indications displayed in figure 2a and 2b.
25
24
20
24
3500000
3500000
3000000
20
2500000
10
15
12
2000000
10
1110000
712500
1500000
10
1000000
5
1
0
2005
41220
500000
1
2006
b3
2005
b1
2008
a)
12
21
22
b2
1
2007
16
2006
33
49
2007
52
33
b
a
2008
b)
Fig. 1 A presentation of the number of traffic accidents

(a) along with material damage (b) based upon one city in Serbia, a vehicle cause,
motorcycle, period 2005 2008, [1]
The support structure of a motorcycle provides an open space for a driver and
co driver, figure 2a, in distinction from the closed space of a passenger car, figure 2b,
the common indication, 1. It means that, in case of a traffic accident, such as
motorcycle overturning, one cannot speak of preserving the integrity of its internal
space.
Masses of a driver and motorcycle represents the values of the same kind in
distinction from a passenger car. This is the reason why each alteration of a drivers
mass has a significant influence upon its dynamic characteristics and the behavior of
the entire system driver motorcycle, indication 2, in figure 2a and 2b.
Unlike a double track vehicle with elastic pneumatic tyres, which does not
possess its own stability as regards course keeping without the driver's assistance,
indication 3 NPr, in Picture 2b, motorcycle as a single track vehicle, does not possess
not only the above mentioned stability but the stability of keeping a vertical, uprigt
position as well, indication 3 NPr and 3 NPo, respectively in figure 2a, without the
driver's assistance. This is one of the reasons why the role of a motorcycle driver is
very complex and difficult.
Owing to its very specific construction, contour and contact environmental
conditions, a motorcycle is more sensitive to excitations due to road unevenes's
indications 4 znz, , as well as to aero effects, indications 4 Fap, MapFa, Ma, than a
double track vehicle, a comparison of which is given in figure 2a and 2b.
A complex function of a driver maneuvering, his/her position as regards the
766
The Research of Correlations Between Motorcycle Oscillatory Processes During the Nonsteady
Modes of Motion
motorcycle and controls, unfavourable contact conditions that entail arms, legs, the
entire body, indications R, N, S, respectively, in figure 2a, lead to the high levels of
his/her mental and physical load. Therefore, a motorcycle ride is a typical example of
exposure of a human operator to a complex influence of vibrations, which carries a
high risk not only when the aspect of his/her own health is taken into account but when
the aspect of traffic safety is considered as well.
In accordance with the above mentioned annotations, it appears that a driver,
as a human operator, has to be observed as a complex physico biological system
during the ride. Taking into consideration the physical aspect, the anatomy of a driver's
body is observed in relation to the presentations of adequate oscillatory systems with
masses connected by elastic and damped equivalents. The biological aspect of a
driver is also complex and it involves psychophysical effects. Therefore, every
approach to the research of possibilities leading to an increase of the level of safety of
the system driver motorcycle, has to be based on specific characteristics of a driver's
behavior, motorcycle technical characteristics, potential influences of the environment
and their mutual interaction.
a)
b)
Fig. 2 A comparison of technical properties of a single track (a) and double track (b)
vehicle
Realizing the significance of the emphasized issues, what we have presented
in this paper is one approach to the identification and evaluation of oscillatory
processes during the nonsteady modes of motorcycle motion.
3. BASIC MODEL, EXPERIMENTAL SYSTEM AND RESULTS
Typical nonsteady modes of motion are accelerating and braking, which are
treated as transient time processes. On the basis of the presentation of the physical
model of a motorcycle and introduced indications, given in figuure 2a, it is possible to
form adequate mathematical models for a description of motion in a longitudinal plane,
for the two observed cases, acceleration and braking, (1), (2):
X2
F0 R f 2
R f 1 Rv R j o F0
K
fGu 0.5C x UAv 2 Gmu (dv / dt ) d MZ (1)
K 1 K 2 | G ' mu ( dv / dt ) d M d Gu
(2)
767
where:
F0 traction force,
X2 corresponding tangential reaction,
Rf=Rf1+Rf2 =fG total rolling resistance,
Rv aerodynamic drag,
Rj inertia force produced from the motorcycle acceleration,
Gu =mug gross weight,
Z vertical reaction on the drive wheel,
K- total braking force,
Cx, , A indicators of motorcycle aerodynamics,
, coefficients of rotary mass participation,
, d the boundary values of the coefficient of adhesion of the pneumatic tyre
road,
(dv/dt) acceleration,
(-dv/dt) deceleration.
a)
b)
Fig. 3 Models for identification of motorcycle oscillatory processes (a) and measuring
system (b)
Basic model for a motorcycle acceleration (1), indicates the expected
acceleration flow which represents the function of the engine speed characteristic,
Me=f1(ne), F0=f2(v) and the vehicle resistance alteration law, R=Rf+Rv=f3(v). On the
other hand, the deceleration flow observed during the braking process, the model (2),
has been defined by means of the realized braking force, on one wheel or both of
them, that is, due to the transmission characteristics of the braking system, the mode
of operation of a driver, that is, ABS system, depending on the applied conception.
The experimental system intended for the research of a motorcycle's oscillatory
processes has been formed on the basis of the above mentioned theses and analyses.
In our previous papers, we have developed and presented the two typical experimental
systems [5], [6]. The first one is based upon the use of a compatible system, oscillation
sensors 3+1 analyzer. The second is conceived for the research of complex
motorcycle dynamics, with a significant increase in the number of sensors and
measuring channels, according to the relation, Nlog x 8. Therefore, the number of
measuring points is obtained as the product of the number of available loggers, Nlog,
multiplied by 8. The capacity of the first measurement system with 3+1 measuring
points, presented in figure 3c, has been enough for the requirements of this particular
research.
Experiments have been conducted on the straightline horizontal sections of an
asphalt road which is in good condition, by combined modes of motion, accelerating
768
The Research of Correlations Between Motorcycle Oscillatory Processes During the Nonsteady
Modes of Motion
constant speed motion braking. At the same time what has been varied is the
following: accelerating intensity final velocity, starting velocity braking intensity, until
the moment of reaching stable boundary modes of motion.The measurement signals of
acceleration, that is, deceleration in the characteristic system points have been
registered during the experiments, and this has been done by combining the
measuring points for the purpose of a simultaneous measurement of the three normal
components, including the measuring points with one measuring component one
measuring direction. Illustrative examples of obtained results are shown in figure 4,
according to the following sequence: peak and mean square values of longitudinal
acceleration deceleration, evaluated in the frequent domain based on the aspect of a
drivers influence, expressed within the time domain with the sliding time interval, for
the combined mode of motion a), for braking b). Then, direct signal relation has been
presented in figure 4c, correlative relation has been given in figure 4d, and partial
correlative signal dependencies for motorcycle accelerating and braking have been
presented in figure 4e, 4f respectively.
Starting signals for the analysis of correlations of motorcycle oscillatory
processes, in figure 4a and 4b, are part of the database which is made of total 54 data
file which are obtained for each of reference motion velocity during the process of
conducting the experiments. In this particular segment of our research, what has been
identified is the correlation between peak signal values, the upper curve given in figure
4a and 4b, and the mean square values, lower curve presented in these figures,
therefore, the relations between individual signal parameters on the one hand, and
time history with signal energy on the other.
a)
b)
c)
d)
e)
f)
Fig. 4 The experimental results of motorcycle oscillation research:

a), b) measurement signals, c) direct relations, d), e), f) correlations
769
According to figure 4d, 4e and 4f, respectively, y = 0.39x 0.28, Rxz = 0.9369;
y = 0.44x 0.49, Rxy = 0.9228; y = 0.37x 0.19, Rxy = 0.9362, the identified relations
and correlation coefficients show that the relation of the observed signal characteristics
converge the linear trends.
4. CONCLUSION
Increasing number of motorcycles in traffic leads to the increase in risk of traffic
accidents which have serious consequences in most of the cases. Oscillatory
processes and inertia loads during the motorcycle accelerating and braking regimes
has an unfavourable influence upon the system preformances, the driver's behavior,
his/her tiredness and health. As regards the study of unfavourable effects, it is
necessary to choose the appropriate evaluation parameter of these short duration but
very intensive processes. According to the research presented in this paper, the mean
square value of ponderable longitudinal accelerations decelerations, defined in the
sliding time interval of a narrow range, can be efficient evaluation parameter of
measuring signals of the above mentioned motorcycle modes of motion, due to the
high correlation with the characteristics of the original signal.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
770
Jankovi, A., Aleksandrovi, B., Jokovi, N. (2009). Traffic accidents in the area of
the city of Kragujevac in the period 2005 2008. Conference - Traffic accidents,
Proceedings, p. 228 236, Zlatibor, May 21 23.
Official Gazette of the Republic of Serbia, Belgrade. (2009). The road traffic
safety law. The second edition.
Radonji, R., Jankovi, A ., Aleksandrovi, B. (2010). The Study of the Single
Track Vehicles Dynamics. Mobility & Vehicle Mechanics International Journal
for Vehicle Mechanics, Engines and Transportation Systems, vol. 36, no. 3, p.2134.
Aleksandrovi, B., apan, M., Jankovi, A. (2010). Experimental Research of
Dynamic Stresses of Motorcycles Frame. Mobility & Vehicle Mechanics
International Journal for Vehicle Mechanics, Engines and Transportation
Systems, vol. 36, no. 4, p. 21-36.
Radonji, R., Jankovi, A., Aleksandrovi, B. (2011). Identification of the
Motorcycle Steering Properties. Mobility & Vehicle Mechanics International
Journal for Vehicle Mechanics, Engines and Transportation Systems, vol.37, no.
2, p.57-68.
Aleksandrovi, B., apan, M., Radonji, R., Jankovi, A. (2011). Experimental
Research of Motorcycles Frame Vibration during Acceleration and Braking.
JUMV XXIII Science and Motor Vehicle 2011 Automotive Engineering for
Improved Safety, paper AETT09.
Stoffregen, J. (2006). Motorradtechnik. Grunglagen und Konzepte von Motor,
Antrieb und Fahrwerk. ATZ MTZ Wiesbaden.
TRENDS IN DEVELOPMENT OF CATALYTIC CONVERTER OF

INTERNAL COMBUSTION ENGINES (ICE)
Izudin Deli1, Izet Ali2
Summary: In the course of operation of the catalytic converter certain problems arise
which directly affect the efficiency and durability of the catalytic devices. The
researchers made so much effort to find technical solutions that will enable smooth
operation of the catalytic converter. This paper brings an explanation of the two basic
problems in the operation of the catalytic converter and an overview of possible
technical solutions which have resulted from scientific and researching activities. The
proposed technical solutions show direction for further developments of the catalytic
converters.
Key words: catalytic converter, emissions of exhaust gases, cold start
1. INTRODUCTION
In order to reduce emissions of exhaust gases from cars in the EU, the EURO
norms have been introduced, requiring each new car, with the aim of exhaust gases
treatment, to have a built-in catalytic converter as a basic piece of its equipment.
The catalytic converter has the task to convert harmful gases (CO, HC, NOx)
into less harmful gases (CO2, H2O, N2). In order for the catalytic converter to
successfully complete its task it is necessary to provide basic operating conditions
such as:
the ICE equipped with the catalytic converter must be powered by unleaded
petrol,
in order for conversion of exhaust gases to be optimal, a blend of fuel/air
should be regulated in the region = 1,
the catalytic converter has to be heated to the optimal operating
temperature which ranges from 400C and 800C, and
the internal combustion engine (ICE) must operate in a stationed working
area as long as possible.
During operation of internal combustion engine, in exploitation conditions,
operational conditions have changeable nature, which results in an non-stationed
operation of the catalytic converter.
Two main problems in operation of the catalytic converter which directly affects
1
2
Mr. sc. Izudin Deli, Tuzla, Mainski fakultet (izudin.delic@untz.ba)

Dr. sc. Izet Ali, Tuzla, Mainski fakultet, (izet.alic@untz.ba)
771
Izudin Deli, Izet Ali
its efficiency are: cold operational mode of the catalytic converter and uneven
distribution in the flow of exhaust gases through the monolith of the catalytic converter.
In order to solve and mitigate the consequences of non-stationed operation of the
catalytic converter some theoretical and experimental researches have been done. In
this paper we have presented an overview of some researches which affect further
development of the catalytic converter.
2. COLD MODE OF OPERATION OF CATALYTIC CONVERTER
The start of the internal combustion engine generates emission of exhaust
gases. However, the catalytic converter can not immediately carry out conversion of
harmful gases because it is necessary for it to get heated to a certain temperature, at
which catalytic reactions start to appear. So, a certain time is needed for the catalytic
converter to be heated to a temperature of about 400 C at which the efficiency of the
catalytic converter is at a satisfying level. Therefore, in the first few minutes of the
engine combustion, the catalytic converter does not have satisfying efficiency.
Several studies have shown that emission of the exhaust gases from a car
during a cold start of an internal combustion engine is about 60% of the total share of
emission of the exhaust gases. The amount of emission during cold has been very
motivating for further development of catalytic converters. A few researchers have
been researching this problem.
Particular studies have determined the effect of thermal insulation of the
catalytic converter on its operation and its ability to retain heat after an internal
combustion engine has been turned off.
Fig 1. Image of catalytic converter with a system for thermal control

In this study, the vacuum insulation of catalytic converter (Fig. 1) has been
applied which allows retention of the temperature inside the catalytic converter so that
it operates with high efficiency once cold engine is re-started.
Thus, the catalytic converter with a system of controlling the thermal insulation
can keep the temperature of the catalytic converter within the range of operating
temperature for several hours, which allows of internal combustion engine to continue
its operation with smaller intervals of cold start.
772
Trends in development of catalytic converter of internal combustion engines
One of the solutions to the problem of cold mode of operation can be additional
electrical heating of the catalytic converter. Mathematical model of heat transfer from
exhaust gases to monolith in the catalytic converter has been developed through some
researches. It has made a contribution to the development of a new design of electric
heater of the monolith (Fig. 2) and a new approach to electric heating of the catalytic
converter.
Fig. 2 Catalytic converter heated by electric heater

In order to reduce emissions of exhaust gases in cold mode of operation of the
catalytic converter the mass transfer in monoliths based on platinum/ceramic and
accumulation of oxygen in these monoliths have been explored. With the monolithic
structures based on platinum/ceramic, with canals design different from the
conventional ones, the temperature of the beginning of operation of the catalytic
converter is lowered to the limit of about 250 C. In addition, influence of large value of
Reynolds number has a positive impact on the conversion in the catalytic converter.
Certain researches show results of experimental research of direction of fluid
flow which reduces emissions during operation in cold mode. Specifically, an additional
element that directs exhaust gases at the central section of the monolith structure of
the catalytic converter has been used. The result is faster heating of the monolith
because of the flow through the same part of the cross section. The developed system
uses 20% of the active surface of the catalytic converter during starting of the internal
combustion engine.
3. DISTRIBUTION OF THE FLOW OF EXHAUST GASES THROUGH CATALYTIC
CONVERTER
2
Monolith structure of the catalytic converter has more than 60 cells of cm

(Fig. 3) which are coated by catalytic elements and through which exhaust gases flow.
Due to huge number of openings necessary for conducting catalytic reaction, cross
section surface of the catalyst converter monolith is much bigger than surface of cross
section of exhaust pipe. Exhaust gases flow through exhaust pipe to reach monolith
structure. The difference in diameter results in occurrence of uneven filling of cross
section with exhaust gases. The consequence is uneven warming of monolith and
773
development of catalytic reaction with lesser degree of conversion of harmful gases in

less harmful gases as well as destruction of the catalytic converter in the middle cross
section area which is the most exposed to the reaction with exhaust gases.
Fig 3. Monolith structures of different cell densities in cross section

Several researchers have been researching this problem and they have
achieved the results whose implementation enables the catalytic converter to be more
efficient in cold mode operational period and extension of the durability of the catalytic
converter.
The study on the effectiveness of the catalytic converter using infrared
thermography and the heat transfer theories show the results obtained by extensive
experimental researches. Due to the changing load of the internal combustion engine,
filling the cross-section monolith of the catalytic converter with exhaust gases is of a
changeable tendency.
Fig 4.The scheme and diagram show the progressive filling of the section monolith with
increasing mass flow rate
The flow of exhaust gases and its distribution through the catalytic converter
(Fig. 4) is not even. The flow mainly goes through the central part of the monolith of the
catalytic converter. This phenomenon results in accelerated aging of a catalytic
substrate in the central part.
Numerical simulations have been used as tools for a quick identification of the
weak points in the flow of exhaust gases. On the basis of displayed distribution of
temperature (Fig. 5) and concentration in the monoliths of the catalytic converter it can
be concluded that the efficiency of the catalytic converter decreases by increasing
heating time of the monolith structure and that conversion decreases with increase in
the speed of the exhaust gases flow through the monolith structures.
774
Trends in development of catalytic converter of internal combustion engines
Fig. 5 Distribution of temperature in the monoliths of the catalytic converter

The dynamics of the flow of the exhaust gases through the catalytic converter
monolith directly affects the performance of the catalytic converter such as pressure
drop, emission of exhaust gases and durability. In addition, different temperature zones
appear in the catalytic monolith, which cause aging of the monolith in the cross section
that is most exposed to exhaust gases.
Numerical simulations are used to investigate the influence of the position of
the catalytic converter, which must be in the vicinity of internal combustion engine
(ICE) due to the rapid heating of the catalytic converter and achieving operating
temperature. The position of the catalytic converter is not suitable due to the car design
and the space around the engine because of which the flow of exhaust gases into the
catalytic converter is unfavorable. This is particularly the case in the changing load in
operation of the internal combustion engine.
4. CONCLUSION
The catalytic converter has greatly reduced emission of exhaust gases from
the internal combustion engine, but the review of the above stated considerations
shows that al lot of improvements can still be made. Therefore, it is necessary to
continue working on improving the existing solutions and on developing new technical
solutions.
In the future, we can expect the emergence of new more strict norms on
emission of exhaust gases, and thus the greater progression in the development of the
catalytic converter for conversion of exhaust gases from the internal combustion
engine.
LITERATURE
[1] Kirchner, T., Eigenberger, G. (1996). Optimization of the cold-start behaviour of
automotive catalysts using an electrically heated pre-catalyst, Chemical
Engineering Science, vol. 51, no 10, p 2409-2418
[2] Karkanis, A.N., Botsaris, P.N., Sparis, P.D. (2004). A Catalyst Surface Control
Automation System for Emission Reduction During Cold Start, ASME 2004 Internal
Combustion Engine Division Fall Technical Conference ICEF2004
775
[3] Pastor, R.R. (2003). Characterization of the Exhaust Flow in the Catalytic
Converter of a Spark-Ignition Engine Using Infrared Thermography, InfraMation
2003, Infrared Camera Applications Conference
[4] Guojiang, W., Song, T. (2005). CFD simulation of the effect of upstream flow
distribution on the light-off performance of a catalytic converter, Energy Conversion
and Management, vol. 46, no. 13-14, p 2010-2031
[5] Bassem, H.R. (2005). Simulation of an Automotive Catalytic Converter Internal
Flow, ASME 2005 Internal Combustion Engine Division Fall Technical Conference
(ICEF2005)
776
MONITORING OIL FOR LUBRICATION OF TRIBOMECHANICAL ENGINE

ASSEMBLIES
Pantelija Daki1, Sreten Peri2
Summary: Different technical systems require for the appropriate lubricant to be used
at the appropriate place, at the appropriate time and in an appropriate quantity.
Determination of technical systems condition has very important role in development of
friction theory and practice, wear and lubrication. Lubricant is, as the contact element
of the tribomechanical system, carrier of information about the state of the whole
system, from the aspect of tribological and other ageing processes. Due to that,
analysis of oils, based on properly defined program, represents a very effective method
for monitoring the state of technical systems, which ensures early warning signals of
potential problems that could lead to failure and break down of the technical systems.
Using Oil Analysis programs for engine oils has several benefits: reduction of
unscheduled vehicle downtime, improvement of vehicle reliability, help in organizing
effectiveness of maintenance schedules, extension of engine life, optimization of oil
change intervals and reduction of cost of vehicle maintenance.
Key words: monitoring, maintenance, Oil Analysis.
1.
INTRODUCTION
It is not always simple to determine a type of lubricant, frequency of lubrication

and the quantity of the lubricant to be used. Optimal recommendation would be to
follow the specification of technical system manufacturers, experiance, lab research or
professional recommendation of lubricant suppliers. Rationalization of lubricants
consumption represents significant task wich can be obtained by on time oil
replacement. That allows maximal period of use and on the same time appreciable
quantity of lubrication. In regard with primary role of lubricant to reduce the negative
effects of tribological processes related to friction, wear and increase of temperature in
the tribomechanical systems, all types of maintenance include lubrication as a very
important part of the whole procedure. On the other hand, lubricant is, as the contact
element of the system, carrier of information about the state of the whole system, from
the aspect of tribological and other ageing processes. Due to that, analysis of oils,
based on properly defined program, represents a very effective method for monitoring
the state of technical systems, which ensures early warning signals of potential
1
2
Prof. dr Pantelija Daki, Elektroprivreda RS, Banja Luka

Dr Sreten Peri, dipl. in., Faculty of Millitary Sciences, Serbia
777
Pantelija Daki, Sreten Peri
problems that could lead to failure and break down of the technical systems. In the
systems structure, besides the mechanical components, the state is also changed of
the lubricant itself, what leads to loss of lubricating properties.
Analyzing great number of failures of complex tribomechanical systems we can
conclude that there have been certain changes as in system that failed as in lubricant.
Namely, failure of tribomechanical system can appear due to changes in lubricant
characteristics, or changes in lubricant characteristics can appear due to failure of
other elements of tribomechanical system.
Numerous tests and instruments have been developed to help monitor and
diagnose machinery lubrication problems. A great deal has been written about the
development of new techniques for lubrication analysis and enhancing the ability to
determine technical system, lubricant, and contaminant condition from oil.
2.
DIAGNOSIS OF TRIBOMECHANICAL ENGINE ASSEMBLIES
The essence of diagnosis is based on the prognostication (recognition) of

damage and/or failure through the characteristic diagnostic parameters. This allows
prevention of the occurrence of delays, and increase reliability, economy, and usage life.
Diagnosis of tribomechanical system can provide validation, working capacity and the
functionality of the system, and to point out the place, form and cause of failure. Through
the detection of symptoms to determine the value of the characteristic parameters and
their comparison with the limit values it is possible to diagnose the system.
If the engine assemblies are considered from the aspect of tribomechanical
systems (e.g. assemblies piston-piston ring-cylinder, cam-valve lifter, bearing-journal
bearing) defined by tribological processes, it can be shown that the determination of
the content of wear products, content of contaminants, state of lubricants and
lubrication conditions have a significant influence on the implementation of
maintenance of these systems.
We should emphasize the importance of monitoring oil for lubrication of
tribomechanical engine assemblies, which provides that in the early stages of the
functioning of the system identification of potential causes and phenomena that lead to
damage and failure. Prognostication and detection of potential and/or current damage
and failures in the system, checking the functionality of oil and determination of usage
life are the main factors of the implementation of monitoring oil.
Therefore, because the mobile components of tribomechanical system engines
necessarily wear, and that contaminants and wear products are collected in the oil for
lubrication, and because of the need to monitor changes in fluid properties during
exploitation, the conclusion can be drawn that of all the techniques of monitoring the
key to maintain the condition and to achieve certain techno-economic effects is the
monitoring of lubricants.
Analysis of the contents of different metals that are in the lubricant is very
important. Metal particles are abrasive, and act as catalysts in the oxidation of oils. In
motor oils, the origin of the elements may be from the additives, the wear, the fuel, air
and liquid for cooling. Metals from the additives can be Zn, Ca, Ba, or Mg and that
indicates the change of additives. Metals originating from wear are: Fe, Pb, Cu, Cr, Al,
Mn, Ag, Sn, and they point to the increased wear in these systems. Elements
originating from the liquid for cooling are Na and B, and their increased content
778
Monitoring Oil For Lubrication Of Tribomechanical Engine Assemblies
indicates the penetration of cooling liquid in the lubricant. Increased content of Si or

Ca, which originate from the air, points to a malfunction of the air filter.
Wearing of the parts is the main cause of inaccuracies in the process of
exploitation of the mechanical components of vehicles. Characteristic of wear is the
change in shape and dimensions of working area parts. Because of friction the
spending of surfaces occurs which reflects in the increase of moving parts gap and
change their mutual relations, and this resulted in violation of the prescribed
relationship between components, and vehicles in general. Spending of surface
assembly cylinder-piston is shown in the figure 1.
With the increase of tactile area spending, the tightness of engine working
space violates which results in a decrease in the value of effective engine parameters
(power and torque of the engine).
Fig. 1 Tactile-wear surface assemble piston-cylinder:

1 piston ring cylinder, 2 piston ring piston, 3 cylinder piston, 4 piston pinpiston and 5
piston pin connecting rods
More intensive than the piston and cylinder piston rings are worn, because their
work conditions are extremely unfavorable. They are exposed to thermal action and
effect of corrosion which causes the actions of combustion products and high pressure
and at the same time insufficient lubrication of tactile area. Consequences of spending
are increasing gap between the piston ring and the side surface groove piston, and
also reduction of their elasticity. This causes the penetration of oil in the engine
working space, and also the penetration of burnt gases in the engine crankcase.
Due to penetration of combustion products in the crankcase engine the
occurrence of rinsing oil from cylinder walls, increasing the oil temperature,
acceleration of the process of degradation and aging of oil, increase engine
temperature and faster spending cylinder walls become evident.
Increasing gap between piston and cylinder results in intensification of wear
parts of the cylinder and piston group, bearing and journal crankshaft, gear camshaft,
cams camshaft and more.
779
Wearing of valve train parts (figure 2) affects negatively the process of making
changes to the working substances. The violation of the kinematics valve train occurs
which causes strokes in valve train and reducing engine power.
Fig. 2 Areas of valve train that are exposed to wear:

1 valve seat, 2 valve guide, 3 valve followervalve stem, 4 journal valve follower, 5
valve seat valve pushrod, 6 valve lifter valve pushrod, 7 valve lifter valve guide, 8 valve
lifter cam.
Areas that are particularly exposed to wear during contact are shown in the
figure 2. Spending of bearings and journal crankshaft affect the deterioration of the
lubrication regime, the occurrence of stroke in the bearings and engine vibration.
Increased smoking engine due to greater consumption of oil, knocking
(strokes) bearings, spending of assembly cylinder-piston group and valve train are the
main reasons for referral to the engine repair.
Diagnostics of tribomechanical systems in motor vehicles is part of the overall
process of managing maintenance. It provides an opportunity for the user to predict the
damage and/or failure, and thus prevent delay in the work and extend usage life of
motor vehicles.
Dependence of the state of moving parts of the process friction and wear
shows that determination of the look, shape and size of particles of wear products,
lubricants and conditions of the state of lubrication are of vital importance in the
process of maintenance.
The oil for lubrication monitoring during exploitation is one of the most
important diagnostic procedure, which includes state of tribomechanical system, based
on the functions and importance of which should be satisfied in this system. The
advantage of the above mentioned procedure is reflected in the fact that information
about the functionality of components of the system is obtained without stopping and
dismantling the vehicle. Modern trends of diagnosis in recent years, go to the
affirmation of the monitoring of oil, which has resulted in growth of interest of producers
and users of oil. The reasons lie primarily in increasing the reliability, effectiveness,
economy, and recently more and more present protection of the environment.
780
3.
RESULTS OF MOTOR OIL INVESTIGATION DURING EXPLOITATION
In this part the results of experimental testing of motor oil in the Laboratory for
fuels and lubricants VTI Belgrade are presented. The physico-chemical characteristics
of oil in accordance with standard methods are examined, shown in table 1. The
analysis was done on the fresh (new) oils and oils that are used in the motor
assembles of vehicles. Testing of used samples were carried out in accordance with
common criteria defined by the quality of used oil.
Allowable values of deviation limits of individual characteristics of the oil are
conditioned by the type of oil, working conditions and internal recommendations of the
manufacturer of lubricants and users. Limited value characteristics of oils that condition
the change of oil charging from engine are given in table 2. They represent the criteria
for the change of oil charge. Deviation of only one source changes characteristics of oil
charge, no matter of what characteristics is about.
Table 1 Implemented tests and methods for examining the physico-chemical characteristics of oil
Characteristic
Density, gr/cm3
Kinematic viscosity, mm2/s
Viscosity Index
Flash Point (C)
Pour Point (C)
Foaming, ml/ml: 24C; 94C; 24C
Water Content, mas.%
Total Base Number (TBN), mgKOH/g
Insoluble substances in pentane, %
Insoluble substances in benzene, %
Fe Content, %
Cu Content, %
Method
JUS B.H8.015
JUS B.H8.022
JUS B.H8.024
ISO 2592, ASTM D 92
ISO 3016
ASTM D892
ASTM D 95
ASTM D 2896
ASTM D 893
ASTM D 4055
ASS
ASS
Table 2 Allowed values deviation of physico-chemical characteristics of new and used oil
Physical-chemical characteristics
oil and products wear
Viscosity at 40C and 100C, mm2/s
Viscosity Index, %
Total Base Number (TBN), mg KOH/gr
Flash Point, C
Water Content, %
Indissoluble substances in pentane, %
Indissoluble substances in benzene, %
Products wear Content Fe,
ppm( g/gr)
Products wear Content Cu,
ppm( g/gr)
Maximum allowed variation

Motor oil
20%
5%
pad do 50%
20 %
0,2 %
3,5
2,5
100 ppm
50 ppm
781
The origin of several elements in the used motor oil (table 3) may be from
additives (Zn, Ca, Ba and Mg), the wear products (Fe, Pb, Cu, Cr, Al, Mn, Ag and Sn)
and contaminants originating from fuel, air and liquid cooling (Na, B, Si and Ca).
Table 3 The origin of certain wear elements in the motor oil.
Elements
Fe
Al
Ag
Cr
Cu
Pb
Sn
B
Na
Ca
Si
Zn,Mg,Mo
Wear Metal Source

cylinder liners, piston ring, journal bearing, valve lifter, camshaft,
crankshaft
piston, Al-Sn bearings, turbocharger
silvered parts, bearings, journals
hard chrome plated parts, pistons, cylinders, valve lifter, exhaust valve,
connecting rods
Cu-Pb bearings, bushings, oil coolers, camshaft, valve train (valves
with system for opening and closing), fuel injector, regulator
Cu-Pb bearings, gasoline, additives
bronzed parts, bearings, pistons
antifreeze
antifreeze
from atmosphere
particles from atmosphere
from additives
The research was carried out in three vehicles (buses MERCEDES O 345)
which have embedded engine Mercedes-Benz, type OM 447HLA. This is four-stroked
engine with six cylinders arranged in line, turbo diesel, liquid refrigeration and with a
combined lubrication, which meets Euro 2 emission standards related to exhaust
gases. Technical data of engines are given in table 4.
Table 4 Technical data for Mercedes-Benz engine, type OM 447HLA
Engine Mercedes-Benz, type OM 447HLA
Engine type
four-stroked, turbo diesel (EURO 2)
Number and spacing cylinders
6, linear
Cylinder bore, mm
128
Engine capacity, litres
11,97
Compression ratio
18 : 1
Nominal output, in 2200 min1, KW
220
Maximum torque, in 1100 min1, Nm
1100
Minimal number of revolutions idle stroke, min1 600
Rated speed, min1
2500
Maximum permissive temperature coolant, C
105
Pressure motor oil (Rated speed), bar
2,5
Pressure motor oil (idle stroke), bar
0,5
The amount of oil with filter, litres
25
782
Characteristics of zero samples of motor oil, VALVOLINE, SAE classification

SAE 10W-40 and API classification CF and ACEA E4 are shown in table 5, and the
results used oil samples in table 6.
Table 5 Results of zero samples of oil from the engine
Motor oil VALVOLINE, SAE classification: SAE 10W-40; API classification: CF ACEA E4
Viscosity at
Viscosity at
Viscosity
Pour
Flash Point
TBN
Foaming
100C(cSt)
40C (cSt)
Index
Point (C)
(C)
14,37
94,96
156
- 36
210
10,7
passage
The research was conducted through periodic sampling oil from engine
vehicles listed above. Apart from the fresh oil (,,zero sample), samples are taken after
10.000 km, 20.000 km and 30.000 km. After 30.000 km the replacement oil charging
engines in all three vehicles.
Also, special attention is paid to the preservation of samples from
contamination, as in the phase of identifying the sample and phase of manipulation,
which is fully met by applying the prescribed procedures. In accordance with the time
provided a very high level of purity of all elements in the chain of systems for sampling,
as well as the separation of samples in a way that is not perturb the integrity of data
that it carries on the state of components of vehicles from which sampling was done.
Table 6 The results of testing samples of used oil from engines examined vehicles
Characteristic
Color
Viscosity
40C, mm2/s
Viscosity
100C, mm2/s
Viscosity index
Flash
point, C
TBN,
mg KOH/g
Sample
0
1
2
3
0
1
2
3
0
1
2
3
0
1
2
3
0
1
2
3
0
1
2
3
Engine O 345 1
3,0
black
black
black
94,96
85,87
84,16
84
14,37
12,67
12,64
12,61
156
156
149
145
210
206
202
200
10,7
10,1
9,42
8,6
Engine O 345 2
3,0
black
black
black
94,96
83,61
83,59
83,04
14,37
13,15
12,99
12,55
156
155
149
147
210
204
201
199
10,7
10,3
9,65
8,7
Engine O 345 3
3,0
black
black
black
94,96
84,81
82,35
81,76
14,37
12,78
12,64
12,38
156
153
148
146
210
205
202
201
10,7
10
9,87
7,1
783
Insoluble
substances in
pentane, %
Insoluble
substances in
benzene, %
Content Fe,
(ppm)
Content Cu,
(ppm)
1
2
3
1
2
3
1
2
3
1
2
3
0,09
0,22
0,31
0,13
0,19
0,37
13,5
13,7
37,4
1,3
1,5
1,9
0,08
0,18
0,27
0,11
0,16
0,32
11,8
13,6
19,8
1,3
2,3
2,5
0,05
0,15
0,25
0,10
0,15
0,30
9,6
12,5
14,8
1,3
3,3
3,9
Engine used motor oil, VALVOLINE quality API CF and ACEA E4, gradation
SAE 10W-40. For testing buses are exploited in the conditions of city driving. The task
was to check the interval to replace motor oil after 30.000 km. It is found that the
changes characteristic for motor oil were expected and within the allowable limits.
During exploitation increase viscosity lubricants is expected to. Reasons for the
increase of viscosity lubricants are as follows: oxidation of lubricants, cavitation due to
foaming lubricants, dissolution of lubricants with water, pouring and charging system
viscosity fat greater than recommended and contamination of solid particles and
products wear lubricants. On the other hand, the reasons for the reduction of lubricants
viscosity are: lubricants contamination of fuel (for motor oil), shearing additive for
reclamation viscosity, drop point of flash, grinding molecules, lubricants contamination
without solubility with water, pouring and charging system viscosity less fat than
recommended, and the impact of liquid cooling. Also, the causes may be high
temperature, load, uncontrolled long interval use, insufficient amount of oil in the oil
system, inefficient cooling systems and the like.
The change of VISCOSITY of motor oil from the vehicle
Mercedes O 345
100
Viscosity at 40 C, mm /s
Viscosity of "0 sample" 94,96
Oil from
the engine
O 3451
95
90
Oil from
the engine
O 3452
85
Oil from
the engine
O 3453
80
Maximum allowed variation 20% 75,97

75
0
10000
20000
30000
Crossed kilometers, km
40000
Fig. 3 The change of viscosity at 40C of motor oil from the vehicle Mercedes O 345
784
The change of VISCOSITY of motor oil from the vehicle

Mercedes O 345
Viscosity at 100 C, mm /s
15
14,5
Oil from
the engine
O 3451
Viscosity of "0 sample" 14, 37
14
13,5
Oil from
the engine
O 3452
13
12,5
12
Oil from
the engine
O 3453
Maximum allowed variation 20% 11,5
11,5
11
0
10000
20000
30000
40000
Fig. 4 The change of viscosity at 100C of motor oil from the vehicle Mercedes O 345
Figure 3 shows the changes viscosity at 40C motor oil during generation. It is
during the fall viscosity after the first 10.000 km of oil from all three engines, and after
this period, viscosity remains approximately constant until the end of the interval
changes oil charge. Fall viscosity for the entire period of exploitation of oil is 11,54 %
for the first vehicle, 12,55% for the second and 13,9% for the third vehicle. This is far
below the allowed limit of 20% (table 2).
Figure 4 shows the changes viscosity at 100C motor oil. Fall viscosity for the
entire period of exploitation of oil is 12,24% for the first vehicle, 12,66% for the second
and 13,84% for the third vehicle, which is also below the allowed limit of 20% (table 2).
The change of TBN of motor oil from the vehicle
Mercedes O 345
Total base number, mg KOH/g
12
TBN of "0 sample" 10,7
Oil from
the engine
O 3451
10
8
Oil from
the engine
O 3452
6
4
Maximum allowed variation down to 50% 5,35
Oil from
the engine
O 3453
0
0
10000
20000
30000
40000
Fig. 5 The change of TBN of motor oil from the vehicle Mercedes O 345
785
Figure 5. shows the changes of total base number (TBN). TBN value is the
largest in the new (fresh) oil, and reduces during the time spent in the work. Low TBN
value indicates that working life of oil is near the end. TBN is usually used in engine
lubricants, a sudden drop points to the poor quality of fuel (high sulfur content). The
importance of TBN is the possibility of evaluating the degree of additive quality, which
is performed on the basis of differences in alkalinity between used and fresh oil. Until
30.000 km TBN value does not exceed the allowed limit. The fall of TBN for the entire
period of exploitation of oil is 19,61 % for the first vehicle, for the second 18,69 % and
33,64 % for the third vehicle, which is below the allowed limit of 50 % (table 2).
Viscosity Index, as it is mentioned, is the tendency of viscosity changes with
the temperature and is expressed empirical, using non-dimensional number. During
the exploitation it is desired that the viscosity changes as lesser as possible with the
change of temperature. If during work temperature modes are changeable and cause
major changes of viscosity that may cause disruptions in the functioning of the system,
which is a manifestation of increased friction, wear and damage.
Change of motor oil Viscosity Index from the bus Mercedes O 345 is shown in
the figure 6. The decrease in the Viscosity Index oil is evident: for the first vehicle 7,5
%, for the second vehicle 5,76 % and for the third vehicle 6,41 %, which exceeds the
limit of 5 % (table 2).
The change of Viscosity Index of motor oil from the vehicle
Mercedes O 345
158
156
Viscosity index
Oil from
the engine
O 3451
Viscosity index of
"0 sample" 156
154
152
150
Oil from
the engine
O 3452
148
Maximum allowed
variation 5% 148,2
146
144
Oil from
the engine
O 3453
142
140
138
0
10000
20000
30000
40000
Fig. 6 The change of Viscosity Index of motor oil from the vehicle Mercedes O 345
Flash point represents data that shows what temperature leads to open fire
ignition by the steam created by oil heating. In motor oil analysis the flash point
determines the presence of fuel oil, which is a consequence of poor motor (bad work
injectors). The reduction of flash point is due to the penetration of fuel.
786
The change of Flash Point of motor oil from the vehicle

Mercedes O 345
212
210
Flash point, C
Oil from
the engine
O 3451
Flash point of "0 sample" 210qC
208
206
204
Oil from
the engine
O 3452
202
200
198
Oil from
the engine
O 3453
196
194
Maximum allowed variation down to 20% 168
192
0
10000
20000
30000
40000
Fig. 7 The change of flash point of motor oil from the vehicle Mercedes O 345
Figure 7 shows the change of flash point for motor oil from the bus Mercedes O
345. The decrease in the flash point is noticeable, and by the end of exploitation
testing does not exceed the allowed limits (20%, table 2) for none of the observed
vehicles. The fall flash point is: for the first vehicle 4,76%, for the second vehicle
5,23%, and for the third vehicle 4,28%.
This indicates that there was no significant penetration of fuel in the engine
lubrication system for the above mentioned vehicles.
Insoluble substances in pentane from vehicles
Mercedes O 345 motor oil
Insoluble in pentane, %
0,35
Maximum allowed
ammount of insoluble
substances in pentane
up to 3,5%
0,3
0,25
Oil from
the engine
O 3451
Oil from
the engine
O 3452
0,2
0,15
0,1
Oil from
the engine
O 3453
0,05
0
0
10000
20000
30000
40000
Fig. 8 The change of amount of insoluble substances in pentane from vehicles

787
Insoluble substances in benzene from vehicles

Insoluble in benzene, %
0,4
Maximum allowed
ammount of insoluble
substances in benzene
up to 2,5%
0,35
0,3
0,25
Oil from
the engine
O 3451
Oil from
the engine
O 3452
0,2
0,15
Oil from
the engine
O 3453
0,1
0,05
0
0
10000
20000
30000
40000
Fig. 9 The change of amount of insoluble substances in benzene from vehicles

Insoluble residues after treatment in pentane are oxidant products and
mechanical impurities, while insoluble residues after treatment in benzene are
insoluble materials such as coke, scale, dust, soot, particles originated from wear
contact area of tribomechanical systems of engines and other mechanical impurities.
Graphical display of changes in the value of insoluble substances in pentane
and benzene is given in the figure 8. and 9. The content of insoluble substances in the
oil is negligible compared to the allowed deviation values (maximum insoluble in
pentane 0,31 %, and it is allowed up to 3,5 %, maximum insoluble in 0,37 % benzene,
and it is allowed up to 2,5 %).
In order to view the degree of wear of elements in tribomechanical system
engines the atomic absorption spectrophotometry (AAS) is applied for oil sampled from
examined engines. This method is determined by the type and concentration of metals
(Fe and Cu) in the oil charge.
Atomic absorption spectrophotometry is a very sensitive, simple, fast and
reproductive method, used for the detection of most elements. It is based on
measurement of light absorption by the atoms that are in the non-excited state.
Reduction of missed light intensity increases proportionally with the increasing of
number of atoms that are in the non-excited state.
Absorption is determined by measuring changes in intensity before and after
the leakage of light through the atomic gas and it is proportional to the concentration of
measured element. If the concentrations of elements are small, light absorption is
linearly dependent on concentration. Metals like iron (Fe) and copper (Cu) were
selected for identification because they are typical elements contained in the examined
engines. On the basis of changes in their concentration in the oil charge it can be
determined their origin from engine elements and the degree of wear.
788
Content of Fe in vehicle Mercedes O 345 motor oil

40
Maximum content of Fe
up to 100 ppm
Content of Fe, ppm
35
Oil from
the engine
O 3451
30
25
Oil from
the engine
O 3452
20
15
10
Oil from
the engine
O 3453
5
0
0
10000
20000
30000
40000
Fig. 10 The change of content Fe in motor oil from vehicle Mercedes O 345
Iron content (Figure 10), as a product of wear, in the oil charge to the end of
exploitation testing have a growing trend and in the last sampling, after 30.000 km, was
37,4 ppm (37,4%) of the maximum allowed value for the first vehicle, 19,8 ppm
(19,8%) for second vehicle and 14,8 ppm (14,8%) for the third vehicle. Operational
criteria for a replacement of motor oil charge to the end of exploitation investigation of
iron is significantly below the allowed limits for all three vehicles. Iron content indicates
that the wear in the engine tribomechanical systems of all three vehicles are within the
allowable limits. Established iron content (figure 10) indicates that the wear in the
engine tribomechanical systems from which sampling was done in the oil is permitted
limits.
Content of Cu in vehicle Mercedes O 345 motor oil
4,5
Maximum content of Cu
up to 50 ppm
Content of Cu, ppm
Oil from
the engine
O 3451
3,5
3
Oil from
the engine
O 3452
2,5
2
1,5
Oil from
the engine
O 3453
1
0,5
0
0
10000
20000
30000
40000
Fig. 11 The change of content Cu in motor oil from vehicle Mercedes O 345
789
Copper content (figure 11), as a product of wear in the oil charge to the end of
exploitation testing have a growing trend and in the last sampling, after 30.000 km, was
1,9 ppm (3,8%) of the maximum allowed value for the first vehicle, 2,5 ppm (5%) for
the second vehicle, and 3,9 ppm (7,8%) for the third vehicle. It can be concluded that
the copper content is far below the allowed limit of 50 ppm.
4. CONCLUSION
On the basis of all mentioned above next conclusions can be drawn:
motor oil VALVOLINE, API CF and ACEA E4, gradation SAE 10W-40 is
analyzed during the exploitation, and it achieves its primary function and
meet the prescribed replacement interval of 30.000 km engine EURO 2
categories, which is found the characteristic analysis of physico-chemical
properties of oil products and wear (Fe and Cu) during the exploitation;
the fall of viscosityis evident during the first 10.000 km, and after this
period, viscosity remains approximately constant until the end of the interval
changes of oil charge. Maximum fall viscosity during the exploitation of oil
from all three engines is significantly below the allowed limit of 20 %;
after 30.000 km TBN value has not exceeded the allowable limit for oil
samples from all three engines;
the content of insoluble substances in the oil is negligible in comparison to
the limit value, because there is no significant presence of oxidation
products and mechanical impurities, insoluble substances such as coke,
scale, dust, soot, particles originated from wear contact area of
tribomechanical system in engines and other mechanical impurities;
small decrease of flash point values shows that there was no significant
penetration of fuel into the system for lubrication;
content of iron and copper is significantly below the allowable limits for all
three vehicles;
the appearance of water in the samples is not found,
after 30.000 km oil is replaced, only by the recommendation of the
manufacturer about how to change the oil charge.
The main focus of experimental research, realized in this paper, was put on the
grease, as the information on the system as a whole. Tests conducted on the vehicle
engines, which have been identified as tribomechanical systems, showed that in most
cases the change of the elements functionality and the entire system is expressed
through changes in lubricant characteristics. This confirms that the change of physicochemical characteristics of oil charge can be adopted for the assessment of the system
state.
During realized investigations it was concluded that changes appear in
physico-chemical characteristics of lubricant in engine. These changes directly depend
on condition of all elements of tribomechanical system, from their functional
characteristics.
Extension of the interval using motor oil without the monitoring of the state is
very risky and can have the following consequences: mutual attachment of piston
rings, burnt and mild covered pistons, quick spending of beds, burnt valves and finally
jam engines.
790
Metal particles, physico-chemical processes and contaminants, detected

through laboratory analysis and ,,in-situ methods, is the appropriate base to identify
possible disfunctionalities in tribomechanical systems, as well as to determine the life
of usage of oil and its functionality in oil systems.
Confirming the basic causes of failures and their elimination, control of certain
phenomena, is defining proactive maintenance, as a new method that reduces
maintenance costs and prolongs the life of assets.
5. REFERENCES
[1]
[2]
[3]
[4]
[5]
B. Ivkovic, A. Rac, Tribology, Jugoslav Tribology Society, Kragujevac, 1995.

M. Babic, Lubricants monitoring, Kragujevac, 2004.
R. I. Taylor and R. C. Coy, Improved fuel efficiency by lubricant design, 2001,
Troyer D., Fitch J., Oil Analysis Basics, Noria Corporation, 1999.,
Peri Sreten Influence of tracked vehicle transmission gear exploitation method
upon physical and chemical characteristics of lubricants, Magistarski rad, Faculty
of Mechanical Engineering, University of Belgrade, 2006.
791
E. MECHATRONICS

ANALYSES THE DAMAGE CAUSED BY ATMOSPERIC DISCHARGE

AND OVERVOLTAGE PROTECTION
Milan Paripovi1
Summary: Devices for power supply of electronic systems must provide the necessary
operational voltage to all permissible variations of input voltage. In a cases of
protection against voltage transients, additional protection of power supply devices is
required. Damage to electronic installations and devices caused by overvoltage are
larger and larger because of the wider use of electronic devices and systems (computers with a modem connection to the Internet, ADSL devices, ...). Decrease of of the
device operational signals level increases device susceptibility to overvoltages. The
paper analyzes the damage done to the Radio-relay facility due to by atmospheric
discharge and overvoltage protection.
Key words: analysis of damage, atmospheric overvoltages, overvoltage protection,
switching operations, devices for protection against overvoltage, standard test pulse,
risk assessment of the surge, induction of conductors
1.
INTRODUCTION
Surges can cause different types of damage depending on what kind of space
the electrical installations are located and why the pool was intended. Damages, which
are related to the repair or replacement of electrical appliances or electrical equipment,
but damage can occur due to decreased function of the drive. If the surge was
particularly high, and the protection it was not consistently carried out, except for repair
and replacement of electrical appliances and electrical equipment may incur additional
costs due to plant outages and the time needed for the reconstruction of any such
business According one assessment reactivation of system failure in information
networks in the bank is 2 days, the sales-oriented companies 3.3 days, the production
facility 4.9 days, with insurance companies 5.6 days [1].
In business offices, it is inconceivable to work without a computer with all its
peri-fern devices connected to telecommunications networks and networked with each
other over an internal computer network, fax, copier, ... Too bad only the devices for
replacement or repair can be considerable, but infinitely more because of the interruption of business.
B. Sc. EE, Milan Paripovi , Banja Luka, m:tel, (milan.paripovic@mtel.ba)
795
Milan Paripovi
2.
DAMAGE ANALYSIS OF TO RADIO RELAY OBJECT
A mobile telephony base station, which consists of an equipment container, on

a certain Telekom RS location is situated. The antenna system is situated on a 70 m
high tower, whereas tevevision (TV) object is situated in the near vicinity. At a distance
of 1000 m of the base station, a Telekom RS object containing equipment is situated
as well as a relay antenna tower. Near the Telekom object, a power line with a tower
mounted transformer station (TMTS) is situated, whose purpose is electrical power
supplying of Telekom object.
Fig. 1 Single line diagram power telecommunications facility

Electric power is supplied via a 80 m long buried cable from TMTS to
distribution cabinet In the Telekom object, there exists an backup generator. Electric
power supply of the base station is realised by a buried cable from the Telekom object
from aggregate closet. From the Telekom object, two telecommunication cables (one
multipair and one optical) are placed for TV and mobile neccesities. Their routes are
separated from the base station power supply cable route. The power supply of the TV
796
Analizes the damage caused by atmosperic discharge and overvoltage protection
is placed on a special route via an overhead power line. All objects are lightning
protected by a classic lightning protection system with grounding and equipoitential
bonding of all metal parts. The grounding system is designed as a combination of ring
shaped Fe/Zn conductor 25x4 mm2 as well as multiple radially placed conductors. One
of these radial conductors is placed in the same channel as the buried power line
which leads to distribution cabinet. The latticed relay tower is connected with the metal
construction of the container via the waveguide metal racks and the antenna system
and the grounding conductor [2].
On a certain Telekom object a lightning strike has occured which stopped the
traffic on the Radio Relay System as well as mobile telephony and TV program. We
have photographed the mentioned situation. It was concluded that the damage on
certain cables and the melting of a number of fuses especially on the base station
power supply cable. A large number of miniature circuit breakers was found to be in
the off position. Aggregate control system and rectifier did not serve the function and
purpose they were meant to serve. On the basis of the described event, the inspected
condition and the places at which the malfunction occured as well as their marks the
following can be concluded: most probably there was a lightning strike of large
magnitude in the large TV tower with GSM antenna system. A strong lightning current
flowed along the metal parts and grounding system conductors toward base station
cottage and remote earth. Since one grounding conductor is situated in the same
channel as the power line buried cable (parallel on a small distance), strong lightning
current of grounding conductor can induced surge wave in power supply cable [3].
The developed surge wave propagates along the cable on both sides. In
distribution cabinet, surge wave penetrates to the earth through the varistors. In that
section, all fuses are blown, whereas varistors are damaged. The second part of the
surge wave, which was propagating along the cable towards the RR object, was
connected to ground through the miniature circuit breakers and fuses. Fuses are blown
and their housings are damaged (burned). An overvoltage of lesser magnitude, which
was transmitted through the distribution cabinet, caused blowing of certain circuit fuses
and damage of thinner cables. Since the consequences of the overvoltage are visible
on the ends of the telecommunication cable, it is logical to assume that a similar
overvoltage was induced in the telecommunication cable, but of lesser magnitude.
Since the route of the telecommunication cable is not reliably known, it is expected that
telecommunication cable is near to power supply cable, near to radio relay tower or
parallel to the grounding conductor.
3.
SURGE PROTECTION OF POWER LINES
At a power cable entrance into the container ( for base station with external
lightning protection implemented ), SPD instaled between power cable and ground
should be characterized with high energy withstand capability.
According to the classification defined by IEC 61643-1[3], SPD installed at the
service entrance is denoted as Class I for protection against direct lightning currents.
The same docu-ment introduces SPD of Class II intended for protection against
indirect lightning effects and which are tested with impulse 8/20 s. Therefore overvoltage protection of the power supply system must be ensured through the effective coo-
797
Milan Paripovi
rdination of surge protectiondevices class I and class II (Figure 2) decoupled by

concetrated or distributed inductance L.
Fig. 2 Two stage coordination of SPDs

If the spark-gap is used in the first class (SPD1) low impedance connection to
the ground and low voltage drop is ensured. The ignition of the spark gap is guaranteed by concentrated indictance L. SPD2 is usualy based on MOV technology. The
spark-gap shows advantage in the coordi-nation, but it suffers from disadvantage that
limits spark-gap application in power supply networks. Electrical arc on spark gap cant
be extinguished until power-freqency voltage cross-es the zero and current from power
supply, known as follow-up current continues to flow through air-gap causing short-circuit in the power network even when surge current has been disappeared [4].
Follow-up current doesnt occur when MOV is used as SPD1. New MOVs with
high energy withstand capability allowed their application for protection against direct
lightni-ng stroke. Current sharing in the case when SPD1 and SPD2 is MOV is
Fig. 3 Current sharing: through SPD!, through SPD2

798
Analizes the damage caused by atmosperic discharge and overvoltage protection
SPD1 and SPD2 can be realized as serial connection of MOV and gas discharge tube (GDT). One of the basic tasks of the varistor in such connection is to
extinguish conduction in the gap after overvoltage pulse disappears. Intiated GDT wil
ensure low impendance and lower voltage drop in the first branch than with single
MOV. This connection eliminates leakage current through MOV. It is well known that
MOV has leakage current when it is connected between phase and neutral line
Fig. 4 Single line diagram with surge protection power telecommunication facility
799
Milan Paripovi
4.
CONCLUSION
Radio-relay centers and base stations are extremely sensitive point

communication system when it comes to exposure to surge impulses caused by
atmospheric electrical discharges, and voltages from the network. Proper performance
of the external protection reduces the induced current in the communication and power
lines in the object. Particular attention should be paid to surge protection energetic
lines and proper selection of elements in these systems, surge protection. Risk
assessment of the surge in the electrical system shows that it is justified installation of
surge protection.
LITERATURE
[1]
[2]
[3]
[4]
800
I. Matekovi,: Risk assessment of the surge in electrical installations ,Energy

2003.
Milan Paripovo ,Protection against lightning and surge protection mobile phone
base stations, 16 Telecommnications Forum TELFOR 2008, Serbia
IEC 61643-1 Surge protective devices connected to low-voltage power
distribution systems Performance reqiirments and testing
V. Murko,A Stagoj, A. Mujcic, N. Suljanovic, Coordinating Surge Protective Devices with Metal-oxide Varistors at Direct Lightning Stroke. Application Note, International Conference on Lightning Protection ICLP,Kanazawa, 2006.
AUTOMATION AND OPTIMIZATION OF PROJECT OPERATIONS IN

THE BRIDGE CRANE DESIGN PROCESS
Slavia Todorovi1, Miroslav Rogi2
Summary: In an era of marked competition production companies are forced to reduce
development and production costs, while maintaining high technological and functional
level of the new product. In addition to modern production technologies, higher levels
of product quality management and work organization, use of Computer Aided
technologies is indispensable, which allow hardware and software engineering tools
throughout all phases of a product life cycle. In this paper, using the software CATIA
V5 is realized automation of a large number of project operations in the design process
of bridge crane. This procedure will speed up product development process, the
occurrence of errors bring to a minimum, reduces the number of different product
variants, which helps to reduce development costs, increase product quality and
improved competitiveness of production company in the market.
1.
INTRODUCTION
Bridge cranes are complex crane machines intended to carry cargo in the
horizontal and vertical planes. Bridge cranes are mobile cranes with one or two major
beams, depending on size and load range. Can be performed standing or hanging. For
the design and construction of bridge cranes are of special importance of the following
characteristics: functionality, operational safety, technical protection, low cost, easy
interchangeability of damaged parts, esthetic form, easily inspected and handling, the
use of the simpler technologies in production and installation. Using CATIA V5
software package was developed families of standard parts catalog, of mechanical
parts which are typical of crane machines, which are not found in existing libraries in
CAD/CAM softwares. Automation of the design procedure for 3D models of mechanical
parts and assemblies, can be achieved by parameterization methods, based on input
values needed for a few basic parameters. This greatly reduces the time of
development of new bridge crane project, which also results in substantial reduction in
product development costs.
1
2
dipl. ing. ma., Slavia Todorovi, Banja Luka, Mainski Fakultet, (slavisa.todorovic@gmail.com)
redovni profesor, prof. dr Miroslav Rogi, Banja Luka, Mainski Fakultet, (rogic@urc.rs.ba)
801
Slavia Todorovi, Miroslav Rogi
2.
DEVELOPMENT AND APPLICATION OF BRIDGE CRANE STANDARD

PARTS LIBARY IN CATIA V5 SOFTWARE
Parametrization of standard parts family in CATIA V5 begins by defining the

Excel design table for the family. Excel design table of the standard parts family shall
contain the names and length parameters of all members of the family. Whole family of
standard parts should be dimensioned by these parameters. Excel design table is
imported into CATIA V5 Part Design workbench, and from Excel design table are
added length parameters into set of parameters of 3D model of standard part. Using
these parameters are dimensioned 3D model of standard part of family.
Fig. 1 Generating parametric 3D model for member of standardized hook family of

bridge cranes
This 3D model is imported into the library of standard parts of bridge crane,
using CATIA V5 Catalog Editor workbench. The big advantage of this procedure is that
a single parametric 3D model provides to the library information on all family members.
Therefore, it is not necessary to model all the family members of standard parts of
bridge cranes.
Fig. 2 Use of catalog in Assembly Design CATIA V5 workbench

802
Automation and Optimization of Project Operations in the Bridge Crane Design Process
Image 2 shows the procedure of using CAD catalog of standard parts in CATIA
V5 Assembly Design workbench, where one*.CATProduct contains several members
of the families of hooks, traverses and hook-nuts, which were added from catalog into
the assembly.
Using of CAD catalog of standard parts achived great speed of modeling
assemblyes that contain a large numbers of standard parts. Occurrence of errors is
low. It is possible to change the shape of standard parts by adding new features into
the parametric structure of parts. Catalog of standard parts is easy and simple to adept
by editing Excel design tables of families of standard parts.
3.
PARAMETERIZATION OF BRIDGE CRANE PARTS AND ASSEMBLIES
All the characteristic of 3D modeles in CATIA V5 are defined by parameters.

These include: part numbers, dimensions, geometric constraints, tolerance, material
properties, area, volume, mass, inertia moments, etc.
The parametrization represents use of tools and designing techniques which is
achieved simplicity and flexibility in manipulation of dimensions and other features of
parts or assmbly 3D modeles. Parametrization provides automatic dimensioning and
positioning of parts in assembly, based on values of few main assembly parameters.
For example, dimensions and position of all parts of rope drum assembly are
determined by the values of few parameters: rope diameter, crane tackle ratio, lifting
height, rope drum sheet metal thicknes etc.
Fig. 3 Defining of parameters via Excel design table for dimensioning crane tackle
parts
Crane tackle assembly contains considerable number of standard parts, which
are added from developed standard parts libary: hook, hook nut, traverse, axial
bearing, plain bearing etc. That is why other parts of crane tackle are parameterized
via Excel design tabeles. Image 3 show how one Excel design table can be used for
parametrization of more crane tackle parts. In such a way it is posible to use Excel
design tables of standard crane tackle parts for parametrization of non-standard crane
tackle parts. This procedure allows automatic dimensioning of non-standard parts
depending on the values of parameters of standard crane tackle parts, and of
parameters of crane tackle main parts: tackle axis and tackle wheel.
During parameterization of the other bridge crane assemblies, parameters are
803
created internally in CATIA V5, via Formula dialog box. Formula dialog box allows the
generation of new parameters, deleting and editing, add formulas in values of
parameters etc.
The folowing techniques and tools in CATIA V5 were used for achieve
automatic dimensioning and positioning parts in bridge crane assemblies, in depending
on values of main parameters:
formulas
parametric links
geometric links
laws
rules
measurements
Most of these tools are part of CATIA V5 Knowledgeware workbenches that
are among the KBE (Knowledge-based Engineering) Computer Aided Technology.
Using these techniques, we created a special datbase of parametric 3D models of
bridge crane assemblyes. Images 4 and 5 shows how these assemblyes adapt and
implement into the current bridge crane product. Parametric 3D model of current
project bridge crane is completed in four phases in image 4.
Fig. 4 Modeling the wagon assembly of current bridge crane project
804
Automation and Optimization of Project Operations in the Bridge Crane Design Process
Fig. 5 Modeling parametric 3D model of current bridge crane project

4.
ACCELERATE OF DEVELOPMENT OF DRAFTS, FEM ANALYSIS, CAM

PROCESSES AND OTHER PROJECT TASKS
High level of standardization and unification of bridge cranes parts and

assemblyes allows the use of the same 3D modeles in multiple bridge cranes products.
CATIA V5 allows associativity between diferent CATIA V5 extensions via external
references links. This fact allows the use of the one extension (*.CATDrawing,
*.CATAnalysis or *.CATProcess) in multiple bridge crane projects.
Once made a drawing of part is transferred to multiple folders of bridge crane
projects, via CATIA V5 Save Management. Then, in the copy of drawing that is located
in current project folder, done link replace. New link must lead to *.CATPart which is
located in current project folder.
During update of drawing, software doing changes of a contour of part and
generate new values of dimensions, depending of dimensions of current project
*.CATPart. All annotations, tolerances, notes, symbols etc. that are created at first
time in drawing were kept in copy of drawing. Designer of drawing has less work that
is related to amending of existing elements of drawing.
Procedure is identical for other CATIA V5 extensions: *.CATAnalysis or
*.CATProcess.
5.
CONCLUSION
In addition to the described advantages of associativity between diferent

extensions of CATIA V5 software, associativity between software extensions allows
realize of concurrent engineering. Teams of experts responsible for different phases of
product lifecycle management can work parallel in the same time. This reduce the
elapsed time required to bring a new product to the market.
Use of the CAD/CAM/CAE software brings many advantages at economic805
marketing field. Rendered photorealistic images and animations, 3D print or 3DVIA

online enviroment can represent new product without providing an real prototype of
product.
Fig. 6 Photorealistic image of bridge crane

LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
806
Dragutin ap: Prenosila i dizala, podloge za konstrukciju i proraun II; Zagreb,

1988.
Miroslav Rogi: Transportni ureaji i fabrika postrojenja; predavanja, Banja Luka
Simo Jokanovi: CAD/CAM/CAPP sistemi; skripta, Banja Luka
Fred Karam i Charles Kleismit: CATIA V5; Kompjuter biblioteka
Nader G. Zimani i Jonathan M. Weaver: CATIA V5 dizajn mehanizama i njihova
animacija; Kompjuter biblioteka, SDC Publications
Standardi: ISO, EN, JUS, DIN
DC STABILIZER WITH DIGITAL CONTROL

Corina Daniela CUNAN1, Ioan BACIU2, Loredana GHIORGHIONI3
Abstract: Power for electronic circuits with good performance is made through stabilized DC
voltage. This paper presents a numerically-controlled voltage source with short circuit
protection. Command output voltage control is obtained from a computer system via an
interface circuit, containing a digital to analog converter and an electrical isolation circuit.
The digital signal obtained from the parallel port is converted into a stream of digital to analog
convertor, which is transformed into a control voltage with a current-voltage converter made
with an operational amplifier.
To protect the computer from voltage or current shock was achieved the electrical isolation
circuit, the circuit structure containing two optocouplers to obtain an output voltage as linear.
1. INTRODUCTION
Voltage stabilizer is an electrical installation that provides constant voltage to
consumer terminals in condition that the supply voltage or the consumer impedance change
in a range of values.
Electronic stabilizers are composed of a control circuit that has the role to feel voltage
variation at the consumer terminals and to correct the value of that voltage when the input
voltage modify or the voltage value.
Integrated regulators are circuits most commonly used, after the operational
amplifiers, in the electronic machine. They are easy to use and the sizing calculations are
simplified in most cases.
Voltage stabilizer parameters must be known to calculate total variation of the
stabilizer output voltage under certain condition of changing the rectifier output voltage, the
current through the load, the ambient temperature and the junctions temperature. Therefore,
the performance parameters of the integrated rectifier, give in catalog, are important in
ensuring the stabilizer quality. Stabilized voltage sources adjustable from zero are necessary
in special for the equipments that exists in the electronic or electrotechnic laboratories, as for
the try laboratories for the electronic or electric machine.
For the voltage stabilizers with a serial regulation element must be provided shortcircuit protection because that current passes through power transistor of the regulation
1
Lecturer Ph.D. eng.Corina Daniela Cunan, Hunedoara, Romania, Politehnica University of Timioara,
(corina.cuntan@fih.upt.ro)
2
Assistant Ph.D.eng.Ioan Baciu, Hunedoara, Romania, Politehnica University of Timioara, (baciu.ioan@fih.upt.ro)
3
Eng. Loredana Ghiorghioni, Hunedoara, Romania, Politehnica University of Timioara,
(loredana.ghiorghioni@fih.upt.ro)
807
Corina Daniela Cunan, Ioan Baciu, Loredana Ghiorghioni
element in full control of the item. Protection with fast or ultrafast fuse is not possible exlusive
because their break time is much longer that the time it reaches the maximum power for the
serial regulation element. It thus requires the use of the electronic protection circuits or mixed
type in both cases.[1]
2. WORKS PRESENTATION
LM117 integrated circuits are precision voltage regulators and have a low dispersion
of the reference voltage. As constant voltage regulators, they include all necessary protection:
over current, short-circuit (thermal protection) and against secondary breakthrough.
Stabilizers output voltage made with these integrated circuits can be adjusted easily
in a wide range: 1.2 ... 35V. As a "floating" regulator (no weight), it can be used for
stabilization of high voltage or of current.
Typical scheme of an adjustable voltage stabilizer is presented in figure 1.
Negative
feedback
Fig. 1 Typical scheme of stabilizer
I
H
9U
5 5
5M
,D
I
H
9U
5 5
Capacitors used in the scheme have the following effects:

C1 - decoupling perturbations for high frequency noise;
C2 - improving ripple voltage rejection,
C3 - improve the transient response from the output stabilizer
Figure 1 includes two diode for the integrated circuit protection in case of power
supply failure or of input short-circuit (D1) respectively in case of output short-circuit(D2),
when the C3 or C2 capacitors are downloading through the integrated producing damaging of
it, especially for high capacity(in case of high output voltage). [1]
R1 and R2 divider is primarily intended to fixed output voltage value (on the load). But
through this divider must close the minimum current of the integrated regulator when sheme
operates in no-load.
Resistance R1 should be mounted as close to Vs terminal, and the end of the table
from the resistance R2 should be very close to the load weight to improve stabilization in
relation with the load (reducing stabilizer output resistance). For figure 1, the output voltage is
expressed as:
(1)
Having three terminals one for control (ADJ), can take a variable voltage between
the other two terminals ('IN' i 'OUT) and having a reduce voltage, approximately constant
808
DC Stabilizer with Digital Control
(1.25V) between the output terminal and the control one, integrated circuit LM117 (217,317) is
like a transistor.
Linear operation for the integrated regulator is similar with the one of a transistor. In
the regulator scheme with two serial transistors, transistor T reaches in the blocking situation.
Replacing transistor T with a LM117 (217,317) integrated regulator type is necessary because
the last one should can block it when between its pins, OUT and ADJ, there is a voltage
greater than 1,25V.
An experiment made to known the regulator behavior in those circumstantial shows
that it is blocking, but for a 7,5V voltage order, appears a breakthrough phenomenon. So the
current that get out through the ADJ terminal must be limited though a resistance (Radj).
That breakthrough appears between the integrated resistor connected at ADJ pin and the
resistance zone (n) that is connected to the OUT pin. Because we want that current through
ADJ terminal dont cause damage to the regulator or complications in the stabilizer scheme
with two serial regulators, it will be limited through a serial resistance, appropriately sized.
Using a regulator with two serial integrated circuits LM117(217,317), lsmax=f(Vie)
characteristic for that new regulator extends much in the current zone and large voltage
difference(compared to one single regulator characteristic, given in fig. 1 ), because of the
mode that acts the new regulator.[1]
Fig. 2 Voltage current dependencies on ADJ pin

For the stabilizer composed in two serial regulators 117 are necessary minimum
power supply failure as follows:
On the second regulator: a drop as in graphic from Fig. 2, plus a reserve of several
tenths of volt (0.20.3 V) witch has to cover deviations that appears because of
the components dispersion on the loop that maintain reduce Vie of the control
element composed.
On the first regulator: a drop as in graphic from Fig. 2, plus a reserve of several
tenths of volt (0.40.5V) witch has to cover dispersion saturation voltage of the
circuit 117 and especially rectified voltage deviations as the calculated value, plus
half of the peek to peek value for the ripple rectified voltage (with the scope to
avoid saturation of the regulator at the inside peek of the ripple and transmission of
it to the load).
Adjustable voltage stabilizer scheme requires dialing to the precision integrated
regulator LM723. That will require all its skills to stabilize entire scheme.(Fig 3)
809
D2 and D3 diode protects those two regulators in case of short-circuit at the input
scheme if at the output is used an electrolytic capacitor greater or equal with 10 F.
As controlled transistor (T3) is used here the output doublet from integrated 723.
Regulators scheme with two integrated circuits LM217 (Fig 3) controlled by LM723
circuit allow to obtaine adjustable voltages 1.2-35 V.
Common radiator
Isolation
Fig. 3 Slabilizer with two serial regulators

Necessary current to open the D1 diode and the final transistor (from 723) is provided
by the R2 resistace. R3 resistance close to ground current from final transistor and ADJ pin
current of regulator 2.
When at the output regulators is sets maximum voltage, currents through R2 and R4
increase them too, must be retrieve by the divider from the output P-R7, together with
necessary minimum current through regulator 117 (5mA la LM117, 217 or even 10mA la
LM317) fot the stabilizer to work properly on no-load. But that would lead to a reduce total
resistance of the divider and to an exaggerate increased of the current through that when
output voltage increase to the maximum value.
Therefore, at the adjustable voltage in a various range, instead of a divider with
reduce resistance will use to take currents upper mention at the no-load operation a transistor
T and resistance R8. Transistor T is polarized by the constant voltage on the slider
potentiometer P, so it provides a constant current source.
Stabilizers performance from fig 3 are practically the one required by the precision
integrated regulator 723 so very good accuracy:stabilization in rapport with network: <2 mV,
stabilization in rapport with load: <10 mV,thermal derivative of the output voltage: < 0,53
10~ /C.
810
DC Stabilizer with Digital Control
)
(
,5

N
)
(
,5
,N
N
,
In Figure 3 reference element that is brought to the positive input of the error amplifier
from the circuit LM723 can replace with a prescribed value from a calculation system through
separation circuit. Through that method is obtained an output voltage with two adjustable
possibility of its value.Interface circuit by the PC and integrated stabilizer 723 comprising two
blocks: one for digital to analog conversion with DAC 08 circuit and one for galvanic
separation.(figure 4)[2],[3]
DAC 08 converter output voltage is determined by relationship:
(2)
$
P

)
(
5
)
(
,5
9 5
where I REF represent reference current is obtained by relationship :

)
(
5
(3)
Common radiator
Isolation
Fig. 4 Electic scheme of a voltage source numeric ordered

Its used a galvanic circuit that contained 2 optocouplers; one is in reaction load of the
input of the operational amplifier.
Isolation amplifier transfer function is basses on assuming the identity of the
characteristic of the two optocouplers. We will use relationship (4).
811
( 5
XL5
,
(4)
9 5
,
(5)
( 5
And for I 2 is obtained:
where: I 1 and I 2 represents collector currents corresponding to the two phototransistors.

It appears that this transfer function is linear and independent of the optocouplers
characteristics, with condition that refers to the characteristics identity for the two
optocouplers.
Following experimental verifications was noted output voltage from the voltage source
with Uies, for the input values between 0 255.[4] Measured values are listed in table 1
Table 1
V
H
8L
In
digital
10
25
50
75
100
125
150
175
200
225
250
255
1,6
2,8
4.7
6,7
8,4
10,3
12,3
14
15,9
17,7
19,6
20
3. CONCLUSION
The circuit is controlled by a sequence of 8 bits that are enrolls at the parallel port
through a program written in C++ language, values that can be enrolled through the keyboard
or with help of mouse. By using C4 capacitor connected to the no inverted input of the circuit
LM723 are eliminated sudden control changes for regulators 1 and 2 witch make that at the
output appears lower voltage variation respectively of the current through load circuit.
Analyzing experimental dates from table 1 it can observe a linear functionality of the
voltage source.
Control circuit allows obtaining at the output, directly proportional with the numeric
value entered from keyboard, a maximum value that not passes over 20V.
To obtain a maximum voltage range up to 35 V can modify the galvanic separation
circuit to increase its output voltage without exceeding the maximum input voltage of the
circuit LM723.
LITERATURE
[1] Mircea A. Ciugudean, Voltage stabilizers with linear integrated circuits, West Publishing,
Timioara, 2001
[2] Toma L., Acquisition systems and numerical processing of signals, Western Publishing
House, Timioara, 1996;
[3] Cristea D., Pnoiu C., Interfaces and peripherics, Mirton Publishing House, Timioara,
2007;
[4] Iordan A., Pnoiu M., Object-oriented programming. C++ Language. Laboratory guide,
Mirton Publishing House, Timioara, 2007;
812
DEVELOPMENT OF DISTRIBUTED CONTROL SYSTEM FOR

ROBOTS CONTROL BASED ON REAL-TIME LINUX PLATFORM
Marija Milievi1, Vladimir Kaplarevic2, Zoran Dimi3, Vojkan Cvijanovi4, Mirko
Buan5
Abstract: Development of a new control system for Lola institutes robots control is
based on Real-Time Linux platform on which is set OROCOS open architecture
software system, designed specifically for creating applications of this type. The
system consists of components of various structures and role in the system. They can
be distributed to different hardware platform with the same operating system, and to
communicate through the CORBA protocol. The system is designed as a modular
solution that allows easy connection and disconnection of components of different role
in the system, depending on the needs of applications, including those needed to
communicate with different types of hardware. That way the structure of the system
remains the same.
1. INTRODUCTION
Development of control systems in the world, these days depends on the
concept of open architecture and open source software platforms. The need for
cheaper, available development tools and requirements for high reliability systems set
the standards in the development of such software control systems.
Development of a new control system for Lola institutes robots control was
started based on a previous version of Lola Institutes robot controller, developed on
the OS-9 real-time platform. Unlike previous version, which now requires a specific set
of hardware, new version of the system is developed on Real-Time Linux platform,
where Linux kernel is patched with Xenomai. On this platform, OROCOS (Open Robot
Control Software) libraries are installed, designed specifically for the development of
applications for robot control. This way, a real-time part of the control system is
completely changed from the real-time part of a previous version. This control system
is modernized according to the development of similar control systems in the world,
achieving greater efficiency and speed of the system. Also, the price of the system
1
Marija Milievi, M. Sc. E. E., Belgrade, Lola institute d.o.o, (marija.milicevic@li.rs)

Vladimir Kaplarevi, Dipl. Math, Belgrade, Lola institute d.o.o, (valadimir.kaplarevic@li.rs)
3
Zoran Dimi, M. Sc. E. E., Belgrade, Lola institute d.o.o, (zoran.dimic@li.rs)
4
Vojkan Cvijanovi, Dipl. Eng. Inf, Belgrade, Lola institute d.o.o, (vojkan.cvijanovic@li.rs))
5
Dr Mirko Buan, Belgrade, Lola institute d.o.o, (mirko.bucan@li.rs))
2
813
Development of distributed control system for robots control based on Real-Time Linux platform
itself and its development becomes much smaller [1].

2. OROCOS
OROCOS is a European project, started on September 1st, 2001. Three
laboratories are participating: the Katholieke Universiteit Leuven, the Laboratory for
Analysis and Architecture of Systems (France) and Kungl Tekniska Hgskolan
(Sweden). Many other European laboratories participate in the discussions and
designin. Goal of this project is to develop software with publicly available code (open
source) that provides a functional basis for robot and machine tools control. This
intended to be platform independent, but it is also thought as application independent
software [2].
OROCOS code base is divided into modules or libraries. There are generally
three types of modules:
1. Supporting modules. That is, the software without functional robotics
contents, that is needed to build a working robot control system. For
example, 3D visualization and simulation; numerical library; software
component configuration tools; real-time operating system; inter-process
communication; documentation writing tools; etc.
2. Robotics modules. That is, the software that implements specific robotics
algorithms: kinematics and dynamics of kinematic chains; servo controllers;
Bayesian and Neural Network estimators; motion planners for mobile
robots, serial and parallel manipulators; etc. The Robotics modules make
use of one or more of the Supporting modules.
3. Components. That is, the CORBA (Common Object Request Broker
Architecture) objects with their IDL (Interface Description Language)
descriptions. These components are constructed out of the previous two
types of modules, and they are the building blocks with which users
assemble their robot control software environment.
0RVWLPSRUWDQWVWDQGDUGVWKDWWKH252&26SURMHFWVXSSRUWVDUH:
1. The CORBA IDL for object embedding. Through the IIOP (Internet Inter-Orb
Protocol), all different implementations of CORBA are interoperable.
CORBA also has a real-time extension, which is very important for robot
control.
2. The structured document language XML for configuration and data files.
3. Modelica, for modeling of dynamical systems.
4. DocBook and LATEX for documentation.
OROCOS software system uses support from other open source software
projects:
1. Octave, is a very rich numerical library, that uses the same foundations as,
and in its scripting form is very compatible with, Matlab.
2. Real-Time Linux and/or RTAI, which are real-time kernel extensions to
Linux; or the Linux-independent eCos.
3. Xenomai is a real-time development framework cooperating with the Linux
kernel, in order to provide a pervasive hard real-time support to user-space
applications.
4. Comedi, , is a library for real-time device
5. drivers, such as AD/DA cards.
814
Marija Milievi, Vladimir Kaplarevi, Zoran Dimi, Vojkan Cvijanovi
6. Various packages and open formats for 3D visualization: VRML, Java3D,

OpenGL with OpenInventor or Coin3D, etc.
The Orocos project supports four C++ libraries:
1. OROCOS RTT (Real-Time Toolkit), is not an application in itself, but it
provides the infrastructure and the real-time functionalities to build robotics
applications in C++.
2. OCL (OROCOS Component Library) provides some ready to use control
components.
3. The Orocos Kinematics and Dynamics Library (KDL) is a C++ library which
allows to calculate kinematic chains in real-time.
4. The Orocos Bayesian Filtering Library (BFL) provides an application
independent framework for inference in Dynamic Bayesian Networks, i.e.,
recursive information processing and estimation algorithms based on Bayes'
rule, such as Kalman Filters, Particle Filters (Sequential Monte methods),
etc.
Orocos applications are composed of software components, which form an
application specific network. When using Orocos, you can choose to use predefined
components, or build your own component, using the Orocos Real-Time Toolkit.
Components work together, cooperatively. For example, the application template for
motion control contains components for path planning, position control, hardware
access and data reporting. The components are chosen as such that their interfaces
are compatible.
A single component may be well capable of controlling a whole machine, or is
just a small part in a whole network of components, for example an interpolator or
kinematic component. The components are built with the "Real-Time Toolkit" and
optionally make use of any other library (like a vision or kinematics toolkit). Most users
will interface components through their (XML) properties or command/method interface
in order to configure their applications. Besides defining the above component
communication mechanisms, Orocos allows the Component or Application Builder to
write hierarchical state machines which use these primitives. This is the Orocos way of
defining your application specific logic.
3. CONTROL SYSTEM ARCHITECTURE
The system is designed as a modular solution that consists of components.
Components are basic units which executes one or more action with a specific activity.
They are created using RTT OROCOS libraries [2]. This way, OROCOS provides an
interface for components design. Component can also use the native libraries of the
operating system. There are four policies for the control of component activities:
NonPeriodicActivity, PeriodicActivity, SequentialActivity and SlaveActivity.
Each component inherits a public interface from its base class (TaskContext),
which defines primitives for component interactions: Events, Methods, Commands,
Properties and Data Port, figure 2.
815
Fig. 1 OROCOS component interface

In the development of our system, basic components of OROCOS OCL library
are used: TaskBrowser, DeploymentComponent, ReportingComponent [2].
The TaskBrowser is a component used to interact with other components. Each
OROCOS component inherits a standard interface through the TaskContext base class,
which is used by the TaskBrowser to interact. It works like a text console, receiving
command lines and executing them.
The DeploymentComponent is a component used to load and configure other
components from descriptions in XML files. The system configuration is performed by
connecting the components with their peers and their Data Ports. With all connections
done, the Properties of each component, as well as the component activity are updated
from that file.
The ReportingComponent is used to monitor and capture data exchanged by
other components. It works by connecting as a peer of the components to be monitored
and attaching to the specified Data Ports .
Components that are designed by the developer of the system make the basic
architecture of the system. It is based on the components model for different independent
functional blocks of a system. Each component model can then be instantiated in a
component by using the specific parameters of a given robot. The basic block model of
the system is given in Figure 2.
The Sampler is the component that generates the sampling rate of the control
loop and synchronizes the other components. It generates an Event which is received by
the other components, automatically triggering a control cycle. Sampler is configured in
the XML file to generate periodic Event on every 1 ms.
The Sensor model abstracts the sensors of the system. It has a Data Port where
it writes the values which are read from the sensors. This Data Port is represented by a
vector which size depends on the number of joints of the robot.
The Actuator component model abstracts the system actuator. Similar to the
Sensor component model it has a virtual member function which is called in response to
an Event from the Sampler. It has a read-only port where the component can read the
vector of actuator values from. Those values should be applied to the robot.
816
Marija Milievi, Vladimir Kaplarevi, Zoran Dimi, Vojkan Cvijanovi
Fig. 2 Blok model komponenata za upravljaki deo sistema

The Controller component model abstracts the system controller. This
component on every 5 ms sends reference value that has to be reached, writing it into
one of his Data Ports. Write from Sensor component to the Data Port triggers a call to
a virtual function called which should compute the control signal and write it to the act
Data Port.
The above models are the basic components of the control system. In addition
to these systems include more models. One of them is Joint component model. This
model represents a component that communicates with the hardware of joint of the
robot, and the number of components that are derived from this model depends of the
number of them. Each of them is associated with components Sensor and Actuator
which Data Ports give signals of all robots axes in the vector form, and because of that
it is necessary to create components for multiplexing the signals from the N Joints, and
similar, for demultiplexing the vector output, so the components can be connected to
each other.
Component that calculates the reference value that is sent to the Controller
component is called Interpolator. It is one of the most important components in the
system. In the development of our new system Orocos KDL library is not used. This
component implements the algorithms developed in the Lola Institute [3]. The interface
of this component to the rest of the system required a change in the original form of
these algorithms, so this component can be included into system that is designed on
Orocos platform.
4. DISTRIBUTION OF THE COMPONENTS
As the presented system is designed as modular and built from components,
they can be distributed to different hardware platforms with the same operating system
and to communicate with each other over the network. Components are linked using
817
CORBA deployer application. Connecting data flow ports of components is done by

defining connections. When components are distributed using the CORBA deployment
component, it is necessary to declare a proxy component in one of the XML files and
connect to a port of that proxy.
5. CONCLUSION
The main idea and the goal of new Lola Institutes control system development
is to provide all necessary functionality for robots, machine tools and similar devices
control in a more efficient way. Also, the idea was to create modular solution so that
the different functionality could be included or excluded in accordance with the needs
of applications without changing the structure of the system. This allows faster and
easier possible system upgrades. The system can be distributed to different hardware
platforms. During its development modern methods and open-source packages are
used that makes development easier and not expensive.
6. ACKNOWLEDGMENTS
This work was created within the research project "Development of the
devices for pilots training and dynamic flight simulation of modern combat
aircraft: 3 DoF centrifuge and 4 DoF spatial disorientation trainer" that is
supported by the Ministry of Science and Technological Development, Republic of
Serbia.
LITERATURA
[1]
[2]
[3]
818
Milievi, M., Vidakovi, J., Dimi, J., Trgovevi, S., Modern open architecture
control systems for machine tools and robots control, 36th JUPITER conference,
32th symposium NU-Roboti-FTS, Proceedings, ISBN ISBN 978-86-7083-696-9,
str.4.41-4.46, Faculty of Mechanical Engineering University of Belgrade, May
2010.
OROCOS- Open Robot Contol Software web site - http://www.orocos.org
Kvrgi, V., Development of intelligent systems for industrial robots control and
programming, PhD thesis, Faculty of Mechanical Engineering University of
Belgrade, 1998.
DEVELOPMENT OF THE CONCEPT OF INTERACTIVE EDUCATION

IN MECHATRONICS
Miroslav Rogi, Bojan Kneevi, Branislav Risti 1
Summary: This paper presents the methods of students' education at the Faculty of
Mechanical Engineering in Banja Luka in the fields of mechatronics and automation of
manufacturing processes. A didactic approach that has been made is a combination of
conventional teaching with an innovative system dominated by the practical work of
students. The final goal is to enable students to acquire new knowledge, develop
creativity, practical skills and to prepare them for future professional challenges in a
creative and innovative way. Since the implementation of the new approach we have
already implemented two projects with two generations of students. Both projects are
described in this paper.
Key words: industrial processes; automatic control; didactics;
1. INTRODUCTION
Automation of industrial processes, as a scientific discipline, is taught at many
universities. It is a standard field of study at mechanical and electrical engineering
faculties coupled with a high interest of the students. In the teaching process, which by
modern standards of studying has been goverened by the following principle: the more
knowledge acquired during as less time as possible, it is necessary to focus the work
with students on mastering practical skills. Access to real industrial processes, in which
students could extend their knowledge and become familiar with new technologies, is
not possible because the industrial production can not stop for the sake of training,
testing or research. The real answer to the question of how to prepare future engineers
for the design of industrial automation systems is the design and implementation of
laboratory apparatus with full engagement of students. This paper will demonstrate
practically implemented laboratory equipment and devices that are being used at the
Faculty of Mechanical Engineering in Banja Luka.
2. PROJECT I: A DIDACTIC BOARD FOR AUTOMATION PROCESS CONTROL
In order to master the training segments that are important for the control of
operating machines, we have created a project called: open architecture learning
1
University of Banja Luka, Faculty of Mechanical Engineering (rogic@urc.rs.ba, bojan.knez@blic.net,

bristic@blic.net)
819
Miroslav Rogi, Bojan Kneevi, Branislav Risti
board for automatic control of machines (Fig. 1). The project task required
development of didactic boards with the following functions:
control of DC motor for driving machines,
control of temperature of working environment with heater and fan,
sequential control of machine operations by different sensors and actuators
(switches, limit switches, relays, signalization devices, etc.)
Fig. 1 Students practice on didactic board project and didactic board (appearance)
The main element of this apparatus is a programmable relay [2]. In addition to
mastering of programming PLC, the goal is to achieve the possibility of connecting
various sensors with inputs and actuators with outputs of programmable relay. In order
to achieve this in a simple manner, terminals of all installed components are connected
with a series of line-up terminals. All implementations of electrical schemes are
conducted by connecting line-up terminals with each other (Fig. 2).
In general, programmable relay inputs are digital, but all of them are not the
same. Some of them can be used as analog, current and voltage inputs. The other
ones are the digital inputs, but for faster changes (for fast counters). The device has a
display for displaying programmed messages, the desired values and possible
adjustments [2]. For exercise regulation, the temperature control model is used which
consists of temperature sensors, an electric heater and a fan. The temperature sensor
changes its resistance. To obtain the voltage or current signal, the signal converter is
used as an interface between the sensor and inputs of programmable relay. The
software is applied in implementation of regulator using a hysteresis control of
temperature.
Apart from measuring temperature, one segment of the apparatus is measuring
the speed of rotation of the motor. The system is made up of a drive motor which is
equipped with speed control from an independent source of variable voltage. Tachogenerator, which has a solid connection with the shaft of drive motor, generates a DC
voltage at its terminals that is proportional to the number of revolutions. Generated
voltage leads to analog input of programmable relay, and the speed is calculated by
the software. There is also an incremental encoder on the same motor shaft. Pulses
from the encoder are conveyed to rapid digital inputs through the interface.
There are voltage dividers with potentiometers which adjust the voltage in the
820
Development of the concept of interactive education in mechatronics
range of 0V to 10V that are used to simulate the change of analog values, except
signals from the temperature sensor (its converter) and the tacho-generator. A set of
simple buttons and switches (S1-S6) is used to simulate the switches and limit
switches of the machine. Colored lights (H1-H4) that are powered by line voltage are
used to visualize the state of outputs. Starting of the motor in the system of speed
measuring as well as the fan and the heater is performed by additional electromagnetic
relays because they are larger power consumers. Also, mushroom switch is installed.
Fig. 2 Electric scheme of didactic board

Described apparatus has been already in use for two years at the Faculty of
Mechanical Engineering in Banja Luka in the course named Automation and
Visualization. In order to provide students with a quality presentation materials in the
teaching, the emphasis is placed on the active involvement of students in the design
and realization of devices that would come to have practical application in practice or
as laboratory equipment to be used in teaching future generations the same or other
subjects.
In the selection of a project task, attention was paid to the following factors: the
technical demands of the devices to be realized; the time needed for implementation;
the available number of teachers and students; the use-value of the final device; the
level of interdisciplinarity; the total costs of implementation.
The device itself should not be too demanding. The level of complexity needs
to match the prior knowledge of students. Proper assessment of the time required for
implementation is crucial. The number of students and teachers who are the holders of
the design and implementation determines the complexity of devices that must be
implemented in one teaching cycle or one semester. The advantage is given to
projects that connect different disciplines which combine the experience and
knowledge.
3. PROJECT II: AUTOMATIC CONVEYOR
The next project is an example of a device that has practical application and
821
can be used as a multipurpose automated transport device - paternoster (Fig. 3).

The initial idea was to realize a device that will serve as a mobile platform that
will be used for robotic welding. At the front of the device, a worker who serves
automated system fixes the position for welding and removes positions that have
already been welded.
The transport system transmits positions that are ready for welding to the point
where the robot performs welding. At the same time, welded elements go back to the
front of the unit where they are taken off and new ones are placed. In order to realize
this idea, a circular elevator - paternoster seemed to be the best solution.
Fig. 3 Students practice on automatic conveyor project and control and drive part of
conveyor
Unlike the previously described laboratory equipment which is predominantly
based on electronic components, the transportation device has prominent mechanical
elements. Platforms for carrying the positions are made of metal sheets, whereas the
portable mechanism is realized by a pair of roller chains.
The basic elements of automation of this device are based on a drive with
induction motor with frequency converter and a programmable control relay [3,4].
Electrical scheme shown in Fig. 4.
Automation is implemented as follows: Parking position of the platform where
welding is performed needs to be fixed. The platforms must be stopped in the same
place, determined by the position of a magnetic sensor. The magnetic sensor is
activated by permanent magnets which exist on all platforms, and its output leads to
the input of the programmable relay which realizes automatic control. A worker who
serves a machine calls the preferred platform in the parking position with buttons on
the instrument panel.
Calling is like that of standard lifts. The speed at which the platform moves is
the maximum possible. When the target platform is entering in the parking position, it is
necessary to reduce the speed of movement in order to ensure accurate stopping.
Speed should be reduced approximately 100 mm before stopping. A signal from the
inductive sensor which detects the passing of chainring teeths serves to determine the
time of deceleration. Speed control is realized by using the frequency converter [4]
822
Development of the concept of interactive education in mechatronics
which is controlled by intelligent input associated with programmable output relays.

The buttons for calling elevator's platforms are not connected to the digital
inputs of programmable relay directly because of their large number that is greater
than the available number of inputs.
To overcome the problem of an insufficient number of digital inputs of the
programmable relay, it was necessary to make the coder as a separate device. Its task
is to turn each active call into logical combination of three states that are related to the
inputs of the programmable relay. Software is used to decod them. In this way, the
number of necessary inputs for the reference platforms is reduced from seven to three.
The interface between a man and a machine is in the form of the control panel with
display and available function keys which is mounted on the control box (Fig. 3) and is
available for reading information and program management.
Fig. 4 Electric scheme of automatic conveyor
Fig. 5 Electric scheme of coder

823
4. CONCLUSION
In this paper we presented methods of acquiring knowledge which are realized
in the teaching process at the Faculty of Mechanical Engineering Banja Luka, the
study program of Mechatronics. Incentives for the development and application of the
presented methods of education and teaching follow the development trend of
industrial production and needs of the companies in the industry for young quality
personnel. Current and future development of industry in our region requires
engineering personnel that will adapt to the conditions of production in a very short
time and become an active part of the engineering team that will carry the company's
development. The main goal is to raise education of young engineering personnel in
the field of industry to a higher level with a significant share of practical work which
encourages creativity, skill development, teamwork and establishes the necessary
experience in the implementation of practical applications.
LITERATURE
[1] M. P. Papoutsidakis, M. P, Andreou, I. K, Chamilothoris, G. E. (2008). Challenging
Educational Platforms In Mechatronics: Learning Framework And Practice. 5th
WSEAS / IASME International Conference on Engineering Education (EE'08).
[2] Zelio Logic 2 Smart Relay (2007), Users Manual, Schneider Electric.
[3] LOGO! Manual Edition (2003), Siemens AG, Nuernberg.
[4] Frequency converter series KN 50 and KN 100plus (2006). Users Manual, KonarMes, Zagreb.
824
EFFECTIVENESS DETERMINATION OF ELECTRONIC DEVICES PREVENTIVE

MAINTENANCE
Ivan B. Krsti1, Boidar V. Krsti2, Dragan I. Milosavljevi3
Summary: Proper form of maintenance systems allow as to make optimization in
choice of the most suitable maintenance system. Such problem may be solved if it is
possible precise determination of all important requirements and constrains. The base
of presented methodology make reliability parameters of analyzed technical system,
obtained by monitoring of its behavior, considering real exploitation conditions, and
cost of maintenance.
Key words: electronic devices, maintenance, optimization, availability, reliability,
costs.
1.
INTRODUCTION
Preventive maintenance represents set of arrangements aimed to prevent

failure of devices and to extend their life cycle. Set up of preventive operations requires
contradict requirements: On the one hand one should increase reliability of equipments
by employing preventive arrangements, which require significant time consumption,
and to another hand prolonged delays of expensive devices caused with preventive
maintenance are not desired from economical point of view. That is the reason that we
consider here optimal conditions to manage preventive maintenance with minimal
production delays.
Research in the field of preventive maintenance enable not only maintain of
reliability level of radio electronic devices, provided in design and production phases,
but make possible increase of that level with increasing number of prevented defects
by timely detection of reasons for defects and its removal. In scientific approach to
preventive maintenance is very important study of development and multiplication of
defects.
2.
CHARACTERISTICS OF DEFECTS OF PREVENTIVE MAINTAINED

ELECTRONIC DEVICES
Analyzing reasons for which delays arise shows that before failure there are
changes in physical chemical structure of elements, but device continue to operate in
1

Faculty of Mechanical Engineering, Serbija, bkrstic@ kg.ac.rs, Serbija
3
Faculty of Mechanical Engineering, Serbija, dmilos@kg.ac.rs
2
825
Ivan B. Krsti, Boidar V. Krsti, Dragan I. Milosavljevi
spite of some their parameters do not fulfill proposed technical conditions. Such
defective parts in many cases do not cause device failure, but may be potential failure
carrier.
There is possibility that in device working process one can detect and remove
defects which possibly may cause failure, but which may be prevent. Some defects is
not possible to detect and they cause unavoidable defects.
Defects that may be prevent are the most often defects that follow visible
physical chemical changes of element structure. The main parts of these failures are
those connected with relative slow changes of parts parameters and may be called
gradual failures. However it is not possible to prevent all gradual failures, because it
is not always possible to take control of inner structure and parameters of parts
changes. On the other hand some sudden failures may be predicted. There are in
principle possibilities to prevent some number of sudden failures on the basis of
statistical laws of their appearances. Failures that may be prevented have two
properties:
- Prediction parameters of part are known, and gradual changes of their
structure follow certain laws. Predictable parameters are those which, in any moment,
describe changes of physical chemical structure of elements and allow control of their
reliability.
- Prediction parameters of part are unknown, but statistical law of time
distribution of failure appearance is known.
For parts having first property, process of multiplication of defects may be
described with specific distribution of random getting of parameter D t out of
tolerance
D kr :
f p1 D t =
where
>D t D 0 K D t @2
1
exp
2V 2
V 2S
(1)
V mean quadratic deviation,
KD coefficient of time change of parameter D t .
D t is considered as random variable, which initially has

specific distribution f D 0 , and may be described with normal law of mathematical
expectation D 0 and with mean quadratic deviation V , in which mathematical
expectation follows linear law: D t = D 0 K D t .
In that case
In second case, when forecasting parameter is not available, detection and

amendment of defect elements should be achieved under assumption that statistical
law of probability distribution is known. It is well known that for some parts of motor
vehicles Weibulls distribution is suitable for use. Specific Weibulls distribution may be
expressed in following way:
tk
k k 1
f p2 t = t exp
t0
t0
(2)
where k and t0 parametric law of distribution.

Choosing values of parameters k and t0 lead to needed statistical probability
826
Effectiveness determination of electronic devices preventive maintenance
distribution of work of different parts, which is for engineering calculations with

satisfactory accuracy.
For many other elements such as electromechanical or mechanical insulators
shortened normal law may be used as follows
f p2 t =
wWhere
t Tsr 2
s
exp
2V 2
V 2S
Tsr mean time of work without failure,
(3)
s normalization multiplier.
The normalization multiplier may be calculated as follows
1
s=
t T
t T
) b sr ) a sr
V
V
where t a , t b boundaries of interval of random variable t change.
In the case when t a = 0 and t b = f it may be obtained
s=
(4)
(5)
T
0.5 ) sr
V
Function of distribution of shortened normal law has following form
F t
t Tsr
)
0.5
T
0.5 ) sr
V
where
Tsr
! 2 and s | 1
(6)
As examples of failure that can be prevent we may use failures of electro

vacuum devices which are caused by gradually damage of oxide film on cathode,
failure of electro motors brushes because deterioration, changes of nominal values of
resistance, capacitor, inductance and leakage of electrolytic capacitors caused by
damage of insulating material.
For estimation of equipments in light of efficiency of applied preventive
arrangements it Is necessary to introduce index which indicate character of equipment
failures. As such index is suitable to use rate of failure that may be prevent and failure
that may not be prevented, for certain period of exploitation Te , which is called
coefficient of failure character
A Te
n p Te
A Te :
n p Te nn Te
(7)
827
where
n p Te , nn Te number of failures which may be and which may not be
prevented, respectively.
Failure character of radio electronic devices and their parts determines
- Properties of material structure of done parts and technology of their
production. These properties are featured first or second type of radio elements taking
account on resistance and other influences.
- Exploitation process influence: period of parts use in device, influence of real
electrical loading, climate conditions, maintenance quality etc.
Failure character coefficients vary not only for different parts, but for different
influences on the same part. If one suppose that in long term exploitation influences
vary slowly than failure character coefficients will depend on real period of exploitation
f Te only. Failure character coefficients A To , which determines number of
failures that can be prevent, may be calculated on the statistical data basis obtained by
analysis of failures of devices which do not have preventive maintenance.
Data given in tables 2 and 3 show that preventive maintenance may keep from
defect large number of failures. Thereby probability of failure prevent Ppv depends on
quality of detection of defect elements, and in general may be determined as:
n pv
Ppv
np
(8)
where
n pv n pn - total number of prevented failures that initiate defects arose
np
before application of routine preventive measures,

n pv , n pv - number of defects detected and undetected during preventive
maintenance, assuming that all undetected defects lead to failure.
Main aim of preventive maintenance is to exclude defected elements which
cause failure.
3.
EFFICIENT OF PREVENTIVE MAINTENANCE OF EQUIPEMENTS
Effect in reliability of preventive maintenance application may be evaluated

through effectiveness parameter of preventive maintenance, which is represented with
rate of achieved working interval up to failure of preventive treated equipment to
equipment without such treatment, as:
W
where
TO pf
TO
(9)
TO operational interval up to failure of equipment without preventive
maintenance,
TOpf operational interval up to failure of equipment with preventive

maintenance.
Effectiveness of preventive maintenance W displays quantitative increase of
level of reliability as results of preventive maintenance. If one assume that probability
828
Effectiveness determination of electronic devices preventive maintenance
of reliable work, without failure, of equipment with preventive maintenance and without
it varies exponentially than:
n
(10)
n pf
Here n number of failure of equipment without preventive maintenance,
W
which may be expressed as
n
where
n pf
n pf n pv
(11)
n pn nn - number of failure of equipment witht preventive maintenance,
nn number of failure of equipment with did not fixed with
preventive
maintenance, that is unavoidable failures.

Use of expression (11) with help of expressions (9) and (10) leads to
1
1 A Te Ppv
(12)
Expression (11) gives quantitative dependence between effectiveness of

preventive maintenance, character of equipment failure and probability of failure
detections. Dependence of effectiveness of preventive maintenance on failure
probability and on coefficient of their character W
f A Te , Ppv gives estimation
of justification of preventive maintenance.

For effectiveness estimation of preventive maintenance coefficient of
effectiveness K ep may be employed. It is necessary to determine dependence of
effectiveness coefficient K ep and effectiveness of preventive maintenance
W.
Effectiveness of preventive measures may be estimated according to tame

saving up to failure, in comparison to failure of equipment without preventive
maintenance.
Increase working life is achieved with prevention of defects. In that case
effectiveness preventive coefficient may be estimated by rate of number of prevented
failures n pv to total number of defects n , developed in equipment:
K ep
n pv
n
(13)
Use of expressions (10) and (13) following expression may be obtained:
1
1 K ep
(14)
Expressions (13) and (14) lead to quantitative estimation of preventive steps

without detailed examination of prevention process of analyzed equipment.
4.
CONCLUSION
In planning of preventive operations, technical staff is faced to contradictory

requirements: on the one hand one should increase reliability of equipments by
employing preventive arrangements, which require significant time consumption, and
829
to another hand prolonged delays of expensive devices caused with preventive

maintenance are not desired from economical point of view. That is the reason that we
consider here optimal conditions to manage preventive maintenance with minimal
production delays.
Research in the field of preventive maintenance enable not only maintain of
reliability level of radio technical systems, provided in design and production phases,
but make possible increase of that level with increasing number of prevented defects
by timely detection of reasons for defects and its removal. In scientific approach to
preventive maintenance is very important study of development and multiplication of
defects.
Analyzing reasons for which delays arise shows that before failure there are
changes in physical chemical structure of elements, but device continue to operate in
spite of some their parameters do not fulfill proposed technical conditions. Such
defective parts in many cases do not cause device failure, but may be potential failure
carrier.
There is possibility that in device working process one can detect and remove
defects which possibly may cause failure, but which may be prevent. Some defects is
not possible to detect and they cause unavoidable defects. Part of these failures,
regardless on reasons which cause them, is not possible to detect causing equipments
unavoidable failures.
Effect of achievement of necessary reliability level by application of preventive
operations may be estimated with coefficient of effectiveness of preventive operations,
which is represented by rate of accomplished time up to failure of preventive
maintained equipments end equipments without prevention.
Effectiveness of preventive operations may lead to qualitative estimation of
preventive operations, without examination of preventive process, which was main aim
of present paper.
5.
REFERENCES
[1]
Bass. M.S., Kwakernak H.: Rating and Ranking of Multiple Aspect Alternatives
Using, Fuyy Sets Automatics, Vol.13, No. 1, 1977, p. 47-58,
S. Vukadinovi: Elements of the Probability Theory and Mathematical Statistics,
Economy Review, Belgrade, 1981. (In Serbian)
B. Krsti, V. Lazi, R. Nikoli, V. Raievi, I. Krsti, V. Jovanovi: Optimal strategy
for preventive maintenance of the motor vehicles clutch, Journal of the balkan
tribologikal association, Vol.15, No 4, (2009), 611-619
B. Krsti, V. Lazi, V. Krsti: Some views of future strategies of maintenance of
motor vehicles, Tractors and power machines, Vol.15, No.1, 2010, p.42-47
Krsti, B., Tehnical serviceability of motor vehicles and engines, University of
Mechanical Engineering, Kragujevac, 2009. (In Serbian)
[2]
[3]
[4]
[5]
830
FURTHER RESULTS ON MODELING, INTEGRATED DESIGN AND

SIMULATION OF A MECHATRONIC SYSTEM WITH FPGA
Mihailo P. LAZAREVI1*, Vasilije VASI, PhD*, Ale HACE2, Karel JEZERNIK
Summary: The basic idea of modeling, integrated design and simulation of
mechatronic product is to combine benefits of the mechanical multibody system model
with the features of advanced drive and control.
The starting point for multiphysics modeling and simulation is multibody system
dynamics with the practical case washing machine. The motion equations are
obtained using modified symbolic Lagrange equations in a covariant way, applying the
Rodrigues method. In this paper it will be presented a novel means of designing a
simple and effective torque control for PMSM based on the Field Programmable Gate
Array (FPGA) technology. Finally, the effectiveness of both suggested approaches is is
demonstrated on mechatronic system simulation (e.g. co-simulation approach) on
planar washing machine model. There are given also practical hints for promising
hardware-in-loop simulation for experimental evaluation.
Key words: mechatronics, rapid-prototyping, multi-body dynamics, FPGA, drive control
1
INTRODUCTION
The design of mechatronic product is very complex and comprehensive task,

since there are combined solutions and technologies from intersecting fields of
mechanical, electrical, control and software engineering. Although, there are present
some industrial guidelines as design methodology for mechatronic systems e.g. Vmodel, where the product designer decompose the product into the hierarchical
sequence (modules). These modules are then treated as independent items with own
modeling and simulation procedure. To implement mechatronic product development
process, the product designer often faces numerous challenges e.g. lack of crossfunctional knowledge, difficulty implementing an integrated product development
solution for all disciplines involved in mechatronics product development and inability
to understand the impact a design change will have across disciplines. Additional
complexity is the expectation for real-time coordinated response from the mechatronic
product (e.g. robot) and its subsystems. The proper answer for the parallel and
1
full prof., Mihailo P. LAZAREVI, Department of mechanics, Faculty of mechanical engineering, University
of Belgrade,Serbia, Kraljice Marije 16, Belgrade 35 , SRB-11120 Belgrade, Serbia, e-mail:
mlazarevic@mas.bg.ac.rs
2
assoc. prof., Ale HACE, Institute for robotics, Faculty of electrical engineering and computer science
University of Maribor, Smetanova ulica 17, 2000 Maribor, Republic of Slovenia, e-mail: ales.hace@uni-mb.si
831
Mihailo P. Lazarevi, Vasilije Vasi, Ale Hace, Karel Jezernik
distributed process control as reliable (industrial) solution is enabled by the FPGA

(Field-programmable Gate Array) technology. In this article, we are going to present
the case of the mechatronic system multibody system dynamics and the drive, where
the simulation is considering mechatronic systems dynamic with various drive
properties.
1.
THE MECHATRONICS SYSTEM AND MODULAR APPROACH
The design of mechatronic systems could be roughly distinguished in four

stages: modeling, analysis, controller synthesis, and hardware-in-the-loop-simulation
(HILS) to test the system in the lab (Figure 1a).
a) Mechatronic system design

b) Modular modeling and simulation
Fig. 1 Mechatronic system modular decomposition and modeling description
The model description of each mechatronic module could be done also
independantly as physical, mathematical and behavioral description (Figure 1b). On
the market could be found some very capable software tools, which enables coupling
the simulation on different physical domains (e.g. LMS Virtual.Lab, MD Adams,
PSpice).
1.1 The multibody system dynamics and modelling
Main goal by analysis of multibody system dynamics or dynamics of
mechanical systems is therefore to develop such formalism procedure (modeling,
formulation, numerical) that automatically generate the equations of motion for
multibody systems.
The multibody dynamic system (mechanical part) could be presented on many
different ways e.g. tree structure. Modeling mechanical systems is mostly done by
numerical methods, but the application of symbolic equations provide numerous
advantages. One of them is providing more efficient models for parametric evaluation
as well as for real time-simulation and control.These theoretical advantages, this
method could be useful for certain practical engineering tasks e.g. easily obtaining
and modeling 3D models of MBSs.
832
Further results on Modeling, Integrated Design and Simulation of a Mechatronic System with FPGA
1.2 The multibody system dynamics and modelling

The household appliances (washing machines) are multibody dynamic systems
with changing loads, which require advanced control techniques such as use of notch
filters to avoid exciting the system at resonant frequencies, or gain scheduling to adjust
for changing load inertia. This also requires customizing the torque loop in addition to
position and velocity, while others may require custom drive electronics including
custom commutation algorithms for optimal performance. An optimal washing process
has to take into the account key factors (or Sinners factors) - temperature (heating),
chemical action (detergent, bleach, water), mechanical action (spin rotations by the
drum of washing machine), and time (duration of washing cycle). All that requires
multiple input and output (control) signals for multiple drives and sensors , which have
to be managed by the control unit in very short response time in very common way
with DSP (Digital Signal Processing)/MCU (Microprocessor Control Unit) or advanced
way - example the Permanent Magnet Synchronous Motor (PMSM) and FPGA (Field
Programmable Gate Array). Very promising solution is hybrid FPGA/DSP technology,
where configurable platform offers more performance and I/O adaptability, while still
providing low cost for high volume production. An important advantage of using an
FPGA for the controller is also that some additional functions like protections, steering,
monitoring etc., can be added with no additional resources and almost no drawback in
performance showing in (Figure 2a) - conceptual representation of a hybrid switching
controller. The FPGA concurrency allows executing the logic dedicated to the
proposed hybrid direct torque control (DTC) and any additional logic denoted to
protections simultaneously. In fact, any algorithm can be added to control one while
there are available resources. As presented on the Figure 2a - the decision logic
selector that controls the switches, or more precisely, that generates the switching
control input vector S (S1, S2 , S3). Its purpose is to monitor the signals that can be
measured or estimated and decide, at each instant of time, which candidate input
switching vector S should be put in the feedback loop with the process. In the
supervisory control, the supervisor combines continuous dynamic with discrete logic
and is therefore hybrid system. Motor torque Te is expressed as imaginary part of
complex product between rotor flux and stator current. The FPGA based DTC PMSM
system is meant to be operated according to certain established rules and principles,
which are presented in Figure 2b as infinite state machine (ISM) in three domain
principle. The rectangular ISM describes normal machine operation including
start/shut/up/down operation, hybrid controller normal operation. The triangle ISM
represents a potential risk for the PMSM with inverter if a major driving parameter drifts
into it, or if a failure occurs. The circle domain is out of the limits of the safe field of
variation of the parameters and system automatically switches off.
833
a.) Hybrid control od AC motor
b.) State CAD diagram
Fig. 2 FPGA Supervisory control of AC motor and state CAD diagram

1.3 The mechatronics system modelling and FPGA
Main issue to conduct the mechatronic system modelling is to assemble the
simulation model of mechatronical system as a multiphysical system - to interconnect
the simulation models of particular subsystems of certain physical nature into one
resulting multiphysical (multidisciplinary) modeling. There are two different approaches,
co-simulation (tight and weak coupling) and uniform modelling with different formalism
- equations (algebraic, differential), dynamic blocks, multipoles and bond graphs.
Nevertheless, the natural basis of simulation models of mechatronic systems are the
multibody models that are feedback controlled, where the mathematical models and
corresponding simulation models are being developed for systems from one physical
domain. After modeling, simulation and control synthesis - the designer has to proceed
with the final stage, hardware in loop (HIL) simulation. Namely, without the real time
simulation with actual analog and digital signals the mechatronic product could not be
properly validated. In the HILS the stepwise transition from the pure model
representation of the mechatronic system to actually mounted mechanical, hydraulic
and electrical components takes place.
2
THE MECHATRONICS SYSTEM DYNAMICS CASE STUDY
The case study of mechatronics system dynamics is the (planar) model of

washing machine, the tub assembly without the cabinet (Figure 3a). The 2-D
multibody system model, which could be presented in classical approach (Figure 3b)
or robotical approach , consists of two rigid bodies which are supported by cabinet of
washing machine. The first rigid body, with its centre of gravity at point T, is called the
basic body. Its links to the surroundings consist of a bilinear spring support, linear
viscous damper and element with implemented dry friction. All of the elements of vibroisolation are collinear with the axis of the co-ordinate system xy. The second rigid
body, with its centre of gravity at point Te, is called the rotor. It is driven by torque M(t)
and attached to the basic body by bearings at point S. The model has four degrees of
freedom (DOFs) where coordinates x, y and M are needed to determine the motion of
the basic body.
834
Further results on Modeling, Integrated Design and Simulation of a Mechatronic System with FPGA
a) Tub assembly of washing machine with

suspension
b) Planar model - classical approach
Fig. 3 2D model of analyzed washing machine

The x and y describe the horizontal a vertical translation of the basic body,
respectively, and M describes its rotation. To characterize the rotation of the rotor the
additional coordinate \ is used. The M and the \ start at x-axis.
2.1 Stationary system response
The reference state for comparison is displacement at steady rotation (Z =
const), which is taken for the angular speed Z = 1000 rev/min = 104,72 rad/s
(frequency 16,67 Hz ) for given drive's torque profile (Figure 4a).
2.2 Drive properties change and system response
The drive properties change should be adjusted to the changeability or
variableness load, which is quite common phenomenon by certain mechatronic product e.g. washing machine, walking prosthesis. Very common DSP system with central MCU
or non-FPGA solution has very limited possibility to follow up and also quick adjust the
drive properties to achieve better system's stability. This issues could be tackled with the
hybrid, DSP/FPGA approach and solution depends solely upon the mechatronic product.
In the following diagram (Figure 4b) is presented system response upon the drive
change in both planar axis x as well as rotational speed M for the chosen mechatronic
product washing machine (Figure 4a) for 10% (M1) and 20% (M2).
On all three figures above is presented a comparison among the reference
state of the drive property with the drive modification. The modification of the drive
properties for 10% (M1) and 20% (M2) cause the raise of deflection for up to the 7%.
Consequently the accelerations and forces cause the raise of unnecessary vibrations
and noise.
3
COMMENTS
In this paper, equations the motion of a mechanical system are written down in
s.c. robotic way (applying the Rodriguez method), modified symbolic Lagrange
equations in a covariant way.
835
a) Drive's torque profile
b) Drive modification - displacement in x-axis
Fig. 4 Steady state response displacement in XY plane and reference drive profile
Presented is co-simulation approach of mechatronic system with the drive
modification and multibody dynamic system response, which is very good basis for
further experimental evaluation (e.g. FPGA hardware-in-loop simulation). Even the
slightest changes of drive properties could cause the substantial increase of system's
deflection and further vibrations. This only proves the necessity of advanced control in
very short response time. Main issue to conduct the mechatronic system modelling is
to assemble the simulation model of mechatronic system as a multiphysical system - to
interconnect the simulation models of particular subsystems of certain physical nature
into one resulting multiphysical (multidisciplinary) modelling. The mechatronic (washing
machine) product's control unit should act in very short (response) time. Rational and
efficient application FPGA technology and related hardware-in-loop simulation (HILS)
demand also efficient mechanical model - in our case modified, symbolic Lagranges
equations of motion mechanical system in covariant way or robotical (Rodriguez)
approach.
5
ACKNOWLEDGEMENT
This work is partially supported by the Ministry of Science and Environmental

Protection of Republic of Serbia as Grant No- 35006, EUREKA project E!4930 AWAST
(Advanced Wavelet Analysis for Structural Testing), Bilateral Scientific cooperation
Slovenia and Serbia for 2010 and 2011.
LITERATURE - only the selected
[1]
[2]
[3]
836
Vasi,V., Lazarevi,M.P.: Standard Industrial Guideline for Mechatronic Product

Design, FME Transactions (2008) 36, p. 103-108
Vasi, V., Hace, A., Lazarevi, M. P., Jezernik, K.: Some aspects of drive actuator
control in contemporary washing machines, Avtomatizacija v industriji in
gospodarstvu, E-Conference, Drutvo avtomatikov Slovenije (www.aig.si), 2009;
Jezernik, K.: Nonlinear torque control for PMSM, , MELECON 2010,15th IEEE
Mediterranean Electrotechnical Conference, 26-28 April 2010, p. 784 789,
Location: Valletta, ISBN: 978-1-4244-5793-9;
IMPACT OF SENSOR FAILURE ON WORK OF ELECTRONICALLY

CONTROLLED DIESEL ENGINES
Miroslav Grubii1, Snjeana Rezi2
Summary: In modern diesel engines, control of the quantity and the start of fuel
injection perform a complex electronic system, which in addition to these regulations,
manage the entire engine operation. This control system has sensors, engine control
unit (ECU) and actuators. Engine control unit with integrated microprocessor,
analogue-digital converter and program memory, converts, analyzes and processes a
large amount of computer input data obtained from sensors, and then sends voltage
control signal to activate the actuators.
This paper tackles the problem of lack of influence of the sensor signal due to a defect
in the overall accuracy of electronically controlled Diesel engines. The aim of this study
was to look into the importance of input values of physical quantities, obtained from the
sensors, the overall accuracy of electronically controlled Diesel engines. The study
results are based on experimental data obtained by conducting experiments on the
four-stroke Diesel engine of Volkswagen Company, power 103 kW, 2 liters volume,
with pump-injector fuel injection system, Variable geometry turbocharger, sensors,
actuators and EDC16 engine management device connected via CAN data bus in the
powertrain. Communication with the engine control unit is performed via OBD socket in
a vehicle diagnostic device using VAS 5051B integrated instruments for the
measurement technique.
Key words: sensors, electronic control, Diesel engine.
1. INTRODUCTION
In Diesel engines, there are two basic parameters that need to be regulated in
order for engine to work, and those are the quantity and the beginning of fuel injection.
These variables directly affect the essential characteristics of the engine such as
torque, power, fuel consumption, emissions of exhaust gases and mechanical and
thermal load on the engine. Thanks to the regulation of the amount and the start of
injection, the engine under all operating conditions works with the optimum fuel
combustion. In modern Diesel engines, quantity control and the start of fuel injection,
1
Mr.sc. Miroslav Grubii, Mostar, Faculty of Mechanical Engineering and Computing,

(miroslav.grubisic@gmail.com)
Prof.dr.sc. Snjeana Rezi, Mostar, Faculty of Mechanical Engineering and Computing,
(snjezana.rezic@gmail.com)
837
Miroslav Grubii, Snjeana Rezi
performs a complex electronic system which, in addition to these regulations, manage

the entire engine operation. This control system has sensors, engine control unit (ECU)
and actuator. Engine control unit with integrated microprocessor, analogue-digital
converter and program memory, converts, analyzes and processes a large amount of
computer input data obtained from sensors, and then sends voltage control signal to
activate the actuator. Engine control unit must always have information about the state
of the engine, so within the management system of modern Diesel engines there are a
number of different sensors. This paper tackles the problem of lack of influence of the
sensor signal due to a defect in the overall accuracy of electronically controlled Diesel
engines, with the aim of determining the significance of sensory signals to the work of
Diesel engines.
2. EXPERIMENTAL SURVEY SETUP
In order to achieve understanding of the importance of input values of physical
quantities, obtained from the sensors, the overall accuracy of electronically controlled
Diesel engine, experiment was conducted on the Diesel engine based on which
research results derived.
Fig. 1 Engine Management VW 2.0 l TDI

Experiment and measurements were conducted on an experimental setup that
includes a Volkswagen Company four-cylinder Diesel engine, power 103 kW, volume 2
liters, with pump-injector fuel injection system, Variable geometry turbocharger,
838
Impact of Sensor Failure on Work of Electronically Controlled Diesel Engine
sensors, actuators and EDC16 engine control unit connected to the CAN bus in the
powertrain. Sensors and actuators are via bundles of electrical conductors, connected
to the controller via multi-contact connectors. Voltage supply of the controller is made
trough this connector. Sensors convert various physical quantities into electrical
signals suitable for the measurement which are then forwarded to the EDC engine
control unit. Sensors mounted on the engine, on which the experiment was carried out,
forward signals to the engine control unit in analog or digital form. Digital input signals
are directly taken by the EDC engine control unit in the processing of input signals [1].
At the reference engine, a digital signal to the engine control unit sends only the brake
pedal switch. Input values received from sensors in analog form are first digitized into
the engine control unit, and next step is the processing of received signals and a
comparison with the values stored in the memory of the controller. After making these
activities, EDC engine control unit activates the actuator and thereby controls the
engine performance.
Communication with the engine controller is carried out by communication
protocol KWP 2000 [2], defined by ISO 14230-4 [3], through the OBD socket in a
vehicle diagnostic device using VAS 5051B with integrated instruments for the
measurement technique. Scheme of the test engine management is shown in Fig. 1
[4], and a list of its sensors and actuators is given in Table 1.
Table 1 Engine sensors and actuators VW 2.0 l TDI
Sensors
1 Engine speed sensor
2 Hall sensor
3 Accelerator pedal position sensor
4 Air mass meter
5 Coolant temperature sensor
6 Coolant temperature sensor
radiator outlet
7 Fuel temperature sensor
8 Intake air temperature sensor and
charge air pressure sensor
9 Brake pedal switch
10 Clutch pedal position sensor
Actuators
11 Fuel injector valves
12 Solenoid valve block with:
- exhaust gas recirculation valve
- exhaust gas recirculation
cooler valve
- charge pressure control
solenoid valve
13 Intake manifold flap motor
14 Fuel pump relay and fuel pump
15 Radiator fan control unit with
left and right fan
16 Glow plug control unit, glow plugs
3. SENSOR IMPACT ON DIESEL ENGINE PERFORMACE ANALYSIS

The experiment is based on turning of all single sensors at tested engine. After
the absence of a necessary signal after turning off a particular sensor, the significance
of each signal was experimentally determined and the results are shown in Table 2. In
order to analyze and complete results, the actual engine operating parameters were
read from the engine control unit at the time of the absence of signals from sensors
and are compared with the required parameters of the analyzed engine, which is
defined by the manufacturer.
839
Table 2 Results of the experimental survey
Sensor Name
Consequence of the signal absence
Effect of the sensor signal absence

on proper operation of electronically
controlled Diesel engines
Insignificant
Medium
Significant
Engine speed
sensor
Engine is hard to start, then working in the

regime of the necessary work with a limited
speed and reduced amount of injection
Camshaft
position sensor
(Hall sensor)
Engine is hard to start, then working in the

regime of the necessary work with a limited
speed and reduced amount of injection
Accelerator
pedal position
sensor
Engine running at increased idling speed

and not responding to pressing the
accelerator pedal
Air mass meter
Significant reduction in engine power
Coolant
temperature
sensor
Increased fuel consumption and maximum

preheating time of the engine compression
compartment
Fuel
temperature
sensor
Inability to calculate the density of fuel as a

corrective size to determine the amount of
the injection, slight reduction in engine power
Charge air
pressure sensor
Charge air pressure control is off with

complete loss of engine power
Intake air
temperature
sensor
Loss of engine power
Brake pedal
switch
Partial reduction of engine power due to the

decrease of fuel injection, turning off the
cruise control system
Clutch pedal
position sensor
Inability to start the engine, and with started

engine there is a twitch when changing
gears
By turning of engine speed sensor, engine control unit has no information

about the exact position of the crankshaft as well as the current engine speed, and
takes longer time to start the engine. Due to the absence of this signal, after starting,
engine works in the necessary work with limited speed and reduced amount of the
injection.
In the absence of signal from the camshaft position sensor (Hall sensor),
engine control unit has no information on the position of the camshaft and the engine
result in the necessary regime, limited speed and reduced amount of the fuel injection,
identical to the previously described failure of engine speed sensor.
840
Impact of Sensor Failure on Work of Electronically Controlled Diesel Engine
By interruption of signal from the accelerator pedal position sensor, engine

control unit has no more information about the position of the accelerator. Therefore, it
does not perform calculation of fuel injection and in this case engine works with a
slightly elevated idle speed and not responds to pressing the accelerator.
By absence of signal from the air mass meter, engine control unit has no
information about intake mass of air into the engine, so for calculation of the amount of
injection uses the default replacement value [5]. The experiment showed reducing
amount of fuel injection and air charging pressure in the intake pipe, which directly
results in a significant reduction in engine power.
With interruption of signal from the coolant temperature sensor, engine control
unit has no information on the current engine temperature, so as an alternative value is
taken fuel temperature sensor signals. The consequence of the absence of this signal
is significantly increased fuel consumption and maximum possible preheating time of
the engine compression compartment.
By turning of the fuel temperature sensor engine control unit can not calculate
the density of fuel, which serves as a corrective method for determining the amount of
fuel injection [6], [7]. The engine then has a slightly decrease of the power and the
controller takes a replacement value of the coolant temperature sensor.
Absence of signal from charge air pressure sensor engine control unit switches
off the charge air pressure control in the intake pipe. Turbocharger in this case is not
effective, leading to complete loss of engine power.
Engine control unit uses intake air temperature signal sensor as a correction
value for calculating the charge air pressure, because temperature affects the density
of charge air. In the absence of this signal, engine control unit performs calculation, on
the basis of a fixed default temperature, which leads to a reduction in engine power.
Absence of signal from the brake switch, engine controller no longer
recognizes the position of the brake pedal and therefore reduced the amount of
injection causing a partial reduction in engine power. Due to the absence of this signal
the cruise control system for controlling vehicle speed is turned off.
Interruption of the clutch pedal position sensor causes engine control unit not
to recognize whether the clutch pedal is activated or not. Due to this signal failure
motor can not run from the sleep mode, and if the failure occurs in the run of the
engine, convulsions occur when changing the gears.
4.
CONCLUSION
After the experiment, measurements and analysis of the results, it is clear that
the input values of physical quantities, which are obtained from the sensor, significantly
impact on the overall accuracy of electronically controlled Diesel engines. The
experiment showed that the absence of a single input signal from the sensor leads to a
disruption in the work of the entire engine system and a significant limitation of its
functionality. In case of failure of multiple sensors, engine controller is deprived of input
parameters, and can not carry out the calculation in order to control the engine, which
results in its failure and limited work. Furthermore, the result of malfunctioning sensors
and actuators activate the wrong side of the engine control unit or causes not
activating when necessary.
841
The values that the engine control unit receives from the sensors are of
paramount importance when diagnosing a fault in an engine, regardless of the
diagnostic method used. Absences of signal or illogical values of sensor readings over
a diagnostic device, as compared to the required values are defined in specific modes
of electronically controlled Diesel engine directly indicate the existence of defects in
engine operation. By comparing these readings with the pre-defined values required, in
certain cases can be a simple way to diagnose a malfunction in the engine.
In situations where the engine control unit doesn't receives signals from
corresponding sensors, backup signal takeover from other sensors is carried out
automatically, however, generally this reserve signal can not maintain proper engine
system operation, but it allows the regime of the necessary work. Therefore, great
importance should be given to the construction of sensors and their reliability in all
kinds of bad working conditions such as extreme temperatures, sudden temperature
changes, and work in aggressive media, moist environment and mechanical vibrations
produced by the engine.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
842
Krger, M. (2008). Grundlagen der Kraftfahrzeugelektronik Schaltungstechnik.

Carl Hanser Verlag, Mnchen, 2008.
Stuphorn, J., Constapel, R., Powolny, S. (2007). Diagnosesnittstellen. Robert
Bosch GmbH. (Editors). Autoelektrik, Autoelektronik. Friedr. Vieweg & Sohn
Verlag, Wiesbaden, p. 144-155.
ISO 14230-4 (2000). Road vehicles Diagnostic systems Keyword Protocol
2000 Part 4: Requirements for emission-related systems. International
Organization for Standardization. Geneva.
Volkswagen AG (2003). Der 2,0 l TDI-Motor. Selbststudienprogramm, vol. 316,
p. 20-21.
Palma, A., Palladino, A., Fiengo, G., De Cristofaro, F., Garofalo, F., Glielmo, L.
(2008). Air Mass Flow Analysis for SI Engine: EGR and Scavenging. Proceedings
th
of the 17 World Congress The International Federation of Automatic Control, p.
8495-8500.
Michalske, A., Lorente, M., Walther, M. (2004). Elektronische Dieselregelung
EDC. Robert Bosch GmbH (Editors), Dieselmotor-Management. Friedr. Vieweg &
Sohn Verlag, Wiesbaden, p. 354-406.
Reif, K. (2007). Automobil-elektronik, Friedr. Vieweg & Sohn Verlag, Wiesbaden.
IMPLEMENTATION OF PLANTWEB ALERTS IN A DELTAV SYSTEM

Miroslav Kostadinovi1, Zlatko Bundalo2, Duanka Bundalo3
Summary: This paper describes and provides information on field device alerts in a
DeltaV system. PlantWeb Alerts are available for both the FOUNDATION fieldbus and
the HART devices. PlantWeb Alerts are implemented in a DeltaV system as a separate
alarm subgroup. This enables the system to target alarms to the right personnel. There
are many different operational philosophies with respect to the alarming. PlantWeb
Alerts enable users to adjust the system behavior to meet their specific needs. This
paper considers and describes the DeltaV field device definition and the alarm
management system, including the definition of alarm priorities, as well as the function
of plant areas within the alarm management scheme.
Key words: PlantWeb Alerts. DeltaV system. HART devices.
1. INTRODUCTION
PlantWeb Alerts are applied in the management of DeltaV alarm systems and
devices. Most control systems use a mechanism to report an alarm associated with the
process and different mechanisms for reporting on the state of the system hardware,
and have expanded the range to cover systems to field devices.
Fieldbus devices detect and report on different states host system, a range of
state of the hardware failures to the sensor. Manufacturer of the device determines
whether a particular state of the device to be pushed, disabled or configured through
the host system. Emerson FOUNDATION fieldbus devices support and PlantWeb
Alerts with DeltaV system software can determine the precise status. The system can
also track lower priority alerts to record and allow for follow-up. Depending on the
individual PC and certain conditions can be enabled or disabled, due to a different
threshold or alert type. DeltaV system to report on these through the system alerts and
warnings about the types of maps priority alarm which allows the working condition of
the device notified the operators, maintenance personnel and so on. FOUNDATION
fieldbus FF912 has introduced a standard that defines a hierarchy of types of
warnings. Emerson follows the development of new standards in the expectation that it
will result in a future update DeltaV system and PlantWeb comply with warnings.
PlantWeb Alerts (Failed, Maintenance, Advisory and Communication) are
1
Dipl. ing. elektrotehnike, mr Miroslav Kostadinovi, Doboj, Saobraajni fakultet, kostadinovicm@gmail.com

Dipl. ing. elektrotehnike, Prof. dr Zlatko Bundalo, Banja Luka, Elektrotehniki fakultet, zbundalo@etfbl.net
3
Dipl. ing. elektrotehnike, Prof. dr Duanka Bundalo, Banja Luka, Filozofski fakultet, zbundalo@etfbl.net
2
843
Miroslav Kostadinovi, Zlatko Bundalo, Duanka Bundalo
offered for all HART devices based on standard HART status condition. However, in
contrast to the FOUNDATION fieldbus devices, HART devices do not support local
alarms and warnings not to report the situation and must determine which state the
device are active or deleted. In HART devices, the configuration state of alert,
publication and reporting DeltaV control system, not the device.
2. PLANTWEB ALERTS WITHIN THE DELTAV SYSTEM
DeltaV system has built-in default settings that help to take advantage of
PlantWeb Alerts with minimal effort. Below, information about priority and distribution of
alerts through the system, helping to adjust the settings in accordance with operational
requirements and management strategies alarms. The implementation of PlantWeb
Alerts should be included in the philosophy of the alarm systems so that all devices are
configured according to warnings and to be significant for those who receive them.
Philosophy definition alarm priorities and their criteria for use in the system. DeltaV
system defines three priority alarm by default and supports up to twelve different
priorities. To understand the PlantWeb Alerts, it is important to understand some of the
basic functions of alarm control in the DeltaV system.
Plant Areas - alarm systems use factory space in order to alarms and warnings
of specific workstations define the scope of control to the user on each workstation.
Each control module, device and component DeltaV hardware is connected inside the
factory premises, so that alarms or complete information may be directed to the user.
Alarm Priorities - Each alarm and/or warning required by the definition of
priorities. Priority is an indication of the importance of alarms and usually reflects
several factors such as the consequences of inaction, and response time. Alarm
systems using priorities to determine which of the alarms or warnings are displayed on
the banner and the order of these alarms.
Alarm Annunciation - DeltaV alarm system groups based on their source
(process, device, hardware). In the particular workstation is the banner alarm, the
alarm of this group are filtered by priority so that the compliant primary task of the
current user. Only alarms with specific priority is published by the operator through an
alarm banner.
DeltaV system provides a default scheme alarm management PlantWeb Alerts.
Operators have the ability to change reporting PlantWeb Alerts that are based on the
criticality of individual devices or services have all the warnings posted by some of the
working procedures. Whatever the reason, DeltaV system allows modification of the
present PlantWeb Alerts to meet the needs of operators.
3. TYPICAL ALARM WIRING
The impact of artificial intelligence in the field of development of today's digital
architecture that optimizes the layout of machinery offers a new wealth of information,
including diagnostics that allows it to anticipate and prevent problems before they
affect the process at the plant. Connection and communication between devices based
on HART protocol and open applications to provide PlantWeb network to obtain
information on equipment and processes where needed within the plant.
Control networks can adjust its size DeltaV system, and in Figure 1 shows the
844
Implementation of Plantweb Alerts in A Deltav System
basic configuration of the control network which consists of network switches,

workstations and controllers.
Fig. 1 Redundant Control Network with DeltaV Controller

When the DeltaV Controller installed in the system with new installations,
routing must include protection for the network cable from the destructive
environmental conditions such as corrosive materials, evaporation, moisture and
electrical interference. Also, when added to existing installations, should be considered
the previously listed risk factors for cables.
Cabling can be installed underground, aboveground, or both, and there is also
a primary and secondary network cables, which routes the various routes to ensure
uninterrupted communications in case one of the network becomes physically
damaged.
Most control networks using relay output for alarm systems, and Figure 2
shows the distribution wiring for alarm system located at the bottom of the DeltaV
Controller with the contacts connected to the power input connectors on the discrete
card. Related contacts are normally open when no power.
Fig. 2 Typical alarm wiring for DeltaV system

845
4. IMPLEMENTATION OF THE DELTAV SYSTEM TO CONTROL FLOW

Hypothetical management process and equipment shown in Figure 3, which
controls the flow of draining the fluid from the reservoir, will be presented using the
DeltaV Controller.
Fig. 3 An example of process control

To implement a DeltaV system to control flow as in Figure 3, requires four
control module for managing the process and equipment, and their names are LI-101,
MTR-101, XV-101 and FIC-101, while the following tags LT 1, FT-1, XI-1 and so on are
the tags used devices such as valves, transmitters and other I / O devices.
In Table 1 contains information on control modules and tags I / O devices
necessary for the implementation of DeltaV system.
Table 1 Control modules and tags I/O devices Tags
Control
Module
Description
LI-101
Level indicator
MTR-101
2-State Motor
XV-101
FIC-101
Block Valve
Flow Control Loop
Purpose
Monitor level of
product in tank
Start and stop
pump
Open/close valve
Regulate flow
Input Device
Tags
Output
Device Tags
LT-1
XI-1
ZX-1
LSC-1
FT-1
XV-1
FY-1
Once you have set up a control module, it is using Sequential Function Chart
SFC create an algorithm to automate emptying the tank. The following lists the
necessary steps to implement the algorithm:
x
Place on an automatic control valve flow (FIC-101).
Specify the desired values for the flow.
Define the timing control valve, then open the block valve (XV-101).
Turn on the pump motor (MTR-101).
846
Implementation of Plantweb Alerts in A Deltav System
Confirm start the engine.

Also, it is necessary to implement a graphical display system for controlling the
flow that will facilitate the DeltaV system operators monitor and control processes. In
this paper we will use DeltaV Operate mode configuration that includes dynamic
process control objects such as pipes, pumps or valves, which are placed on the
image of the original or a new set of objects. In order to realize a graphical display
system for flow control is necessary to use several different objects and modify their
dynamic properties, and following is a brief summary of placing objects in the image:
x Add dynamic characteristics of pumps and animate color, so to change from red to
green when the pump signal goes from off to on.
x Add dynamic properties of the reservoir so as to allow graphical representation of
discharge tank.
x Add dynamic properties of regulation valves to change the color shows their status
(closed or open).
x Add dynamic properties of tubes by changing the color of the connection with the
above listed facilities
Using the steps for placing dynamic objects in the configuration mode DeltaV
Operate realized is a graphical representation of the system for controlling the flow as
in Figure 4
Fig. 4 Graphical display system for flow control

5. CONCLUSION
DeltaV system and PlantWeb HART devices implement proactive alerts that
warn the operator and / or technicians to maintain, so that includes an alarm system
that is based on assigning priorities alarm. DeltaV system provides the operator the
possibility of advanced settings to set alerts for each device in accordance with their
847
own philosophy of alarm at the plant and practical needs.

As the number and sophistication of the alerts increase, there is an increased
need to understand what each device type is capable of, and which of these alerts are
appropriate for a particular installation. It is also important to keep documentation on
the alerts used/not used for each device and the reasons. That way, when device
configurations are reviewed in the future there is good information for accomplishing
consistent and effective device setup.
PlantWeb Alerts should be introduced methodically to a DeltaV system, to
avoid inappropriate notifications to operators who either have no action to take when
alerted or have not been trained to respond. When properly deployed, PlantWeb Alerts
can significantly improve system availability and reduce unscheduled outages.
PlantWeb digital architecture has been proven that uses the power of predictive
intelligence to improve plant performance. PlantWeb reduces engineering costs
compared to traditional DCS architectures and provides even greater operational
benefits by enabling enhanced throughput, availability, quality, cost of conversion,
which resulted in improved performance and efficiency of the plant. PlantWeb is not
just a product or a specific automation control system, it is a proven strategy forfor
building a digital architecture that optimizes the layout of machinery.
LITERATURE
[1] Deji, C., Mark N., Aloysius M., (2010). WirelessHART: Real-Time Mesh Network
for Industrial Automation. Springer. ISBN: 978-1-4419-6046-7.
[2] Gutierrez, J., Callaway, E., Barrett, R., (2007). Low-Rate Personal Area Networks:
Enabling Wireless Sensors with IEEE 802.15.4. Second Edition, Published by
Standards Information Network IEEE Press.
[3] Sheldon, M., Chen, D., Nixon, M., Mok, A., (2005). A Practical Approach to Deploy
Large Scale Wireless Sensor Networks. IEEE International Conference, Mobile
Adhoc and Sensor Systems. Washington DC. Page(s): 8pp - 250.
[4] Kostadinovi, M.,(2008). Application of Hart protocol for communication needs in
process industry, Proceedings of INFOFEST Conference, Budva, Montenegro.
[5] Kostadinovi, M., Stojev, M., Bundalo, Z., Bundalo, D., (2009) Application of
modified TrueTime simulator in control of WirelessHart networks, Proceedings of
Intrenational Symposium Ee 2009., Novi Sad, Serbia.
[6] Kostadinovi, M., Stojev, M., Bundalo, Z., Bundalo, D., (2009). Design,
Implementation and Simulation of WirelessHart Network. Proceedings of TELSIKS
Conference, Ni, Serbia.
848
IMPLEMENTATION OF RFID TECHNOLOGY AND SMART PARTS IN

WIRELESS MANUFACTURING SYSTEMS
Vahid Bagher Poor1, Majid Hashemipour2
Summary: Wireless Manufacturing leans meaningfully on wireless devices such as
RFID (Radio Frequency Identification) or Auto ID (Automatic Identification) sensors,
and wireless communication networks such as WiFi for the accumulation and
synchronization of manufacturing field data. In this article the state-of-the-art and
recent developments in RFID-based wireless manufacturing systems are reviewed.
The study is managed by examining related case reports and research articles
available in the literatures in order to explore the industrial and consumer applications
of RFID. Typical motivations are presented to emphasize why and how manufacturers
can benefit from applying RFID solutions in addressing shop-floor challenges and
facilitating current manufacturing strategies.
Key words: Wirelwss Manufacturing, Smart Cards, Radio Frequency Identification
(RFID)
1. INTRODUCTION
Personalized products and tailor-made solutions are taking over large shares of
the marketplace from mass produced goods and standardized solutions respectively.
The so-called smart parts based manufacturing system addresses these concerns
well. The smart part carries operating instructions for the operating workers or
automatic machines. Uniquely identified individual parts can be processed according to
their specific requirements based on individual customer preferences. Therefore, there
is the need to correctly identify every part to ensure reliable process control in such a
flexible and customer oriented manufacturing system.
Wireless technologies (WT) have evolved dramatically in recent years. Rapid
developments have been achieved in wireless sensors; wireless communication and
information network technologies (e.g. radio frequency identification RFID or Auto-ID,
Bluetooth, Wi-Fi, GSM, and ZigBee). They have nurtured the emergence of wireless
manufacturing (WM) as a key advanced manufacturing technology (AMT) in nextgeneration manufacturing systems (NGMS).
1
PhD. Candidate and research assistant, Eastern Mediterreanean University, Mechanical Engineering
Department, (vahid.bagherpoor@cc.emu.edu.tr)
2
Professor of Mechanical Engineering Department and Vice Rektor of Eastern Mediterreanean University,
(majid.hashempoor@emu.edu.tr)
849
Vahid Bagher Poor, Majid Hashemipour
Automatic identification, or auto ID for short, is the broad term given to a host of
technologies that are used to help machines recognize objects. Auto identification is
often coupled with automatic data capture. That is, companies want to identify items,
capture information about them and somehow get the data into a computer without
having employees type it in. The aim of most auto-ID systems is to increase efficiency,
reduce data entry errors and free up staff to perform more value-added functions, such
as providing customer service. There is a host of technologies that fall under the autoID umbrella. These include bar codes, smart cards, voice recognition, some biometric
technologies (retinal scans, for instance), optical character recognition (OCR) and
radio frequency identification (RFID).
Radio frequency identification is a technology that has been in use for some
time. It offers features that are well suited to be adapted for such flexible smart-parts
manufacturing. As the use of this technology grows, it will come into the industrial
mainstream just as it has already done in retail outlets where it is used for electronic
surveillance.
Wirelessly networked sensors facilitate the automatic collection and processing
of real-time field data in the manufacturing processes, and reduce and eliminate the
error-prone, tedious manual activities. Manufacturing decision support systems are
networked, wirelessly whenever appropriate, to complement wired networks. Real-time
information visibility and traceability closes the loop of production planning and control
for adaptive decision making. WM provides a networked enterprise environment
without excessive and difficult wiring efforts in manufacturing workshops. WM provides
unprecedented opportunities for the manufacturing industry to improve the quality and
productivity and to speed up decision processes throughout the entire product lifecycle.
RFID or Auto-ID (automatic identification) is a typical wireless sensor
technology. It has received increasing attention in supply chain logistics (SCL scenario)
since 2004. This is particularly true as a result of the recent aggressive promotion and
developments, mainly through the Auto-ID Labs. The rapid drop in the costs of higher
frequency RFID tags and readers has made automated data collection without the
limit of line of sight economically viable and affordable. It is now possible for RFID
technology to eliminate a very large number of manual business process transactions
and associated manual data collection/entry in supply chain management (SCM).
Major retailers and buyers have been conducting pilot projects. While testing the
technical feasibility and demonstrating operational boundaries, they expect to achieve
major benefits in supply chain logistics. For example, the speed of order fulfillment is
dramatically increased and the accuracy is improved while the on-going operating
costs are reduced. Collected data become value-adding enterprise assets. In this
sense, data collection is no longer considered as non-value adding.
2. RFID TECHNOLOGY
2.1 Operating Principles
RFID is a technology which allows remote interrogation of objects using radio
waves to read data from RFID tags which are at some distance from an RFID reader.
This has several advantages over manual scanning using optical barcodes, since
many tagged items could be simultaneously identified in an automated manner, very
850
Implementation of Rfid Technology and Smart Parts in Wireless Manufacturing Systems
quickly and without the need for line-of-sight to each item [1].
Generally speaking, an RFID system typically comprises the following three
components [2]:
An RFID device (tag);
A tag reader with an antenna and transceiver; and
A host system or connection to an enterprise control system.
Fig. 1 illustrates the relationship of three components and working mechanism
in an RFID system.
Fig. 1 A typical RFID system

Today the vast majority of 13.56 MHz systems operate passive, without the
need for an integrated battery. They have positive implications on cost, lifetime and the
environmental situation. The basic operating principle of passive 13.56 MHz and below
135 KHz RFID systems is energy and data transmission using inductive coupling (Fig
2.a). This is exactly the same principle as used in transformers. By changing one
parameter of the transmitting field (amplitude, frequency or phase), the data
transmission from the reader to the tag can realize. The return transmission of the tag
concerns the load (amplitude and/or phase).
Fig. 2 (a) Inductive coupling (b) Propagation coupling

UHF and MW (e.g. 400-1000 MHz, 2450 MHz & especially 5.8- GHz) RFID
systems make use of conventional electromagnetic wave propagation (see Fig 2.b) to
communicate their data and commands, and in the case of battery-less tags also to
power the RFID transponders. The basic operating principle of this RFID system is
851
energy and data transmission using propagating radio signals (E field transmission).
2.2 Operating distance
The operating zone of passive inductive RFID systems (13.56 MHz and below
135 KHz) is in the near field of the read transmission antenna, which results in
achievable operating distances of approximately the diameter of the transmission
antenna. Differences are mainly given by the output power of this RF-module and by
the sensitivity and the selectivity of its receiver. The ranges are denoted as proximity
(below 100mm), medium range (below 400mm), vicinity (long range - 1.5m), far
field (0.5 to 12 meters - 2450 MHz, passive power), and up to 30 meters (active power
tags depending on microwave frequency).
3.
LITERATURE SURVEY
Different uses of the RFID technology were reported in recent years in the
manufacturing industry. Ford Motor Company has successfully implemented an RFIDbased JustIn-Time (JIT) manufacturing model at its facility in Cuautitlan, Mexico [3]. In
the manual coding system, the identification sheets were manually updated at every
stage in the production line. In the RFID-based system, however, updates are
automatically written on the tag as the vehicle advances on the production line without
the risk of operator error.
IBM has transformed chip production at its Fishkill plant with a semiconductor
manufacturing system that leverages real-time information to automatically control the
fabrication process, enabling employees to work more productively and be more
responsive to customers' product status inquiries [4]. IBM has accomplished this using
IBM SiView Standard, a manufacturing execution system that the company integrated
with its own wireless e-business technology. SiView Standard leverages information
from IBM DB2 Universal Database to automatically control each step of the fabrication
process. DB2 manages information about the fabrication processes that need to be
applied to every wafer containing chips, and supports data analysis tools that provide
production-related statistics. IBM WebSphere MQ provides the messaging platform
that enables DB2 to exchange information with the production tools and other
application programs used to run the plant.
BMW and Vauxhall use RFID tags to enable accurate customization of
customer orders [5]. A read/write smart tag is programmed in the customer order. The
tag is then attached to and travels with the car during the production process. This
tracking ensures that the car is manufactured with the correct color, model, interior,
and any other option the customer specifies.
Extensive studies were also performed on the various types of systems that
RFID technology can be used in. [6] discusses an architecture where communication
between RFID-based data acquisition system and various monitoring terminals is
performed using RFID, Bluetooth and Internet channels. [7] addresses various
architectures related to wireless/wired communications in manufacturing environments.
[8] proposes "Intelligent Tracking Technologies", or IT2, which comprises of Global
Positioning Systems (GPS), Geographic Information Systems (GIS), wireless
communications, and RFID to enable dynamic scheduling in manufacturing and supply
chain management. [9] And [10] develop "Wireless Manufacturing" (WM) technology to
852
manage job shop Work-In-Progress (WIP) inventories in real time. The emphasis is
placed upon how to avoid changing from functional (or "walking-worker fixed-position"
in their terms) to cellular layouts "in order to retain existing operational flexibility while
improving efficiency and capacity".
By taking advantage of data capacity stored on an RFID tag, critical
manufacturing information on a product can be locally stored with the product. [11]
offers a framework to enable the instant delivery of pertinent data and information on a
uniquely identifiable job/product at point-of-need across factories. Alternatively, [12]
discusses the application of RFID in construction production field where information
related to a product is carried by the product itself and can be handled to manage the
whole system.
The interested reader is referred to [13] and [14] for more examples and case
studies on how RFID technology was successfully implemented in real-world
manufacturing projects.
4.
MOTIVES FOR RFID-BASED WIRELESS MANUFACTURING
Motivations for product manufacturers to consider applying RFID technology

arise both externally and internally. The major external motivation is pressure from an
end customer, who imposes a mandatory requirement that RFID tags should be
attached to products and/or their packages at different levels (pallet, box and/or item
level). In order to meet such mandatory requirements, product manufacturers must
invest in tagging processes.
Product manufacturers do consider this as a value adding activity, and it can
then generate internal motivations to expand RFID and other WM systems, which
enable online collection of real-time shop-floor data to close the manufacturing control
loop, improve work-in-progress inventories, and adaptive manufacturing planning,
scheduling and execution decisions. Three main motivations are include:
Online collection of real-time shop-floor manufacturing data
Improvements in inventory control
Adaptive planning, scheduling and execution
5.
APPLICATION AND DEVELOPMENT
5.1 Recent developments in rfid-based wireless manufacturing

This section addresses where and how RFID technology applies in
manufacturing environments. The sources of materials used for discussions are mainly
selected from the research and development (R&D) activities of RFID manufacturing
applications reported in the literature, some well-known or major R&D projects, and
some direct R&D experiences undertaken by a HKU (Hong Kong University) research
team in collaboration with companies. The discussions are however categorized
differently according to application scenarios, e.g. part fabrication, product assembly,
just-in-time manufacturing, mass customization, manufacturing asset management and
maintenance and product lifecycle management. These individual cases or projects
deal with one or more, but so far not all, manufacturing decision levels among
advanced production planning and scheduling, manufacturing execution and control,
853
line and machine maintenance, technical support and customer service, and product
recycling and disposal. However recent development in RFID based WM include:
RFID applications in part fabrication
RFID applications in product assembly
RFID applications in JIT manufacturing
RFID applications in mass customization and reconfigurable manufacturing
RFID applications in manufacturing asset management and total productive
maintenance
RFID applications in product lifecycle management (PLM)
5.2 Some current and possible application areas
Almost 100 million contacts less 13.56 MHz cards have been sold worldwide.
They are being used in: It is used in library field, carton marking, airline baggage
sector, express parcels and high value/big ticket item management, advanced shipping
label and item level identification where the primary identifier, barcode, suffers from a
line of sight problem. The normal application areas are: Transportation and logistics
management; Security; Waste management; Postal tracking; Electronic article
surveillance - clothing retail outlets being typical; Protection of valuable equipment
against theft, unauthorized removal or asset management; Controlled access to
vehicles, parking areas and fuel facilities - depot facilities being typical; Automated toll
collection for roads and bridges - since the 1980s, electronic Road-Pricing (ERP)
systems have been used in Hong Kong; Controlled access of personnel to secure or
hazardous locations; Time and attendance - to replace conventional slot card time
keeping systems; Animal husbandry - for identification in support of individualized
feeding programs; Miniature tags can be placed within tool heads of various types such
as block or Cat V-flange, or even within items such as drill bits where individual bits
can be read and selected by reader guided robot arms. Automatic identification of tools
in numerically controlled machines - to facilitate condition monitoring of tools, for use in
managing tool usage and minimizing waste due to excessive machine tool wear;
Identification of product variants and process control in flexible manufacture systems;
Sport time recording; Electronic monitoring of offenders at home; Vehicle anti-theft
systems and car lock; etc.
6.
CONCLUSION
The focus of this paper is to underline why and how manufacturers can benefit
from applying RFID solutions in addressing shop-floor challenges and facilitating
contemporary manufacturing strategies.A review of existing RFID technology is
provided and a number of RFID applications in manufacturing-related systems have
been highlighted.
LITERATURE
[1]
[2]
854
Parlikad, A.K., McFarlane, D (2007), RFID-based Product Information in End-oflife Decision making. Control Engineering Practice 15, 13481363
Roberts, C.M(2006), Radio Frequency Identification (RFID). Computers &
Security 25, 1826
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Johnson, (2002), RFID tags improve tracking, quality on Ford line in Mexico
Control Engineering 49 (11), pp. 16.
IBM Fishkill semiconductor plant an example of agile, real-time operation,
available online http://www.306.ibm.com/software/success/cssdb.nsf/CS/CGES
Brewer and T. Landers (1997), Radio frequency identification: a survey and
assessment of the technology University of Arkansas, Department of Industrial
Engineering Technical Report.
Z. Shouqin, L. Weiqing, and P. Zhongxiao (2007), An RFID-based remote
monitoring system for enterprise internal production management, International
Journal of Advanced Manufacturing Technology, Volume 33, Issue 7-8.
K. Koumpis, L. Hanna, M. Andersson and M. Johansson (2005), Wireless
industrial control and monitoring beyond cable replacement, PROFIBUS
International Conference.
Brewer, N. Sloan, and T. L. Landers (1999), Intelligent tracking in manufacturing
Journal of Intelligent Manufacturing 10, pp. 245- 250.
G.Q. Huang., Y.F. Zhang, P.Y. Jiang (2008), RFID-based wireless manufacturing
for real-time management of job shop WIP inventories, International Journal of
Advanced Manufacturing Technology, Volume 36, Issue 7-8.
G.Q. Huang., Y.F. Zhang, P.Y. Jiang (2007), RFID-based wireless manufacturing
for walking-worker assembly islands with fixed-position layouts, Robotics and
Computer-Integrated Manufacturing 23, pp. 469-477.
R.G. Qiu (2007), RFID-enabled automation in support of factory integration,
Robotics and Computer-Aided Manufacturing, Volume 23, Issue 6.
J. Yagi, E. Arai, and T. Arai (2005), Parts and packets unification radio frequency
identification application for construction, Automation in Construction 14, pp. 477490.
Zhekun, R. Gadh, and B.S. Prabhu (2004), Applications of RFID technology and
smart parts in manufacturing Proceedings of ASME 2004 Design Engineering
Technical Conferences and Computers and Information in Engineering
Conference, DETC 2004.
Baudin and A. Rao, RFID applications in manufacturing, available online at
http://www.mmtinst.com/RFID%20applications%20in%20manufacturing%20_Draf
t%2007.pdf
855
INTELLIGENT APPROACH FOR MOBILE ROBOT SIMULATOR WITH

ROBOSIM SOFTWARE
Tihomir Latinovic1, Sorin I Deaconu2, Remiquez Labudski3, Marcel Topor4
Summary: In this paper is presented an intelligent approach for mobile robot
simulation. One of the basic idea is how to modeling of robot manipulators. It has an
expanding area of interest because it can aid the design and usage of robots. We
usually need design and testing of manipulators, robot action planning, on line control
of robot manipulators, user interface, and training and education. RoboSim is a very
good simulator developed for simulating the interaction, cooperation, and
communication among various robots existing in a complex virtual environment.
RoboSim is evaluated from 1995 year in very respectable tool. Usually Robosim used
to perform simulations of many types of robots for working within a single, virtual
environment. It usually provides a better way to develop various robotic applications in
different enviroment. This work analysing the performance of Robosim simulator,
various design changes applied to the simulator, and the implementation various
architecture on the existing Simulator. All of these capabilities of the graphical
simulation approach to the study of robotics and automation, and provide the student
basic knowledge about simulation in area of Robotics.
1. INTRODUCTION
Robot programming is a very critical task in robotics education becouse it is a
bridge between robotics theory and physical implementation. There are basically two
types of robot programming, namely, on-line, and offline. Some programmers prefer
on-line programming because their system is appropriate to the programming
interface. [1] Others prefer off-line programming to see the results before manipulating
the robot. Off-line programming is that the mechanical robot and other equipment are
not connected during the programming; everything worked on a host computer. The
main advantage of off-line programming is that a robotic production cell can be
designed, programmed, and its operations may be simulated before the cell is actually
built. One can determine which type of robot that should be used, or how the
equipment in the cell should be arranged according to the simulation results.
1
Ass. Prof, Tihomir Latinovic, Banja Luka, Faculty of Mechanical Engineering, (tiho@inecco.net)
Ass. Prof, Sorin I Deaconu, Hunedoara, Engineering Faculty, (sorin.deaconu@fih.upt.ro)
Ass. Prof, Remiquez Labudski, Poznan, Faculty of Technology , (remigiusz.labudzki@put.poznan.pl)
4
Phd Student, Marcel Topor, Hunedoara, Engineering Faculty, (marcel.topor@fih.upt.ro)
2
3
857
Tihomir Latinovic, Sorin I Deaconu, Remiquez Labudski
ROBOSIM bassicaly has been written to assist students of robotics and

mechatronics in developing algorithms for the control of mobile robots. The simulation
provides a good virtual environment, which closely resembles the real environment of
the laboratory. The students can develop their robot control algorithms off-line, without
the need for access to the actual hardware . The resulting algorithms can then be
verified using real robots.[2]
In addition to the advanced features of collision detection and intelligencebased collision avoidance routines. Robosim includes other features such as
configuration management and composite objects. With ROBOSIM , students
construct a robot manipulator link by link and assign coordinate frames to the joints of
each link. Links of the manipulator have been constructed , a single command
automatically assembles the links into the complete manipulator. [3]
2. TECHNICAL OVERVIEW OF ROBOSIM
ROBOSIM is a 32 bit application that is running on Windows 2000 and
Windows 7. ROBOSIM application is done using the MicroSOFT version of Visual C +
+, . ROBOSIM application is programmed for WINDOWS applications, including the
Toolbar, Menu, Statusbar. The Graphical User Interface (GUI) is implemented in Java
which supports the development of graphical interfaces. The RoboSiM GUI consists of
the robot arm and a command field with buttons and scroll bars.
Applications supports dot, wire and a volume model. Also, the simulation can
be done with each of these types of models. The application supports the so-called.
Culling, hiding the last area of the model, the effect of smooth surfaces and lines, and
light. Application is targeted for users who do not really need to know the deeper
Robotics, since computer interaction. Therefore Robosim stable application that can be
used for educational purposes for example. in lectures and seminars in subjects
robotics. [7] [8]
The purpose of Robosim, was to provide a constructing and viewing three
dimensional models of robot manipulators and various objects in their workspace, and
simulating action sequences performed by the manipulators.
Education and research for simulation realistically substitutes a robot.
Demonstrations can be done with the virtual robot within lectures. Virtual robots can be
used in combination with real robots. It is possible to build and test virtual production
lines with one real robot which is monitored and other collaborating robots which are
simulated. Real and virtual devices can be blended. The virtual robot can be examined
in extreme situations without damaging an expensive real robot arm. [4]
In education and training real and virtual devices can be controlled by the same
control system. Therefore the number of expensive real robots can be reduced.
Trainees are beginning to work on virtual robots. At the end of the course they test
their experience on a real system without changing the used control system. Therefore
they already get a very realistic impression while training with the simulation. It is even
possible to compare real and virtual devices while they are controlled by RoboSiM at
the same time.[10]
858
Intelligent approach for mobile robot simulator with ROBOSIM software
Fig. 1 The Design of robot simulation system

Geometrical modeling of robot manipulators has a number of ways:
Design and testing of manipilators

Robot action planning
On-line control of robot manipulators
Telerobotics user interface
Training and education
The ROBOSIM language for modeling specify complex geometric shapes

which are used as manipulators links or as passive objects in models of robotic
systems. [5] [9]
The modeling of robot arms consist of robot arms and objects and geometric
scenarios in two steps:
Modeling the geometric shapes, robot links to build the scenario
Creating one or more named , object instances of these shapes in the
ROBOSIM workspace
859
Fig. 2 ABB IRB 2000 robot

The Robosim software is open source project in JAVA language. It is an
interactive simulation enviroment where every command entered by the user is
immediately executed and the results are displayed on a graphics screen. The
commands can be grouped as follows:[6]
Environment configuration
Manipulator control
Status reporting
The command language has been designed with two goals in mind:
To provide an interactive user interface
To be usable as the interface to a higher-level task planner program
2.1 Functions for robosim aplication
Robosim are builded from Java applet and needs a reserved place of 920x670
pixels. [7]
2.2 Basic Limitations [9] [10]
Control programs can only be done within the Applet Viewer because there
exist a lot of security restrictions
For the control real robot we need special RoboSim version and with
integrated Microsoft C++ and with Java., This integration would make it
860
Intelligent approach for mobile robot simulator with ROBOSIM software
impossible for this program to run on every computer system.[10]

2.3 The Commands
x
x
x
x
x
Manual Control: This command controls the robot and its rotation links
modifying the angles corresponding to these links. They are as follows:
Angle 1: 180.0 degree to -180.0 degree
x Auto Control: If you need to can control the robot by giving it a target
position (x-, y-, and z-value) with this command.
x Determination of the arm lengths: This command allows you to vary the
length of the arms of the robot. Also there exists a crash test. They are as
follows:
Arm 1: 15.0 to 4.5
Arm 2: 15.0 to 2.5
Arm 3: 15.0 to 2.5
x Angle weights with this angle obtain the best solution of the inverse
cinematic. The values have to be positive integers.
x Point of view: You can change the point of view by dragging the robot with
the mouse to the left and to the right side.
x Speed: You can select your required speed with the corresponding
scrollbar.
Define/Go to positions: Three positions can be defined simply by pushing the
define pos button of your choice.
Virtual/Real robot: You can choose between the virtual robot as shown in the
middle of your screen and the real robot [11]
Open File: By selecting this button, you can load a file which either has been
written by hand or has been produced with the Teach button.
Teach: After having activated this button, a file dialog window appears in which
you can choose the name and the directory of the file which will be produced and
written automatically.
Writing own control programs: You can write control programs, but you have
got more command power.
3. CONCLUSION
The development of computer hardware and components has led to a stage
where an engineer can build your ideas and design solutions to see even before the
application. This enables the modeling and simulation problems. Also the development
of numerical methods, which are again caused by the development of information
technology lead to solving those problems which was almost impossible to solve.
Today, there is almost no scientific discipline which is modeled on a software
package strictly from a given area. Software crisis was overcome through the emergence
of object-oriented programming languages. The Robosim application is created using the
861
mentioned object-oriented languages Visual C + +. And JAVA. Program with its output
value that simulation satisfy all the requirements of a modern application for simulating
robots. The program provides all the output parameters of the simulation, homogeneous
transformations, Euler - these angles, controlled coordinates, current location pins
Robots or TCP (center pins) and so on. Application is very easy to operate and can be
used in exercises to demonstrate the simulation of robots. Communication is done
through the dialog boxes so that the mishandling reduced.
The ability of industry to compete globally will depend on an adequate supply of
engineers and technologists trained in the application of robotics and automation to the
problems of industry. The wide use a Robot simulation programs like ROBOSIM will
allow all schools to provide students with the ability to study and develop advanced
robotic systems.
LITERATURE
Journal Papers:
[1] Mirolo, C. A Solid Modeling System for Robot Action planning, IEEE Computer
Graphics and Applications , January 1999, pp 55-69
[2] Fernandez, K..R., Robotic Simulation and a Method for Jacobian Control of a
Redundant Mechaninism with Imbedded Constraints, NASA Technical Paper
2807, 1998, 54 pages
[3] Kucuk S. and Bingul Z., An off-line simulation package for robotics education and
industrial purposes, 11th IEEE International Conference on Methods and Models
in Automation and Robotics, Poland, 2005.
[4] Huang B. and Milenkovic V., Kinematics of major robot linkages, Robot Int SME
2 (1993), 16_31.
[5] Craig J. J., Introduction to robotics: Mechanics and control, Addison-Wesley,
New York, 1989, 22-34
[6] Nethery J. F., and Spong M. W., Robotica: A mathematical package for robot
Analysis, IEEE Robot Autom Mag 1, 1998, 13_20.
Proceedings Papers:
[7] Fernandez, K.R., Use of Computer Graphic Simulation Techniques for Robot
Control Proceedings. III Computer Society No. 996, pp 433/438
[8] Wenrui D., Markus K., PIN-A PC-Based Robot Simulation and Offline
Programming System Using Macro Programming Techniques, IECON '99
Proceedings, IEEE Industrial Electronics Society, Vol. 1, 1999, pp. 442-446.
WWW pages:
[9] University of Western Australia, from http://robotics.ee.uwa.edu.au/robosim/,
[10] Peter Ullrich http://www.arrickrobotics.com/robomenu/robosim.html accessed on
2011-02-10
[11] Jef Mangelholc,
http://sites.google.com/site/jefmangelschots/home/robotics/links/software
accessed 2011-01-10
[12] Eggshell Robotics http://www.eggshell-robotics.com/blog/257-the-world-byindustrial-robot-density accessed 2011-02-11
862
MACHINING ROBOT CONTROLED AND PROGRAMMED AS A

MACHINE TOOL
Milutinovic D. 1, Glavonjic M. 2, Slavkovic N. 3, Kokotovic B. 4,
Milutinovic M. 5, Zivanovic S. 6, Dimic Z. 7
Summary: Industrial robots are promising cost-effective and flexible alternative for
certain multi-axis milling applications. The paper describes machining robot for
complex parts of light materials with lower tolerances having freeform surfaces. For the
experimental configuration of a 5-axis machining robot modelling approach and
prototype of developed control system with programming in G-code are shown. The
experimental prototype of low-cost control and programming system has been verified
by successful machining of several test work pieces of light materials.
Key words: Robot modelling, Control and programming system, Machining
1. INTRODUCTION
Industrial robots are promising cost-effective and flexible alternative for certain
multi-axis milling applications. Compared to machine tools, industrial robots are
cheaper and more flexible with potentially larger workspace. For these reasons, the
researchers, robot and CAM software manufacturers as well as people from machining
shops are enthusiastic to replace machine tools by robots for some machining
applications. These include milling materials, such as clay, foam, wax etc. for new
product design, styling and rapid prototyping projects. Robotic machining of work
pieces of traditional materials, such as wood, stone, aluminum etc. in which
dimensional tolerances are low or even middle also produce satisfactory results [1,2,3].
It is well known that poor accuracy, stiffness and complexity of programming are the
most important limiting factors for wider adoption of robotic machining in machine
shops [1,4,5].
1
prof dr Dragan Milutinovi, University of Belgrade, Faculty of Mechanical Engineering,

dmilutinovic@mas.bg.ac.rs
2
prof dr Milo Glavonji, University of Belgrade, Faculty of Mechanical Engineering,
mglavonjic@mas.bg.ac.rs
3
Nikola Slavkovi, University of Belgrade, Faculty of Mechanical Engineering, nslavkovic@mas.bg.ac.rs
4
Branko Kokotovi, University of Belgrade, Faculty of Mechanical Engineering, bkokotovic@mas.bg.ac.rs
5
Milan Milutinovi, Tehnikum Taurunum - High Engineering School Vocational Studies,
mmilutinovic@gavrogroup.rs
6
dr Saa ivanovi, University of Belgrade, Faculty of Mechanical Engineering, szivanovic@mas.bg.ac.rs
7
Zoran Dimic, LOLA Institute, Belgrade, zoran.dimic@li.rs
863
Milutinovic D., Glavonjic M., Slavkovic N., Kokotovic B., Milutinovic M., Zivanovic S., Dimic Z.
In order to contribute to efficient use of robots for machining applications,

research and development of reconfigurable robotic machining system were initiated
[6,7]. The research and development comprise two groups of problems. The first
relates to the realization of a specialized 5-axis vertical articulated machining robot with
integrated motor spindle in order to improve robotic machining accuracy. The second
refers to the development of the machining robot control and programming system
which can be directly used by CNC machine tool programmers and operators [8].
This paper describes the concept of machining robot for complex parts of light
materials with lower tolerances. For the experimental configuration of a 5-axis
machining robot, the modeling approach and prototype of developed low-cost control
system with programming in G-code are shown. The experimental prototype of control
and programming system has been verified by successful machining of several test
work pieces.
2. CONCEPT OF THE MACHINING ROBOT
A basic module of the proposed
concept of the robotic machining
system consists of a specialized 5-axis
vertical articulated robot, Figure 1, with
integrated motor spindle, similar to [9]
and with larger workspace, higher
payload and stiffness. Due to its
advantages in respect of stiffness [1]
and singularities [10], such robot would
operate as a specific vertical 5-axis
milling machine (X, Y, Z, A, B) spindletilting type [11]. In addition to the
mechanical structure shown in Figure 1,
whose realization is in progress, the
concept of this machining robot is
based on:
Low-cost control system based on
Fig. 1 Conceptual model of 5-axis
PC real-time Linux platform and
machining robot
EMC2 (Enhanced Machine Control)
software system [12,13] with programming in G-code;
The possibilities of using the existing CAD/CAM systems and reverse engineering
methods with implemented 3- to 5-axis machining for vertical milling machines (X,
Y, Z, A, B) spindle-tilting type;
Virtual machining robot configured in object-oriented programming language
Python implemented in the control system for program simulation and verification.
As it can be inferred the basic idea of developing the control and programming
system is that it should be a low-cost system directly applicable in machine shops by
the personnel experienced in CNC technology and programming in G-code that is still
widespread in industry [14]. For the development of experimental prototype of low-cost
control system with programming in G-code, a 6-axis vertical articulated robot with
payload of 50kg, Figure 2, was used as a testbed in a way that the sixth axis was
864
Machining robot controled and programmed as a machine tool
blocked. The robot is equipped with high speed motor spindle with maximum speed of
18000 min-1.
3.
Fig. 2 The experimental 5-axis

machining robot in reference position
MODELING APPROACH
As it was mentioned, the

experimental 5-axis machining robot from
Figure 2, will be considered in this paper
as a specific configuration of the 5-axis
vertical milling machine (X, Y, Z, A, B)
spindle-tilting type [11]. Figure 3
represents a geometric model of the
robot.
The robot reference frame {M}
has been adopted according to the
standard of this machine type [15]. The
tool frame {T} is attached to the milling
tool tip T in a way that axis zT coincides
with tool axis and the frame {W} is
attached to the work piece. The thus
configured machining robot, where
machining is performed on a work table in
front of the robot as well as limited
motions in joints relative to the reference
position allows for: taking into account
only one solution of inverse kinematic,
avoiding
the
robot
singularities,
conveniences related to the stiffness.
3.1 Joint and world coordinates
Fig. 3 Tool position and orientation in

robot reference frame {M} and work
piece frame {W}
Joint coordinates vector for this

vertical articulated robot is represented as
>i @T where
Ti , i
1, 2 , ..., 5 are scalar
joint variables controlled by actuators.

The description of world coordinates is based on tool path calculated by
CAD/CAM systems defined by the set of successive tool positions and orientations in
the work piece frame {W}, Figure 3. The thus calculated tool path is machine
independent and is known as a cutter location file (CLF). A tool pose is defined by the
position vector of the tool tip T in the work piece frame {W} as
pT and tool orientation
is defined by unit vector of the tool axis as W kT . In the general case, the tool tip
position vector and tool axis vector in robot reference frame {M} can be expressed as
M
M
pT
kT
>X M
>kTx
YM
M W
Z M @T M pOw W
R pT ;
kTy
kTz
@T
M W
W R
kT
(1)
865
where
pOw is the position vector of the origin of work piece frame {W}. Determining
the position vector M pOw and the orientation of the work piece frame {W} is conducted
according to the standard procedure for 5-axis CNC machine tools. It should be noted
M
that determining the orientation matrix W
R in equations (1) is determined and
executed later in control system without changing G-code. To complete the vector of
world coordinates, it is also needed to determine the tool orientation angles A and B
which define direction of tool axis zT that also coincides with axis of the last link of the
robot to which motor spindle is attached. Given that the robot has 5 DOF, only the
direction of tool axis zT is controllable, while axes xT and y T will have uncontrollable
rotation about it. The position and orientation of the tool frame {T} relative to robot
reference frame {M} can be described by homogenous coordinate transformation
matrix 4x4 [16-19] as
iTx jTx kTx X M
i
M
M
pT
M
TR
Ty jTy kTy YM
T
(2)
T
1 iTz jTz kTz ZM
0 0 0
0
0
1
0
M
pT represents
where the rotation matrix M
T R represents the orientation, while vector
the position of the tool frame {T} with respect to the robot reference frame {M}. To bring
the tool axis zT to a desirable orientation with respect to frame {M}, the tool frame {T}
must be rotated first about axis X M by angle A, and then about axis YM by the angle
B, as prescribed by the convention for 5-axis vertical milling machine (X, Y, Z, A, B)
spindle-tilting type. As it is known, the rotation matrix
M
TR
specifying the orientation of
tool axis zT can be derived as

M
TR
RYM ,B R XM , A
(3)
where R XM , A and RYM ,B represents basic rotation matrices [17]. Using the equations
(2) and (3) the angles A and B are determined as
kTx
k
(4)
, Tz )
cos( A ) cos( A )
Although for the angle A the second solution exists, by using the positive square root
the single solution for which 90q d A d 90q is always computed [11,18]. This way, the
2
A tan 2( kTy , 1 kTy
)
and
world coordinates vector can be expressed as x
A tan 2(
>X M
YM
ZM
A B @T .
3.2 Direct and inverse kinematics

To model the robot, the Denavit-Hartenberg (D-H) notation [16-18] was used.
Figure 4 shows D-H coordinate frames and link kinematic parameters for the
experimental 5-axis robot from Figure 2 i.e. Figure 3 in the reference position taking
into account the ranges of joint motions.
866
Substituting D-H parameters

of the links the transformation
matrices i 1i A, i 1, 2 , ...,5
are
obtained
first.
As
noticeable from Figure 4 the
frame {T} can be described
relative
to
the
frame
( x 5 , y 5 , z5 )
by
homogeneous transformation
matrix T5T .
Now, as it is wellknown [16-18], the tool
position and orientation i.e.
the position and orientation of
frame {T} with respect to the
robot reference frame {M},
Figure 4, for the given joint
coordinates vector and
specified link parameters can
be determined as
Fig. 4 D-H link coordinate frames and kinematic

parameters
M
0 1 2 3 4 5
T T 1 A2 A3 A4 A5 AT T
(5)
where for further consideration are of significance only

kTx sT1 sT 23 cT 4 cT 5 cT1 sT 4 cT 5 sT1 cT 23 sT 5
kTy
cT1 sT 23 cT 4 cT 5 sT1 sT 4 cT 5 cT1 cT 23 sT 5
kTz
XM
cT 23 cT 4 cT 5 sT 23 sT 5
a5 kTx sT1 ( d 4 cT 23 a2 sT 2 )
YM
a5 kTy cT1 ( d 4 cT 23 a2 sT 2 )
ZM
a5 kTz d 4 sT 23 a2 cT 2
and where T ij
(6)
Ti T j .
Through the vector M kT from equations (6) the angles A and B can be
determined using equations (4). This way, the world coordinates vector has been
completed i.e. direct kinematics problem is solved.
As noticeable from Figure 4, the last two joint axes z3 and z4 intersect at
point C (wrist center). This fact and the above presented remarks and constraints
make possible to consider this 5-DOF robot as a special case of 6-DOF robot with the
last three joint axes intersecting at a point to which Piepers method is applied [16,19].
As it can be concluded from Figure 4, the position of wrist center C is influenced only
by joint coordinates T1 , T 2 and T 3 . For the specified world coordinates the position
vector of the wrist point C can be calculated as
M
pC
pT M pTC
pT a5 M kT
(7)
867
Based on the calculated components of position vector M pC from equation (7), joint
coordinates T1 , T 2 and T 3 are solved using the geometric approach [8]. As it can be
seen from equation (5)
M
RT
M
3 R( T1 ,T 2 ,T 3
)3 RT ( T 4 ,T 5 )
(8)
As T1 , T 2 and T 3 are solved geometrically from equation (8) T 4 and T 5 are solved
analitically.
4. CONTROL AND PROGRAMMING SYSTEM
One of the distinctive features of the concept of machining robot, described in
Section 2, is that it is applicable directly by CNC machine tool programmers by using
the existing CAD/CAM systems and programming in G-code. Among several proposed
OAC solutions, the development of the first low-cost control system prototype is based
on PC real-time Linux platform with EMC2 software for computer control of machine
tools, robots, parallel kinematic machines, etc. EMC2 was initially created by the NIST
(National Institute of Standards and Technology) and is a free software released under
the terms of the GPL (General Public License) [12,13]. The development of the
machining robot control system prototype comprised a number of stages. For testing
the functions of inverse and direct kinematics, robot off-line programming, control
system behavior testing in real-time and collision detection, a virtual robot is
configured.
Figure 5 shows a simplified structure of the first prototype of low-cost control
and programming system where EMC2 software as a basic component is indicated.
EMC2 software system [13] is composed of four modules:
Motion controller (EMCMOT) is a real-time module. It performs trajectory
planning, direct and inverse kinematics calculations and computation of
desired outputs to motor drivers;
Discrete I/O controller (EMCIO) handles all I/O functions, which are not
directly related to the actual motions of machine axes;
Task coordinating module (EMCTASK) is a task level command handler
and program interpreter for the RS-274 NGC machine tool programming
language, commonly referred to as a G-code;
Graphical user interface (GUI). Among several user interfaces AXIS is the
most advanced GUI, featuring interactive G-code previewer. It is expanded
to specific application needs of the proposed robotic machining system.
Hardware abstraction layer (HAL) provides transferring of real-time data from
EMC2 to robot control hardware or to virtual robot. During control system start up a
choice is made for a corresponding configuration between real and virtual robot
control, Figure 5. It is common to start up first the control system configuration for a
virtual robot to visually detect possible collisions and to make final verification of the
program. Connections to drivers from PC side are done via appropriate machine
control interfaces including ADC, DAC and I/O channels. The virtual robot is configured
using several predefined Python classes in EMC2. Based on inverse and direct
kinematics equations, kinematic module is programmed in C language and is
integrated in EMC2 software system, Figure 5.
868
Fig. 5 Structure of the control and programming sytem

As obvious from Figure 5, the part programming is very conventional with the
use of a postprocessor to convert CLF into G-code. This means that the programmer
starts from the work piece CAD model in a common way, in this case in CAD/CAM
869
system Pro/Engineer, generating CLF. The generated tool path is tested through the
NC check (animated display of tool path and material removal). Using the configured
postprocessor for the vertical 5-axis milling machines (X, Y, Z, A, B) spindle-tilting type,
post-processing of CLF is done to obtain the robot program in G-code, which is
transferred to the robot control system. If the robot is initialized and tool and work piece
setting is done, the program can be tested in two ways. First, during G-code loading
EMC2 software displays the programmed tool path. However, the second way is of
crucial importance, because it employs a virtual robot. The virtual robot enables final
verification of G-code. After these verifications, the program can be safely executed on
the real robot.
5. EXPERIMENTS
Experiments
were
conducted
to
machine several test
work pieces of light
materials, Figure 7 using
experimental robot from
Figure 2. The setting of
robot reference position
and
elementary
calibration
was
performed
by
robot
manufacturers experts
as their contribution to
this research. The main
goal of the experiments
was to test capabilities of
the developed control
system prototype. The
CAD/CAM
system
Pro/Engineer was used
for the experiments with
the
idea
that
the
programming
of
Fig. 7 Examples of test work pieces
machining robot and
machining itself is done
in exactly the same way as it is done on a 5-axis vertical milling machine (X, Y, Z, A, B)
spindle-tilting type. Prior to machining, the programs are tested in two ways:
by graphical simulation of tool paths in EMC2, and
on virtual machining robot to perform the final program verification.
The experiments were organized to embrace 3-axis and 5-axis machining of
analytical and freeform (or sculptured) surfaces. For these xamples of test work
pieces, Figure 7, styrofoam and high-density polyurethane based material where used.
For these experiments, specific flat endmills and ball-endmills [20] designed by one of
the coauthor are used. These experiments confirmed that it is possible to realize
machining robot with low-cost control and programming system for the complex870
surfaces parts of light materials and lower tolerance, which can be directly used by
CNC machine tools programmers and operators.
6. CONCLUSION
The paper describes the concept of machining robot for complex parts of light
materials with lower tolerances. The 5-axis vertical articulated robot is considered as a
specific configuration of 5-axis vertical milling machine (X, Y, Z, A, B) spindle tilting
type. Robot modeling approach is shown in detail as well as the prototype of the
developed low-cost control and programming system based on EMC2 software
system. Verification of the experimental prototype of developed low-cost control and
programming system is presented using the examples of machining of several test
work pieces of light materials. The shown examples of test work pieces comprised 3axis and 5-axis machining of analytical and freeform surfaces, where programming and
the machining itself were performed according to the procedure applied for CNC
machine tools. The developed and investigated experimental prototype of low-cost
control and programming system indicates that such commercial system may be
superior to the compatible robotic machining solutions, considering the G-code is still
very widely used in industry. The subsequent stages of research will involve the
development of specialized 5-axis vertical articulated machining robot and control
system that will enable 3-axis and 5-axis machining by various combinations of robots
axes and additional rotational and translational axes.
7. ACKNOWLEDGEMENT
The authors would like to thank the Ministry of Science and Technological
Development of Serbia for providing financial support that made this work possible.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
Abele E., Kulok M., Weigold M. (2005). Analysis of a machining industrial robot.
Proceedings of 10th International Scientific Conference on Production
Engineering-CIM2005, p. II 1-11.
Abele E., Weigold M., Rothenbucher S. (2007). Modeling and identification of an
industrial robot for machining applications. Annals of the CIRP, vol.56, no.1,
p.387-390.
Shin-ichi M., Kazunori S., Nobuyuki Y., Yoshinari O. (1999). High-speed end
milling of an articulated robot and its characteristics. Journal of Materials
Processing Technology, vol.95, no.1-3, p. 83-89.
Pan Z., Zhang H. (2008). Robotics machining from programming to process
control: a complete solution by force control. Industrial Robot: An International
Journal, vol. 35, no. 5, p. 400-409.
DePree J., Gesswein C. (2008). Robotic machining white paper project-Halcyon
Development, http://www.halcyondevelop.com.
Milutinovic D., Glavonjic M., Zivanovic S., Dimic Z., Slavkovic N. (2009).
Development of robot based reconfigurable machining system, Proceedings of
33rd Conference on Production Engineering of Serbia, Belgrade, p.151-155.
871
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
872
Milutinovic D., Glavonjic M., Zivanovi S., Slavkovic N. (2009). 5 axis robot based
reconfigurable machining system, Proceedings of 9th Intermational Conference
on Accoplishments of Electrical and Mechanical Industries, Banjaluka, p.273-280
Milutinovic D., Glavonjic M., Slavkovic N., Dimic Z., Zivanovic S., Kokotovic B.,
Tanovic Lj. (2011). Reconfigurable robotic machining system controlled and
programmed in a machine tool manner. Int J Adv Manuf Technol, vol.53, p.12171229.
STAUBLI. High Speed Machining (HSM) robot, from
http://www.staubli.com/en/robotics/robot-solution-application/high-speedmachining-robot/, accessed on 2011-04-07.
Affouard A., Duc E., Lartigue C., Langeron J.M., Bourdet P. (2004). Avoiding 5axis singularities using tool path deformation. International Journal of Machine
Tools & Manufacture, vol. 44, no.4, p.415-425.
Lee R.S., She C.H. (1997). Developing a postprocessor for three types of fiveaxis machine tools. Int J Adv Manu Technol, vol. 13, no.9, p. 658-665.
Real-Time Control Systems Library Software and Documentation,
http://www.isd.mel.nist.gov/projects/rcslib/, accessed on 2011-04-07.
LinuxCNC, EMC's webpage, http://www.linuxcnc.org/, accessed on 2011-04-07.
Shin S.J., Suh S.H., Stroud I. (2007). Reincarnation of G-code based part
programs into STEP-NC for turning applications. Computer-Aided Design, vol.39,
no.1, p. 1-16.
ISO 841:2001 Industrial automation systems and integration - Numerical control
of machines - Coordinate system and motion nomenclature.
Paul R.P. (1981). Robot Manipulators: mathematics, programming and control.
The MIT Press.
Fu K.S., Gonzalez R.C., Lee C.S.G. (1987). Robotics: control, sensing, vision,
and intelligence, McGraw-Hill.
Craig J.J. (1989). Introduction to robotics: mechanics and control, 2nd ed.,
Addison-Wesley.
Spong M.W., Vidyasagar M. (1989). Robot Dynamics and Control, John Wiley &
Sons.
Gavro Group, http://www.gavrogroup.rs/, accessed on 2011-04-10.
NEW APPROACH FOR DESSIGNING ROBOT PROGRAMING

SYSTEM BASED ON L-IRL PROGRAMING LANGUAGE
Vladimir Kaplarevi1, Marija Milievi2, Jelena Vidakovi3, Vladimir Kvrgi4
Abstract: L-IRL is robot programing language based on the procedural paradigm.
Originally, L-IRL was Pascal-based programing language, but in the new version there
are functionalities based on other modern programing languages. New language
parser is formed as LR(1) type parser and written with tools such as Bison and Lex
with the C++ progaming language. This aproach enables logical and functional
separation between different phases of parsing and compiling. L-IRL is using XML as
one of the main comunication tools between different elements of the system. XML is
used as meta language for system specification file and as object code of the compiler.
1. INTRODUCTION
L-IRL (Lola Industrial Robot Language) is programing language based on the
procedural paradigm and it is basis of the offline part of the robot programing system.
Language itself is a Pascal-based programing language with some syntax and
functional modifications. Main idea for developing this and previous versions of the LIRL is to design compact, portable and programing language easy to learn and use[1].
For its main use, robot programing, handling many languages constructs such as:
dynamic arrays, parallel executions blocks, system variables definition using XML is
simplified. Portability is enabled by developing interpeter and object code compiler
based on XML. Aside from basic elements which exists in other programing languages,
L-IRL contains specific structures and language contstructs based on DIN 66312
standard. Special language constructs proposed by the standard are special geometric
datatypes, geometric expressions and move statements which are used for robot
movement controll. Currently defined robot movements are PTP(Point to Point)
movements and movements along mathematicalty defined paths(CP movements) as
well as aproximate PTP and CP motions. Currently present mathematicaly defined
paths are line and the circle. By specifying parameters of the move statement it is
posible to change path, speed, acceleration, orientation of the end-effector and other
characterists of the movement.
1
Vladimir Kaplarevi, Dipl. Mat. Belgrade, Lola Institute, vladimir.kaplarevic@li.rs

Marija Milievi, Dipl. Eng. E.E. Belgrade, Lola Institute, marija.milicevic@li.rs
Jelena Vidakovi, Dipl. Eng. M. E. Belgrade, Lola Institute, jelena.vidakovic@li.rs
4
Dr. Vladimir Kvrgi, Dipl. Eng. M. E. Belgrade, Lola Institute, vladimir.kvrgic@li.rs
2
3
873
Vladimir Kaplarevi, Marija Milievi, Jelena Vidakovi, Vladimir Kvrgi
In this new system XML is extensively used as one of the main ways of
comunication between different parts of the system. XML is used in the system
specification file which introduces new variables into working memory of the program,
specifies behavior of some built-in languages structures and defines special
record(structure) called Robot which defines robot characteristics, constraints, etc.
Object code based on XML contains instructions for the virtual machine and it is
interpreted in real-time.
Language parser is developed using software tools Lex and Bison which
generate lexical and syntax analyzer of the source code. In previous version, parser
and compiler were developed using recursive descent approach. In this version of LIRL parsing is based on LALR(Look Ahead Left to Right) algorithm which is directly
incorporated into Bison. Using these software tools together with methodologies of the
object oriented programming, efficient structural and logical refactoring of the source is
achived. RapidXML parser for C++ was used for parsing and generating XML code.
2.
PROGRAM STRUCTURE IN L-IRL LANGUAGE AND BASIC STRUCTURE OF

THE XML OBJECT CODE
L-IRL is procedural language with rigid syntax and every program obeys certain
structural rules. Program in L-IRL comprises of these parts: linking with extern files
part, for declaration of contstants part, part for declaration of user defined types,
variable declaration part, part for defining procedures and functions and main
execution block[2].
Beside the source code in L-IRL language, in order to interpret and compile the
source code, system specification file - sysspec.xml needs to be present included in
the execution folder. Compiler translates source code of the program in L-IRL together
with system specification file into XML based object code. Basic structure of the object
code is divided in three basic parts. Object code starts with root node <program>
which has three child nodes:
<system spec> - part which matches content of the system specification file,
<symbols> - part which provides information about defined symbols and
<seq> or <par> - part which defines sequential or parallel main execution
block.
3. SYSTEM SPECIFICATION AND DTD

System specification file is used for inclusion of system variables inside
memory of the program in execution as well as describing behaviour of certain built-in
structures of the language. System specification filesysspec.xml is written in XML
874
New Approach for Dessigning Robot Programing System Based On L-IRL Programing Language
which is defined by special syntax rules i.e. meta language. Since this file can be
changed by L-IRL programer it is necessary to define rules of the XML syntax used in
this file. Approach used for describing the rules of the XML syntax is called Document
Type Definitions[3] or DTD.
A document type definition provides a list of the elements, attributes, notations,
and entities contained in a document, as well as their relationships to oneanother.
DTDs specify a set of rules for the structure of a document. DTDs can be included in
the file that contains the document they describe, or they can be linked from an
external URL. Such external DTDs can be shared by different documents and Web
sites or other software. DTDs provide a means for applications, organizations, and
interest groups to agree upon, document, and enforce adherence to markup standards.
4. XML BASED OBJECT CODE
Basic statements of the object code are statements which define structure of
generated XML file. Abstract syntax tree(AST) matches DOM tree of the XML object
code document. Interpretation of the object code by real-time virtual machine is
achived by traversing a DOM tree of the XML document and executing statements
and expressions coded within tree nodes. This way the resulting object code has same
semantics as source code of the program. Generated XML file has node <program> as
root node. As mentioned earlier, root node has three child bodes: <system spec>,
<symbols> and <seq> or <par>. Each of these nodes defines separate section of the
generated XML object program file. Node <system spec> defines section of XML
object file that contains content of the system specification file. Content of the this node
is defined in compilation phase by adding the root node of the system specification file
to the root node of object program. Second node <symbols> defines section which
contains definitions of variables, constants, procedures and functions. Third child node
of the root node, defines section which contains translated statements, jump
statements, etc. Element <seq> defines sequential execution of the main execution
block while <par> defines parallel execution of the main execution block.
Syntax check of the object code is enabled with the use of XML Schema. This
way the syntax check is same as validation of the object code using XML Schema
validation.
4.1 Basic Statements
Basic statements of the object code are statements which define structure of
generated XML file. Generated XML file has node <program> as root node. As
mentioned earlier, root node has three child bodes: <system spec>, <symbols> and
<seq> or <par>. Each of these nodes defines separate section of the generated XML
object program file. Node <system spec> defines section of XML object file that
contains content of the system specification file. Content of the this node is defined in
compilation phase by adding the root node of the system specification file to the root
node of object program. Second node <symbols> defines section which contains
definitions of variables, constants, procedures and functions. Third child node of the
root node, defines section which contains translated statements, jump statements, etc.
Element <seq> defines sequential execution of the main execution block while <par>
defines parallel execution of the main execution block.
875
Vladimir Kaplarevi, Marija Milievi, Jelena Vidakovi, Vladimir Kvrgi
4.2 Jump Statements

In order to generate object code for some statements(loops, if statement, )
labels need to be defined. Labels define target address for jump instructions and they
are defined with tag: <label>. Unconditional jump instruction is defined with tag <jmp>.
As required attribute, jump instruction takes value of the target label. Conditional jump
is performed with the instruciton <jmpif>. Similar to unconditional jump instruction, label
value has to be provided in the form of attribute. Jump condition is provided as a child
node of the <jmpif> node.
4.3 Special Statements
Special statements are move statements and statements that manage parallel
execution(wait, signal, etc...). Move statements are provided with special syntax for
defining all move parameters. Move statements are robot motion statements. They are
provided with special syntax due to define parameters that represent all necessary
information for robot movement control. They consist of parameters for determining the
type of path-PTP, LIN, CP which are followed by data of geometric type describing the
target and, if necessary, extra point of the path. With these parameters, there are
additional parameters (ACC, ACC PTP, C ORI, C SPEED, SPEED etc.) that are used
optionally and they define the characteristics of the movement [1].
5. CONCLUSION
This paper introduced a new approach for designing and developing offline part
of the robot programming system with special emphasis on development of interpeter
and compiler and use of XML. XML is extensively used in web programming [5], but its
use is found in other ares as well. One use of XML as object code is shown in this
paper. Further work is focused on development of real-time virtual machine which
interprets XML object code.
LITERATURE
[1]
[2]
[3]
876
Vladimir M. Kvrgi, Razvoj inteligentnog sistema za upravljanje i programiranje

industrijskih robota, doktorska distertacija, Univerzitet u Beogradu, Mainski
fakultet, 1998.
Marija A. Pavlovi, Vii programski jezik za multirobotske operacije, magistarski
rad, Univerzitet u Beogradu, Elektrotehniki fakultet, 1994.
Elliotte Rusty Harold, XML Bible, IDG Books Worldwide, Inc W3C, Extensible
Markup Language (XML) 1.0 (Fifth Edition), http://www.w3.org/TR/REC-xml/
PRACTICAL STRATEGIES FOR STABILISATION OF ALGORITHMS

BASED ON SECULAR EQUATIONS OF RSPDTM
Kostic Aleksandra1, Velic Melisa2, Bektesevic Jasmin3
Summary: From Pisarenkos work [9] the problem of finding the smallest eigenvalue
of a real symmetric, positive definite Toeplitz matrix (RSPDTM) plays an important role
in signal processing. For this reason there are more methods for determination of
smallest eigenvalue. These methods can be divided in three large groups: methods
based on characteristic polynomial, methods based on secular equation and hybrid
methods. It can happen that the smallest eigenvalue of matrix and the smallest
eigenvalue of matrix are very close to each other, which can result in very large speed
reduction of the second group of methods and in the extreme cases it can result in
interruption of algorithm which is based on second group of methods. In this note we
bring new strategies for overcoming this problem and we compare algorithms based on
these strategies with already existing algorithms. These new strategies are based on
mathematical modeling.
Key words: Toeplitz matrix, eigenvalue, practical strategies, secular equation
1.
INTRODUCTION
From Pisarenkos work [9] the problem of finding the smallest eigenvalue of a
real symmetric, positive definite Toeplitz matrix (RSPDTM) plays an important role in
signal processing. The computation of the minimum eigenvalue of Tn was studied in, e.
g. [1,2,4,5,6,7,8,10,11]. Cybenko and Van Loan [1] were the first to present an
algorithm which is a combination of bisection and Newtons method for the secular
equation, which converges monotony and quadratically to O1(n). However, their
algorithm often ad problem because the smallest root of secular function and pole of
secular function were too close to each other. In some cases the problem was
extremely slow algorithm and in some cases the problem was brake of algorithm. This
approach was improved considerably in [2,5] by replacing Newtons method by a more
appropriate root finding methods for the secular equation. Taking advantage of the fact
that the spectrum of a symmetric Toeplitz matrix can be divided into even and odd
parts the methods based on the secular equation were accelerated in [7,8,10].
Doc. dr., Aleksandra Kostic, Sarajevo, Faculty of Mechanical Engineering, (kostic@mef.unsa.ba)

BSc., Melisa Velic, Sarajevo, Faculty of Mechanical Engineering, (melisa_velic@hotmail.com)
3
Ass., Jasmin Bektesevic, Sarajevo, Faculty of Mechanical Engineering, (bektesevic@mef.unsa.ba)
2
877
Aleksandra Kostic, Melisa Velic, Jasmin Bektesevic
The paper is organized as follows. In Section 2 we present the basic properties

of Toeplitz matrices and the notation we will use. In Section 3 we summarize properties
of underlying methods using secular equation. In Section 4 we suggest two practical
strategies for stabilization of algorithms based on secular equations of RSPDTM. The
paper closes with concluding remarks and suggestions for further research.
2. PRELIMINARIES
The Teoplitz matrix is quadratic matrix which has the same elements in its
respective diagonals, which means that . Because we take in
consideration symmetric matrixes it also means that . The Teoplitz matrix is
defined with vector 1, t1 , , t n 1 R n so the (i,j)th element of an n x n symmetric
Toeplitz matrix Tn is given by
t i j . We can conclude, form above given, that Teoplitz
matrixes are centrosymmetric and satisfy JTnJ= Tn. We use for the identity matrix and
J for the exchange, or flip matrix. For simplicitys sake, our
notation will not explicitly indicate the dimension of the matrices and J.
The vector x is an eigenvector of Tn if and only if Jx is. If
eigenvalue of Tn then from x

2
Jx we say that
Jx 2 we obtain x
Jx or x Jx . If x
is a simple
an eigenvector x symmetric and the corresponding eigenvalue O is even. In an

analogous manner, we define skew-symmetric eigenvector and odd eigenvalue.
In what follows, an important role is played by the so-called Yule-Walker
equations. For an n x n symmetric Toeplitz matrix Tn, defined by 1, t1 , t 2 , , t n 1 , this
(n)
system of linear equations is given by Tny =-t where t=(t1,,tn) . There exist several
methods to solve these equations. Durbins algorithm solves them by recursively
computing the recursively computing the solutions to lower-dimensional systems,
provided all principal sub matrices are non-singular. This algorithm requires 2n2+O(n)
flops. There are too: Split-Durbin algorithm, Split Levinson algorithm, Even-Odd
algorithm, Even-Odd Split Levinson algorithm and superfast methods
3.
RATIONAL APPROXIMATION
Firstly, we will define secular function. Let
Tn
t
i j
i , j 1,, n
R ( n ,n ) be a
RSPDTM. We denote by T j R ( j , j ) its j-th principal sub matrix, where t 0 1 . Let
O1( j ) d O(2 j ) d d Ojj be the eigenvalues of T j then the interlacing property
O(jk)1 d Ojk11 d Ojk ,2 d j d k d n, holds. If O is not in the spectrum of Tn 1 then block
Gauss elimination of the variable x2 , , xn of the system
we get 1 O t T Tn1 OI 1 t x1
878
1 O
tT
x
t
Tn 1 OI
0 that characterizes the eigenvalues. We assume that
Practical Strategies for Stabilization of Algorithms Based on Secular Equations of RSPDTM
O1( n) O1( n1) . Because
is eigenvector of Tn, than x1 z 0 . From this follows that all

eigenvalues of Tn are roots of the function qO : 1 O t T Tn1 OI 1 t . Function is
called secular function. Secular function may be rewritten in modal coordinates as
q (O )
n 1
1 O
j 1
t v
T
( n 1)
j
j 2
O
,where v
denotes the eigenvector of Tn 1 corresponding to
O(jn 1) . From the last equation is obvious that the poles of the secular function are
actually eigenvalues of RSPDTM Tn 1 . From explained above we can see that there is
a significant problem if
, especially if
.
Methods based on finding the roots of the secular functions are presented in [1,2,5,]. It
is interesting that the secular function q(O ) may be written as q(O ) p n O ,where
p n 1 O
p n O and p n 1 O are characteristic polynomials of Tn and Tn 1 respectively.

But it can happen that the smallest eigenvalue of matrix Tn and the smallest
eigenvalue of matrix Tn-1 are very close to each other, which can result in very large
speed reduction of the second group of methods and in the extreme cases it can result
in interruption of algorithm which is based on second group of methods.
Next figure shows and in interval
respectively. It can be seen that and

are very close to each other, but they
are far enough from

. Scaling factors are taken for the purpose of better figure
quality.
-9
10
x 10
-2
0.002
0.004
0.006
x
0.008
0.01
0.012
Fig. 1
Figure 1 gives motivation to deliberate two practical strategies for stabilization
of algorithms based on secular equations of RSPDTM.
4.
PRACTICAL STRATEGIES
We bring two practical strategies for stabilization of the algorithms based on

secular equations of RSPDTM and they are exploiting symmetry in rational
interpolation and non-symmetric strategies.
4.1 Exploiting symmetry in rational interpolation
Exploiting symmetry in rational interpolation is very important in the Teplitz
matrixes. This problem was discussed by [3,7,8,10]. Their specific structure, as we
879
have seen, enables us to divide spectrum in odd and even parts. With this, it is now
clear that there is even and odd secular function, whose roots are odd and even
eigenvalue respectively. For determination of the odd and even secular function we will
consider
where .
To take into account the symmetry properties of the eigenvector we eliminate
the variables from the system given above.
Than, every eigenvalue of Tn which is not in the spectrum of is an
eigenvalue of two dimensional nonlinear eigenvalue problem

(1)

Hence, if the smallest eigenvalue is even, than it is the smallest root of

rational function
(2)

And if is an odd eigenvalue of than it is the smallest root of
(3)

The last two functions are called even and odd secular functions, respectively.
If the symmetry class of the principal eigenvector is known in advance than a straight
forward generalization of the scheme in [2,5] can be based on (2) and (3), respectively.
In the general case it is the minimum of the smallest roots of and , and the
symmetry class must be detected by method itself.
The functions and can be written as

From last equation follows that poles of even and odd secular functions are
eigenvalues of the matrix Tn-2, which means that poles of even and odd secular
functions are more distant form its roots than in secular function. The existence of
rational approximations of even and odd secular function is given in [10].
It is interesting that where is even and odd
characteristic polynomial. This result is given in [3,7].
Table 1 contains the average number of flops and average number of Durbin
steps needed to determine the smallest eigenvalue in 100 test problems with each of
dimensions and . The iteration was terminated if the
error was guaranteed to be less than 10-6. For comparison we added the results for the
nonsymmetrical method [2].
880
Table 1 Rational Hermitian interpolation

Dimension
non-symmetric
method from
[2], flops
1.086 E04
4.639 E04
1.804 E05
7.837 E05
3.512 E06
1.531 E07
32
64
128
256
512
1024
non-symmetric
method from
[2], steps
4.34
5.14
5.25
5.84
6.62
7.26
symmetric
method, flops
9.087 E03
3.653 E04
1.407 E05
6.046 E05
2.597 E06
1.065 E07
symmetric
method, steps
3.75
4.12
4.14
4.55
4.92
5.08
Algorithm was improved because the smallest pole of even and odd secular
function is more distant from the smallest eigenvalue than pole of secular function.
4.2 New strategies
In this section we will generalize strategies given in [9], which is based on pole
moving of the secular function. In the following determinant is defined the characteristic
polynomial
pn ()
of matrix Tn where A 1 O t1 and BT= We denote w ( n 2 ) t2 ,t3 ,,tn 1 T
t1 1 O
, Tn 2 OI z
( n 2 )
w
( n 2 )
y ( n 2 ) , y n 2 , , y ( n 2 )
2
3
n 2
, y
and t t2 ,t3 ,,tn 2 .

We get the equation
multiplication. To find recursion for Tn 2 OI z( n 2 )
get
w
from matrix
we will analyze next equation
( n 2 )

We get

For above proposed calculation we spent 6(n-3) flops.
In equation (2) by using block elimination matrix B is eliminated. This way we

pn()

where is

We get following recursion

881
Secular equation is now
. Relation between old and new secular
function as well as its appearance is given in next figure.

0.8
0.04
0.7
0.035
0.6
0.03
0.5
0.025
0.4
y
0.02
0.3
0.015
0.2
0.01
0.1
0.005
0
-0.1
-0.2
-0.005
0.002
0.004
0.006
x
0.008
0.01
0.012
0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009

x
0.01
Fig. 2
New idea is based on the inductive concluding. To generalize presented by
making block separation of matrices
pn ()=
,

while matrix A and matrix B are dimensioned and respectively.
Because of interstep prices which is 6(n-k+1)(k-1) this makes sense in case of
1<<k<<n. In this way we get new secular equation

whose smallest pole is quite distant from smallest eigenvalue which gives us more
freedom in modeling new secular equation.
To generalize use of the symmetry properties as in first strategy, but now on
new generalization, we will consider problem of eigenvalues written in next manner
where and matrixes A, B, C are defined in harmony with idea presented in

this paper. Than from using symmetry properties we get a nonlinear eigenvalue
problem

which will be the subject of further research.
5.
CONCLUSION
We have presented new secular equation of RSPDT matrix Tn and theoretically

constructed its rational approximation. Our goal is to improve already existing
algorithms which are based on secular equation [2]. With new algorithm we try to
overcome the situation when minimal eigenvalues of matrix Tn and Tn-1 are too close to
each other. In this note, we presented two practical strategies for stabilisation of
algorithms based on secular equations of RSPDTM.
In further research it is necessary to practically confirm suggested algorithm,
compare it with previous algorithm and use symmetry properties of eigenvector.
882
LITERATURE
[1]
Cybenko, G., Van Loan, C.F. (1986). Computing the minimum eigenvalue of a
symmetric positive definite Toeplitz matrix. SIAM J. Sci. Stat. Comput, vol. 7, no.
1, p. 123-131
[2] Kostic, A., Voss, H. (2002). A method of order 1 3 for computing the smallest
eigenvalue of a symmetric Toeplitz matrix. WSEAS Transactions of Mathematics,
vol. 1, p. 1-7
[3] Kostic, A., Voss, H. (2004). Recurrence relations for the even and odd
characteristic polynomials of symmetric Teoplitz matrix. Comp. Appl. Math.
173:p.365-369
[4] Mackens , W. , Voss, H. (2000). Computing the minimum eigenvalue of a
symmetric positive definite Toeplitz matrix by Newton-type methods. SIAM J. Sci.
Comput., vol. 21.no. 4, p. 1650-1656
[5] Mackens , W. , Voss, H. (1997). The minimum eigenvalue of a symmetric positive
definite Toeplitz matrix and rational Hermitian interpolation. SIAM J. Matr. Anal.
Appl, vol. 18. p. 523-534
[6] Mastronardi, N., Boley, D.(1999). Computing the smallest eigenpair of a
symmetric positive definite Toeplitz matrix. SIAM J. Matr. Anal. Appl.
[7] Melman, A. (2003). Computation of the smallest even and odd eigenvalues of a
siymmetric postive definite Teoplitz matrix. SIAM J. Sci. Comput. vol. 20, no. 5,
p. 1921-1927
[8] Melman, A. (2006). Computation of the Newton step for the even and odd
characteristic polynomials of a symmetric positive definite Toeplitz matrix.
Mathematics of computation. vol. 75, no 254, p. 817-832,
[9] Pisarenko, V. F. (1973) The retrieval of harmonics from a covariance function.
Geophys. J. R.. astr. Soc. vol. 33., p. 347-366
[10] Voss, H. (1999) Symmetric schemes for computing the minimum eigenvalue of a
symmetric Toeplitz matrix. Lin.Alg.Appl.387:359-371
[11] Kostic A., Velic M., Hadiabdic V.(2010). New secular equation of RSPDT d its
rational approximation, Proceedings of 21th International DAAAM Symposium,
p0043-0044
883
REMOTE MEASUREMENTS BY ZIGBIT WIRELESS MODULE

Platon Sovilj1, Nenad abrilo2, Vladimir Vujii3, Ivan upunski4
Summary: System which demonstrates remote measurements, using ZigBee wireless
communication protocol, has been developed.The system is consisted of one remote
microprocessor measurements system with ZigBit communication module (transmitter
module), a receiving module and PC with adequate software for receiving and
processing of signal sent by transmitting module.
Key words: remote measurements, ZigBee, sensors, transducers
1.
INTRODUCTION
Modern measurement systems are often based on transducing physical

quantity to electrical signal. Conditioning of the signal, its processing, remote
transmitting and results presentation have been improved by time. Physical quantitives
are transduced to electrical signals by various measurement transducers, which
nowdays are usually called sensors[1-3].
ZigBee is wireless communication protocol based on standard IEEE 802.15.4
2003. IEEE 802.15.4 defines physical and MAC layer of OSI referrent model for LR
WPAN (Low Rate Wireless Personal Area Network), and ZigBee is the extension of
this standard, defining the rest of layers up to the level of application[4].
This paper presents the development platform consisted of:
transmitting microprocessor measurement system with ZigBee communication module (actually ZigBit module by Meshnetics is used)
receiving station microprocessor system with identical ZigBee module and
PC with software application for receiving and processing measurement
results.
Transmitting module is conected with next measurement transducers: digital
sensor for measurement of light intensity (TSL230BR), digital temperature sensor
DS1820 and analog transducer for measuring concentration of oxygen and methane
1
Dr, Platon Sovilj, Novi Sad, Fakultet tehnikih nauka, (platon@uns.ac.rs)

Nenad abrilo, Novi Sad, Fakultet tehnikih nauka
Dr, Vladimir Vujii, Novi Sad, Fakultet tehnikih nauka
4
Dr, Ivan upunski, Novi Sad, Fakultet tehnikih nauka
This work was supported in part by the Ministry of Science and Technological Development, Republic of
Serbia, under Research Grant TR32019 Measurement in Concept of Smart Grids, which started on
January 1, 2011.
2
3
885
Platon Sovilj, Nenad abrilo, Vladimir Vujii, Ivan upunski
KE-50 and TGS2611 [5-6]. Communication between transmitting and receiving station
is performed by ZigBee protocol, using digital communication chips ZDM-A1281-A2.
Both microprocessor modules (transimtting and receiving station) are implemented
with 8-bit PIC 18F8520 microcontroller, and firmware is developed by Mikroelektronika
IDE MikroC. Receiving station communicates with PC by series connection, and
software application for acquisition, processing and presenting of data is developed by
software package LabVIEW 7.1.
2.
TRANSMITTING MICROPROCESSOR MODULE
As stated above, transmitting module is conected with next measurement

transducers: digital sensor for measurement of light intensity (TSL230BR), digital
temperature sensor DS1820 and analog transducer for measuring concentration of
oxygen and methane KE-50 and TGS2611. Digital temperature sensor DS1820 and its
connection to microcontroller is presentd in more details in this paper.
DS18B20 enables 9-bit and 12-bit temperature measurement. Digital
information is sent to/from the DS1820 over a 1Wire interface, so that only one wire
(and ground) needs to be connected from a central microprocessor to a DS1820.
Power for reading, writing, and performing temperature conversions can be derived
from the data line itself (without an external power source). Each DS1820 contains a
unique silicon serial number, so multiple DS1820s can exist on the same 1Wire bus.
This allows for placing the sensors in many different places. Applications where this
feature is useful include HVAC environmental controls, sensing temperatures inside
buildings, equipment or machinery, and in process and environment monitoring and
control [7].
Fig. 1 DS18B20 block diagram.

The block diagram of figure 1 shows the major parts of the DS1820. The
DS1820 has three main data parts: 1) 64bit lasered ROM, 2) temperature sensor, and
3) nonvolatile temperature alarm triggers TH and TL. The device derives its power from
the 1Wire communication line by storing energy on an internal capacitor during
intervals of time when the signal line is high and continues to operate off this power
886
Remote measurements by zigbit wireless module

source during the low times of the 1Wire line until it returns high to replenish the
parasite (capacitor) supply. As an alternative, the DS1820 may also be powered by an
external 5 volts supply.
Communication to the DS1820 is via a 1Wire connection. With the 1Wire
port, the memory and control functions will not be enabled before the ROM function
protocol has been established. The master element must first provide one of five ROM
function commands: 1) Read ROM, 2) Match ROM, 3) Search ROM, 4) Skip ROM, or
5) Alarm Search. These commands operate on the 64bit lasered ROM portion of each
device and are able to single out a specific device if many are present on the 1Wire
line as well as indicate to the Bus Master how many and what types of devices are
present. After a ROM function sequence has been successfully executed, the memory
and control functions are accessible and the master element may then provide any one
of the six memory and control function commands.
One control function command instructs the DS1820 to do a temperature
measurement. The result of this measurement will be placed in the DS1820s
scratchpad memory, and can be read by issuing a memory function command which
reads the contents of the scratchpad memory. The temperature alarm triggers TH and
TL is consisted of one byte EEPROM each. If the alarm search command is not
applied to the DS1820, these registers can be used as general purpose user memory.
Writing TH and TL is performed using a memory function command. Read access to
these registers is enabled through the scratchpad. All data is read and written, having
least significant bit first.
The core of the module is microcontroller PIC18F8520, used in BIGPIC5
microprocessor development board. The main characteristics of this microcontroller
are [8]:
C compiler optimized architecture/instruction set: Source code compatible
with the PIC16 and PIC17 instruction sets
Linear program memory addressing to 128 Kbytes
Linear data memory addressing to 3840 bytes
1 Kbyte of data EEPROM
Up to 10 MIPs operation:
x DC 40 MHz osc./clock input
x 4 MHz 10 MHz osc./clock input with PLL active
16-bit wide instructions, 8-bit wide data path

Priority levels for interrupts
31-level, software accessible
Address capability of up to 2 Mbytes of external memory
16-bit interface for external memory
High current sink/source 25 mA/25 mA
Four external interrupt pins
Five timer/counter modules
Secondary oscillator clock option Timer1/Timer3
Five Capture/Compare/PWM (CCP) modules:
x Capture is 16-bit, max. resolution 6.25 ns (TCY/16)
x Compare is 16-bit, max. resolution 100 ns (TCY)
887
x PWM output: PWM resolution is 1 to 10-bit

Master Synchronous Serial Port (MSSP) module with two modes of
operation:
x 3-wire SPI (supports all 4 SPI modes)
x I2C Master and Slave mode
Two Addressable USART modules:

Supports RS-485 and RS-232
Parallel Slave Port (PSP) module
10-bit, up to 16-channel Analog-to-Digital Converter (A/D)
Low-power, high-speed Flash technology
Fully static design
Wide operating voltage range (2.0V to 5.5V)
Industrial and Extended temperature ranges
ZigBit stands for ultra-compact, low-power, high-sensitivity 2.4 GHz

802.15.4/ZigBee OEM modules from MeshNetics, based on Atmels mixed-signal
hardware platform. They are designed for wireless sensing, control and data
measurement and acquisition applications. The ZigBit modules lower the need for
costly and time-consuming RF development, and shorten time to market for a wide
range of wireless applications.
Two types of ZigBit modules are available: ZDM-A1281-B0 module with
balanced RF port for applications where the benefits of PCB or external antenna can
be utilized and ZDM-A1281-A2 module with dual chip antenna satisfying the needs of
size sensitive applications.
ZigBit features standards-based networking stack, based on IEEE802.15.4
PHY and MAC layers, and ZigBee NWK/APS/ZDO layers[9]. It enables multipoint,
multihop communications over an area of thousands of square meters at moderate
data rates with relatively cheap infrastructure support. The architecture of the Wireless
Sensor Networks (WSN) allows for implementation of low powered devices.
The key features of ZDM-A1281-A2 chip are[9]:
Ultra compact size (24 x 13.5 mm)
Balanced dual chip antenna design with antenna gain of approximately 0 dBi
High RX sensitivity (-101 dBm)
Outperforming link budget (104 dB)
Up to 3 dBm output power
Very low power consumption (< 6 A in deep sleep mode)
Ample memory resources (128K bytes of flash memory, 8K bytes RAM, 4K
bytes EEPROM)
Wide range of interfaces (both analog and digital):
x
x
x
x
888
10 spare GPIO, 2 spare IRQ lines

4 ADC lines
UART with CTS/RTS control
2
I C, USART/SPI

Up to 30 lines can be configured as GPIO
Capability to write own MAC address into the EEPROM
3.
RECEIVING MICROPROCESSOR MODULE
Receiving station (figure 2) communicates with PX by RS232 hardware

connection and UART serial protocol. Microcontoller PIC18F8520 and chip for wireless
communication ZDM-A1281-A2 are also used in receiving station. Communication
between microcontroller and ZDM-A1281-A2 is implemented by serial protocol.
Fig. 2 Receiving station: ZigBit module is connected from right side to

development board, and further with receiving microcontroller. From upper
side of development board, USB connection is for power supply and
programming of receiving microcontroller, and RS232 connection is for
transmission of data to PC.
Firmware of receiving station microcontroller is structured in such way, that
after necessary initializations, data are received from ZigBee module (which actually
forwards data received from transmitting module). As these data are mesured results
reperesented in form of characters, it is necessary to convert the data into numerical
values.
889
Fig. 3 Panel of virtual instrument, with configuration fields and indicators.

First byte is the ASCII character as indicator of type of measured value
(temperature, light intensity or gas transducers output voltage). The other bytes are
characters respresenting the measured value sent by transmitter. If-else condition
statements recognize the type of measured value and perform the conversion of
received data to final measurement result.
PC software application for acquisition,processing and representing of received
data is the virtual instrument developed by software development package National
Instruments LabVIEW 7.1[10]. Application (panel of virtual instrument is presented at
figure 3) can be executed through LabView interface or as standalone executable
application. During application initialization default settings are loaded, but parameters
of RS232 communication and measurement callibration may be configured.
First segment of virtual instruments block diagram implements VISA driver for
serial communication, with adequate fields necessary for configuring parameters of
serial communication. If there is any data at serial port main loop of program is
entered; on the contrary,after two seconds the state of data at port is checked again.
Inside the loop, excpected three bytes are read from port and further processing of
data is performed depending on first byte. At the end of measurement and acquisition
session, the port is closed and enabled for other applications.
4.
CONCLUSION
The structure of implemented measurement and acquisition system for

networking of measurement transducers is presented. In the measurement transducers
module, digital temperature sensor is described in more details. Receiving station and
transmitting microprocessor system are implemented by 8-bit microcontroller
890

PIC18F8520. The firmware is developed in Integrated Development Environment
MikroC. ZigBit modules, using ZigBee communication standards, are based on ZDMA1281-A2 chip characterized by ballanced double antenna and very low energy
consumption. The main data of hardware, firmware and software structure are
described.
LITERATURE
[1]
J. Fraden, Handbook of Modern Sensors: Physics, Designs, and Applications,

Springer-Verlag New York. Inc., 2004, ISBN 0-387-00750-4.
[2] M. Popovi, Senzori i merenja, Zavod za udbenike i nastavna sredstva, Srpsko
Sarajevo, 2004.
[3] Practical design techniques for sensor signal conditioning, Published by Analog
Devices, Inc.,1999, ISBN-0-916550-20-6.
[4] ZigBee, http://www.zigbee.org, accessed on 2010-03-15.
[5] TGS2611 for the detection of Methane, Product information, Figaro Engineering,
rev 02/05.
[6] KE-50 GS Oxygen Sensors, Product information, Figaro Engineering, rev 08/03.
TM
[7] DS1820 1Wire Digital Thermometer, Production information, Dallas Semiconductor, Document no. 021497.
[8] PIC18F8520,
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en010319,
[9] ZigBit OEM Modules, download.maritex.com.pl/pdfs/wi/zigbit.pdf, accessed on
2010-03-15.
[10] LabVIEW 7.1 Documentation Resources National Instruments,
http://digital.ni.com/public.nsf/allkb/06572E936282C0E486256EB0006B70B4,
891
SOME ASPECTS OF KNOWLEDGE ENGINEERING

K. Abharyaa, D. Djukicbb, H-Y. Hsua, Z. Kovaciccc, D. Mulcahya,
S. Spuzica*, F. Uzunovicdd
Summary: Knowledge Engineering is one of the most significant factors contributing to
sustainability of Engineering Systems. Principal concepts present fundamental bricks
of ontological structures. Organisational function, mission and scope impose branching
and hierarchy on information, and much of knowledge is categorised into sub-domains
understood only by experts. This has raised barriers to cross-disciplinary
communication. In this 'age of knowledge' in which the information can be stored,
retrieved and transferred by the speed of electro-magnetic waves, the impediments
such as prolixity, ambiguity and jargon are amplified to critical levels. The analysis
presented suggests that the root causes for these impediments include knowledge
gaps, discrepancy of intentions, and differences in beliefs. Teams can perform quite
differently depending on how they communicate knowledge. Authors hypothesise that
the conflicts in interpretations of some basic engineering concepts significantly
contribute to multiplication of ambiguities, which hinders knowledge sharing and
application. Examples of ambiguous concepts such as 'technology', vector and
frequency, are presented, and a disambiguation strategy discussed.
Key words: Knowledge engineering, homonymy, technology, disambiguation
1. INTRODUCTION
Motivated by the need for improving knowledge engineering, authors have
initiated an investigation into the causes of the ambiguities observed in engineering
applications and sharing of scientific theories and hypotheses.
Engineering systems can perform quite differently depending on how they
communicate knowledge. Working towards a common goal requires sharing
perceptions, intentions, beliefs and knowledge. For example, an industrial system
suffered financial losses due to the lack of dialogue between the antipode experts who
were each protecting their territory of knowledge. Such policies of intellectual
protection and confidentiality safeguards result in the loss of the expertise. [1]
a
PhD, Division of Information Technology, Engineering and the Environment, University of South Australia,
Adelaide, Australia
b
PhD, School of Engineering & Advanced Technology, Massey University, Wellington, New Zealand
c
PhD, School of Information and Social Sciences, Open Polytechnic, Wellington, New Zealand
d
PhD, Faculty of Metallurgy and Materials Science, University of Zenica, Zenica, Bosnia and Herzegovina
*Corresponding author: Dr Sead Spuzic, e-mail: sead.spuzic@unisa.edu.au
893
Abhary, Djukic, Hsu, Kovacic, Mulcahy, Spuzic, Uzunovic
Within the institutional fences, the team work and other modes of collaboration
are officially strongly promoted. Yet in the practice, the achieved synergy is too often
below the claimed levels. The declarative commitments are only the superficial
remedies when compared to addressing the core issues such as the underlying beliefs
and the resulting intentions. The belief that the knowledge should be kept confidential
and intellectually protected instigate isolationism that blocks the exposure and
verification of important concepts, theories and hypotheses. [1]
The above state of the art provides fertile ground for growth and multiplication
of misconceptions, misunderstandings and otherwise ambiguous manifestations in
knowledge engineering.
A diametrically opposite belief is that the system does not depend on the
competition between the peers as much as on competition with wider ambient (e.g.
global climate change). This realisation triggers a recognition of constructive motives,
which in turn can dramatically change our intentions. Another progressive belief is that
the efficiency, growth and verifiability of knowledge increase with the sheer quantity of
the informed participants. An example of this can be found in the recent increase in
collaborative knowledge sharing such as Wikipedia. [1, 2].
Stock of our knowledge continues growing at a rate that encourages
considerable optimism. Development and maintenance of transparent and unambiguous
definitions of fundamental concepts is particularly important, since any inconsistency in
basic scientific notions triggers a chain reaction causing accumulation of misconceptions.
However, a trend in disseminating homonymous, synonymous, and otherwise
ambiguous definitions of a whole range of scientific concepts, can be observed in
publications distributed by maintainers of scientific databases, such as ScienceDirect
(Elsevier), SpringerLink, InterScience (Wiley) and others. It is certainly important to
discuss possible causes and remedies for this kind of hindrances, at least for some of
the key concepts such as technology, vector and frequency.
The concepts technology, vector and frequency are chosen because of their
importance in knowledge engineering. These concepts are affiliated with differing
definitions, henceforth associated with inevitable ambiguity when otherwise verified
knowledge and theories are shared between differing disciplines. Moreover, when
differences in these key concepts are ignored in publishing scientific advances, one is
prompted to sense certain disrespect for others. Finally, the recipients and users of
knowledge, such as the students and engineers, can fall into traps of misinterpretation.
Growing evidence indicates that there can be a significant disruption to a learners
progress when, confronted by misalignments of meaning [1, 3]. Reasoning modality is
influenced by perceptions of similarity, which can be blunted by conceptual
inconsistencies and terminological contradictions [3].
In this treatise three examples of ambiguous concepts technology, vector and
frequency are presented along with the disambiguation strategy. The intent is to
demonstrate how these key concepts can be defined in less ambiguous way.
Underlying research and axioms are presented in references [4-6].
2. TECHNOLOGY
Homonymous usage of the term technology is notorious in engineering and
scientific publications [3]. According to [7], 'technology' is the science of techniques.
894
Some Aspects of Knowledge Engineering
However, other sources embrace within the concept 'technology' actual 'techniques',
including the pertinent materials, tools, equipment and other means (e.g. lubricants,
and energy providing media - fuels) for achieving a certain pre-defined aim.
'Technique' and 'tool' are defined in [8].
These differing definitions allow for differing interpretations of the contexts that
include the concept of technology. Numerous international sources (e.g. Wikimedia
Foundation and Merriam-Webster Dictionary) use the term technology to address a
technique or a system of techniques. The University of Technology, Sydney (2008)
commits to pursuing best practice in management and technology systems. The
Massachusetts Institute of Technology publishes that their mission is to advance
knowledge and educate students in science, technology, and other areas of
scholarship. Therefore one can conclude that technology is one of the scholarship
areas that can be distinguished from sciences. However, the Delft University of
Technology commits to offering education and research within the technical sciences
along with developing technologies for future generations. Here it appears that
technologies means technical assets and techniques. [3]
There is a significant difference in offering new chemo-physical materialised
resources (e.g. tools or equipment) along with the instructions for usage, compared
with only offering specific knowledge (e.g. definitions, intelligence and instructions)
without supplying any materialised tools, energy or other chemo-physical resources.[3]
3. VECTOR
A most general definition provided in mathematics is as follows: 'Vector' is an
ordered i-tuple of numbers, where i = a natural number. More special formulation is
that vector is a geometric object, or a straight line segment whose length is magnitude
and whose orientation in space is direction [9,10].
However, in medicine, biology, ecology and genetics, vector is defined as:
any vehicle used to transfer foreign genetic material into another cell [9];
an organism which transmits infection by conveying pathogens from one
host to another [10].
a transmitter any agent, person, animal, microorganism, DNA molecule
that carries and transmits a genetic material (e.g. causing an infection) [9].
This conceptual mismatch could create confusion e.g. when using vector
mathematics in communication involving civil, epidemiology, biology or environmental
engineering. Concrete walls are constructed usaing statics, where the idea of vector
is a fundamental concept. Such walls can be built with cement containing an agent that
breaks air pollutants through a photocatalysis. A photocatalyser, when incorporated in
cement, hastens the decomposition of pollutants (vectors) and prevents their build-up
on surfaces. In this context the term vector can imply mismatching concepts to civil
engineers who are trying to communicate with environmental engineers.
4. FREQUENCY
Knowledge engineering often requires counting occurrences of phenomena
(relations, events, systems) of particular interest. In this context, the concept of
'frequency' is often put to use. However, a simple statement that the frequency is the
895
number of observed occurrences is ambiguous. Many authors use the term

frequency with differing special meanings, for example when related to periodic
phenomena (e.g. in so-called time series analysis in the frequency domain) in relation
to the concept of duration.
Frequency is defined as:
(a) The number of occurrences per given duration, e.g. the frequency of
modulation is 40 cycles per second (also known as frequence) [9].
(b) The rate of change of phase of a sinusoidal waveform [10]. In this case,
ambiguity is increased due to the fact that the term phase is a homonym as well [5].
(c) The number of times the event occurred in the experiment or the study [10].
The frequency is how often something occurs [20].
(d) The number of observations in a given statistical category (also absolute
frequency) [9].
(e) The ratio of the number of observations in a statistical category to the total
number of observations (also called relative frequency) [9].
Definitions under (a) and (b) refer specifically to duration. In the case of
-1
periodic events, the unit for measuring frequency is hertz (Hz = s ). The use of a
separate unit, becquerel (Bq) for measuring the frequency of aperiodic events
strengthens the notion of relation between the terms frequency and duration (Bq = s-1).
The major concern in this definition is the relativity of the duration as a measure.
In a cross-disciplinary communication, reference to the statistical frequency of
observations in connection to the frequency of impacts in a dynamic test would be
surely an ambiguous reflection.
5. SOME ASPECTS OF DISAMBIGUATION STRATEGY
Strategies for improving transparency and transferability of knowledge should
draw on the disciplines of didactics, epistemology and ontology. Producing definitions
is one of the major tasks in building an ontologically valid knowledge repository. The
disambiguation strategy proposed in this transcript is to promote the use of established
generic definitions of concepts that are broadly used in various disciplines by adopting
a nomenclature hierarchy already established in mathematics, chemophysics and
other fundamental sciences. Other concepts can be denoted by introducing new terms
to avoid homonymy, synonymy and other ambiguities.
It is unlikely that one language, with its unique vocabulary, can hope to cover
the full diversity of human interests. Since English is the lingua franca of science it is
clear that an organised system of extension of the scientific nomenclature should be
adopted. The multiplication in homonyms can be avoided by using terms from the
vocabulary provided by other languages. We propose that trend of adopting the foreign
words in English language be strategically continued.
5.1 Technology
With regard to the concept of technology, it is suggest hereby to follow an
analogy with the nomenclature accepted in numerous other disciplines: biology,
geology, tribology, psychology, sociology, anthropology and toxicology describe
different branches of science.
The English suffix -logy or -ology denotes a scientific field. Hence:
896
Some Aspects of Knowledge Engineering
technology is science of techniques. A hypernym for technology is science.

We propose avoiding the use of the term technology to denote techniques.
In special cases such as denoting the branch of engineering that deals with the
use of computers and telecommunications to process, store and transmit information,
rather than using the term information technology we propose using information
technologie (adopting terms from Dutch, French and German).
5.2 Vector
The concept of vector is well established in mathematics. Definition Vector is
an ordered i-tuple of numbers, where i = a natural number is concordant with the use
listed by the United Nations Mutilingual Terminology. This definition is also promoted in
the ontological approach to engineering disciplines.
The concept of a transmitter that carries and transmits a genetic material can
be conveniently denominated by using the term vettore (Italian).
5.3 Frequency
The confusing homonymy related to the concept of 'frequency' can be
eliminated by adopting the generic definition of the term frequency irrelevant of time.
Frequency f is defined as a ratio of the count x and the count y: f = x/y,
where x is the count of actual specific occurrences and y is the total count of all
possible events of relevance.
Events of relevance are the observations of the phenomena that are
classifiable within the analysed scope (within some boundaries defined by the purpose
of observations).
For example, if we observe a large container filled in with a mix of large number
of yellow, green and red balls, we can count how many red balls have been found in
small boxes filled each time with ten balls chosen at random from the above container.
If all colours have been well mixed within the large container, then the frequency of red
balls in each box can vary from 0 to 1.
Hypernyms for term frequency are statistics and number.
Furthermore, it is necessary to distinguish this concept from its homonyms by
introducing a term frequence (French) as a tentative solution. It is proposed therefore
to consider allocating the term frequence to the following concept: frequence (f) is
the ratio of the count of the observed cycles, e.g. wavelengths (n), and a multiple of
the count of wavelengths of some convenient, measurable, stable and standardized
radiation (N1s) that occurs simultaneously.
For the convenience, rather than counting N1s each time we need to measure
the frequence of some cycles, a metrological device termed a clock is developed.
Such a device enables value N1s to be expressed in seconds. [10]
Frequence is measured in units termed hertz (Hz): 1 Hz = 1 s1
Bearing in mind that N1s is subject to some level of uncertainty, the value of the
derived frequence f is always just an approximation characterized by some estimate of
an average value and variance. The current state of the art in sciences allows for
reducing this relative uncertainty to the order of 1015. [10]
In the above definition, the use of the concepts of duration (and especially that
of the time) was deliberately avoided or at least delayed, until it becomes necessary
to introduce some observable metrics, namely the second. It is instructive to
897
emphasise that a second is defined in terms of the count N1s.

Hypernym for frequence is attribute.
6. CONCLUSIONS
Information systems are omnipresent in knowledge engineering and today
human-computer interaction lies at the crossroads of many disciplines. In this new
ambient the leading sources promote knowledge transparency, shared beliefs and
constructive intentions, which all leads to a decrease in the information entropy.
The significance of establishing and maintaining unambiguous concepts and
non-homonymic nomenclature is in defining knowledge in terms of its application and
sharing. Difficulties associated with ambiguous definitions add to increasing
misconceptions and decreasing readability.
Clear and consistent definitions are knowledge probability intensifiers.
LITERATURE
[1]
Abhary, K., Adriansen, H.K., Begovac, F., Djukic, D., Qin, B,, Spuzic, S., Wood, D. and
Xing, K. (2009) "Some Basic Aspects of Knowledge" Procedia - Social and Behavioral
Sciences, Volume 1, Issue 1, 2009, Pages 1753-1758.
[2] Tapscott, D., & Williams, A. D. (2006) Wikinomics: how mass collaboration changes
everything. New York: Penguin.
[3] Abhary, K., Adriansen, H.K., Begovac, F., Kovacic, Z., Shpigelman, C.N., Stevens, C.,
Spuzic, S., Uzunovic, F., & Xing, K. (2009) A Contribution to Transparency of Scientific and
Engineering Concepts. The International Journal of Knowledge, Culture and Change
Management, Volume 9, Issue 5, 93-106.
[4] Spuzic, S., and Nouwens, F. (2004) "A Contribution to Defining the Term Definition",
Issues in Informing Science and Information Technology Education, Volume 1 (2004) pp
645 662.
[5] Spuzic, S., Abhary, K., Stevens, C., Fabris, N., Rice, J., Nouwens, F. (2005) Contribution
to Cross-disciplinary Lexicon (Editors: D. Radcliffe and J. Humphries) Proceedings 4th
ASEE/AaeE Global Colloquium on Engineering Education, Sydney, 26-29 September
2005.
[6] Spuzic, S., Abhary, K., Stevens, C., Uzunovic, F. (2006) Contribution to Knowledge
Management: Cross-disciplinary Terminology; the 5th Pan-European Conference on
Planning For Minerals and Transport Infrastructure: PEMT'06 - Sarajevo 18-20 May 2006.
[7] Strawn, G.W. (1982) Technology transfer: What is it? What can it do for me?, Journal of
Technology Transfer, vol. 6, no. 2, pp. 45-51.
[8] Spuzic, S. Definiton of the term Technique (accessed on 10th April 2011)
http://tablea.yolasite.com/technique.php.
[9] WordNet lexical database for the English language, Cognitive Science Laboratory at
Princeton University (accessed on 10th April 2011)
http://wordnetweb.princeton.edu/perl/webwn.
[10] Wikipedia, Wikimedia Foundation (accessed on 10th April 2011) http://en.wikipedia.org.
898
STACKING AISLE WIDTH FOR FORKLIFT TRUCKS IN PALLETIZED

STORAGE AND HANDLING SYSTEMS
Nenad Miloradovi 1, Rodoljub Vujanac 2, Blaa Stojanovi 3
Summary: Forklift trucks are restricted by the size of aisle widths in which they can
operate. Stacking aisle width is the aisle width required under the most favourable
conditions for stacking operations with the forklift truck positioned at right angles to the
aisle. Accordingly, the load is swung into position only within the stacking aisle width,
with exception of the forklift trucks that can only pick up loads from the floor and lay
them down. For these trucks, the swinging operation is planned to be carried out above
the lay-down location and has therefore been taken into account in the calculation of
the stacking aisle width. Calculations are done using FEM regulation.
Key words: stacking aisle width, forklift truck, warehouse
1. INTRODUCTION
There is a wide range of equipment available for moving pallets around a
warehouse, from simple manual aids to sophisticated computer-controlled equipment.
Some of the most common types are: hand pallet trucks, powered pallet trucks,
forklifts, conveyors, automated guided vehicles (AGVs). In general, the above types of
equipment are used solely for horizontal movement. For placing pallets into storage
positions, some form of lifting mechanism is required. However, it should be noted that
many of these lifting trucks are also commonly used for horizontal movement around
the warehouse.
The most significant unit load in warehousing is probably the wooden pallet,
and storage/handling systems specifically designed for this type of load are examined
in this paper.
Wooden pallets made to standard sizes, although there are different standards,
which can cause some problems with international movement. For example, the
dominant pallet in continental Europe is the Europallet (1,200 mm by 800 mm), whilst
the dominant UK pallet is slightly larger (1,200 mm by 1,000 mm), similar in size to that
in the United States (48 inches by 40 inches). In addition to pallet size, other variables
include the pallet construction: two-way or four-way fork entry, reversible or nonreversible (double-sided or single-sided), open- or close-boarded. Some pallets are
1
Mr Nenad Miloradovi, M.Sc., Kragujevac, Faculty of Mechanical Engineering, mnenad@kg.ac.rs

Mr Rodoljub Vujanac, M.Sc., Kragujevac, Faculty of Mechanical Engineering, vujanac@kg.ac.rs
3
Mr Blaa Stojanovi, M.Sc., Kragujevac, Faculty of Mechanical Engineering, blaza@kg.ac.rs
2
899
Nenad Miloradovi, Rodoljub Vujanac, Blaa Stojanovi
made from other materials, such as plastic and fibreboard. The wooden pallet is
probably the most commonly used of all the unit load types.
2. CALCULATION OF THE STACKING AISLE WIDTH
Fork trucks are restricted by the size of aisle widths in which they can operate.
The minimum aisle width for 1.2 metre pallets is defined in [5]. Accordingly, if different
sized pallets are being handled, then these need appropriate adjustments. The major
problem with pallet sizes is where mixed pallet sizes are handled, for example with
Euro pallets of 1200 by 800mm. Here the option taken is often one where the larger
size of pallet being used will then determine the racking aisles width. Alternatively,
separate areas for different sizes pallets will be required.
With the space saving of narrow aisle/very narrow aisle equipment, it could be
envisaged that all warehouse should have them. However, wider aisle options are still
viable, especially in warehouses that require much movement. As the aisle widths
narrow, then speed slows. Wide aisle trucks enable the picking of pallets from racking
and loading to a trailer immediately, combined with fast travel speed. They also
operate at lower cost. So, wide aisles give more flexibility. Moving to narrow aisles
gives greater storage density.
Therefore, the balance has to be determined between the speed of operation,
the storage density and the cost. It is often not a easy decision but should be one on
which time is taken, before erecting relatively permanent fixed racking structures and
perhaps also committing to semi specialised and expensive handling equipment.
The general principles for internal warehouse layout include [1]:
logical flow patterns with minimal cross-flows or backtracking of goods,
based on analysis of movements, generally in a rectilinear layout,
minimizing the amount of movement required for staff and for handling
equipment,
making the best use of building volume,
good access to inventory (e.g. through adequate aisle widths),
safe systems of work including the provision where possible of separate
movement aisles and access doors for people and for mobile equipment,
elimination of dead areas in which operators could be trapped, e.g. no
aisles with closed ends.
3. INFLUENCE OF FORKLIFT TRUCK DESIGN AND LOAD DIMENSIONS
Manoeuvring allowance is the dimension by which the aisle width calculated
from the dimensions of the forklift truck and the load is increased. Accordingly,
inaccuracies in steering operations, as well as in the load dimensions and load
application are taken into account.
200 mm are uniformly used as the manoeuvring allowance. This makes it
easier to compare a variety of forklift truck designs and sizes with one another. Due to
other factors, it may be required to increase the manoeuvring allowance. Other
important factors are: floor not skid-proof, load units not rigid, large load dimensions
which make precise driving approaches more difficult, operator skills.
According to measurement circumstances of truck and load, the aisle width is
created from four possible geometrical situations [5]:
900
Stacking Aisle Width for Forklift Trucks in Palletized Storage and Handling Systems
regarding the truck side either radius Wa or with wide loads the truck load
corner radius R;
regarding the load the length as far as to the outer load side b12 or the
diagonal of outer load corner;
on three wheel trucks the diagonal usually lies outside the pivot point. Since
the deviation is only minimal the centre is used;
for low lift trucks, the normal case is the use with standard pallets. Therefore
the same dimensions as for the standard pallets apply for the load
dimensions.
Counterbalanced fork-lift trucks carry the payload forward of the front wheels
(Figure 1), so there is always a turning moment tending to tip the truck forward. To
balance this, a counterbalance weight is built into the rear of the machine hence the
name. Heavy components like engines and batteries are also positioned as far back as
possible to help counter the overturning moment.
l6 x 2 b12 b13
Ast
Wa x l6 a
Ast
Wa R a
if
Ast
b12
! b13
2
if
b12
b13
2
and 12 b13 Wa
2
b12
b13 R a
2
if 12 b13 ! Wa
2
Fig. 1 Four wheel forklift truck (point of rotation outside of axle centre)
Since the load is always in front of the front wheels, counterbalanced four
wheel forklift trucks are long, necessitating a wide turning area. They therefore need to
operate in wide aisles (e.g. for putting pallets into or out of racking) of about 3.5 metres
or more. Consequently, although these trucks are very robust and fast, and are very
good as 'yard trucks' and for vehicle loading and unloading, they are less appropriate
for many activities inside the warehouse because of the space required for access
aisles.
Most forklift trucks have rear-wheel steering (Figure 1, 2). However, there are
some trucks that are articulated so that the truck body 'bends' in the middle during
turning. These forklift are designed to operate in aisles of as little as 1.7 - 1.8 metres.
Reach trucks (Figure 3) are designed to be smaller and lighter than
counterbalanced trucks and to operate in a smaller area. This is achieved by having a
mast that can move forward or back in channels in the outrigger truck legs. When
picking up or setting down a load, the truck is turned through 90 to face the load
location; the mast reaches forward, places or retrieves the load, and is retracted back
901
into the area enclosed by the wheels. The truck travels with the mast in the retracted
position. This virtually eliminates the need for a counterbalance weight, and reduces
the truck length. Reach trucks are battery-powered. They are widely used in
conventional, i.e. non-automated, warehouses. Aisle widths for trucks placing standard
pallets into storage are typically in the range 2.8 m to 3 m, which is considerably less
than for standard counterbalanced trucks.
l6 x 2 b12
Ast
Wa R a
Ast
b
if 12 Wa
2
b12
b
R a if 12 ! Wa
2
2
Fig. 2 Three wheel forklift truck (point of rotation in centre of axle)
l6 x 2 b12
2
2
Rh
Ast
Ast
b
x 2 12
2
Wa R a if Rh Wa
Rh R a if Rh ! Wa
Fig. 3 Reach truck (point of rotation in centre of axle)
l6 x 2 b12
2
2
Rh
Ast
Ast
b
x 12
2
Wa R a if Rh Wa
Rh R a if Rh ! Wa
2
Fig. 4 High lift pallet truck

902
Stacking Aisle Width for Forklift Trucks in Palletized Storage and Handling Systems
Narrow and very narrow aisle trucks should be used to reduce overall
warehouse area and therefore shorten distance within the storage area. Shorter
distance travelled reduce handling costs. Automatic height selection, or cameras at
fork height, may be used to assist the driver when locating pallets at high level (Fig. 4)
The layout of the low lift pallet truck is presented in Figure 5.
Rh
Ast
Ast
b
x 12
2
Wa l6 x a if Rh Wa
Rh l6 x a if Rh ! Wa
2
Fig. 5 Low lift pallet truck

The load length l6 shall be used as 2xc, whereby c is the standardised load
centre distance. If several load centre distances are standardised, the smaller value
must be selected. If no load centre is specified, the value mentioned for the rated
capacity shall be used.
The width of the load predominantly intended for each type of forklift truck shall
be used as the load width. Accordingly, the standardised dimensions of the load units
(pallets, containers) and the specifications for the load length shall be observed.
Fig. 6 Design of pallet racking layout [2]

An efficient design of pallet racking layout which should provide safe
warehousing operation considers recognition of basic principles of warehousing as well
as methods for their realization. Based on previous principles, a design solution for
warehousing in Phermaceutical Institution of Montenegro "Montefarm" in Podgorica is
carried out [3]. Warehouse area has dimensions 33000x13000x8000 mm and it is fitted
with Ponteggi Dalmine equipment [4]. The solution builds on the long standing
experience gained in design and construction of warehousing systems as well as on
usage of the actual FEM regulations.
903
4. CONCLUSION
This paper has set out the basic storage and handling systems for palletized
goods, and the essential characteristics of the different systems. A key aspect for
determining the most appropriate system for a particular application is to select one
whose characteristics most closely match the overall requirements of the warehouse
within which it is to work. In most warehouses more than one system is used.
There is a range of other specialist pallet stacking equipment for use with
specific storage types. Typically, space inside a warehouse is more expensive than
space outside a warehouse. Therefore, 90 docks are more popular than finger docks
because they require less inside space.
The basic objectives for determining the most appropriate storage and handling
system for any application are likely to include:
effective use of space - building height, building area and access aisles,
good access to pallets for taking out and replenishing,
high speed of throughput,
low levels of damage,
high levels of accuracy,
integrity and security of inventory,
personnel safety,
minimum overall system cost.
There are often compromises to be made between these objectives. For
example, storage systems that utilize building space most effectively often do not give
particularly good access to the stock, and vice versa. On the other hand, where a
system offers both excellent space utilization and individual access (as with powered
mobile racking), then speed of throughput is compromised. Thus, a trade off often has
to be made between these factors when deciding on the use of any storage and
handling system.
LITERATURE
[1] Rushton A., Croucher P., Baker P.,(2006). The handbook of LOGISTICS and
DISTRIBUTION MANAGEMENT, 3rd edition Kogan Page Limited.
[2] Vujanac, R., Slavkovi, N., Miloradovi, N., (2006). Warehousing solution in the
Pharmaceutical institution of Montenegro Montefarm Podgorica, International
scientific-technical meeting, GNP 2006, Conference Proceedings, p.p. 953 958
[3] Contractor project of self-supporting pallet racking warehouse at Pharmaceutical
institution of Montenegro - Montefarm Podgorica, IC Engineering Kragujevac
and Faculty of Mechanical Engineering from Kragujevac, 2003
[4] Ponteggi Dalmine, Gruppo Marcegaglia, Product Range, Rapid - Block Catalogue,
Milano
[5] FEM 4.005 (2005) Industrial trucks 90 stacking aisle width. FEDERATION OF
EUROPEAN MATERIALS HANDLING. Product Group Industrial Trucks
904
THE CONTROLLER DESIGN FOR TRACKING TRAJECTORY WITH

CONTROLLED JERK
Mihajlo J. Stoji1, Bojan Kneevi2
Summary: This paper describes the design of robust controllers for linear object which
will provide asymptotic tracking of the required trajectory. Trajectory is determined by
the need that jerk in the movement of an object is changed by the function that is
defined in the paper. Characteristics of the obtained controllers are tested by computer
simulation. A model of elevator with induction motor drive is used as an object.
Key words: jerk, robust tracking, the controller design
1. INTRODUCTION
The object movement in the positioning systems is divided into three phases:
acceleration, constant velocity movement and deceleration. Duration of acceleration T1
and deceleration T2 phases are in most cases is the same, T1 T2 T . This time
depends on the maximum acceleration A and the desired speed V while an object
has in motion with constant velocity. This speed is the also maximum speed of object.
Maximum acceleration A is set and it depends on the power drives and loads.
Keeping in mind the above times and if the desired time of positioning is W , then the
time when the object is moving with constant velocity is W 2T . In this paper we
consider only the case when the parameters are chosen so that is W t 2T . The case
when W 2T is analyzed in [1].
Since the jerk is speed of change of acceleration and if the acceleration have
step form then it causes a big jerks (theoretically infinite) which have a negative effect
on the mechanics and dynamics of the system. The accuracy of the system is reduced
and the overshoot and settling time increases. This is the case with the frequently used
trapezoidal form of changes in speed, because then the acceleration in the phase of
acceleration and deceleration have step form. Significant improvement can be
achieved if the acceleration changes by a smooth curve. This provides that the jerks
and their harmful effects is less. Because of that, designing of position systems starts
from the prescribed smooth change of jerks [2,3]. However, smooth changes of jerk
implies that the control of the positioning system is more complex, because it increases
the order of the trajectory reference of the positioning system.
1
2
PhD, Mihajlo J. Stoji, Banja Luka, Faculty of Mechanical Engineering, (stojcicmihajlo@gmail.com)

MSc, Bojan Kneevi, Banja Luka, Faculty of Mechanical Engineering, (bojan.knez@blic.net)
905
Mihajlo Stoji, Bojan Kneevi
Now, some word on the structure of the paper. For the proposed function of
jerk, corresponding change of referent trajectory are carried out in the second section.
In the third section, for obtained referent trajectory and the known object, a controller is
designed and its robustness and simulation are carried out at fourth and fifth sections.
2. TRAJECTORY WITH CONTROLLED JERK
A case is analyzed when the jerk in the process of acceleration and
deceleration has the sinus function [2]:
STA sin 2TS t ,
t > 0, T @
j 0,
t [T , t2 ] ,
(1)
S A sin 2S (t t ), t [t ,W ]
2
2
T
T
where t2
W T is a moment when the object starts deceleration phase,
SA
T
j0 the
Z is a circular frequency of jerk.

Based on (1) the acceleration in particular intervals is obtained based on the
relation a j dt Ca , velocity as v adt Cv and the displacment as d vdt Cd .
maximum amplitude and
2S
T
From those relations we obtain that the displacements d , for every time interval, are:
A t 2 AT22 1 cos 2S t ,
t > 0, T @
T
8S
4
AT
d 4 2t T ,
t [T , t2 ] .
(2)
AT
2
2
2S
A
AT
4 2t T 4 t t2 8S 2 1 cos T t t2 , t [t2 ,W ]
Constants Ca , Cv and Cd are determined from initial conditions. The changes
of j, a, v i d are presented graphically in Fig. 1. From (2) , for t W , the total
displacement is determined:
AT
d (W ) D
t2 Vt2 .
(3)
2
Maximum value of displacements D , velocity V , acceleration A and the
desired positioning time W are known, so that from (1) and (3) times T and t2 are
calculated. If the condition W ! 2T (or t2 ! T ) is not satisfied we can reduce velocity V
(or time T
2V
A
) until it is satisfied.
3. CONTROLLER DESIGN
Design of a robust controller that will provide asymptotic tracking of
displacements given by (2) will be observed on the object which is taken from [1]. It is a
lift whose drive is the induction motor, and whose transfer function is:
P(s) K
Go ( s )
(4)
, K 16.52 .
Q( s) s 2
Displacements reference given by (2) is a square function of time and can be
seen as an acceleration type of reference. An object follows the quadratic reference
without an error if the polynomials in the numerator and denominator of the total
transfer function
906
The controller design for tracking trajectory with controlled jerk
j=jerk, a=acceleration, v=velocity, d=displacement
6
4
displacement
j, a, v, d
acceleration
velocity
0
-2
jerk
-4
0
0.5
1.5
2.5
3.5
4.5
time
Fig.1 Changes of jerk, acceleration, velocity and displacements

G( s)
N (s)
D(s)
have the same coefficients with s0, s1 i s2, ([4,5]). This paper discusses the
case that besides these three, the same coefficients are with s3 as well. Recognizing
these conditions, the transfer function of the object (4) and the feasibility conditions of
the transfer function of the whole system (see [5]), we can obtain transfer function of
the whole system:
d3 s 3 d 2 s 2 d1 s d 0
N (s)
N (s)
.
(5)
G( s)
5
D( s ) s d 4 s 4 d3 s 3 d 2 s 2 d1 s d 0 s 5 d 4 s 4 N ( s )
Since this is a system in which it is necessary to adjust both the numerator and
denominator of transfer function, it generally requires two compensators - two degrees
of freedom. Realization is possible with the system as on the Fig. 2a or the equivalent
configuration as on the Fig. 2b, where the compensators are marked. The transfer
function of a closed system (from reference r to output y) from Fig. 2 is:
Y ( s)
L( s ) P ( s )
N (s)
.
(6)
G( s)
R ( s ) A( s )Q ( s ) M ( s ) P ( s ) D ( s )
Compensators design procedure, C1 ( s ) and C2 ( s ) , or the determination of the
polynomials L( s ), A( s ) and M ( s ) is described in [4] and [5]. The first step is to
establish the transfer function:
G ( s)
N ( s)
E ( s)
,
(7)
P( s ) D( s) P( s) F ( s)
from which it is to be checked whether the condition deg F ( s ) t 2n 1 is true, where
n deg Q( s ) . If the condition is satisfied, then L( s ) E ( s ) , and polynomials A( s ) and
M ( s ) are obtained as a solution of the Diophantine equation:
A( s )Q( s ) M ( s ) P ( s ) F ( s ) .
(8)
In our case, E ( s ) and F ( s ) are: E ( s )
N (s)
P(s)
1
K
(d3 s 3 d 2 s 2 d1 s d 0 ) and
F ( s ) D( s ) d 0 d1 s d 2 s 2 d3 s 3 d 4 s 4 s 5 . D( s ) is a characteristic polynomial of a
closed system (6) and five poles are chosen for its determination that determine
desired behavior of a closed system: p1 21, p2 23 i p3,4,5 26 .
907
C1 (s )
r (t ) L (s )
w (t ) Go (s )
y (t )
P (s )
Q (s )
A(s )
C1 (s ) w (t )
L (s )
r (t )
A (s )
y (t )
P (s )
Q (s )
C2 (s ) M (s )
M (s )
C2 (s ) A(s )
a)
Go (s )
L (s )
b)
Fig.2 System configuration with two compensators a) and the equivalent configuration b)
Using D( s ) obtained, we have:
L( s )
E (s)
N ( s)
P( s)
N (s)
K
359.75( s 3 24.31s 2 294.9 s 1428) ,
a3 s 3 a2 s 2 a1 s a0 and M ( s )
Polynomials A( s )
(9)
P3 s 3 P2 s 2 P1 s P0 , what
replacing to (8) gives Diophantine equation with 8 unknown coefficients (ai and i,
i=0,1,2,3) and has rank at most 6. This means that we must choose the two unknown
coefficients. According to [5], the values of the coefficients of the lowest degree of
polynomial A( s ) we choose. Thus, let a0= a1=0, so that equation (8) now becomes:
0 0 P0 d 0

0 0 P1 d1
0 0 a2 d 2

.
0 K P2 d3
0 0 a3 122

1 0 P3 1
polynomials
A( s )
and
M (s) :
0
0
0
0
Solving
M (s)
(10)
0
K
0
0
0
0
0 0
0 0
0 K
0 0
1 0
0 0
(10)
A( s )
s 3 122s 2 ,
359.75( s 3 24.31s 2 294.9s 1428) are obtained. The obtained polynomials L( s )
and M ( s ) are identical so compensator is C2 ( s )

controller transfer function is:
L( s)
C1 ( s )
A( s )
359.75
M (s)
L(s)
1 (Fig. 2b). Finally, the
s 3 24.31s 2 294.9 s 1428

.
s 3 122s 2
(11)
4. ROBUSTNESS OF SYSTEM
If the system should be robust, it is necessary to prove that the system follows
the reference without errors and that it is completely immune to the impact of external
disturbance (in this case we consider, step and ramp) and internal disturbances
(changes in the object parameters). When Q( s) s 2 , L( s) NP ((ss)) and A( s) s 2 ( s d 4 )
where d 4
122 , are incorporated into obtained transfer function (6):
Gyr ( s )
L( s ) P ( s )
A( s )Q ( s ) L( s ) P( s )
N (s)
P(s)
P( s)
s (s d4 )s
2
N (s)
P(s)
P(s)
N ( s)
,
s d4 s 4 N (s)
5
(12)
we get transfer function that is identical to the required transfer function (5), which is
enough for proof of asymptotic tracking of reference (2).
908
The controller design for tracking trajectory with controlled jerk
To make the system robust to the effects of step and ramp of external
disturbances and resistant to change parameters of an object, it is necessary to prove
that the response of the system due to these disturbances is zero.
The transfer function Gyw ( s) , (from w(t) to y(t), Fig. 2b) is:
Gyw ( s )
Y (s)
W (s)
1
P(s)
Q(s)
L(s) P(s)
A( s ) Q ( s )
P ( s ) A( s )
,
A( s )Q( s ) L( s ) P ( s )
(13)
then the response action due to disturbances in the frequency domain is:
P ( s ) A( s )
Y w (s)
W (s) .
A( s )Q ( s ) L( s ) P ( s )
If a disturbance in the time domain in the form of w(t ) D t p , p
D p!
the frequency domain is W ( s )

y w (f )
s p 1
lim y w (t )
0,1, 2... , then in
then it follows that:
lim sY w ( s )
t of
(14)
lim s
s o0
For the robust controller y w (f)
s o0
Ks 2 ( s 122) Ds pp1!
A( s )Q ( s ) L( s ) P ( s )
0 which is satisfied for p
(15)
0 and p 1 ,
because the P ( s ) , Q( s ) , L( s ) and A( s ) do not contain common factor s , q t 1 , so that

robustness to external disturbances is proved.
Let P ( s ) and Q( s ) of the object be replaced with P ( s ) p ( s ) and Q( s ) q ( s ) .
Response (14) for p d 1 now becomes:
q
y w (f)
lim sY w ( s )
s o0
lim s
s o0
s 2 ( s 122)( K p( s)) sD2
A( s) Q( s ) q ( s ) L( s ) P ( s ) p ( s )
(16)
which shows that the system is resistant to changes in the object parameters.
5. SIMULATIONS
Simulation of the system with designed controller and step type of disturbances
was performed in SIMULINK, Fig. 3. To generate a reference (2) the S-function was
used.
-K-
SFsinus
Clock
control
disturbance
Step
S-Function
-Kamplifier
num(s)
s3+122s2
16.52
s2
controller
plant
displacement
error
Fig.3 SIMULINK model

The simulation results are shown in Fig. 4. Referent r (t ) and real displacement
y (t ) are presented by [m] while error e(t ) r (t ) y (t ) is presented by [mm]. The
909
following values are selected: D
6 m,W
6s, A 1.35 m/ s and V
2 m/ s .
Desired output, real output and output error

8
reference
real output
output error
e[mm]
r,y [m],
4
2
0
-2
3
time t[sec.]
Fig.4 Simulation results of systems from Fig. 3

6. CONCLUSION
The paper describes the design of robust controllers, which provide asymptotic
tracking trajectory in which the jerk is changed at a predetermined function. The real
elevator with induction motor drive was used as the object model. Simulation of the
entire system with designed controller, which is made in the software package Matlab
and its toolbox Simulink, has shown that the designed controller meets the
requirements completely. Asymptotic tracking trajectory with immunity to external and
internal disturbances was realized. The value of jerks kept the value in all cases.
LITERATURE
[1] Bojan Kneevi, Branko Blanua (2011): Motion control of elevator in the function
of the value of jerks (org. Upravljanje kretanjem lifta u funkciji vrijednosti trzaja),
International Scientific Professional Symposium INFOTEH-JAHORINA Vol. 10,
Ref. A-9, pp. 40-44
[2] H. Z. Li, Z. M. Gong, W. Lin, and T. Lippa (2007): Motion profile planning for
reduced jerk and vibration residuals, SIMTech technical reports, Volume 8 Number
1 Jan - Mar, pp 32-37
[3] Paul Lambrechts, Matthijs Boerlage, Maarten Steinbuch (2005): Trajectory
planning and feedforward design for electromechanical motion systems, Control
Engineering Practice 13, pp. 145-157
[4] i-Tsong en (1999): Analog and Digital Control System Design: Transfer
Function, State Space and Algebraic Methods, State University of New York at
Stony Brook.
[5] Mihajlo J. Stoji (2009): Synthesis of linear automatic control systems (org.
Sinteza linearnih sistema automatskog upravljanja), Faculty of Mechanical
Engineering, Banja Luka.
910
ULTRASONIC WELDING SYSTEM FOR AUTOMOTIVE WIRINGS

INDUSTRY
Deaconu, S. I.1, Opria N.2, Popa, G. N.3, Latinovic T.4
Summary: The ultrasonic welding machines are key elements in the process of
achieving an automotive wiring. By the quality of the performed welds depends the
quality of the entire wiring. From this reason, a special attention should be given to the
correct programming, parameters optimization and replacing and adjusting procedures
of these machines active devices.
These equipment should ensure the reproducibility of the welded joints' quality,
automation possibilities for the manufacturing processes and their monitoring.
Key words: automotive wirings industry, ultrasonic welding machine, maintenance,
adjustments, optimization.
1. INTRODUCTION
Ultrasonic welding is based on transforming the electric power in mechanical
oscillation power of the soundtrode. By this procedure, plastic or metal parts are
perfectly bined by means of high acoustic frequency vibrations, applied locally,
achieving a high quality connection, both from mechanical and electric viewpoint [1].
The working frequency is of tenth of kilohertzs. The acoustic energy is transformed in
thermal energy by friction, the welding period being smaller than one second.
Moreover, is no need of additional material, lowering the costs and giving a special
aesthetical aspect to the weld [2], [3].
The welding process is made automatically, the single problem imposed to the
operator being the correct positioning of the wires to be welded. By this cause, to
mainain the welds quality, the ultrasonic welding machines should properly maintained
and the adjustments to be done according to the specific standards [4], [5], [6].
In this work are treated the component elements, the operation mode, set-up
mode, replacement and adjustment mode of an ultrasonic welding machine of Schunk
Minic-Quick Automatic PC type.
Associate Professor, Deaconu Sorin Ioan, Hunedoara, Romania, Politehnica University of Timisoara,
(sorin.deaconu@fih.upt.ro)
2
Engineer, Opria Nicu, Deva, Romania, Politehnica University of Timisoara, (oprisa.nicu@fih.upt.ro)
3
Lecturer, Popa Gabriel Nicolae, Hunedoara, Romania, Politehnica University of Timisoara, (popa.gabriel@fih.upt.ro)
4
Professor, Latinovic Tihomir, Banja Luka, Bosnia-Herzegovina, University of Banja Luka, (tiho@inecco.net)
911
Deaconu, S. I., Opria N, Popa, G. N., Latinovic T.
2.
CONSTRUCTIVE AND FUNCTIONAL ELEMENTS
The Minic-Quick automatic PC system from Schunk company was conceived

for welding of copper and aluminum electric wires with total cross-section between
0,5y30 mm2. The machine should be connected to a compressed air network with
pressure ranged between 6 and 10 bar and a flow of 900 l/min, and to the electric grid
of 3u400V, 50/60 Hz and a maximum current of 16A. The machine can execute line
welds or end welds (fig. 1) [7].
Fig. 1 Welds executed by means of ultrasounds
Fig. 2 Machines general diagram
The machines main component parts are (fig. 2): welding module - 1, devices - 2,
connectors - 3, process computer - 4, connecting module to the electric grid - 5, pneumatic
drive system - 6, monitor - 7, keyboard - 8, swiveling desk for monitor and keyboard - 9,
work table - 10, high frequency generator - 11, pressure regulator and compressed air filter
- 12, control pedal - 13, module housing -14, table positioning system - 15.
An important component represents the high frequency generator supplied
from the three-phased voltage grid of 3u400V, 50 HZ, maximum current 7A, having at
output the following values: maximum voltage 1400V, frequency 20 kHz and maximum
power of 3000 W.
The welding process is controlled by means of a suitable program called
SutWin that runs under Windows, installed on the process computer (PC) and which
communicates with the PLC by serial interface. The process computer sends to PLC
the data comprising the welding parameters and diverse controls given by keyboard,
and the PLC (the machines microcontroller) manages and controls effectively the
welding operation. The PLC sends to PC data regarding the welding operation, the
measured parameters and the error signals. The PC, by means of the related software,
allows the achievement of some statistics of the performed welding operations, offers
the possibility of results analysis, possibility to edit and configure the welding
parameters and setting various operational parameters of the machine.
The ultrasounds produced by the generator are undertaken by the converter
and transmitted to the soundtrode which comes in contact with the wires to be welded.
This wires are clamped by means of an assembly of mobile parts pneumatically driven,
which is pressing the wires on the main device and executes a vibrating movement
along its axis.
The wires are melting, binding one-to-another due to the temperature obtained
by the sounds propagation and vibrating movement of the base device which
determines also the weld's leveling [1].
912
Ultrasonic welding systems for automotive wirings industry
3.
WELDING PARAMETERS OPTIMIZATION

START
Introducing
cross-section
Automatically preset
welding parameters
Introducing wires
Start welding
Weld checking
Result
checking
NO
Adjusting
parameters
YES
Tolerances checking
Saving the parameters set
STOP
Fig. 3 Optimization chart of the welding parameters to a certain section

Optimization of the welding parameters is made based on the chart from fig. 3
assuming the passing through the following steps: introducing the conductors crosssection and the welding width, introducing the wires and correct positioning in the
welding space, commanding the weld, checking both visual and by capability surveys
in laboratory (traction tests, cross-section microscope survey, measuring the contact
resistance etc.), deciding the parameters adjustment or continuation, modifying
energy, pressure or amplitude by r10%, checking the tolerances for the welding height
and compacting, using the saving functions.
In fig. 4 is presented the menu for the graphic presentation of a 3-equal wire
weld.
913
a)
b)
Fig. 4 Menu of a welds graphic presentation a) selecting the colour; b) selecting the
section
There are standardized values specific to each welding section, given by manufacturer.
Against the standardized values, there is the possibility to make adjustments in
restricted limits, increasingly or decreasingly [1].
4.
The welding energy takes values from 50 Ws to 3000 Ws, and the welding time
ranges between 250y2000 ms. For diverse sections, the welding energy varies.
If the welding energy is too high, the weld becomes brittle. In case a too small
pressure force, the weld is weak, and in case of a pressure too high, the wires are
damaging or breaking-up [1].
In fig. 5a are presented the parameters displayed further a weld and in fig. 5b
the power curve during the welding process.
The statistic information about a certain welding node are stored and can be
displayed (fig. 6).
Also, voltages at idle operation and load operation can be visualized (fig. 7).
a)
b)
Fig. 5 a) welding parameters; b) power curve P = f(t)
914
Ultrasonic welding systems for automotive wirings industry
Fig. 6 Welds statistic into a certain node

5.
Fig. 7 Voltage monitoring
MAINTENANCE AND CALIBRATION OF ULTRASONIC WELDING MACHINES
The access in this submenu can only be done based on a password, because
the incorrect settings or non-respecting the diagnose procedure leads to mistunings or
even damaging of some components of the machine.
After making revisions or after replacing some devices, is imposed to perform
the diagnose procedure for calibrating the machines measuring devices. To perform
the diagnose are necessary a series of calibrated pins supplied by the producer
together with the machine. By diagnose, the machine calibrates its zero and work
positions of the two mobile devices, respectively the welding spaces width and welds
height, based on some calibration algorithms included in the Sutwin software.
a)
b)
Fig. 8 a) diagnose menu; b) operation parameters menu

The following parameters can be set:
- after-welding relaxation time (0,10) is the period of interrupting the vibrations
after finishing the welding, before the time of relaxation delay;
- time of relaxation delay (0,20) is the vibrations restart period to allow the
welds release and removal of possible small residues;
915
- adaption value after storing the reference heights, in the learning process
the machine makes a permanent average of the last n welds (n is a set-up value). In
this way, the average of the last n welds performed is considered reference for the next
weld. The adaptions purpose from one weld to another is to allow the machine to
follow the small modifications due to external factors, in conditions in which the
machine maintains its capacity to detect non-permitted deviations in the welding
process.
In fig. 8a is presented the menu for performing the diagnose and in fig. 8b the
set-up menu of the machines operational parameters.
6. CONCLUSION
The advantages of ultrasonic welding are: high work speed, respectively
productivity, by performing of one single operation, energetic efficiency, aesthetic
aspect, high binding quality level, materials' cost cutting by eliminating the
consumables and high efficiency of the specific welding equipments.
Automotive industry is one of the industries where the production's renewal degree
is continuous and the buyers' demands increase exponentially. The machines' endowments
are personalized, delivery terms are short and the quality level should be faultless.
The automotive wiring represents an important element, even vital for the
automotive's good operation. The hundreds or thousands of connections between
wires should ensure the respective circuits' continuity, regardless the operation regime
and external conditions. Any contact fault compromises the entire assembly.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
916
Deaconu, S.I., Topor, M., Groza, A., Ojog, F., Popa, G.N. (2009). Optimization of
Ultrasonic Welding Joint of Automotive Electric Wirings. SIELMEN 2009
Conference Proceedings, p. 333-336.
Deaconu, S.I., Costineanu, D., Popa, G.N. Popa, I. (2006). Ultrasonic welding
th
system of the conductors for automotive wirings. The VI Multidisciplinary
National Conference with international participation Prof. Dorin Pavel founder
of the Romanian hydroenergetics Conference Proceedings, Sebe, vol. 10, ISBN
973-8130-82-4, p. 335-342.
ora, I., Golovanov, N. (1997). Electrothermics and Electrotechnologies, vol. I,
Technical Publishing House, Bucharest, p. 187-232.
ora, I., Golovanov, N. (1999). Electrothermics and Electrotechnologies, vol. II,
Technical Publishing House Bucharest, p. 77-91.
ora, I., Nicoar, D., Vekas, L. (2002). Electroacoustic equipments for performant
processing, University Horizons Publishing House Timioara, p. 121-135.
Popescu, C., Popescu, M.O. (1997). Electrotechnologies. Principles and
applications, Matrix Rom Publishing House Bucharest, p. 96-153.
Schunk Ultraschalltechnik Operation Manual Minic-Quick Automatic PC, from
http://www.sonosystems.eu/en/shunck 01.c.30329.de, accessed on 2011-01-12.
WIRELESS SENSOR NETWORKS IN HOME AUTOMATION

Draen Paali1, Zlatko Bundalo2, Duanka Bundalo3, Miroslav Kostadinovi4
Summary: Increased combination of wireless sensor networks and actuators in home
automation systems results in efficient control of applications and energy saving. This
paper considers and desribes the infrastructures and concepts of such wireless
networks in home automation. Overall architecture and design consideration of
wireless sensor networks in home automation are described first in the paper. Main
features and possibilities of implementing wireless sensor networks in home
automation are given. Application and integration of actuators in wireless sensor
networks to monitor and control various physical and environmental conditions in home
automation are given. Simulation of a sensor-based system for specific home
automation application using certain software is given and desribed in detail.
Keywords: wireless sensor networks, home automation, applications, sensor nodes
1.
INTRODUCTION
Wireless sensor networks (WSN) are considered as an essential field in Home

Automation (HA), which combines usage of different technologies and devices. There
is a strong aim to develop and deploy a wireless networked control systems, where a
control system communicates over a wireless network, in factory and HA [1]. In order
to realize such wireless communication infrastructure, variety of wireless technologies
such as ZigBee, Bluetooth and WLAN can be used in an efficient manner in HA. These
technologies use different devices such as sensors, controllers and actuators. Such
devices in WSN act as decentralized system in which every sensor or controller is selforganized and responsible for local acquisition and processing of data. Therefore,
WSN including these devices provide a communication infrastructure which has the
ability to monitor physical environment and implement embedded data processing,
which are further sent to the base station or control unit.
Draen Paali, BS., Agency for Intermediary, Informatical and Financial Services, Banja Luka.
(pasalic.drazen@gmail.com)
2
Zlatko Bundalo, PhD., Faculty of Electrical Engineering, Banja Luka. (zbundalo@etfbl.net)
3
Duanka Bundalo, PhD., Faculty of Philosophy, Banja Luka. (dusbun@gmail.com)
4
Miroslav Kostadinovi, MS., Faculty of Traffic Engineering, Doboj. (kostadinovicm@gmail.com
917
Draen Paali, Zlatko Bundalo, Duanka Bundalo, Miroslav Kostadinovi
2. WIRELESS SENSOR NETWORKS

The term wireless networked control system or wireless sensor and actuator
network (WSAN) refer to a control system, which communicates over a wireless
network [1]. A WSN consists of tiny resource-limited devices called sensor nodes.
Each sensor node has some sensing ability [2]. In home automation WSN, sensor
nodes measure and process physical data, which are further distributed to the control
unit. Using these data, the control unit controls the actuators in order to implement
specific task. An example of WSN infrastructure for HA is shown in Fig.1.
Fig.1 Wireless sensor network architecture example in home automation

These low-power sensor nodes in WSN system are cheap, can use different
energy sources (battery, solar, etc.) and have minimal energy consumption. To
decrease maintenance costs these sensor nodes are designed to operate as
autonomously as possible in terms of energy consumption [3]. Therefore, such sensor
nodes can operate as self-organized and self-optimized mobile devices. Energy of
these nodes is used to power the sensor electronics, the data processing and routing,
and the radio transmission [3]. Because the sensor nodes use low-power energy
sources, they can change their activity modes. General sensor node architecture is
shown in Fig. 2.
Fig. 2 General single node hardware architecture

918
Wireless Sensor Networks in Home Automation
2.1 Sensor network topologies

A major consideration during the WSN system design in HA represents the
usage of appropriate network topology. Estimating the performance requirements and
the building environment is a key task before deciding which network topology to use.
Wireless sensor networks in HA commonly use three different network technologies:
star, mesh and hybrid. The main difference is the use and behaviour of the network
components [2].
Star topology uses sensor nodes as end-point devices which are usually
battery driven. This type of network topology commonly consists of small number of
nodes which communicates with the sink (relay, router) node.
In comparison to the star topology, mesh topology has greater coverage area
and can consist large number of nodes. All nodes in wireless mesh network have
routing capabilities and message can hop from node to node until they reach their
destination [2]. This can increase the communication complexity, probability of packet
drop and data latency, but coverage more building area and reduce sensor errors.
A hybrid network topology combines the features of star and mesh network. As
with star network, sensor nodes are battery powered and communicate directly with
routers, which extend the network range. This kind of sensor network topology
combines the benefits of the multi-hoping mesh network and the minimal low-power
sensor nodes energy consumption in star topology.
2.1.1
Node Localization
Because sensor nodes have limited communication area coverage, an

important task during the WSN design phase is the node placement, especially when a
large number of sensor nodes are used. Node localization is a technique to define the
sensor nodes position in an area. Some sensor nodes are aware of their own position,
these nodes are known as anchors or beacons. All other nodes localize themselves
with the help of location references received from the anchors [4]. In general, the basic
approach is to measure the distance between the nodes which can be established
using different localization methods.
The most popular methods can be classified as follows: Received Signal
Strength Indicator (RSSI), Time based methods (ToA, TDoA), Angle of Arrival (AoA),
Hyperbolic trilateration, Triangulation and Maximum Likelihood (ML) [4]. These
localization methods lead to improved data forwarding from sensor nodes to the control
unit. In such data transmission, the energy consumption is minimized, sensor fault
errors and packet drop probability is decreased.
2.2 Energy saving and consumption in wireless sensor networks
As previous mentioned, decreased energy consumption is accomplished using
different localization techniques, but there also exist other energy consumption
sources. In a sensor node, there are three activities which represent main sources of
energy consumption: sensing, computation and radio operation. Out of those three
sources, energy loss due to radio operation is the maximal one [5]. The processing
energy is spent by the processor with memory, which on active state has constant
power consumption. The embedded sensors and A/D converter (Fig.2) are adapted as
there is any sensing task, and the corresponding power consumption is constant.
919
For wireless communication the reception and transmission energy is derived

from the RF circuits [6]. This energy consumption results from the need of running the
radio frequency (RF) circuits at the transmission and receiving phase, in which m bits
can be sent or received. The value of the energy consumption at the transceiver side
depends also on the distance d between two sensor nodes, as well as on the energy
needed by the electronic circuits. A general packet transmission scenario of m bits is
shown in Fig.3. Therefore, when sensor node i transmits data to a sensor node j, the
power consumed by the transmission portion is [6]:
Pt
Eel m Eamp m d
where m represents number of bits to send, Eel is electronic energy consumption per
one bit sent, Eamp is a constant which represents the energy dissipation value and
depends on the RF circuit architecture, and d is the distance between sensor nodes.
Fig. 3 Transmission of m bit packet between sensor nodes

In addition to the previous mentioned, essential amount of energy loss exist in
the three most important layers of the WSN protocol stack: physical layer, MAC
protocol and network layer. In general, the physical layer ensures the communication
through physical link, implementing modulation and data coding. In case of channel
noise appearance, the receiver can efficient decode the data.
Energy saving in the physical layer can be accomplished by controlling the
transmission power or data rate. Power control can rely on the receiver observations
which are fed back to the transmitter or channel transmission probed by the
transmitter. The receiver feedback loop is achieved through the predictive controller, or
simple algorithms [3]. The receiver observation include the bit error rate.
The MAC protocol defines how access the medium by a node, and how to
transfer the data, when there is more than one node accessing a single wireless
channel simultaneously [5]. The MAC layer also implies channel assignment for sensor
node and power control. The channels can be reused which implies that a sensor node
get a channel at the data transmission phase. The purpose of the MAC layer in WSN,
is to reduce sensor node energy consumption which can be established using different
operative sensor node states. Idly waiting wastes huge amounts of energy during
communication. Major energy wastes in MAC protocol for WSN include collision
(packet drop), idle listening (listening on possible traffic which is not sent) and control
packets [5].
The network layer is in charge of choosing a path in the network to transmit the
packets between the transmitter and the receiver, in an energy-efficient manner. The
usual metric for this layer is the network (node) life-time [3].
920
Wireless Sensor Networks in Home Automation
3. SIMULINK MODEL TO CONTROL PROCESSES IN HOME AUTOMATION

We present here a wireless sensor control network Simulink model for
controlling processes (etc. ventilation, temperature) in home and building environment.
This model uses sensor nodes for data transfer and actuator to control the process.
The model was designed using PiccSIM Toolchain which has support for modelling,
control design and simulation. In combination with the PiccSIM Toolchain, we used the
ns2 network simulator to configure the network and simulate our model. The wireless
sensor network Simulink model is shown in Fig. 4.
Fig.4 Wireless control Simulink model

In order to control a process in home or building environment with a control
device, we used reference temperature for this device. The temperature response of
the control device is shown in Fig. 5.
Fig. 5 Resulting simulation of process control device in a building
921
4. CONCLUSION
This paper proposed the wireless sensor network infrastructure deployment
considerations in home automation. Choosing the appropriate network topology and
node placement leads to improved wireless sensor network performance. In restricted
communication area coverage, this improved network performance can be
implemented using node localization methods. This is a key issue especially in large
buildings with unconventional interference and the need for real-time responses.
Furthermore, this leads to improved network lifetime and stability, minimized node
communication failures and energy wastes.
5. LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
922
Bjrkbom, M. (2010). Wireless Control System Simulation and Network Adaptive

Control. PhD Thesis. Aalto-Print, Helsinki.
Akbari Nejad Zajad, A., Akbari Nejad Zajad, K., Javid, A., Shayban, P. (2010).
Design the multipurpose and low-power wireless sensor board for building
management system. Proceedings of the 9th WSEAS International Conference
on Electronics, Hardware, Wireless and Optical Communications (EHAC 10). pp.
32-36.
Cardoso de Castro, N., Canudas de Wit, C., Johannson, K. H. (2010). On EnergyAware Communication and Control Co-design in Wireless Networked Control
Systems. Proceedings of the 2nd IFAC Workshop on Distributed Estimation and
Control and Networked Systems (NecSys'10).
Pal, A. (2010). Localization Algorithms in Wireless Sensor Networks: Current
Approaches and Future Challenges, Network Protocols and Algorithms, Vol.2,
No.1, pp. 45-73.
Roy, A., Sarma, N. (2010). Energy Saving in MAC layer of the Wireless Sensor
Networks: A Survey, National Workshop in Design and Analyses of Algorithm
(NWDAA 10).
Wang, X., Wang, S., Ma, J.-J., Bi, D.-W. (2008). Energy Efficient Organization of
Wireless Sensor Networks with Adaptive Forecasting. Sensors - Sensors. Vol. 8,
No. 4, pp. 2604-2616.
SOME APPLICATIONS OF NEUROARM INTERACTIVE ROBOT AND

WEBOTS ROBOT SIMULATION TOOL
Mihailo Lazarevi1, Petar Mandi2, Vasilije Vasi3
Summary: In this paper they are presented possibilities an advanced 7 DOF robotic
arm research platform (NeuroArm) equipped with various sensors, and controlled by
seven 16 bit processors and one 32 bit Core Embedded PC running 2.6 Linux. The
NeuroArm comes with a suite of software including a MATLAB configuration and a
suite of C, API libraries and low level protocols for all joint level controls and sensors.
Moreover, the robot incorporates, as standard, a novel human-robot safety interaction
system to facilitate robot operation in close proximity to people. Also,software tools
including the Cyberbotics Webots Simulation package i.e. Webots is a development
environment used to model, program and simulate robot and it is well suited for
research and educational purposes related to robotics. Webots also contains a number
of interfaces to real robots, so that once your simulated robot behaves as expected,
you can transfer its control program to a real robot like e-puck, Khepera, Hemisson,
LEGO Mindstorms, Aibo, etc. Finally, the effectiveness of webots will be illustrated on
suitable model of Neuroarm and inverted pendulum.
Keywords: NeuroArm robot, Cyberbotics Webots Simulation package, properties of
robot, 7dofs
1.
INTRODUCTION
It is well known that robotic systems are more and more ubiquitoused in
industrial applications as well as in the field of direct interaction and helping people socalled friendly home environment. As one of these robotic systems capable of
operating in such environments is NeuroArm robotic system. It is an integral part of the
Laboratory of Applied Mechanics, Mechanical Engineering in Belgrade, (Fig.1) and
serves the educational purposes within the course Mechanics of robots, the Master
works as well as for scientific research in the field of robotics to doctoral studies.
Within NeuroArm Manipulator System - there are a rich set of options that
1
Full prof., dr Mihailo P. LAZAREVI, Department of Mechanics, Faculty of Mechanical Engineering,

University of Belgrade, Serbia, st. Kraljice Marije 16, Belgrade 35, SRB-11120 Belgrade, Serbia, e-mail:
(mlazarevic@mas.bg.ac.rs)
2
MsC student, Petar Mandi, Faculty of Mechanical Engineering, University of Belgrade, Serbia, Kraljice
Marije 16, Belgrade 35, SRB-11120 Belgrade, Serbia, e-mail: (mandicpetar84@gmail.com )
3
PhD student, Vasilije Vasi, Faculty of Mechanical Engineering, University of Belgrade,Serbia, Kraljice
Marije 16, Belgrade 35, SRB-11120 Belgrade, Serbia, e-mail: (vvasic@mail.amis.net)
923
M. Lazarevi, P. Mandi, V. Vasi
enable scientists and engineers to configure your robot that will meet the needs. From
the mechanical point of view NeuroArm a robotic arm has 7 degrees of freedom (6 for
rotating and translating 1), including gripper, (Fig. 2).
Mechanism of griper has the possibility of a parallel movement and control the
position and speed. The maximum load that can raise the robot is up to 2-3 kg
depending on the additional equipment. Control is based on the applying of robotic
processors Atmega 128 with I2C or SPI high-speed highway,(Fig. 1B). Processor
Atmel Atmega 128 Robot - integrate other coprocessors that are used to control on
the executive level, surveillance sensors, robot status, while for the security of the
robot uses NeuroBackdrive & NeuroArm Protocol. In the main joints there are
sensors for measuring position, velocity, acceleration and torque. This robot has a
highly configurable advanced PID controllers, the actuators torque high intensity, i.e. 3
Amp Maxon DC motor with planetary gearing and high-resolution encoders for each
robot joint, respectively. These provide a good/smooth control in closed feedback loop
in respect to position, velocity, acceleration of a given profile, as well as control which
based on the torque (torque control). Besides, there are inertial sensors on the endeffector in the purpose of determining and taking into account the inertial effects
caused by the inertial forces. Also, the part of the equipment is haptic joystick with
force feedback. It allows us to control a robot remotely over the web, with the ability of
the operator "feel" what the robot "feels. "
Fig. 1 A) left-Laboratory Robot - NeuroArmTM, Department of Mechanics, Faculty of

Mechanical Engineering, BU, Belgrade B) right - hardware structure of control system
of given robot
Also, Powerful Robot Host High Level is used to sophisticated real-time control
with simultaneously using of inverse kinematics, dynamics, trajectory, and control by
force, adaptive control, control based on artificial intelligence, etc. In addition, it is
enabled the internet connection with the accompanying web server as well as the USB
host controller. Robot includes as standard equipment, the new man-robot security
system in the process of interaction to facilitate the operation of robots in the vicinity of
people. NeuroArm Manipulator System in conjunction with the Human-Robot
Interaction Safety System is used to solve this problem.
924
Some applications of NeuroArm Interactive Robot and Webots Robot Simulation Tool
Fig. 2 Model of Neuro Arm with 7 degrees of freedom

In particular, on the top of the robotic arm, near the gripper it is built a camera,
that is, computer vision sensor CMU cam3 which allow us to visually tracking given
objects, or object recognition enabling the implementation of various telerobotic
applications. CMUcam3 is fully programmable built-in computer vision sensor-where
the processor is connected to the Philips LPC2106 OmniVision CMOS camera sensor
module. Common C code can be developed for the GNU port CMUcam3 using the
number of open source libraries and various ready-made programs.
2.
CONTROL OF NEUROARM ROBOT
There are four basic methods of control given robotic system that can be used
to set requirements and programmatic experience. The first of these is the use of
Windows GUI Tool - Windows graphical interface (Fig. 3) which allows for
easy use and control of robot educational purposes
Fig. 3 Graphical Interface in Matlab environment
925
Another way is the use of open source serial protocol & C library so that it
can provide built-in programmable controller PC. This level requires a good
knowledge of C programming language, C++, Java, Python as well as good
understanding of the robotic system in real time and is intended more for
scientific purposes, ie, researchers and PhD students.
It can also be used so URBI's (universal real-time behavior interface)
specially developed interface that enables control of any robot in a very
short time and in any language.
Details will be displayed here last, the fourth way, which includes the use of
Webots simulation package Cyberbotics company.
3.
SOME APPLICATIONS WEBOTS SIMULATION PACKAGE
Webots is a professional simulation software package for robots, with

emphasis on the use of mobile robots and it can be used for educational purposes and
for research in robotics. It offers a rapid prototyping environment robot, which allows
the user to create 3D virtual world, taking into account the real properties such as
mass, the corresponding coefficients of friction, etc. The user can now add a simple
way active / passive objects (stationary or mobile robots). In addition, the Webots
contains a large number of robot models and examples of control programs that can
help users in the initial stage. Webots also contains a large number of interfaces for
real robotic systems, which enables the developed program is transferred to a real
robot, such as e-puck, Khepera, Hemisson, LEGO Mindstorms, Aibo, etc.
If they are not used to use existing models of the robot one may create their
own models of robots in simulated environments, using appropriate software packages
for 3D modeling application with the VRML97 standard. Webots-alone application and
consists of three phases. In the first phase it is formed file of Webots world which
consists of one or more robots and the corresponding 3D virtual environment forming
so called .wbt files. The second phase one may define the control algorithms for given
robot using C, C + +, Java, Python, Matlab program, while in the third phase one can
optionally use supervisor. Supervisor is a privileged driver type robot that can perform
operations that are normally carried out by the operator and not a real robot. For
example, this type of monitors controller allows the robot takes a random position in
space, to realize a simulation of video recording, etc.
Here are two examples of application shown Webots, i.e., a modeling,
simulation and control a robotic arm that corresponds to the actual configuration
NeuroArm robotic system (Fig. 4, 5), as well as an example of modeling and control of
inverted pendulum, (Fig. 6). On Fig. 4a it is shown the graphical user interface under
development wbt file where one can see two working windows, the left side of the
window where it is defined the model while in the right window shows following code in
C language. Figure 5 presents a graphical interface that also includes a mask control
as a result of the second phase of the simulation within a webots. In the second
example (Fig. 6) is realized as the modeling and control of inverted pendulum in a
given webots environment.
926
Some applications of NeuroArm Interactive Robot and Webots Robot Simulation Tool
Fig. 4 A) left-user-interface of robotic arm.wbt

B)right- defining the geometry of the robot segments
Fig. 5 Simulation of movement of robotic arms with a display driver interface

A)left- initial position B) arbitrary-set position
Fig. 6 Example of control of inverted pendulum in the Webots environment

A) left- code in C B)right- defining the geometry of given model
927
4.
DISCUSSION
This paper presents the possibilities of advanced robotic systems (NeuroArm),

which is given in the form of a robotic arm with seven degrees of freedom are equipped
with various sensors, controlled by seven 16-bit processors and a 32-bit Core
Embedded PC. NeuroArm robot is supported with appropriate software package that
includes MATLAB and C, API library and the low level protocols for all levels of
management and joint sensors. It also has a development environment - Cyberbotics
Webots simulation package used for modeling, simulation and robot programming,
where the application of appropriate interfaces for real robots, can be applied to give
the control program to real robots. Last, two examples illustrate the application of
simulation in webots environment: robot Neuroarm and inverted pendulum.
5.
ACKNOWLEDGEMENT
This work is partially supported by the Ministry of Science and Environmental

Protection of Republic of Serbia as Grant No- 35006, EUREKA project E!4930 AWAST
(Advanced Wavelet Analysis for Structural Testing)
LITERATURE only the selected
[1]
[2]
[3]
[4]
928
Angeles J.,Fundamentals of Robotic Mechanical Systems: Theory, Methods, and

Algorithms,Book,Springer,2007
ovi V, M Lazarevi, Mehanika robota, Knjiga 2009, Mainski fakultet, Beograd
http://www.neurorobotics.co.uk
Webot user guide from http://www.cyberbotics.com/
F. MAINTENANCE OF TECHNICAL SYSTEMS

AVAILABILITY ANALYSIS OF IRREDUNDANT TECHNICAL

SYSTEMS
Rusmir Bajri1, Enver Omazi2, Fehmo Mrkaljevi3
Summary: Due to continuous growth increasing in demand for energy derived from
coal there is a need to analyze mine site operation and production time and also need
to lower total cost. Availability is directly connected to the system up time and at the
same time profitability of system. In this paper the analysis of redundant conveyor belt
waste overburden transport systems of coal mine Dubrave. Increasing profitability and
market competitiveness of the manufacturing system is often reflected in decreasing
production costs. One method for achieving this objective is to increase the availability
of the production system. This paper seeks to identify and quantify the reasons for
delay of the production system and at the same conveyor belt transport systems at
opencast mine which is irredundant, with respect to reservation. A failure of this
production system is a direct reduction in profitability, or increased losses, which is why
availability analysis is of great importance.
Key words: availability, maintenance, process design, failure
1.
INTRODUCTION
The effectiveness of the production system is backbone of business success.

Availability of production systems within the coal mine is described as a key element of
successful production strategies including strategies for the mine waste overburden.
Availability of equipment is recognized as a key indicator of process control which can
often be directly related to maintenance management. With the competitive market
today, the coal industry is forced to continuously improve production efficiency and at
the same time waste overburden production as precondition successful coal
production. In order increase effectiveness of waste overburden, it is important to be
able to gain valid and reliable conclusions for necessary interventions to be able to
maintain production equipment and establish a proper management. Managing the
production and maintenance should be conducted in such a way that goals are
achieved by lower cost with a relatively high reliability. In technical systems this cannot
be easily achieved due to poor reliability of components and its complicity [5].
Production and usually maintenance targets are often focused on the analysis of
1
Mr. sci., Rusmir Bajri, Tuzla, Coal Mine Kreka, Opencast Mine Dubrave, rusmir.bajric@kreka.ba
Dipl. Ing., Enver Omazi, Tuzla, Coal Mine Kreka, Opencast Mine Dubrave, enver.omazic@kreka.ba
3
Dipl. Ing., Fehmo Mrkaljevi, Tuzla, Coal Mine Kreka, Opencast Mine Dubrave, fehmo.mrkaljevic@kreka.ba
2
931
Rusmir Bajri, Enver Omazi, Fehmo Mrkaljevi
availability and reliability. If critical equipment used in the production process it is

necessary to concentrate more on the availability, which is the case with the waste
overburden production system at surface mine Dubrave as irredundant technical
system. Unfortunately, the increase in availability is not cheap. So, it is important to
increase the availability of specific production system through the consideration of
elements of availability.
2.
THEORETICAL RESEARCH
Systems that are used for achieving the goals of production are called
production systems. It is hard to make a margin of any production system, because in
principle, any system is composed of subsystems, but at the same time is a part of a
larger system. Availability means that a system is on-line and ready for access. A
variety of factors can take a system off-line, ranging from planned downtime for
maintenance to catastrophic failure. The goals of high availability solutions are to
minimize this downtime and/or to minimize the time needed to recover from an outage.
Exactly how much downtime can be tolerated will dictate the comprehensiveness,
complexity and cost of the solution. Redundancy or backup configuration provides the
ability to avoid the delay when a system component fails. The serial system is a
system without reservation, if one component in the serial system fails the whole
system fails [3, 6]. The definition of availability is somewhat flexible and is largely
based on what types of downtimes one chooses to consider in the analysis. As a
result, there are a number of different classifications of availability, such as: inherent
availability AI, achieved availability AA, operational availability AO.
Inherent availability is the steady state availability when considering only the
corrective downtime of the system. It is defined as the expected level of availability for
the performance of corrective maintenance only. Inherent availability is determined
purely by the design of the equipment. Inherent availability is generally derived from
analysis of an engineering design. Inherent availability fulfills the need to distinguish
expected performance between planned shutdowns.
Achieved availability represents probability that an item will operate
satisfactorily at a given point in time when used under stated conditions in an ideal
support environment. It excludes logistics time and waiting or administrative downtime.
It includes active preventive and corrective maintenance downtime. Achieved
availability is defined as the achieved level of availability for the performance of
corrective and preventive maintenance. Achieved availability is determined by the hard
design of the equipment and the facility. Achieved availability is very similar to inherent
availability with the exception that preventive maintenance downtimes are also
included.
Operational availability is a measure of the average availability over a period of
time and it includes all experienced sources of downtime, such as administrative
downtime, logistic downtime, etc. It is the probability that an item will operate
satisfactorily at a given point in time when used in an actual or realistic operating and
support environment. It includes logistics time, ready time, and waiting or
administrative downtime, and both preventive and corrective maintenance downtime.
The operational availability is the availability that the customer actually experiences. It
is essentially the a posteriori availability based on actual events that happened to the
932
Availably analyses of irredundant technical systems
system. The previous availability definitions are a priori estimations based on models
of the system failure and downtime distributions [2, 3].
Table 1 Specific availability calculations
AI
AI=
AA
MTBF
MTBF + MTTR
AA=
AO
MTMB
AO=
MTMB + M
MTMB
MTMB+MDT
Description of tags used in Table 1: MTBM - mean time between maintenance actions,
MDT - mean downtime, MTTR - mean time to repair, M - mean maintenance
downtime, MTBF - mean time between failures.
3.
DISCRIPTION OF IRREDUNDANT TECHNICAL SYSTEM
Portion of waste overburden production system consisting of belt conveyors

that are connected in series, without redundancy, and analysis of its availability are of
great importance because any delay of transport element of the system is direct
suspense of complete production system. This article analyzes the availability of the
system from collecting conveyor belt to spreader, including the spreader. Each of these
production system elements has its own specifics and they require certain technical
conditions that are directly related to system maintenance.
4.
EXPERIMENTAL RESEARCH
Program of collecting information about operation, delays and maintenance

time has been established in order to successfully analysis of availability of irredundant
waste overburden production system at surface mine Dubrave. Collected in formations
are categorized in terms of vulcanization delays, technological delays, mechanical
delays and electrical delays. Covered delay time by the plan of the experiment in this
case, is a period of three years in which records of delay time are categorized by three
shifts during the day. Evidence is guided only for those delays which have a direct
impact on the availability of the waste overburden production system.
AA
AO
2010
2009
2008
0.54
0.56
0.58
0.60
0.62
0.64
Fig. 1 Availability of technical system in observed time

For simplicity of analysis this study was based on the availability of the month
933
as time unit, while the variable as delay time is measured in hours. To complete the
analysis of availability, time when the system does not operate but it is capable to work
has been collected, so called idle time. In this article following tags has been used: AAachieved availability, AO-operational availability, TD-technological delay, VDvulcanization delay, MD-mechanical delay, ED-electro delay, IT-idle time, OT-operation
time, DT-delay time. Because there is no record of mean time between failures
inherent availably has not been considered in analysis. It can be noted the availability
in observed period is relatively low and there is a great unused production capacity,
Fig. 1. Low availability rates may be sign of poor initial system design, poor
technological procedures, poor selection of maintenance strategy and poor
maintenance implementation. It also can be noted that the achieved availability is lower
than the operational availability, which indicates it is necessary to minimize the number
and duration of unplanned delay. Achieved availability can be increased by equipment
capital investments in terms of improving the maintenance [4, 5, 6].
TD
VD
MD
ED
Nov
Sep
Jul
Maj
Mar
Jan
0%
20%
40%
60%
80%
100%
Fig. 2 Time delay structure of technical system in year 2008.

TD
VD
MD
ED
Nov
Sep
Jul
Maj
Mar
Jan
0%
20%
40%
60%
80%
Fig. 3 Time delay structure of technical system in year 2009.
934
100%
Availably analyses of irredundant technical systems
TD
VD
MD
ED
Nov
Sep
Jul
Maj
Mar
Jan
0%
20%
40%
60%
80%
100%
Fig. 4 Time delay structure of technical system in year 2010

Easy access and space for equipment maintenance, increasing the number of
equipment to support and improve the management of spare parts could be some of
the measures to increase the availability. Considering the structure of delays it can be
noted that there is extremely high vulcanization delays in year 2008. The percentage of
their share decreases in year 2009 and 2010, Fig. 2, Fig. 3, Fig. 4. Deducing of
vulcanization delays is directly associated with replacement of rubber conveyer belt on
the conveyor belt transporter in series. This leads to the conclusion that it is necessary
to define a model of optimal replacement of rubber conveyer belt on the conveyor
transporter in order to prevent major delays of this kind with overall availability potential
increase. Considering total observation period it is evident that the technological delays
make up the largest percentage share of total delays. Despite the technological delays,
percentages of other types of delays do not change a lot, which leads to the conclusion
that other types of delays are likely directly related to technological. Analyzing the
existing system of collecting information about delays types, it could be noted the lacks
of systematic information like time between failures as the main argument in
addressing the incidence of failures and the implementation of measures of its
increase. Also, it could be noted there is need to manage idle time, Fig. 5. It is even
more important, knowing the idle time is directly in correlation to system which is
redundant. So, there is need to carry out activities that would reduce the idle time and
have great potential to increase availability of the considered technical, and at the
same time production system. Operational availability for year 2009 and 2010 are
close to achieved but considering time delay structure it is evident in increase in
technological time delay for year 2010 as major factor for low availability. Considering
achieved availability in same period it is noted decrease in year 2010. This is because
of increase mean maintenance downtime which described as total time required either
to restore system to a given performance level or to keep it at that level of
performance. This is indicator for increase effort in making more effective corrective
maintenance, preventive maintenance, and administrative, logistic actions in process
of maintenance of used equipment.
935
2010
2009
2008
TD
IT
OT
DT
0
1000
2000
3000
4000
5000
Fig. 5 Specific operation and delay times of observed period

5.
CONCLUSION
The article analyzes the achievable and operation availability of waste

overburden production system in process of coal opencast mine Dubrave. Structure of
time delays, operation time and outage time has been analyzed. Availability analysis of
the production system has enabled a comprehensive understanding the delay
problems and process effectiveness through the identification of causes that have a
negative impact. Focusing on issues that are identified in the analysis it is possible to
increase the overall availability of production systems and achieve better production
results. It is noted the lack of information about mean time between failures are need
for better management of delays. Measures to increase the overall availability of
observed system are proposed. It is recommended to establish a system for collecting
data for better proceeds understanding. This way it is easier to find out the reasons
that led to delay and be in position to continuously develop strategy in response to
fulfill production goals. Also, there is a need to point maintenance function to
preventive and predictive maintenance that help determine the condition of in-service
equipment in order to predict when maintenance should be performed and avoid
unplanned failures and thus increase the availability.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
936
S. Belak, D. iin-ain, (2009). Razvoj koncepta terotehnologije. Pomorski

fakultet u Splitu, Croatia.
Ramesh G., Terrence O.Ricky S., (2006). Maintenance and Reliability Best
Practices. Industrial Press, New York.
Jovii S., (1990). Pouzdanost, pogodnost za odravanje, raspoloivost
odravanja, logistika podrka. Nauna knjiga, Beograd.
P. Muratovi, (1994). Maksimalizacija operativne gotovosti proizvodnih sistema sa
neprekidnim tokom proizvodnje. FEM, Tuzla.
Fred K. G., Heinz P. B., (2006). Maximizing Machinery Uptime. Elsevier Inc., USA
Keeter B., (2002). Managing Availability for Improved Bottom-Line Results. ARMS
Reliability Engineers, Australia.
COMPUTER AND EXPERIMENTAL DETERMINATION OF THE HUB

UNIT LIFE
Aleksandar ivkovi1, Milan Zeljkovi2, Milorad Rodi3 Milivoje Mijukovi4,
Summary: This paper presents the experimental stand for determination of the hub
unit bearings life and matematical model for determination the same. Experimental
device provides test life of the hub unit in operational conditions. The mathematical
model is defined by the nonlinear change of axial and radial deformation and angle of
contact under the influence of the external axial and radial loads. Deformation and the
contact angle changes are determined based on body size and raceway, Hertz's
contact theory, the conditions of static equilibrium and John Harris's methods of load
distribution. Systems of nonlinear equations were solved by Newton's method. On the
basis of prior is determined by the dynamic load capacity and bearing life.
Key words: bearing life, experimental testing, hub unit bearing
1.
INTRODUCTION
Starting from the need of increasing compactness and reliability, and reduced
mass, in the world the new solutions of the wheel bearing has been developed in mind
the so-called integrated bearing assemblies (or HUB unit bearings) that achieve
reduced fuel consumption and increased stability during the vehicle driving. Three
basic types of integrated circuits bearing: HUB I, HUB II and III are developed, and
appear, in these bearing assemblies, also IV and V generation. From the previous
results stems that is still worked on improving these solutions. This applies in particular
to reduce weight and facilitate the process of assembly and disassembly while
reducing costs. Given the complexity of the research problems, this paper presents the
results of experimental studies and mathematical modeling related to the determination
of the second generation of the HUB unit bearing life (HUB II).
2.
MATHEMATICAL MODEL FOR DETERMINING THE LIFE OF HUB UNIT

BEARING
Under the action of load bearing composite parts are deformed and by Hertz,
contact area has the shape of an ellipse [1].
1
Mr ivkovi Aleksandar, assitent, Novi Sad, Faculty of technical science, (acoz@uns.ac.rs)

Dr Milan Zeljkovi, full proffesor, Novi Sad, Faculty of technical science, (milanz@uns.ac.rs)
3
Dipl. Ing., Milorad Rodi, Temerin, FKL A.D.
4
Dipl. Ing., Milivoje Mijukovi, Temerin, FKL A.D., mile.mijuskovic@fkl-serbia.com
2
937
Aleksandar ivkovi, Milan Zeljkovi, Milorad Rodi, Milivoje Mijukovi
Credible analysis of load sharing and stiffness of ball bearings, should include
non-linear loads and the relationship between contact deformation and loads.
For bearings analysis and calculation a different mathematical models are
used. Analysis of the developed models for bearings static behavior study has been
shown that, from the standpoint of design phase these models can be classified into
two groups:
x Previous calculation (dimensioning and optimization of basic geometric
size);
x Final calculation (check - identifying bearings behavior).
2.1 Calculation of the road resistance
Integrated wheel bearing life is calculated based on the reaction force at the
wheel for various exploitation conditions. Since it's a two-row ball bearing life is
determined for each track rolling. In Figure 1 the distribution of loads on the wheels is
shown.
Fig. 1 Load distribution on the wheels with parameters for life calculation
The forces of the road reaction are determined based on the vehicle weight
and Coriolis acceleration:
Wr
Wt
H
W
f w G W
T
2
G Wr
(1)
where: Wr - radial reaction force; Wt- axial reaction force; fw- load coeficient depending
of the vehicle type; G- Coriolis acceleration
2.2 Bearing load determination
Figure 2 shows a schematic view of bearing with the necessary parameters for
the calculation. Radial load acting on the bearing is determined by the relationship:
Frs
R
b
Wr s Wt
l
l
Fru
R
a
Wr s Wt
l
l
(2)
where: Frs- outer row bearing radial load; Fru- inner row bearing radial load; l-distance
between the effective centers of charge, a-distance between the wheel center point
and the nominal effective area of the outer row of bearing, b-the distance between the
wheel center point and the nominal effective area on the inner row of bearing;
938
Computer and experimental determination of the hub unit life
Figure 2 Geometrical parameters necessary for bearing life calculating
Axial load acting on the bearing in case of straight driving is the result of radial
load and in case the bearing outer row is obtained through the relation:
Fas
Fau
Fru
Wt
2Y
0
(3)
In the case of curvilinear driving, axial load is determined by the relationship:

Fas
Fau
Fru
Wt
2Y
Fas Wt
(4)
2.3 Bearing contact angle determination

At ball bearings under the influence of axial load, the load is distributed equally
to all rolling elements [1]. So:
Fa
Q
; where Z is the number of rolling elements
(5)
Z sin D
If we take into account the external axial load acting on the bearing, then [1]:
1/ 2
cos D 2
i.e.:
1 cos 2 D 2 D
2
F
1 cos D1 D
a

1/ 2
cos 2 D1
ZK n A
2G p
sin D1 D 0
cos D1
sin D 2 D 0
cos D 2
(6)
(7)
Relationships (6) and (7) are solved for 1 and 2 by Newton's method.
2.4 Bearing life determination
A mathematical model for the static behavior of ball bearings analyzing has
been developed using MATLAB software package based on Hertz's contact theory, the
conditions of static equilibrium and John Harris's methods of load distribution. Systems
of nonlinear equations are solved using Newton's method.
As these types of bearings operate with variable loads, for calculating the
939
bearing life is used expression:
Lhna
3,33
3,33
1,39
1,39
0,41
1 J i cos D 0,7 Z 2 / 3 D1,8 J r 2 R r 0,41 1 J i cos D 0,7 Z 2 / 3 D1,8 J r

r
0,7 2 R

98,1i
D rR
1/ 3
1/ 3
D
r
R
J
J

1 J
1 J
1
2
3 t P t P t P
1 I1
2 I2
3 I3
0.3
(8)
where PI1- bearing equivalent load when the vehicle is moving in a straight path, PI2 bearing equivalent load when the vehicle turns left, PI3 bearing equivalent load when
the vehicle turns right, t - time share of straight or cornering driving; p = 3 for ball,
bearing, R-radius of the corresponding raceway, r- raceway radius, D- ball diameter, J
bearing coefficient, i- the number of rows of rolling elements, D-angle of contact. In the
case of the two-row ball bearings with angular contact, angle of contact is determined
from the equation (6) and (7) assuming no effects of centrifugal forces and gyroscope
moment, so the angles of both raceways are identical. Z-number of rolling elements in
one raceway, Jr., J1 and J2 - integrals for bearings load and the matching ring.
2.5 Results view
Developed software solution was verified on the integrated vehicle wheel
bearing (32 128 x), as a special case two-rows ball bearings with angular contact.
In determining of the road resistance the mass limit of the specific vehicle type m =
1600 [kg] is taken, corresponding to the full board. Figure 3 shows the changes of
radial and axial deformation depending on the preload and Coriolis acceleration for the
vehicle mass m = 1600 [kg]
Fig. 3 Change of bearing radial and axial deformation, depending of the preload for different
values of Coriolis acceleration for m = 1600 kg
From the previous figure can be concluded that the increase of preload causes
axial strain decreasing with increasing radial deformation. On the other side, increase
of lateral acceleration causes a rise of the the axial and radial strain.
Based on previous data the dynamic load capacity and bearing life are
determinated. Bearing life is determined by the maximum vehicle weight m = 1600 kg,
with the Coriolis acceleration G = 0,7 g and n = 900 [rpm]. It is also predicted that drive
is, on the right track, 90%, in the right and left curves 5% i.e. in total 10%. The above
input data are defined by the appropriate standard. Based on these data is determined
the bearing dynamic load which is C0 = 69.500 [N], and bearing life at maximum
unfavorable operating wheel vehicles conditions is about 155.000 [km]. In Figure 4 is
shown the change of bearing life, depending on the preload of the previous test
conditions.
940
Computer and experimental determination of the hub unit life
Fig. 4 Bearing life, depending on the preload
3.
EXPERIMENTAL DETERMINATION OF THE BEARING LIFE
Based on the results of mathematical modeling, it has been concluded that is

necessary to design and produce test device for integrated wheel bearing testing that
would simulate the more realistic conditions. Therefore, the concept was defined, and
test device has been designed and built (Figure 5) that allows the simulation of
exploitation conditions of the integrated vehicle wheel bearing. Testing of the
integrated bearing is performed by effect of radial load and at defined time diagram of
axial load. This is the way to simulate the exploitation conditions of the integrated
wheel vehicle bearing (driving on the right track 90% of the curves right and left by
5%).
In examining of the life of the integrated wheel bearing inner ring temperature is
measured by the, the vibration in three directions and speed for the defined values of
radial and axial loads. Radial and axial load, of the tested bearing, are accomplished
by hydraulic power unit and the measurement of the actual loading is done by
measuring the oil pressure in the hydraulic cylinder with pressure transducer. Figure 6
presents the results of parameters measurements through which is possible to monitor
the state, or failure of the tested bearing.
Fig. 5 Experimental device for bearing life examination with schematic view of the load during
testing
941
Fig. 6 Measuring results a) vibrations b) temperatures
4.
CONCLUSION
Summing up the presented research results, it can be concluded that the

paper presents an attempt that, under-developed mathematical model, enables, even
in the design phase, with the reliability to predict the life of integrated bearing in
exploitation.
Also in the the paper test device for bearing life determination is presented.
The test results presented in this paper were obtained by radial load of 10.000 [N],
axial loads up to 6.000 [N], and revs up to 1500 [rpm]. Based on the bearing
calculation model, defined maximum allowed voltage on the tracks of the rings and
rolling elements, conditions of operation of each specific type of bearing, can be
determined the expected time of continuous operation of the device untill bearing fail.
In this case it is cca 155.000 [km]. For bearing failures determination, i.e. test stopping
two criteria are used: appearance of unallowed vibrations and excessive temperature
rise on the inner bearing ring.
REFERENCES
[1]
Tedric, A. H., Michael, N. K.: Rolling bearing analysis: Essential Concepts of

Bearing Technology, Fifth edition, Taylor & Francis Group, ISBN: 0-8493-7183X, 2007.
[2]
Todi, V., Mijukovi, M., Miloevi, M., ivkovi, A.: Prikaz i analiza
konstrukcionih reenja HUB integrisanih leaja, Zbornik radova, VIII
meunarodni struni skup o dostignuima elektrotehnike, mainstva i informatike
DEMI 2007, Banjaluka, 2007., str 107-114, ISBN 978-99938-39-15-6
[3]
ivkovi A., Zeljkovi M., Mijukovi M., Borojev Lj.: Matematiki model za
odreivanje deformacija integrisanog leaja, Zbornik radova - CD ROM, 35.
JUPITER konferencija, 31. simpozijum NU-ROBOTI-FTS, 2009, str. 3.21-3.26,
ISBN 978-86-7083-666-2
[4]
ivkovi, A., Zeljkovi, M., Tabakovi, S.: Matematical Model for the Roller
Bearing Life Determination, Academic Journal of Manufacturing Engineering,
Vol. 8, ISSUE 3/2010, 2010, pp 108-115, ISSN 1583-7904
Acknowledgements
In this paper some results of the project: Contemporary approaches to the development
of special solutions related to bearing supports in mechanical engineering and medical
prosthetics R 35025, carried out by the Faculty of Technical Sciences, University of Novi
Sad, Serbia, are presented. The project is supported by Ministry of education and science of the
Republic of Serbia.
942
DECENTRALIZED WASTEWATER TREATMENT SYSTEMS IN

LARGE SETTLEMENTS
Danijela Nikolic 1, Vanja Sustersic2, Jasmina Skerlic 3
Summary: Up to the considering of wastewater treatment, compared to the size of
settlements, we can select a process type of wastewater treatment. Group of large
settlements include settlements with over 20 000 inhabitants. Many of them are not
covered by the centralized systems of water treatment, so the wastewater treatment of
these settlements uses a special type of waste water processing. This paper provides
an overview of this process through the pre-treatment, primary, secondary and tertiary
treatment, with special attention to each of these procedures. Today, these ways of
wastewater treatment are dominant in world practice.
Key words: wastewater treatment, decentralized systems, large settlements
1.
INTRODUCTION
The principal objective of wastewater treatment is generally to allow human

and industrial effluents to be disposed of without danger to human health or
unacceptable damage to the natural environment. According to the definition of the
Best Available Technique (BAT), term best available wastewater treatment of
settlement includes some processes of waste water treatment with the regard to quality
of raw sewage, requirements quality of treated waste water and convenience of using
wastewater treatment processes in practice. Conventional wastewater treatment
consists of a combination of physical, chemical, and biological processes and
operations to remove solids, organic matter and, sometimes, nutrients from
wastewater.
This paper represents decentralized wastewater treatment system (DEWATS)
in large settlements. Large settlements include settlements with over 20 000
inhabitants. Waste water treatment of these settlements is quite natural choice type of
purification process of municipal wastewater. Today, it is dominant in the global
practice. This is the type of process which consists of:
Preliminary (pre-) treatment - removes materials that can be easily collected
from the raw waste water before they damage or clog the pumps and
skimmers of primary treatment clarifiers.
1
Assistant, Danijela Nikoli, Kragujevac, Faculty of Mechanical Engineering, (danijelan@kg.ac.rs)

Associated professor, Vanja Sustersic, Kragujevac, Faculty of Mechanical Engineering, (vanjas@kg.ac.rs)
3
Researcher, Jasmina Skerlic, Kragujevac, Faculty of Mechanical Engineering, (jskerlic@kg.ac.rs)
2
943
Danijela Nikolic, Vanja Sustersic, Jasmina Skerlic
Primary treatment - temporarily holding the sewage in a quiescent basin

where heavy solids can settle to the bottom while oil, grease and lighter
solids float to the surface.
Secondary treatment - removes dissolved and suspended biological matter.
It is typically performed by indigenous, water-borne micro-organisms in a
managed habitat.
Tertiary treatment - treated water is sometimes disinfected chemically or
physically (by lagoons and micro filtration) prior to discharge into a stream,
river, bay, lagoon or wetland, or it can be used for the irrigation of a golf
course, green way or park. If it is clean, it can also be used for groundwater
recharge or agricultural purposes.
Fig. 1 Process Flow Diagram for a typical treatment plant

The treatment applications are based on the principle of minimal maintenance,
the critical parts of the treatment system work continuously and uninterrupted with low
energy inputs. The technology provides treatment for domestic and industrial (nontoxic) sources. It can treat effluent flows from 1 up to 1000 m per day.
2.
PRELIMINARY (PRE-) TREATMENT
The objective of preliminary treatment is the removal of coarse solids and

other large materials often found in raw wastewater. Removal of these materials is
necessary to enhance the operation and maintenance of subsequent treatment units.
Preliminary treatment operations typically include coarse screening, grit removal and,
in some cases, comminution of large objects. In grit chambers, the velocity of the water
through the chamber is maintained sufficiently high, or air is used, so as to prevent the
settling of most organic solids. Grit removal is not included as a preliminary treatment
step in most small wastewater treatment plants. Comminutors are sometimes adopted
to supplement coarse screening and serve to reduce the size of large particles so that
they will be removed in the form of a sludge in subsequent treatment processes.
Chlorination may be used in preliminary treatment.
944
Decentralized wastewater treatment systems in large settlements
A device called a settler is used for this phase. If needed, a screening device,
for preventing unwanted large materials to enter the system can also be installed. The
settler is an underground constructed tank with one partition wall. Within the settler two
main treatment processes take place, first a sedimentation and second a stabilization
and digestion of the settled sludge through biological treatment. Storage volume is
provided for 18 up to 36 months, this parameter defines the necessary desludging
period.
Fig 2 Typical Section of a Settler

3.
PRIMARY TREATMENT
In the primary sedimentation stage, sewage flows through large tanks,

commonly called "primary clarifiers" or "primary sedimentation tanks." Sedimentation is
often used as a primary stage in modern waste water treatment plant, reducing the
content of suspended solids as well as the pollutant embedded in the suspended
solids. Due to the large amount of reagent necessary to treat domestic wastewater,
preliminary chemical coagulation and flocculation are generally not used.
In the late 1800s, began the use of septic tanks for primary treatment of
wastewater, and discharge of septic tank effluent into gravel-lined subsurface drains
became a common practice during the middle of the 20th century.
Today, the tanks are used to settle sludge while grease and oils rise to the
surface and are skimmed off. Primary settling tanks are usually equipped with
mechanically driven scrapers that continually drive the collected sludge towards a
hopper in the base of the tank where it is pumped to sludge treatment facilities. Grease
and oil from the floating material can sometimes be recovered for saponification. The
dimensions of the tank should be designed to effect removal of a high percentage of
the floatables and sludge. A typical sedimentation tank may remove from 60 to 65
percent of suspended solids, and from 30 to 35 percent of biochemical oxygen demand
(BOD) from the sewage.
The simple septic tank system is the most commonly known primary treatment
method for onsite wastewater treatment because of its considerable advantages.
Septic tanks remove most settle able solids and function as an anaerobic bioreactor
that promotes partial digestion of organic matter. A device called an anaerobic baffled
tank is used often for this phase. Several tanks are constructed in series to digest
degradable substances. Baffle walls or down-flow PVC pipes direct the waste water
945
stream between the chambers from top to bottom and up again. During this process
the fresh influent is mixed and inoculated for digestion with the active blanket deposit
of suspended particles (floating bacteria media) and microorganisms occurring
naturally at the bottom of each chamber. Because of the physical separation (multiple
chambers), various microorganisms are present at different stages, allowing a high
treatment efficiency.
Fig 3 Typical section of an Anaerobic Baffled Tank

In many industrialized countries, primary treatment is the minimum level of
preapplication treatment required for wastewater irrigation. It may be considered
sufficient treatment if the wastewater is used to irrigate crops that are not consumed by
humans or to irrigate orchards, vineyards, and some processed food crops.
4.
SECONDARY TREATMENT
The objective of secondary treatment is the further treatment of the effluent

from primary treatment to remove the residual organics and suspended solids. In most
cases, secondary treatment follows primary treatment and involves the removal of
biodegradable dissolved and colloidal organic matter using aerobic biological treatment
processes. Aerobic biological treatment is performed in the presence of oxygen by
aerobic microorganisms (principally bacteria) that metabolize the organic matter in the
wastewater, thereby producing more microorganisms and inorganic end-products
(principally CO2, NH3, and H2O).
There are many secondary treatment methods for DEWATS. Considering that
sand is the most common and available media for filters, sometimes media filter is
equivalent to sand filter. Generally, in areas with deep, permeable soils, septic tank
soil absorption systems can be used. On the other hand, in areas with shallow, very
slowly permeable or highly permeable soils more complicated onsite systems will be
required.
At the end of the last chamber of the baffled tank reactor one or more
chambers can be fitted out as an anaerobic filter in order to improve further the
treatment efficiency. A filter media allowing widespread contact with the effluent stream
is used which is very efficient in retaining and digesting the left over pollutants. The
problem of encountering clogging is minimized due to the digestion and treatment that
946
Decentralized wastewater treatment systems in large settlements
occurred already in the baffled tank treatment. The process works with fixed bacteria
media. The pre-treatment (settler), first treatment (baffled tank) and second treatment
(anaerobic filter) are constructed below ground level. The different phases can be built
together, or as a separate set-up. The effluent passing out of the anaerobic filters will
have a 90% of the original pollution load removal.
Fig 4 Typical section of an Anaerobic Filter

Since the system works in a closed environment without oxygen supply
(anaerobic conditions) the effluent will continue to smell despite the fact that a major
part of the effluent treatment has occurred already. For this reason an additional
treatment is included in the design lay out, being the third treatment in the form of a
planted gravel filter.
5.
TERTIARY OD ADVANCED TREATMENT
The purpose of tertiary treatment is to provide a final treatment stage to raise

the effluent quality before it is discharged to the receiving environment (sea, river, lake,
ground, etc.). More than one tertiary treatment process may be used at any treatment
plant. If disinfection is practiced, it is always the final process. It is also called "effluent
polishing." Tertiary or advanced wastewater treatment is employed when specific
wastewater constituents which cannot be removed by secondary treatment must be
removed. Individual treatment processes are necessary to remove nitrogen,
phosphorus, additional suspended solids, refractory organics, heavy metals and
dissolved solids. The removal of nitrogen is effected through the biological oxidation of
nitrogen from ammonia to nitrate (nitrification), followed by denitrification, the reduction
of nitrate to nitrogen gas. Nitrogen gas is released to the atmosphere and thus
removed from the water. Phosphorus removal is important as it is a limiting nutrient for
algae growth in many fresh water systems. Phosphorus can be removed biologically in
a process called enhanced biological phosphorus removal. Phosphorus removal can
also be achieved by chemical precipitation, usually with salts of iron (e.g. ferric
chloride), aluminum (e.g. alum), or lime. Chemical phosphorus removal requires
significantly smaller equipment footprint than biological removal, is easier to operate
and is often more reliable than biological phosphorus removal. Another method for
phosphorus removal is to use granular laterite. Once removed, phosphorus, in the form
of a phosphate-rich sludge, may be stored in a land fill or resold for use in fertilizer.
Improvement and upgrading of wastewater treatment units as well as the need to
minimize environmental effects has led to the increased use of tertiary treatment.
947
6.
ADVANTAGES OF DEWATS
There are many advantages of DEWATS, such as: minimum or no

requirement of energy, low-cost and minimum operation and maintenance, treatment
of wastewater flows from 1-1000 m per day, modular design of all components,
tolerant towards inflow fluctuations, reliable and long-lasting construction design, easy
integration to landscaper, involving local communities in the management, etc.
7.
CONCLUSION
The concept of decentralised wastewater management aims at the development of

wastewater systems that are more financially affordable, more socially responsible,
and more environmentally benign than conventional centralised systems. This
approach allows wastewater management to be broken down to the neighbourhood
level and to serve disaggregates of the larger urban areas, resulting in small-scale and
low-cost facilities directly related to reuse of valuable components in the wastewater.
The decentralised concept promises a number of advantages over conventional
practices in the development of new wastewater systems. The flows at any point would
remain small, implying less environmental damage from any mishap. System
construction would also result in less environmental disturbances as the smaller
collection pipes would be installed at shallow depths and could be more flexibly routed.
The system expansion would be afforded by adding new treatment centres rather than
routing ever more flows to existing centres. Financial advantages would result from the
elimination of a great deal of the collection system infrastructure, the use of small
diameter sewers, and the choice of technologies that incur minimal maintenance costs.
LITERATURE
[1]
[2]
[3]
[4]
[5]
948
Bradley, R.B., Daigger, G.T., Rubin, R., Tchobanoglous, G. (2002) Evaluation of

onsite wastewater treatment technologies using sustainable development criteria.
Clean Technologies and Environmental Policy, vol. 4, p. 87-99.
Burkhard, R., Deletic, A., Graig, A. (2000) Techniques for water and wastewater
management: a review of techniques and their integration in planning. Urban
Water, vol. 2, no 3, p. 197-221
Innocent N. (2004) A framework for the decentralised management of
wastewater. Physics and Chemistry of the Earth, Vol 29, issues 15-18, p. 12651273.
Massoud M. A., Tarhini A., Nasr J. A. (2009) Decentralized approaches to
wastewater treatment and management: Applicability in developing countries,
Journal of Environmental Management, Vol 90, p. 652-659
Tchobanoglous, G., Darby, J., Ruppe, L., Leverenz, H., (2004) Decentralized
wastewater management: challenges and opportunities for the twenty-first
century, Water Science and Technology: Water Supply, 4 (1), p. 95-102
DETERMINATION THE PERIODICITY OF MANAGING OF

PREVENTIVE MAINTENANCE OF TECHNICAL SYSTEMS
Ivan B. Krsti1, Dragan I. Milosavljevi2, Boidar V. Krsti 3
Summary: There is given methodology of determination the periodicity of managing
the processes of preventive maintenance of technical systems, within strategy of its
preventive maintenance. This methodology is possibly to apply when there one can
manage the revision which doesnt change intensity failure of analyzed part after every
failure.
Key words: motor vehicle, technical systems, preventive maintenance
1.
INTRODUCTION
Timely managing of preventive maintenance operations and quality of their

conduction represent effectiveness of preventive maintenance of technical systems.
Frequency of failure occurrence in vehicles dictates times of preventive maintenance
operations. Quality of failure identification depends on kind of discovering of defective
elements, applied methods of defects prediction and periods predicted for preventive
maintenance. Efficiency of preventive maintenance depends in essence on skill of
motor vehicles usage.
Some of basic methods of failure prevention are:
- Quality control of devices functionality on the base of outgoing parameters.
This method is based on the fact that change of intake parameters lead to interrupted
functionality, which lead to changes of outgoing parameters. Here it is not
straightforward possible to discover element which caused failure. For identification of
defect is necessary to detect defective element and its maintenance.
- Use of statistical probability of part proper operation until first failure, obtained
on the base of long term operation experience. In this case is possible, with certain
probability, to predict moment of failure and to make steps to prevent it.
- Control of physically chemical changes of structure of considered parts if
prediction devices are available.
Above methods are the most often applied for prevention of failures of electro
mechanical devices and elements of technical systems, for which statistical rules of
failure appearance are established.
1
2
3

University in Kragujevac Faculty of Mechanical Engineering, dmilos@ kg.ac.rs
University in Kragujevac Faculty of Mechanical Engineering, bkrstic@ kg.ac.rs
949
Ivan B. Krsti, Dragan I. Milosavljevi, Boidar V. Krsti
2.
ACUMULATION OF DEFECT TECHNICAL SYSTEMS
Planning of preventive maintenance on time mainly dictate its effectiveness in

technical systems. Recognition of failure regularity of technical systems lead to
schedule of execution of preventive maintenance. Too early maintenance causes
unnecessary and irrational delays of technical systems, but too long period T pr leads
to increase of failures caused with unfixed defects. Thus there is optimal periodicity of
preventive operations T pr opt , which leads to the best results, that is to maximal
effectiveness of system.
Depending on method of defects prevention character of defect accumulation
process in time may be described as follows.
1.
If for some element predicting parameter is known, then probability of
its operation without failure for time t may be estimated according following expression
f
P t
< t dt
(1)
p1
where < p1 t function obtained by calculation in time.

2.
For parts of same type there are failures that can be prevent, as well
as failures that cannot be prevent, and their statistical laws of distribution are known,
leading to probability of work without failure given as follows
Pt
f t dt
(2)
where
f t C1 f p t C2 f n t
(3)
f p t , f n t specific probability of distribution of appearance of unavoidable

failures; C1 , C 2 coefficients that determine contributions of
compose f (t).
fp(t) and
fn(t), that
Lately there are more and more effective objective methods for evaluation of
technical condition of mobile systems, based on implementation of automatic
diagnostic systems. Coefficients C1 and C2 satisfy relation C1+C2 = 1.
In the first case, when prediction parameter is known, it is usual to determine
certain boundaries of part qualities, which may be controlled in working process.
Quality of element gradually decreases, when it approaches to moment after which
failure occurs. Thereby it may be determined preliminary degree of controlled
parameter after which one detects and replace defective parts. This level is called
degree of prognosis. During analyzing of prognosis parameter changes in time
following quantities are employed:
- 0 - mathematical expectation of initial values dissipation,
950
Determination the preriodicitz of managing of preventive maintenance of technical systems
- cr - critical degree of operating ability (functionality) mathematical

expectation of limiting parameter values in which failure occurs,
- pr prediction degree (preventive control),
- T pr - prediction period period between two preventive control,
-
' cr , ' pr
total and preventive reserve of parts reliability, which may be
expressed in the following forms
'D kr
D 0 D kr
'D pr
D pr D kr
(4)
With knowledge of statistical low of change of controlled parameter with time

allows that on the basis of its measuring in the moment t cr t pr T pr , represents
precondition for prevention of failure of technical system.
Forecasting period depends on rates of changes of parameters with time. On
that basis any prediction parameter ( t ) may be evaluated according coefficient of its
change in time:
KD
D 0 D t
t
(5)
That coefficient characterizes process of failure multiplication in time.

Distribution of time of occurrence of unavoidable failure, with acceptable
accuracy, may be approximated according exponential lows [1]:
f N t O exp Ot
(6)
where - intensity of unavoidable failure.
For failures that may be avoided it is possible to assume distribution described
with truncated (partial) normal law [1]. In that case superposition of described laws,
according expression (3), having in minds equations (6) and (2), will lead to expression
f t C1
t Tsr 2
c
exp
C 2 O exp Ot
2V 2
V 2S
(7)
In expressions (3) and (7) coefficient C1, which determines number of failures
which may be prevent, represents coefficient of failure character A(Te).
Having in minds relation
(8)
C1+C2 = 1, it may be obtained that is C2 = 1 - A(Te)
It may be shown that substitution of (7) in (2), using (8), expression (2) may be
transformed as follows
951
t Tsr
)
V
P t A Te
1 A Te e Ot
T
) sr
V
t Tsr Tsr
were )
, ) are tabulated function of probability integral [2].
V V
(9)
Expression (9) characterizes statistic distribution during process of

multiplication of failures without existence of prediction parameter. If, in expression (7),
coefficients C1 and C2 are expressed with coefficient of failure character A(Te), it may
be obtained expression for determination of frequency of failure appearance c (t ) in
case of superposition of exponential and truncated normal law:
t Tsr 2
c
exp
A Te
1 A Te O exp Ot
2V 2
V 2S
. (10)
2
t
t Tsr

c
exp
A Te
dt 1 A Te exp Ot
2V 2
V 2S 0
Oc t
Pre-request for timely undertake of procedure for preventive maintenance

technology of technical systems is knowledge of laws of appearance of failure in time
(rate of change of prediction parameter k in the first case, and statistical distribution
of probability of work without failure in second case).
3.
POSSIBILITIES OF FAILURE DETECTIONS
Efficiency of preventive maintenance works on technical system depends not

only on well timed recognition and quality of maintenance, which in turn depends on
general timing, but on preventive maintenance according to in advance set schedule.
This quality depends on all staff competency, equipment quality and time devoted to it.
In practice is important to determine timing of preventive maintenance, when skill level
of staff and prognostic equipments are known. Time necessary for preventive
maintenance of any technical system consists of time for detection tB, time for repair ty,
and time for subsidiary works, such as tool and accessories preparation, assemblies
etc. Time necessary for detection of defect parts depends on kind of work. These
works are connected to time random processes of defect parts discovery and,
therefore, they have random character. Maintenance time may be calculated as
d
Tpc
pi
(11)
i 1
of work.
952
where tpi time for performance of i-th work, and d number of different kinds
Determination the preriodicitz of managing of preventive maintenance of technical systems
Intervals for necessary works cannot be set in advance. Intervals necessary

for discovering of defect parts usually is much greater than those for amendment, that
is tB >> ty. Remain preventive works such as replacement of defect parts, reassemblages, examinations, cleanings, lubrications, etc, have routine character. Time
necessary for their fulfillment may be calculated as
r
Tpy
(12)
pj
j 1
where r - number of different kinds of work.

Preventive, which include both random and determined works and all working
processes may be set on the basis of amendment intervals. Same as efficiency of any
process, which may be estimated on the basis of number of products in time and time
necessary for production, preventive efficiency may be estimated on the basis of
number of detected and amended failures., Number of detected, checked and
amended elements during preventive works, in general, are not linearly dependent
because of random character of failure detection. Productivity of preventive
maintenance is determined with number of checked elements in time unit. Intensity of
detection of failed elements means number of detected defect elements, that is
prevented failures, in tome unit in comparison with their number in moment t.
Now we are going to establish connection between probability and frequency
of detection of defect parts, and time interval necessary for preventive work Tp. In
random detection of defects frequency n(t), in analogy with failure intensity, may be
determined with rate of number of detected defect parts in unit time, in relation to
undetected defects. Therefore one obtains:
Q t
dn pv
n p Tpr n pv t dt
(13)
where
nP (Tpr) number of preventively fixed defects, accumulated up to
beginning of preventive operation,
nP (Tpr) npr (t) number of defects undetected up to moment t.
After some simple algebra it may be obtained
Q t
Ppv/
1 Ppv
(14)
Integration of expression (14) and easy transformations lead to
Ppv
Tp
1 exp Q t dt .
0
(15)
Therefore in preventive maintenance random process of detection of defected

elements may be approximated with expression (15). In the case of scheduled process
of failure detection probability of failure prevention, mainly, depends on prediction
accuracy and on time used for it. It may be assumed that probability of defect part
overlooking, because of inaccuracy of measuring equipments, mistakes made by
technicians and shortage of time Q(L), independent. In that case probability of failure
detection may be expressed as
953
Ppv
>1 G @ 1 Q L ,
(16)
where
1 LTp , pri 0 d Tp d Tp0 ;

Q L
Tp0 time necessary for preventive
0, priTp ! Tp0 .
1
time norm for preventive operations, D - relative error of
maintenance, L= L
Tp0
real system.
Knowing laws of failure detection process (eqns. 15 and 16), leads to reliable
determination of time for preventive work.
za odreivanje vremena sporovoenja preventivnih radova.
4.
CONCLUSIONS
Effective preventive maintenance means timely determination of necessary

preventive operations, and quality of their performances. Time of preventive operations
is governed with speed of defect appearance. Quality of defect detection depends on
applied prediction methods, on way of defect detection and time planned for preventive
operations.
It is obvious that in both cases efficiency of preventive maintenance depends
on skilled exploitation of equipments. In analysis of preventive measures are often
imposed questions connected with quantitative estimations and comparisons of
different methods of preventive maintenance. Mathematical model, shown here, allows
quantitative estimation of preventive arrangements influence on reliability of technical
systems. In analysis of preventive arrangements often it is not reliable benefit only
criterion, but it should be considered cost of achieving it.
Mathematical model, shown here, may be employed for estimation of quality of
preventive works, as well as for quantitative cost analysis. Usage of obtained
characteristics may lead to estimation of preventive maintenance influence, depending
on cost of its application, on reliability of technical systems.
LITERATURA
[1] Krsti B.: Technical exploitation of motor vehicles and motors (in Serbian),
Mainski fakultet, Kragujevac, 2009., str.488.
[2] Vukadinovi S.:Elementi teorije verovatnoe i matematike statistike, Privredni
pregled, Beograd, 1980. , str 525.
[3] B. Krsti, V. Lazi, R. Nikoli, V. Raievi, I. Krsti, V. Jovanovi: "Optimal strategy
for preventive maintenance of the motor vehicles clutch", Journal of the balkan
tribologikal association", Vol.15, No 4, (2009), 611-619
[4] B. Krsti, V. Lazi, V. Krsti: Some views of future strategies of maintenance of
motor vehicles, Tractors and power machines, Vol.15, No.1, 2010, p.42-47.
954
ENVIRONMENT PROTECTION FOR WELDING AND ALLIED

TECHNIQUES
Mihaela Popescu1, Radu Alexandru Rou2, Carmen Opri3, Ibolyka Bran4
Summary: Welding processes and allied techniques (thermal spraying), by their
nature are generating issues in terms of environment protection. The European
Community has constant preoccupations in approaching these issues. The documents
of the International Institute of Welding (IIS/IIW), and also, the European Welding
Federation (EWF) have also such preoccupations. Synthesizing these concerns, the
mode of intervention in combating undesirable aspects, acting in the context of the
Sustainable Development, and numerous case studies are punctually presented.
Key words: environmental protection, environmental management, welding, thermal
spraying, ergonomics, ventilation
1.
GENERAL ISSUES AT ENVIRONMENTAL PROTECTION
WELD field in the generic sense, by its particularities, is a generator of toxic fumes
gases and toxic vapors. Therefore, besides the technical-technological approach, in
addition for a properly quality assurance, at welding it is necessary to study the
implications of these techniques application [1,10]. Environmental protection field in
welding and allied processes (thermal spraying, brazing, thermal cutting) enter in the
European community involvement and Occupational Safty Health Agencies (OSHA). It
must also mentioned the International Institute of Welding (IIS / IIW), which by the
Commission documents and its working groups have this approach point. Synthesizing
these data, the mode of intervention in combating undesirable issues in the context of
sustainable development and numerous case studies are presented [3].
2.
CONTENT AREA ENVIRONMENT PROTECTION
The main directions of environmental protection field approach to environmental

protection in welding and related techniques are summarized in Figure 1.
Environmental management approach: quality-environment-health-security
integrated system welding companies, the advantages of using integrated
1
2
3
4
Prof.assoc.Mihaela Popescu, Politehnica University of Timioara, e-mail: hela.popescu@yahoo.com

Phd. Eng. Radu Alexandru Rou, Politehnica University of Timioara, e-mail: rrosu@isim.ro
Assist. Phd.Carmen Opri, Politehnica University of Timioara, e-mail: c.opris@gmail.com
Phd.Ibolika Bran, Politehnica University of Timioara, e-mail: ibi.bran@gmail.com
955
Mihaela Popescu, Radu Alexandru Rosu, Carmen Opris, Ibolyka Bran
management systems, companies certification issues. Environmental management

system is in accordance with ISO 14001.
Welding environment protection
Welding environmental management issues
Waste resulted in welding processes
Product life cycle
Welding Environment and Health
Welding Ventilation systems
Principles of sustainable development
Ergonomics in Welding
Fig. 1 Directions of environmental protection
Waste must be considered, taking into account the principles of management,
recyclable materials after welding, which can be customized on the distinct processes,
respectively on basic materials group Product life-cycle, engineering life cycle
process design and execution stages, ecological exploitation are treated as distinct
application of the principles of Integrated Engineering Environmental and
occupational health: definitions, general issues, welding processes impact (thermal
spraying, brazing, thermal cutting) on the operators health in accordance with OHSAS
18001 regarding the management system, safety and health at work. Factors
pollutants: emissions into the atmosphere, wastewater, noise and vibration and
hazardous chemical substances, as well as implications for operators are very
important to take into account Ventilation systems in welding describe the
microclimate conditions, local and general ventilation, sometimes forced, imposed,
possible variants applied in concrete cases Sustainable development principles
applicable to the production of welded structures in the context of environmental
conservation for future generations are a constant concern for environmental
protection Ergonomics in welding has made customization in approach on methods
products and examples, operators welding position depending on the size and
application, furniture, shelves, work tables and seats, with implications for the combat
musculoskeletal diseases, welding torches with design and function tailored to help
operators, ergonomic welding helmets, face masks, clothing, ergonomic gloves.
3
APPROACH DOMAIN ENVIRONMENTAL PROTECTION TO WELDING AND

RELATED TECHNIQUES
3.1 Case study 1-Thermal Spraying

The approach of thermal spraying from different point of view, such as
technology, equipment or its applications, is completed by the data referring to
environmental protection when using these processes. Thus, noxes emissions (table 1)
and radiations, the obligation to ventilate the working area, the requirement to train and
perfect the thermal spraying operators, according to the EWF regulations (European
956
Environment Protection for Welding and Allied Techniques
specialist in thermal spraying, EWF-459r1-06, European thermal sprayer, EWF-507r106, European practitioner for thermal spraying, EWF-592-01), controlled noise level
(table 2) and ergonomic protection recommendations (figure 2) are only a few of the
aspects to be taken into account in this case [6,7,8].
Table 1 Noxes emissions, depending on types of thermal processes and applications
Process
Sprayed material
Metallic materials
Zn, Al
Mo
LVPF FeCr Alloys
wire
FeCrFNi Alloys
NiAl
CuO, Cu2O, Al2O3
FeCr Alloys
LVOF
NiCrBSi
powders
CoCrW
Fe3O4
NiO, WO, SiO2
HVOF CoCrW
oxidic ceramics (TiO2,
Cr2O3, Al2O3)
NiAl
APS
oxidic ceramics (TiO2,
Cr2O3, Al2O3)
CuSn, CuZn
Arc
Zn, ZnAl
FeCr
Critical components
Total fumes, NO2
ZnO2, total fumes, NO2
Total fumes, NO2
Total fumes, Cr2O3, NO2
Total fumes, Ni oxides, NO2
Ni oxides, NO2
Cu Oxides, ZnO, NO2
Cr2O3, NO2
Total fumes, Ni oxides, NO2
Total fumes, Cr2O3
Cr2O3, total fumes
NiO, total fumes, NO2
Total fumes, Cr2O3, NO2
Total fumes, NO2
Ni oxides, UV, O3, noise
Total fumes, UV, O3, noise
Cu and Zn oxides
Total fumes, UV, O3, noise
ZnO, total fumes, UV, O3
Cr2O3, total fumes, UV, O3
Applications
reconstruction
corrosion resistant layers
antifriction layers Mo
wear resistant layers
adherent NiAl layers
bearings, bronze, brass
wear resistant layers,
wear and corrosion
wear and corrosion layers
wear resistant layers, thermal
barriers
adherent NiAl layers
wear resistant layers, thermal
barriers
bearings, bronze, brass
corrosion resistant layers
Stipulations regarding ergonomics (figure 2), ergonomic considerations at

thermal spraying [8].
Table 2 Controlled noise level [6]
Thermal spraying process
Detonation gun
HVOF liquid fuel
HVOF gaseous fuel
FVAF
Wire Flame Spray
Powder Flame Spray
Rod Flame Spray Gun
Electric Arc
Air Plasma Spray (APS)
Decibel, dBA
145
133
125-135
133
118-122
90-125
125
109-119
110-125
Fig. 2. Ergonomics principles
957
3.2 Case study 2-Flux cored arc welding (FCAW)

The usage of flux cored arc welding is justified by the high quality of the welds
and also by the productivity thus obtained. The main disadvantage is that it requires
local forced ventilation, constant measuring of noxes in situ so that it is possible to
intervene in time and to verify their accordance with the PEL (Permissive Exposure
Limits) allowed by OSHA (Occupational Safety and Health Administration). The
composition of the fumes at FCAW is presented in table 3.
Table 3. Fume composition related to welding process and consumables - flux cored
arc welding
Process
Automatic Semi- automatic (flux cored)

Mild-low
Mild-low alloy
Stainless
alloy steel
Filler Metal
steel (GasHardfacing steel (Gas(SelfShielded)
Shielded)
Shielded)
CO2
CO2,
98Ar/2O2
Shielding Gases
90Ar/10CO2
75Ar/25CO2
75Ar/25CO2
Fe, Mn,
Fe, Mn, Si, F, Fe, Mn, Si, Fe, Mn, Si,
Components in
Ba, Si, F,
Na, K, Cr
F, Cr
F, K, Cr
Consumable
Ca
Fe
7.0-42.2
28.0-62.0
4.0-37.0
12.0-19.0
Mn
0.5-7.0
5.0-18.0
1.0-38.0
5.0-8.0
Si
0.1-6.0
1 .0-5.0
<.01-2.0
5.0-8.0
Range
Na
0.2-22.0
0.3-13.0
<.01-6.0
5.0-12.0
of Key
F
4.0-24.0
0.6-17.0
1.0-18.0
3.0-6.0
Components in
K
<.01-15.0
0.1-9.0
0.2-6.0
1.0-3.0
Welding Mg
0.9-28.0
<01-4.0
0.6-25.0
<0.01-4.0
Fume
Li
0.4-8.0
0.1-1.0
Fume
Ni
0.02-2.0
1.0-3.0
Analysis
CrVI
Ba(sol)
CrVI
Wt.(%) CrVI
1.0-24.0
0.4-1.0
0.3-2.0
Cr (total)
Cr
Ba (total) Cr (total) 0.4- Cr (total)
2.0
0.5-11.0
0.8-13.0
(total) 4.0-42.0
Stainless
steel (SelfShielded)
Fe, Mn, Si,
F, K, Cr
13.0
6.0
4.0
5.0
21.0
18.0
<0.01-3.0
3.0
CrVI 4.0
Cr (total)
9.0
The solutions to counteract unwanted effects are: to elaborate flux cored arc
wires that ensure low noxes emissions, to use advanced systems for local forced
ventilation or to use welding guns that ensure immediate absorption of the noxes that
are emitted.
New generation flux cored arc wire have been produced lately, such as low
fume tubular wire, Stardual 208 HP, T 42 3 M M 1 H5 according to EN ISO
17632-A, that ensure low levels of fumes at welding [10]. This new low
fume copper coated metal cored wire generates less fume than similar
standard products. It enables fume emission rate reduction of up to 40%
(standard shielding gas M 21) [9].
958
Environment Protection for Welding and Allied Techniques
Other advantages include:

- it is a slag less metal powder tubular cored wire with outstanding welding
properties in the short-arc and spray-arc ranges.
- almost splatter free when welding in the spray-arc range.
- provides good restriking, even with a cold the wire tip, thus being suitable
for a robot applications.
Characteristics that recommend this type of wire for FCAW are: high deposition
rate and welding speed, good wall fusion, finely rippled welds, without undercutting into
the base metal, not even on contaminated or corroded metal surfaces. Main
applications are for this type of wire are: shipbuilding; rolling stock construction;
structural steelwork, boiler-works; bridge cranes, cranes, earth moving machines.
A classification of the welding guns for noxes emissions at FCAW,
according to the IIS/IIW (International Institute of Welding) Documents
prescriptions is presented in figure 3.
Fig. 3 Welding guns with immediate noxes absorption systems (IIS/IIW DOC)[10]
Variants of efficient welding guns with immediate absorption of noxes offered
by producers with tradition in the field are presented in figure 4.
K289- 500 Ampers at 60% DA, wire diameter 3.0-5.6 mm

Fig. 4. Constructive variants for welding guns with noxes absorption
959
CONCLUSIONS
There are described the environmental protection issues at welding and allied
processes for companies which are producing welded structures in the context that in
the welding domain was done until a decade ago, only by technical-technological point
of view. The issue of environmental protection is treated, referring at the claims and
obligations imposed by the European Community by specialized bodies for health care
operators. The case studies, punctually presented for thermal spraying case and the
use of tubular wires confirm those presented.
Acknowledgement
This work was partially supported by the strategic grant POSDRU/89/1.5/S/57649,
Project ID 57649 (PERFORM-ERA), co-financed by the European Social Fund
Investing in People, within the Sectoral Operational Programme Human Resources
Development 2007-2013.
REFERENCES
[1]
Popescu,M.(2011).Environmental protection and legislation in the welding

processes, Masters Course, Productive Welding Processes in Protective Gas,
UPT
[2] Popescu, M., erban, V.A, Rdu, A., Nicoar, M., Magda, A.V,Locovei, C.,
Demian, C. (2008). Le concept du developpement durable applique aux
processus speciaux du soudage, Conference Cofret 08, Nantes, France, 1113 juin,ISBN 2.6905267.61.5, p. 314-318
[3] Popescu, M., Mocua , E.G., Vartolomei , M.(2009). WELDING: Integrated
quality,environment and safety management system,Annals of DAAAM &
Proceedings of the 20th International DAAAM Symposium, Volume 20, No. 1,
ISSN 1726-9679 ISBN 978-3-901509-70-4
[4] Popescu,M., Marta, C.,Mocua, G.E.,Opri, C. (2010). Mechanical Ventilation in
Welding, Proceedings of the Annual Session of Scientific Papers "IMT Oradea",
ISBN 978-606-10-10-0128-6, p. 3.83-3.86
[5] Popescu,M., Marta, C,Mocua, G.E. (2010). Ventilation systems,used in welding
processes, Proceedings of the Annual Session of Scientific Papers "IMT Oradea"
ISBN 2010 978-606-10-10-0128-6,p. 3.142-2.147
[6] Popescu,M.,Opri,C.,Bran,I. Rou,R. (2011) Thermal spraying according to UEs
requirements for environment protection, Modtech, p.158-160
[7] xxx:www.praxair.com/2010. 02.1
[8] xxx;www.Society(TSS).com/2010. 09.07
[9] xxx:.www. lincolnelectric.com/safety,2010. 12.10
[10] xxx: The environmental management of fabrication by welding(2002),Doc IIS/IIW1593-02,Welding in the World, Vol.46,No.7/8,p.22-31,ISSN 0043-2288
960
FTA AND FMEA IN PREDICTING INCIDENTAL CONDITIONS IN

CABLE CARS AND SKI LIFTS
Milomir upovi1, Desimir Jovanovi2, Bogdan Nedi3,
Summary: Incidental coditions are unwonted events in any socio technical systems.
If there are technical systems for mass human transpor, such as chair lifts and ski lifts
in ski resorts, the importance of predicitons of such events is even greater. The role of
the security systems on chair lifts and ski lifts is to detect and prevent any ilegall
situation on the installations that can cause a possible incident. Inherent properties of
mounted security systems components, detremine the quality of carrying of the task.
Based on previous author's researches about significance of chear lifts and ski lifts
subsystems, from the point of ocurrence on incidental conditions, the paper analyzes
the security subsystems.
Key words: failures analysis and their consequences, fault tree, chair lifts and ski lifts,
security subsystem, rope position detector set , the incidental condition
1.
INTRODUCTION
The nightmare of every Ski Area manager is incidents on the technical means.
The most unpleasant, of all effects could be incidents on the means of en masse
transport of skiers: chair lifts and ski lifts. Large number of users points to possibility of
large consequence such as threat to users health. Incidental conditions follow all
forms of social and technical systems. The human factor is always identified as the
cause of incident conditions. Is it always? Despite the great experience and great
attention of safety issues on chair lifts and ski lifts, incidents are unavoidable during
operational. One of the worst accidents occurred to famous manufacturer of ski lifts,
the French company Poma. During construction installation of cable car in 1989, the
cabin with 11 workers crushed down. Tragic result was 11 dead people.
Researching the causes of incidents in the ski resort of Kopaonik, in the paper
[1], the author has identified: the causes of incident states and subsystems were they
are happened, Table 1: Data were obtained from the complex research of the 16 ski
lifts and chair lifts installations for a period of five ski seasons. The motivation of this
Prof. D.Sc. Eng, Milomir upovi, Raka, State University of Novi Pazar, (e-mail:mimocupovic@gmail.com)
M.Sc. Eng, Dejan Jovanovi, Kragujevac, Zastav arms, (e-mail: j.desimir@gmail.com )
Prof. D.Sc. Eng., Bogdan Nedi, Kragujevac, University of Kragujevac,Mechanical Engineering Faculty, (email: nedic@kg.ac.rs)
2
3
961
Milomir upovi, Desimir Jovanovi, Bogdan Nedi
paper is that the incident states appears as a consequence of own characteristics of

the built-in features (T character failures).
Table 1
CAUSE
Tinherent properties
H0 no
victims
D16/14
D15
2.
PK - errors in the use
MT- meteorological
of
conditions
CONSEQUENCES
H1- with
potential
H0
H1
H0
H1
victims
SUB-SYSTEM IN THE STATE OF INCIDENTAL
D14
D14
D15
D12
D15/14
D10
D15/14
D15
D15/14
D15/14
D15
D15
PO/PR- error staff
H0
H1
D04
PLACE AND ROLE OF SAFETY SUBSYSTEMS
Fig. 1. Principled scheme of safety subsystems

The concept of high security, imposed the existence of so-called safety
subsystems for the chair lifts and ski lifts. The role of the safety subsystem is to identify
and prevent any irregular state of the object. Conceptual, the safety subsystem is the
acquisition of certain parameters of the system, compares them with predicted values
in the time of the illegitimate state gives the command to the executive to stop work.
Figure 1, presented principled scheme of security subsystems. Figuratively speaking,
the safety subsystem is the brain of a chair lifts or ski lifts. Although the incidental
conditions occur because of characteristic of the mounted parts, we keep thinking
about the severity of this situation in part of the safety subsystem.
962
FTA and FMEA in predicting incidental conditions in cable cars and ski lifts
3.
FMEA (FAILURE MODES AND EFFECTS ANALYSIS)
FMEA (Failure Modes and Effects Analysis) is one of the methods for analysis
the way and consequences of technical systems failure. It also represents the first step
in understanding how reliable system operates. The aim of the analysis is to discover
the causes of component parts failure, and also to reduce the consequence of system
failures. Very practical format for FMEA usage is a matrix form [2], which allows
determining the cause of elements failure. Besides it is used to analyze the
consequences of failure in the objective function of the system. In order to implement
FMEA, observed system is separate on subsystems, then on its constituent parts, to
the lowest level of separation. Mentioned method can be applied at each level of
system separation. For the lowest level of separation reliability data must be provided.
This type of data is provided from exploitation research or downloading from the
reference literature for the same or most similar systems and elements. Mostly,
elements reliability data in FMEA security subsystem are obtained from a complex
temporal system condition as Mean Time Between Failures (MTBF) or Failure
Intensity, (t). For elements were we have no data from temporal system condition [3]
and [4] were used.
In the conducted analysis, the ultimate consequences of failure were ranked as:
x
E1 - loss of subsystem functioning control possibility
x
E2 - instantaneous subsystem function cut off
x
E3 - instantaneous subsystem function cut off, with subsequent harmful
effects
x
E4 - failure does not cause a delay, but indirectly contributes to the harmful
effects.
It is very important to determine the elements were is possible appearance of socalled. undefined state, with E1 consequence. There are such elements and events:
Fig. 2. Matrix form of the FMEA safety subsystem

In the case of the safety ramp and "stop" button it is result of phenomenon
called froze elements. Such situations are covered by exploitation procedures, i.e. the
method of treatment and control, before putting the installation into operation. It is very
important to conduct research on rope position detector set.
963
4.
ROPE POSITION DETECTOR SET
Rope position detector set on Ffigure 3, represents the essence of the

existence of the security subsystem. Mentioned figure shows one of several types of
applied rope position detector. It is located close to the balancer, allows the detection
of correct behavior of the rope on the sheave train. In fact if the rope fall off the sheave
train, rope position detector set makes refraction, thus causing breakage of the break
fork, as a result there is interruption of signals from the line and stopping of the chair lift
or ski - lift.
Fig. 3 Rope position detector set

By FMEA method the rope position detector set was detailed researched,
Table 2. The analysis showed that the failures of rope position detector set have
almost exclusively T character.
Table 2
Fig. 4 FMEA of the Rope Position Detector set

5.
FTA OF THE ROPE POSITION DETECTOR SET
Method of failure analysis by the tree of failure was used for the first time in,
"Bell phones" laboratories during 1961 and 1962. year. According Fusell in [2], the
value of the tree of failure contains the following:
the analysis is oriented to finding failure, provides a deep insight into the
working process of the system;
964
FTA and FMEA in predicting incidental conditions in cable cars and ski lifts
method allows the specialists to concentrate on specific failures and

individually study and cleared them up.
Analysis of the rope position detector set by FTA method, was carried out using
the OpenFTA software package. The software is the GNU General Public License, with
very good User Manual [5]. Beside Minimal Cut Sets, also allows the calculation of
Numerical Probability of peak events and Monte Carlo simulation (statistical).
Fig. 5 The failure tree for rope position detector set

The studied event is presented by coherent failure tree, with multiple events,
Figure 5. The tree is constructed of "xor" and "or" logical symbols, which tells about the
events that are mutually excludive (the appearance of any type of mistakes leads rope
position detector into a state of failure).
Existence of so-called undeveloped events has been noticed. Reasons for
event undeveloping are related exclusively to the availability of information in this case.
Failures that occur on the element No. 6, Figure 4 in qualitative analysis showed the
highest impact on the value of the probability of unwanted events (failure 1: 40.25%,
failure 2: 6.70% and failure 3: 10.03%, which is the sum of 56.98%). Large value of the
importance of these events and the fact that it is undeveloped events focused the
attention of maintenance engineers.
Break fork, pos. 6, Figure 4 is the simple element, made of general structural
steel. Surface protection is performed by electrolytic zinc plating and the body of the
break fork is weakened with two holes in order to make process of cracking easier.
Temperature range in which the material must retain its properties is -40 to +10C.
Compression towers of chair lifts and ski lifts have increased vibrations that are
transmitted to the rope position detector carrier and on the break fork itself.
In the case of excessively brittle materials (impacted hardness of steel,
decreases with reduction of temperature, and also the occurrence of hydrogen
965
brittleness is possible); break fork will crack due to vibrations. Any impact could cause
a loss of contact and the lift will stop. The consequence is a large, non-operating
installation. Detection of such failure is extremely difficult.
Second, by its consequences much more serious situation can occur if
material of the break fork is too tough. In this case, when the rope fell off the sheave
train, break fork will not break (depending on the intensity of the impact, or on the size
of the resultant force on that particular tower). Chair lift will continue to operate and the
consequences will be disastrous.
6.
CONCLUSION
Using the matrix form of FMEA, on the safety subsystem of chair lifts and ski
lifts, phenomena of loss of possibility subsystems functioning control are noticed. This
kind of failure can lead to unforeseeable consequences, related to the health of users.
Further analysis is directed on elements whose internal properties can cause specified
condition. OpenFTA software capability are used to identify a break fork as an element
that can cause loss of control of the security subsystem functioning.
Well developed ski centers, strategically linked to famous producers of chair
lifts and ski lifts, probably will not encounter such problems. Spare parts are purchased
from these manufacturers. Small, local ski resorts are tempted, they have problems
with finances and way of purchase of spare parts. They often reach for production in
small local factories. Although it seems trivial, break fork as the analysis showed,
deserves special attention, so it is best not to take a risk, spares followed by certificate
should be mounted.
LITERATURE
[1]
[2]
[3]
[4]
[5]
966
Milomir upovi: Izbor sistema odravanja ski iara u realnim uslovima okoline,
magistarska teza, Kragujevac 1998.g.
Fault Tree Handbook with Aerospace Applications - NASA Office of Safety and
Mission Assurance NASA Headquarters Washington, DC 20546, August, 2002
OREDA, Offshore reliability Data Handbook, 4th Edition
RADC NONELECTRONIC RELIABILITY NOTEBOOK ROME AIR
DEVELOPMENT CENTER
OpenFTA, Version 1.0,User Manual
MAINTENANCE AND LIFECYCLE OF THE EXCAVATION UNITS

Milorad Panteli1, Sran Bonjak 2
Summary: Starting with significance of maintenance function and necessity to execute
specific maintenance tasks, in this paper is pointed that early integration of all phases
of a technical system lifecycle is necessary. Emphasized is priceless value of
ascertainment that it is necessary to have an overall perception of technical system
lifecycle, from the users initial demand, over general design, development, fabrication,
delivery, exploitation and maintenance, all the way to withdrawal from operations,
depreciation and recycling. Next is presented design for maintenance approach, which
originated together with lifecycle engineering, whose introduction changed entire
philosophy of usage and maintenance of technical systems. New paradigm of technical
systems usage and maintenance is: now days, efficient are considered not those
systems that effectively fulfill their goal function, but those with lifecycle ending safely.
Key words: excavation units, maintenance, lifecycle
1. INTRODUCTION
Bucket wheel excavators (BWE) and spreaders, Fig. 1, are the core of the EBS
(Excavator-Band-Spreader) systems. Because of structure complexity and specific
task of its maintenance, BWEs and spreaders, as a fundamental parts of the open pits
technological equipment, are usually named as excavation units [1]. Common
denominators to all failure of mentioned high-capacity machines are very high financial
losses and serious risks to the workers safety and life [2,3]. Financial losses caused
by production delays due to the principal machine failure in a surface mining system
often significantly exceed the financial losses caused by direct material damage. The
size of the negative economic effects caused by failures is remarkably revealed in the
fact that the total cost of failure in USA and Europe is of order of 4% of GNP [4].
2. MAINTENANCE FUNCTION
Literature gives maintenance definitions that differ from each other as well in
including activities that fall into maintenance domain as in treating maintenance
organization within company (subsystem integral part within business system,
business system function, service industry of approved and narrowly specialized
company). Although maintenance of technical systems could be defined in different
1
2
assistant professor Milorad Panteli, Technical Faculty aak, milorad.pantelic@kolubarametal.com

professor Sran Bonjak, Belgrade, Faculty of Mechanical Engineering, sbosnjak@mas.bg.ac.rs
967
Maintenance and lifecycle of the excavation units
ways, the most commonly this term implicates practicing of all necessary measures
for one machine, facility or entire factory to be functional in proper manner, developing
results within reference boundaries i.e. with required performances and quality, without
dismissals and along with regulated environment protection and under assumption that
all conditions are well secured anent by necessary logistic support [5].
Fig. 1 BWE SRs 1200 and spreader ARs 1800/(14+33+60)x20 in the open pit mine
Kolubara (Field B) Serbia
3. MAINTENANCE TASK AND ENGINEERING OF LIFE-CYCLE
Need for performing specific tasks of technical system i.e. functionality that
user must maintain, represents entrance into maintenance task, and successful task
performing represents way out from maintenance task [6], Fig. 2.
Fig. 2 Maintenance task [1]

968
Necessity for very early integration (user request specification, designing) of all
life-cycle stages of technical system was indicated by following related disciplines: lifecycle engineering [7], competitive engineering [8] and industrial engineering [9]. Of
priceless significance is acknowledgement that spotted technical system was observed
through its entire life-cycle, from user request to usage retirement, disusing and
recycling. However, maintenance is often observed isolated, more or less independent
of other stages in technical systems life-cycle. It specially concerns developing
activities which are not corresponding enough to maintenance requirements. Prevailing
opinion is that maintenance has no connections with development (designing,
construction), i.e. that designer does not have to be concerned about maintenance and
that it is system users care, Fig. 3. Consequences of this attitude are much uttered
problems in maintenance of technical systems which considerable decreases its total
effectiveness. For example, because of very unsuitable designing solution, bearing
change on BW shaft of excavator SchRs 350 requires minimum of 7 days [10].
Fig. 3 Designing and exploitation of excavator units as isolated processes [1]

Exploitation with maintenance and usage retirement, as inevitable stages of
life-cycle, are significantly important for excavator units, especially in case of multi
decennium exploitation. Fig. 4. Thereat it especially stands out necessity and delicacy
of diagnostic and functionality as well as justification of revitalization. Rotary
excavators are designed to operate nominally 25 to 30 years. Revitalization and
modernization of rotary excavators on open pit mines RB Kolubara and TE-KO
Kostolac (Serbia), along with Kolubara Metal as holder of these projects, enable
prolonging their exploitation for another 25 years.
969
Fig. 4 Age and capacity structure of BWEs in open pits Kolubara and Kostolac
(Serbia): 1-SchRs 300; 2- SchRs 350; 3,5,8,10,11-SRs 1200; 4-SRs 1201; 6,7,9,12,20SRs 470; 13,17,25,29,30-SchRs 630; 14-SchRs 900;15-SH 630; 16-SRs 1301;
18,27,28-SRs 2000; 19-SRs 400; 21-SchRs C700s; 22,23-SRs 1300; 24-SchRs 1760;
26-SchRs 800; 31-SchRs 1600 [9]
4. DESIGNING FOR MAINTENANCE
With occurance of competitive engineering and life-cycle engineering in few
last years, entire philosophy has been changed within domain of utilization and
maintenance of technical systems. Now not only systems which effectively perform
function of cause are considered to be efficient but also systems which life-cycle was
accomplished safely. Safety issues of system, operator and environment possible is,
among other, being solved through fulfillment of international quality standards,
environment, security, safety at work [5].
When its about excavator units, life-cycle engineering philosophy assumes
integrated approach, Fig. 5, meaning, cooperative work of designers, manufacturers,
users and preservers of excavator units [1]. Thereat it is good to have in mind that
users of excavator units are not only open pit mines but also organisations that are
involved in maintaining. Success criterions of cooperative work are exploitation with
fewer dismissals, halts and operating without breakdowns. At the same time that is
also criterion of excavator units buyers gain. Therefore, such modern, contemporary
approach enables achievement of following basic users` requirements:
Great unit capacity;
Small unit mass;
Exploitation of 24 hours per day (Including holidays);
Approachment of exploitation performances to designed (theoretical)
performances;
Degree of usage within limits of 0.8 0.9;
More plentiful usage of existing recourses for exploitation support
(concomitant equipment, maintenance, trained manpower, spare parts,
technical documentation);
Simpler maintenance and realization of overhauling (investment)
maintenance for 20 days, instead of usual 30 days.
970
Fig. 5 Design and exploitation integrated with maintenance of excavator units [1]
Significance of introducing requests related to maintenance into restrictions
during designing is persuasively testified as well as participation of maintenance within
life-cycle of BWE, Fig. 6. Namely, when it is about BWE it is necessary:
2 years for conceptual and preliminary designing;
1 year for detailed designing;
2 years for manufacturing, delivery and launching;
30 years for exploitation until revitalization;
1 year for revitalization;
20 to 25 years for exploitation after revitalization;
0.5 to 1 year for usage retirement.
Fig. 6 Duration of life-cycle stages of BWE

971
5. CONCLUSION
In general, life-cycle engineering concept, which requires early usage process
consideration as well as maintenance of excavator units, shapes needs, requirements
and priorities of user in stage of designing which results with higher level of project
quality and operative efficiency of excavator units. Just therefore it is necessary to use
new approach to design named serviceable designing [5]. Future lies in option that
resources from next stages of excavator units` life-cycle work not when they are
necessary (during usage and maintenance) but within designing stage.
LITERATURE
[1]
Panteli, M. (2009). Maintenance conception improvement via operational safety

management of excavator units at coal mines. PhD dissertation, University of
Kragujevac, Technical Faculty in aak; aak. (In Serbian).
[2] Wintle, J.B., Pargeter, R.J. (2005). Technical failure investigation of welded
structures (or how to get the most out of failures), Engineering Failure Analysis,
vol. 12, no. 6, p. 102737.
[3] Dreyer, E. (1995). Cost-effective prevention of equipment failure in the mining
industry. International Journal of Pressure Vessels and Piping, vol. 61, no. 2 - 3,
p. 329-347.
[4] de Castro, P.M.S.T., Fernandes, A. A. (2004). Methodologies for failure analysis:
a critical survey. Material and Design, vol. 25, no. 2, p. 117-123.
[5] Papi, LJ., Milovanovi, Z. N. (2007). Odravanje i pouzdanost tehnikih sistema.
Istraivaki centar za upravljanje kvalitetom i pouzdanou, DQM, Prijedor.
[6] Kneevi, J. (1977). Systems Maintainability, Analysis, Engineering and
Management. Chapman and Hall, London.
[7] Carter, D.E., Stilwell Baker, B. (1992). Concurrent Engineering, The Product
Development Environment for the 1990s. Addison-Wesley Publishing Company,
Reading.
[8] Blanchard, B.S., Fabrucky, W.J (1998). Systems Engineering and Analysis.
Prentice Hall, Upper Saddle River, New Jersey.
[9] Bonjak, S., Arsi, M., Zrni, N., Rakin, M., Panteli M. (2011). Bucket wheel
excavator: integrity assessment of the bucket wheel boom tie rod welded joint,
Engineering Failure Analysis, vol. 18, no. 1, p. 212-222.
[10] Bonjak, S., Petkovi, Z., Gnjatovi, N., orevi, M. (2009). Redesign of the
Bucket Wheel Excavating Device. MHCL09 Conference Proceedings, p. 123-128.
Acknowledgements
972
MODELING MAINTENANCE ACTIVITIES ON A WELDED

CONSTRUCTION USING IDEF0 METHODOLOGY
Dusan Jovanic1, Drago Soldat2
Summary: The introduction of information technology and computers in the design
process enables easier, faster and more accurate design of welding technology, as
well as higher quality, repeatability and traceability of processes. This paper presents
the modeling of welded construction maintenance activities by IDEF0 methodology.
Modeling maintenance activities on a welded construction using IDEF0 methodology
(Integration DEFinition), the software implementation of which is the BPWin (Business
Process windows) CASE (Computer Aided Software Engineering) tool, allows
functional decomposition and design at all levels of the system that consists of
workers, machines, materials, computers and information.
Key words: welding, information technology, maintenance, optimization process.
1. INTRODUCTION
Welded constructions control need to be assured for the purpose of their
efficiency and avoiding serious problems occurence in their manufacturing and usage.
The control should involve design, material selection, manufacturing and
accompanying examinations.
For a system which comprises people, machines, material, computers and
information standard IDEF0 provides functional decomposition and design at all levels
along with creation of documentation with revision of ISO 9000: 2000 standard, as well
as reengineering of business processes, better communication between members of
the project team, users and managers. At the same time, this standard enables
functional work performance in management, especially bigger and more complex
projects while providing elements needed for information modeling at the same time.
2. MODELING OF ACTIVITIES OF WELDING TECHNOLOGY DESIGN BPWIN
METHODOLOGY
Development of information system should be used to define as objective
1
Mr Dusan Jovanic, Zrenjanin, Serbia, Technical College of Applied Sciences in Zrenjanin,

(jovanickosta@gmail.com)
2
Prof. Dr Drago Soldat, Zrenjanin, Serbia, Technical College of Applied Sciences in Zrenjanin,
(biznisinvest@open.telekom.rs)
973
Dusan Jovanic, Drago Soldat
image as possible of the past, present and future state of a real system, as well as a
basis for further development and application of information technology. Development
of information system is provided through the following stages: functional modeling,
information modeling, applicative modeling and implementation [1].
Standard IDEF0, i.e. CASE Bpwin tool is used to carry out functional modeling
(methodology from top to bottom) and it is necessary to carry out the following
activities: functional decomposition, definition of users requirements and technical
prerequisites.
Implementation of the research will involve IDEF0 standard application for
modeling welded construction maintenance activity.
2.1 Functional modeling
Functional modeling activity is carried out by CASE Bpwin tool, while IDEF0
standard is used for the process of welding technology design modeling. There follow
detailed functional modeling activities.
2.2 Functional decomposition
Within this activity we will define system limits and activities tree for welding
technology design and perform activity tree verification.
2.3 System limits defining
Process modeling i.e. welding technology designing is adjusted with the ISO
9000 standard requirements for quality assurance which prescribe composition of
accurate documentation, in this case, a welding technology specification [1, 2].
The process output is a documented welding technology specification which
will provide repeatability and traceability of the process, i.e. the same quality for each
subsequent product. It could be said that documentation improves routine and assures
quality.
Thus, a database i.e. welding technology specification base available in every
moment will be created so that its browsing provides knowledge i.e. information about
previous welding procedures.
Context diagram shown in Fig. 1. defines a model framework of welding
technology design. Input values are customer requirements defined in a contract and
by exploitation conditions, required reliability, welding quality of the welded
construction and quality of the base material. When the welded construction is
designed by a manufacturer, another input value is added the customer satisfaction.
The output values are customer response and the welded product. Standards, laws
and regulations present control in welding technology design process, while
mechanisms in this case are technological performances of the manufacturer shown
through equipment and people.
Decomposition diagram shown in Fig. 2 defines levels and details of welding
technology design. The welding technology design activity at the first level is
decomposed into five activities: contract and design analysis, welding process
determination, welding technology definition, quality control and welded construction
maintenance. [3].
974
Modeling Maintenance Activities On a Welded Construction Using Idefo Methodology
USED AT:
06-05-2008
AUTHOR: Dusan Jovanic
DATE: 10/20/2003
WORKING
PROJECT: Doktorska disertacija
REV:
DRAFT
4/1/2011
READER
DATE CONTEXT:
TOP
RECOMMENDED
PUBLICATION
NOTES: 1 2 3 4 5 6 7 8 9 10
Laws and
Regulations
Standard
Specifications of
certificates and
verification
Customer
requirements-contract
Customer
response
Customer
Satisfaction
Quality
records
Exploitation
conditions
Risk assessment and

consequences for human
health and environment
MODELLING WELDING PROCESS
Required reliability
quality of base materials
Subcontracting
quality of welded
structures
Welded product
0
equipment, people, knowledge bases
NODE:
TITLE:
NUMBER:
28-06-2006
A-0
Fig. 1 Context diagram of a welding technology design model

USED AT:
DATE: 10/20/2003
WORKING
REV:
DRAFT
4/1/2011
READER
DATE CONTEXT:
RECOMMENDED
PUBLICATION
NOTES: 1 2 3 4 5 6 7 8 9 10
Laws and
Regulations
Specifications of
certificates and
verification
Standard
Customer
A-0
ANALYSIS
CONTRACT OR
PROJECT
Customer
response
Report of a contract or
project
1
Customer
Satisfaction
Subcontracting
DETERMINATION
WELDING PROCESS
List of
equipment
Welding
procedure
specification-WPS
2
Exploitation
conditions
Reports of welding
process, base and
fillet materials
Required
reliability
DEFINITION
WELDING
TECHNOLOGY
3
Plan proizvodnje
Quality
records
Report of
welding
parameters
quality of
welded
structures
QULITY
CONTROL
Risk assessment
and consequences
for human health
and environment
4
quality of base
materials
Reports of procedure
qualification records
MAINTENANCE
WELDED
CONSTRUCTION
5
Welded product

NODE:
TITLE:
NUMBER:
A0
Fig. 2 Decomposion diagram of welding technology design

2.4 Activity tree definition
Fig. 3 shows welding technology design activity tree. Contract or project
analysis activity is decomposed into the following activities: contract requirements
analysis and construction requirements analysis.
975
Welding process determination activity is decomposed into the following

activities: technologibility analysis, working conditions defining and workers training
analysis.
Welding technology definition activity is decomposed into the following
activities: defining the order of manufacturing activities, welding parameters defining
and equipment selection.
Quality control activity is decomposed into the following activities: defining the
control spot, defining the control parameters, controlling, maintenance of measuringcontrolling devices, development of welding technology qualification and development
of welding technology specification.
Welded construction maintenance activity is decomposed into the following
activities: welded product installation and commissioning, monitoring welded product in
use, welded product servicing and withdrawal of welded product.
USED AT:
AUTHOR: Dusan Jovani c
DATE: 11/1/2003
WORKING
PROJECT: Doktorska disertaci ja
REV:
DRAFT
4/12/2011
READER
DATE CONTEXT:
TOP
RECOMMENDED
NOT ES: 1 2 3 4 5 6 7 8 9 10
PUBLICATION
A-0
M ODELOVANJE INTEGRALNOG
PROCESA ZAVARIVANJA
0
ANALYSIS
CONTRACT OR
PROJECT
DETERMINATION
WELDING
PROCESS
1
ANALYSIS
REQUEST FROM
CONTRACT
ANALYSIS
REQUIRED FOR
CONSTRUCTION
CONTRACT OR
PROJECT
REPORT
DEFINITION
WELDING
TECHNOLOGY
2
COST EFECT IVE

ANALYSIS
DEFINITION OF
TERMS MA KING
SELECTION OF
BASE MATERIAL
SELECTION OF
FILLER MATERIAL
EQUIPMENT
SELECTION
M AINTENANCE
WELDED
CONST RUCTION
QUALITY
CONTROL
3
DEFINISANJE REDOSLEDA
PROIZVODNIH AKTIVNOSTI
(PLAN PROIZVODNJE)
ODREDJIVANJE
PARAMETARA ZAVARIVANJA
IZRADA KVALIFIKACIJE
TEHNOLOGIJE ZAVARIVANJA
IZRADA SPECIFIKACIJE
TEHNOLOGIJE ZAVARIVANJA
DEFINITION PLACES
CONTROL
INSTALLATION
WELDED PRODUCT
DEFINITION WORKING
CASES CONTROL
M ONITORING
WELDED PRODUCT
DEFINITION
PARAMETERS
CONTROL
SERVICING
WELDED PRODUCT
DURING USE
CONTROL
WITHDRAW
WELDED PRODUCT
OUT OF THE USE
STAFF ANALYSIS
MAINTENANCE
MEASUREMENT
EQUIPMENT
ANALYSIS OF
WELDERS
QUALIFICATION
ANALYSIS OF
NON-COMPLIANCE AND
CORRECTIVE ACTION
MAKING QUALITY
RECORDS
NODE:
TITLE:
NUM BER:
A0
Fig. 3 welding technology design activity tree

3. WELDED CONSTRUCTION MAINTENANCE
Welding construction maintenance i.e. monitoring welded product should
indicate to closing the quality loop [4]. A series of operations is done within this activity
which can be shown in decomposition diagram. Fig. 4 shows decomposition diagram
for welded construction maintenance activity which is implemented through the
following activities: welded product installation and commissioning, monitoring welded
product in use, welded product servicing and withdrawal of welded product liquidation.
976
Modeling Maintenance Activities On a Welded Construction Using Idefo Methodology
USED AT:
DATE: 10/20/2003
WORKING
REV:
DRAFT
4/1/2011
READER
DATE CONTEXT:
RECOMMENDED
NOTES: 1 2 3 4 5 6 7 8 9 10
PUBLICATION
A0
Standard
Welded product
INSTALLATION AND
COMMISSIONING
WELDED PRODUCT
5.1
Request for
installationijom
documentation
for
maintenance
Information on
product usage
MONITORING
WELDED PRODUCT
IN USE
Customer
5.2
Informacije za
poslovnog partnera
Customer
complaints
SERVICING
WELDED PRODUCT
5.3
Request for
withdrawal
Report of service
WITHDRAW
WELDED PRODUCT
OUT OF THE USE
Report on works done

5.4
NODE:
TITLE:
MAINTENANCE WELDED CONSTRUCTION
NUMBER:
A5
Fig. 4. Decomposition diagram of welded construction maintenance activity

3.1. Welded product installation and commissioning
This activity is performed by experts provided by the manufacturer or the
customer so the customer could have better insight into the characteristics of the
product. Hence, the instructions are a compulsory part of users documentation where
instructions for the product use and maintenance are clearly described.
In order to perform this activity the following is necessary: preparation for
embedding where the competence regarding product installation is regulated, control
of document compliance with the received product and analysis of installation
instruction, installation according to the instruction with tools and implements
complying with the quality and with control of availability of all resources necessary for
installation (space, energy, etc.), visual control, which provides damage remedying on
the spot and similar, examining and solving of inconformity and documenting a product
delivery.
3.2. Monitoring welded product in use
In order to perform this activity the manufacturer should collect the following:
reports from the customers, failure analyses and data about reliability of welded
construction. Based on the collected data, an analysis can be done of causes of
problems and actions which could be undertaken to solve them. It should be borne in
mind that activity of monitoring of welded product in use comprises: monitoring of the
product in the process of exploitation, suggestions and complaints about the product
and development of relating report.
977
3.3 Welded product servicing

Servicing establishes confidence in manufacturer-customer relationship.
Monitoring of product servicing provides detection of its deficiencies during its shelf life
phases [5]. This activity involves product maintenance, creation of necessary
documentation, training of customers and product quality monitoring.
3.4 Withdrawal of welded product - liquidation
Implementation of this activity involves clearly defined reasons (due to product
deficiencies or people and environmental safety requirements and for providing
product recycling), defining of the withdrawal procedure due to detected deficiency,
monitoring the activity of product withdrawal, defining periodical validations and
assessment of efficiency level.
4. CONCLUSION
Introduction of information technologies i.e. computers in design process, an
easier, faster and more accurate design of welding technology is provided. More
accurate design of welding technology may provide avoidance of undesirable defects
costs which may occur in welded construction development phase.
Welding process modeling provides easier management i.e. management of
welding process. Easier management involves functional decomposition of welding
procedure at all levels and creation of documentation in compliance with requirements
of a series of ISO 9000 standards.
Modeling of welding technology design process was performed by means of
IDEF0 standard i.e. its software realization of CASE Bpwin tool which is approved by
ISO organization as a tool for quality system introduction.
REFERENCES
[1] Veljovic, A., (2001). Kompjuter u sistemu kvaliteta - primena standarda ISO
9000:2000, Savez inenjera i tehniara Jugoslavije, Beograd, p. 111-123.
>2@ Veljovic, A., (2001). Razvoj informacionih sistema i baze podataka, Uprava za
naunu i izdavaku delatnost, Vojno tehnika akademija VJ - Sektor za kolstvo,
obuku, naunu i izdavaku delatnost, Beograd, p. 243-248
>3@ IDEF Family of Methods, http://www.idef.com, january 2011.
>4@ Soldat, D., Gruji, N., (2009). Odravanje maina iopreme,Technical College of
Appplied Sciences in Zrenjanin, Zrenjanin, p.61-93.
>5@ Soldat, D., Mirkovic, R., (2009). Mainsko-tehniki prirunik, Graevinska knjiga,
Beograd, p.III 66-67.
978
NATURAL SYSTEMS OF WASTEWATER TREATMENT IN SMALL

SETTLEMENTS
Jasmina Skerlic1, Vanja Sustersic2, Danijela Nikolic3
Summary: A small group of settlements include settlements with less than 5 000
inhabitants. In these settlements, it is recommended natural treatment systems for the
wastewater treatment. In the most of cases, this system is land natural treatment
systems. It is a complex (biological, chemical and physico-chemical) process of
removing the pollution of waste water, which occurs at the surface and upper levels of
the soil. The land is used as a process of tertiary treatment of waste water, primarily for
the removal of nutrients, or as a secondary treatment process, for removal the organic
pollution of waste water. The most represent treatment systems are: irrigated land with
waste water, rapid infiltration of waste water through the soil, and spreading of soil with
waste water.
Key words: wastewater treatment, decentralized systems, large settlements
1. INTRODUCTION
During the past decades, increasing attention has emerged for environmentally
sound, sustainable, low-cost and effective wastewater treatment technologies based
on ecological principles, namely ecological technology or natural treatment systems.
The environmentally-sound attribute of ecological technology is its capability of
resource recovery and reuse (water and nutrients) while at the same time consuming
little or no energy and chemicals. The different types of natural treatment systems
correspond with the different ecosystems along the land-water gradient, starting from
the land-side with high-rate infiltration fields, overland flow systems, constructed
wetlands and finally waste stabilisation ponds or lagoons.
It is recommended Natural Treatment Systems for wastewater treatment of
small villages, which are mostly cases of so land treatment systems. It is a complex
(biological, chemical and physico-chemical) process of removing wastewater pollution
which occurs at the surface and upper levels of the land. The purification involved
microorganisms that live in soil and plants that the growth or are there planted, then
the interaction between the soil and waste pollution water, such as chemical
precipitation, adsorption, ion exchange. This complex process is held the control by
1
Researcher , Jasmina Skerlic, Kragujevac, Faculty of Mechanical Engineering, (jskerlic@kg.ac.rs)

Associated professor, Vanja Sustersic, Kragujevac, Faculty of Mechanical Engineering, (vanjas@kg.ac.rs)
3
Assistant, Danijela Nikoli, Kragujevac, Faculty of Mechanical Engineering, (danijelan@kg.ac.rs)
2
979
Jasmina Skerlic, Vanja Sustersic, Danijela Nikolic
regulating the amount of waste water used to water the land, to avoid the penetration
of untreated wastewater in groundwater or to dreinage into surface watercourses, and
to their pollution. In addition, wastewater is added the amount that can be "accepted"
the land, and to avoid overcoming anaerobic conditions in the soil, resulting in a
decline in the capacity of wastewater treatment.
2. DESCRIPTION OF PROCESS IN NATURAL TREATMENT SYSTEMS
Land can be used as a process of tertiary wastewater treatment (purified in the
secondary treatment), primarly to remove nutrients; or as a secondary treatment
process for removing organic pollution of waste water. The three most common
treatment system are: (i) irrigated land waste water, (ii) rapid infiltration of wastewater
through the soil, and (iii) spreading of soil with waste water. The lagoons, where water
plants are grown and fish which are then processed into animal feed or used in the
industry, can be used for wastewater treatment, where the focus is, usually, on the
increase biomass and less on the effect of purification of waste water, which is
generally quite weak.
2.1 Irrigation
Irrigation (Fig. 1) is the most commonly used system. It is considered the most
reliable and best system of purification of waste water land and provides the best
quality of processed waste water. Irrigation may be defined as the application of water
to soil for the purpose of supplying the moisture essential for plant growth. Irrigation
plays a vital role in increasing crop yields and stabilizing production. In arid and semiarid regions, irrigation is essential for economically viable agriculture, while in semihumid and humid areas, it is often required on a supplementary basis. It is usually
irrigated arable land, where crops are grown and other crops whose sales reimburse
part of costs of such a system of purification. Selection of plants to be grown also
depends on the number of factors: the capacity of the adoption of nitrogen, water, and
adopted by the tolerance increased soil moisture, resistance to pollution from waste
water, and in this respect suitable are various types of grasses.
Fig.1 Purification of waste water irrigation land

2.2 Rapid infiltration
Rapid infiltration land treatment system consisting of periodic waste water
filling of shallow channels or pools from soil exhumed large permeability (sandy and
ravelly soil), Fig. 2a, where the water purified during trickling through the layer of soil to
980
Natural systems of wastewater treatment in small settlements
the groundwater levels. If purified waste water can not allows into the groundwater, it is
collected by drainage (perforated) pipes Fig. 2b, into the proper channels or pull out
the surface through a system of wells, the Fig. 2c.
Fig. 2 Rapid infiltration through soil: (a) scheme processes, (b) drainage pipes, (c) a
system of wells
The largest part of the wastewater is drainage (usually no vegetation). For the
treatment of effluent septic tanks, in the processes of wastewater treatment of the
smallest settlements, this procedure could come into consideration, provided that the
problem would not be important odor emissions.
Rapid infiltration is less restrictive than other types of land treatment. Rapid
infiltration basins offer the advantage of being less dependent on climate than other
natural treatments systems. Vegetation is not a required design element, so the length
of the growing season does not impact effectiveness unless a high degree of nitrogen
removal is necessary. However, the life spans of rapid infiltration systems have been
reported to be potentially reduced due to saturation of the soil with phosphorus and
heavy metals.
For onsite wastewater treatment, sand mounds are more common than basins,
when rapid infiltration is implemented. Sand mounds are used only when the upper
portion of the soil profile is conducive to rapid infiltration, and may be appropriate
981
where soil permeability is too slow or too fast, or where shallow bedrock or a shallow
water table exists.
2.3 Overland flow
Overland flow is a treatment system where a treasure inclined smooth soil low
permeability, overgrown with vegetation cover, on top of the slope poured with the
wastewater that flows in a thin layer, gradient, down the slope and collects at the base.
Fig. 3 Purification of waste water overflows the land

The choice of plants is essential to in this system because it significantly
influences the capacity and filtration efficiency.
Land treatment systems are characterized by relatively large investment costs
but low operation costs. It is recommended that received biomass is not used for
human consumption, or must first establish whether the biomass has a pathogenic
microorganisms and toxic substances (heavy metals, pesticides, etc.) originating from
waste water. The overland flow process differs from the other land treatment
processes in that treatment of the effluent occurs on or above the land surface, rather
than in the subsurface. In this process, effluent is spread evenly along the top of a
vegetated slope. The slope is gentle enough so that sheet flow occurs over the
vegetated land, where chemical, physical, and biological processes improve the quality
of the effluent. Sedimentation, filtration, and biochemical activity are the primary
treatment processes at work in overland flow. Besides controlling erosion, the
vegetation removes nitrogen and other nutrients from the effluent, and also filters out
suspended solids. Microorganisms such as bacteria and algae, attach to the vegetation
and break down dissolved organics. Additional biochemical activity occurs in the top
layer of saturated soil. Overall, effluent treated by overland flow is of relatively high
quality. Suitability for discharge to streams, however, will be dependent on local
discharge requirements, as well as the characteristics of the receiving water body.
Evapotranspiration and percolation into the subsurface, provides some reductions in
the amount of water that is eventually discharged. Percolation, however, is necessarily
limited by the low-permeability soils that are required at the treatment site.
Overland flow depends on sheet flow across the land and vegetation surfaces,
so this option can be implemented only where underlying soils will restrict infiltration.
982
Natural systems of wastewater treatment in small settlements
For this reason, overland flow can provides an alternative to the other land treatment
options, where site conditions limits the use of drainage fields, spray irrigation, rapid
infiltration, and other methods that require percolation into the subsurface.
2.4 Wet fields
Wet fields, or construcred wetlands can be (i) systems with free water surface
(free water surface systems) and (ii) systems with the flow of water below the surface
(subsurface flow systems) such systems are the advocate in practice.
Analysis should include investment and operating costs during 15 years (cost
cutting, waste deposit plants, planting new plants, control waterproofing, root
replacement filter layer, etc.). The analysis should to cover the costs and benefits of
the community bearing in mind the purpose of fertile land that would be used for such
purposes. In America, for example, wet fields with free surface over, are used rarely for
primary treatment, because of potential exposure to pathogens, but generally for
effluent polishing lagoons, filters etc. Fields with the flow below the surface are used
mostly for the treatment of primary effluent the secondary standards, but it should be
noted that the application of this technique only in development and that it is necessary
for her to do a careful assessment of benefits and costs to in investment and in the
long-term exploitation. In this regard special attention should be given to certain types
of pollution such as nutrients, oil and grease, metals and micropolutants organic
(detergents, various chemicals used in households, pharmaceuticals, metabolites).
Removal of nitrogen that is achieved by wet fields according to the literature is good,
although the natural level is usually higher than 1 mg/l, organic nitrogen accumulate in
vegetation and may later be released or recycled.
Nitrate removal is generally good. Removal of phosphorus by wet fields is not
entirely efficiently. As for the metals data are not enough and they are rare and
diverse. So eg. removal efficiency for cadmium ranges from 75-99%, 40-96% for
copper, 0-86% for lead, 49-88% for nickel, and 33-96% for zinc. The accumulation of
toxic organic and inorganic micropolutants can form toxic fields. Worrying is their ability
to penetrate into groundwater, and even the fact that they are detected in the drinking
water. It was found that irrigation and filtration through the soil can cause water
pollution, certain medications. An important aspect of the assessment the impact of
this technology is the problem with climate factors (eg. the appearance of large rainfall
that cause flooding and major changes in load swampy fields or sensitive plant and
animal species in different physical and chemical factors).
3. BENEFITS
Benefits of natural treatment systems are needed primarily to less energy for
these systems, because most of the energy taken from the natural environment; but
the potential weaknesses of these systems are numerous, as are the systems that
must be design and create exactly the present location, its geography, microclimate,
soil, and other characteristics, with carefully chosen and dimensioned plant for the
pretreatment of wastewater.
Creating such systems requires multidisciplinary knowledge, the experts in
different fields; keeping the process a lot difficult, because of the sensitivity of the
process on the environment, and process control is a demanding and relatively
983
expensive. These are all reasons to possible wider use of natural processes of
purification of waste water treatment of small settlements can not be approached
without a comprehensive and long-term monitoring test facility, from which order to
collect enough data for the relevant assessment acceptance of natural purification
process for wastewater treatment of small settlement.
Selection of any purification systems for a small settlements, is of crucial
importance to the plants in small areas are not left to itself, because it is difficult,
practically impossible to provide all the necessary logistics for the operation of these
systems.
4. CONCLUSION
Natural wastewater treatment systems are simple, cost-effective and efficient
methods to purify the growing amount of wastewater produced by our society. They
can be applied as secondary or tertiary purification treatment, allowing the removal of
most of the bacteria, microorganism and the destruction of the organic matter. Among
them phytodepuration, lagoon purification and storage in tanks gave good results in
terms of yield and are quite diffused all over the world.
The selection of a natural wastewater treatment system requires the
consideration of a number of factors, including wastewater volume and pollutant
characteristics, site soils and geology, and climate.
Their extreme simplicity in building, operation and maintenance make these
systems competitive with the conventional (sewer) wastewater treatment methods.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
984
Selma C . Ayaz, Lu tfi Akc a, Treatment of wastewater by natural systems,

Environment International 26 (2001) 189195
Ayaz C S, Akc a L. Treatment of wastewater by constructed wetland in small
settlements. Water Sci Technol (2000);41:6973.
Bowman M. S., Clune, T.S. & Sutton, B.G. (2002) Sustainable management of
landfill leachate by irrigation. Water, Air, and Soil Pol lution,134, 8196.
Reed SC, Middlebrooks EJ, Crites RW. Natural systems for waste management
and treatment. New York, NY: McGraw-Hill,(1988)
Lars Thrneby, Lennart Mathiasson, Lennart Mrtensson and William Hogland,
The performance of a natural treatment system for landfill leachate with special
emphasis on the fate of organic pollutants, Waste Manage Res (2006): 24: 183
194
Youngchul Kim, D.L.Giokas, Jin-Woo Lee, P.A.Peraskevas, Potential of natural
treatment systems for the reclamation of domestic sewage in irrigated agriculture,
Desalination 189 (2006) 229-242
PROCEDURES FOR RISK BASED MANAGEMENT AND

MAINTENANCE RIMAP PRINCIPLE
Miodrag Milutinovi1, Vladimir Popovi2
Summary: this document shows the most important news for document CWA 15740
(CEN Workshop Agreement), that gives the basic assessment elements of industrial
resources based on risks and according to RIMAP approach (Risk-Based Inspection
and Maintenance Procedures). One of the main goals of this document is to contribute
to the harmony of regulatory requests from European Union which refer to programs
for management and maintenance of industrial utilities, making them rentable, and
simultaneously maintaining or improving security, with influences on health and
environment.
Key words: maintenance, risk, harmonization of regulations.
1.
INTRODUCTION
When we talk about agreement CWA 15740:2008, this universal CWA (CEN
Workshop Agreement) gives the basic assessment elements of industrial resources
based on risks, and also according to RIMAP approach that is developed and
demonstrated by European RIMAP R&D project (GIRD-CT-2001-03008 and the
appropriate RIMAP network:Procedures for risk based maintenance and management
in European industry).
Document is intended for managers and engineers that establish RBIM policy
(Risk Based Inspection and Maintenance) in companies from energy supply,
processing, steel and other important industries. It is assumed that this standard will be
used along with relevant internationally accepted practices, national regulations and/or
company policies. Document provides general instructions for this policies and the
development of appropriate management and maintenance programs in different
industry fields. Every part of this Agreement can be used as a standalone document. It
can be expected that positive influence and exchange in industrial practice, which
results from the application of this document and approach, will be useful for the
European industry and for the enforcement of its competence by applying better
management and maintenance.
The aim of CWA 15740 is presentation of transparent and precise scope for the
application/implementation of RBIM and RBLM concepts (Risk Based Life
1
2
Faculty for Education of the Executives, Alfa University, Serbia e-mail: miodrag.milutinovic@daimler.com
Faculty of Mechanical Engineering, University of Belgrade, Serbia e-mail: vpopovic@mas.bg.ac.rs
985
Management) in industrial organizations. Document defines procedures for the risk

based approach, supporting work optimization and maintenance, as well as managing
resources. The goal of RBIM is to assure that clearly defined and accepted risk level
that relate to:
Security,
Health,
Environment, and
Business/production/work,
are being achieved by the application of efficient methods of management and
maintenance. RBIM methodology that is described here is based on the method
developed in European RIMAP project. RBIM methodology is developed and
confirmed for chemical and petrochemical industries, steel industry and energy supply,
in Instruction manuals, but the methodology itself is intended to be industrially
independent. Methodology refers to the following aspects:
Management and maintenance;
All kinds of equipment;
Technical and managing aspects of planning, management and
maintenance;
Managing resources related to control, maintenance and the assessment of
utility, system and component longevity;
Production and work.
Although RBIM includes RBI (Risk Based Inspection) and RCM (Reliability
Centered Maintenance), this document is primarily focused on RBI. RCM is included
only in the sense of demonstrating applicability in the entire RBIM context.
2
RIMAP SCOPE
2.1 RIMAP and RBIM

Collection of RBIM reports originates from RIMAP projects that is financed by
European commission. Documentation provides guides for planning and executing
maintenance and management based on risks. Term RIMAP, which is being used in
this document, is a synonym for RBIM methods, as are applied in RIMAP project.
RIMAP documentation also provides guides securing quality and the following
activities, tasks and work processes as the scope of an organization. The need for
quality of all elements in work process is emphasized, as well as ongoing
improvement. Also, its important to secure ongoing connection between technical
planning and actual RBIM execution. RBIM shouldnt be considered quick
methodology of cost reduction, but an all-around (complete) philosophy of managing
resources. Therefore, company management must provide and support the execution
of this procedure, as well as encourage its application.
2.2 RIMAP Principles
Since the late 1990s, the approach in maintenance was globally moving from
steady/time based decision making, towards control decisions based on risk. This
trend is based on a clear goal to increase on-stream production time, and with
intention to reduce unplanned standstill due to maintenance interruption or the
986
Procedures for Risk Based Management and Maintenance RIMAP Principle
unknown state of equipment, which undoubtedly can lead to jams. Generally, if the
company wants to apply simple, steady approach to maintenance/ management, it is
necessary to apply strict conservative criteria during decision making.
Contrary to the previous statement, an approach based on risk demands a
detailed, multidisciplinary technical analysis in order to secure not to sacrifice security
elements due to the planning process application of maintenance/control. The
appropriate risk based methodology includes these principles:
Planning RBIM evaluations of primary products and approach to
management in such a way that manages risks at system and/or equipment
level, with emphasis on health/security/ecology and/or economic risks.
Defining RBIM technology in a reasonable scope (as good technical
practice, or in the scope of referent industrial standards), when were
dealing with dangerous materials and substances in industrial plant (utility).
Determining the process of work and jurisdictions required for handling
projects in a certain way.
It is also necessary to define minimum requirements for performing and
documenting RBIM assessments in order to align with legal regulations and guides.
2.3 RIMAP demands
RIMAP demands include the following:
General demands;
Demands related to personnel;
Demands for performing PoF analysis (Probability of Failure) they include
the following: general acceptance, conservative simplified approaches,
possibility of results validation, approaches on multiple levels (qualitativequantitative, in the depth of the plant), procedural character, additional
aspects for consideration;
Demands for performing CoF analysis (Consequence of Failure) they
include: general demands for CoF assessment, CoF security, CoF health,
CoF ecology, CoF work;
Demands for risk assessment.
2.4 RIMAP in the scope of the entire management system
Development and application of RBIM plan requires resources like personnel,
budget, spare parts and documentation. Management should assess RBIMs
effectiveness by observing performance indicators such as reliability, costs and risks.
RBIM planning requires multidisciplinary team with wide scope of engineering
competence. Management should determine and define goals related to acceptable
levels of risk in managing and maintaining activities. Goals should be transparent and
should support a complete set of goals of the company in the sense of health, security,
environment, production, quality, etc. Goals should also be in accordance with local
and other normative requirements and possible contractual demands.
RBIM strategy should provide identification and application of risk mitigation
activities before the risks of equipment failure to health, security and environment
become unacceptable. If this risks are tolerated (they are), some activities that reduce
economic and other business risks can still be necessary.
987
RIMAP scope should be observed as part of overall work process, which

consists of:
Defining goals and requirements
Establishing control and maintenance programs
Planning of tasks and activities in management and maintenance
Performing work errands
Reporting failures and status
Assessment of technical conditions
Preparation for improvement tasks
Performing corrective activities
Active management
Managing changes
Work procedures
Safe work practice
Review of work commencement
Reaction and management in emergencies
Investigation of incidents
Training
Securing quality.
2.5 Limitations
RIMAP scope is also applicable in other industries, not just in those where it is
directly intended (petro-chemical, chemical, energy, steel), however, it is limited to
nuclear application. RBIM scope is applied only in systems and equipment in work
stage. For design or production phase, we need to respect relevant laws and
engineering standards. In case of RIMAP principles and procedures, we need to
assure that all measures are in accordance with local and national regulations. During
RBIM application, we should bear in mind:
1. That RBIM assessment is good only as her input data;
2. That RBIM is not a substitute for quality engineering practice/consideration.
2.6 Compatibility with other known approaches
Generally speaking, the entire RIMAP approach is compatible with the biggest
number of approaches based on risk, such as API, VGB or ASME, and which have a
similar purpose. However, although principles are mostly similar, users shouldnt
expect identical results. There are differences in details that can result in significant
differences when various approaches are used in the same utility, case or system. For
instance, as opposed to other known approaches, RIMAP was primarily designed to be
industrially independent and to provide seamless and easy transfer between various
analysis levels (from rough recording to details).
3
RIMAP PROCEDURE
RIMAP procedure provides directions for the development of management and

maintenance program based on risks, with higher level of quality or managing
environmental risks. Procedure is applicable in many industries and for various type of
988
equipment (for instance, static equipment, security systems and electric

equipment/instruments). Procedural steps are the same for all cases, even if models
and tools for probability assessment or failure consequences can vary from one
application to another.
Procedure includes the following main steps:
1. Initial analysis and planning
2. Gathering data and validation
3. Risk analysis in multiple levels
4. Making a decision and planning activities
5. Performing and reporting
6. Performance review
The following elements are defined for each of the above steps:
1. General description and scope
2. Requirements
3. Input
4. Procedure
5. Outcome
6. Warnings and limitations in application.
Review of RIMAP procedure is shown on Figure 1.
989
Fig.1 RIMAP procedure in the scope of complete system management

3.1 Risk analysis in multiple levels (from basic recording to detailed analysis)
3.1.1 General description and scope
Risk analysis consists of the following steps:
a) Identification of dangers
990
b) Identification of relevant damage mechanisms and forms of failure

c) Determining failure probability (PoF)
d) Determining failure consequences (CoF)
e) Determining risks and equipment classification
Risk analysis in multiple levels defines risk assessment in relation to (i)
analysis complexity (i.e., from the simplified analysis/recording to detailed analysis),
and to (ii) hierarchy level (depth) of utility. Figure 2 shows that in RIMAP approach,
analysis complexity, or in other words component analysis, continually decreases,
recording up to detailed analysis, while it decreases step by step in conventional
approach. Also, in Figure 3 it can be seen that analysis depth can gradually grow from
recording to detailed analysis with RIMAP approach, while it increases step by step in
conventional approach.
Fig. 2 Risk analysis in multiple levels: analysis complexity
Fig. 3 Risk analysis in multiple levels: hierarchical level of utility
991
3.2.1
Risk analysis basic level of recording
3.1.2.1 Description
Basic recording (so called, screening) of risks will be relatively fast, simple and
rentable process, compared to detailed risk analysis. Risk recording is especially
suitable for wide based problems and limited number of points that are being
considered. Risk recording divides systems and equipment groups in two subgroups:
highly risk points and points with average/small risk. Highly risk points should be
analyzed in detail. Points with average risk should be additionally considered in order
to make a decision about the usage with minimal surveillance in order to confirm
assumptions reached during risk recording. For example, it can refer to color,
protection, functionality or correct, unimpaired state of structure. In case that recording
is insufficient to determine the risk for equipment, such equipment should be
considered a high risk and perform detailed assessment.
Work procedure for basic risk recording is detailed in Figure 4.
Fig. 4 Work procedure for basic risk recording.

3.1.2.2 Requirements
We need to meet the following requirements for risk analysis:
1. Define and register valuation criteria.
2. Determine PoF for given (previously defined) time based on forecasting
damages for the operation in the scope of certain work window. Certain
work window should include factors which can be influenced by the
operation (i.e., temperature, pressure), as well as factors which cant be
influenced by it.
3. In order to assess consequences, we need to include at least
consequences for health, security and environment. Furthermore, the
assessment of consequences should be such that the highest valued
992
individual contributions (health, security, environment and/or work

consequences) can influence on final evaluation (accordingly, the is no
average).
4. Methodology should be verified.
5. This task should be carried out by RBIM team.
6. Results should be checked by experts from various fields (independent
assessment); accordingly, methodology, input data, criteria for decision
making and results should be documented (results should be recorded in
the form of official document).
3.1.2.3 Procedure
This level of analysis is often enough to mark fields with the biggest
probability/frequency of failures in the utility (devices/systems). Work process of risk
recording is shown in Figure 4. Main goal of risk recording is to identify low risk
components (see Figure 5) and to exclude them from further analysis. Its very
important not to classify too many components in high risk category, so its useful to
compare the spectrum of assessed PoF, CoF, and risk categories, with the ones
obtained with similar assessments.
Fig. 5 Matrix of basic risk recording.

3.1.2.4 Output data
Typical results of these tasks are:
PoF value or category for the analyzed part of the equipment;
CoF value or category for the analyzed part of the equipment;
Risk value or category for the analyzed part of the equipment (from the
matrix of basic risk recording Figure 5).
3.1.2.5 Warnings and limitations for application
Failure probability assessment usually requires more details, and that makes it
more expensive than the failure consequence assessment. Accordingly, some people
suggest recording systems and component groups only in relation to failure
consequences assessment. This can also be acceptable, even though this report
suggests other type of risk recording.
993
3.2 Making decisions/activity plan

3.2.1
General description and scope
Conservative management and maintenance represents one efficient approach

in risk decrease activities compared to optimized management and maintenance plan
development. In order to manage and maintain on a daily basis, a program with
previously determined time intervals is being determined. Based on projects that were
realized so far, a description is given for suggested way of decision making in order to
determine managing and maintenance strategies.
Several factors directly influence the need to manage and maintain:
Equipment/machines wearing and unreliability;
Human factor unreliability, maintenance or managing equipment/machines;
Laws and other regulations;
External factors (earthquakes, bad weather, etc);
Severity of consequences.
Plan of activities consists of:
Work supervision, and
Conditions tracking.
Management and maintenance program is determined depending on these
unreliabilities and risks, as well as depending on legal requirements. Maintenance that
is introduced due to human error and external factors is considered as part of program
for ordinary management and maintenance.
Termination of some component is related to the cause of her failure, which
can be caused by circumstances that emerge from utilization or maintenance.
Managing and maintenance strategy represents an approach selected to prevent
physical damage and negative influence on the environment, injuries, loss of life or
resources.
3.2.2
Requirements
RBIM plan will create a team that includes experienced experts with following
qualifications:
Enough knowledge about risk levels, failure probabilities, consequences
and expert management, depending on local requirements/regulations;
Qualified knowledge about component adequacy and experience with
managed utility (systems, equipment or components). Generally, knowledge
about reliability in engineering or several years of experience and utility
maintenance are required.
Team should have the approach to all relevant data and risk analysis. RBIM
plan will contain all relevant details based on strategies for carrying out procedures, in
order to achieve a desired risk level decrease that is established by RBIM analysis and
process.
3.2.3
Input
RIMAP project includes documented methods for determination and forecast of

damage mechanism, as well as methods for the assessment of consequences and risk
probabilities. Identified damage mechanisms, CoF, PoF, and related risk are used as
994
input data to determine methods for maintenance and management, protection of

health, life, environment and resources.
3.2.4
Procedure
Suggested scope of decision making is divided on the main level, and the level
of strategy for management and maintenance. Figure 6 shows the main level, which
includes the following factors:
Possibility to eliminate failure causes;
Personnel risk, during execution of management and maintenance strategy;
Risk of introducing new causes and failures.
If the strategy for management and maintenance cant be replaced, technical
(robots) or organizational (training) measures can be introduced in order to decrease
risks for personnel and new failures.
Decision making logic has three important goals:
Assessing the need for preventive activities;
Assessment consistency between various utility systems;
Simplifying documentation which is used to reach conclusions.
Fig. 6 The main level of decision making scope

When management and maintenance strategy is determined, methods,
intervals and management scope should be also determined, so that risks stay
acceptable, and costs optimal. This can be achieved by applying risk reduction
measures for components that go beyond acceptable limits, if its possible to apply
them. We need to determine the effect of alternative measure of risk reduction, as well
as costs of new measures.
995
3.2.5
Output
In general principles, the decision making logic provides directions to establish

desirable management and maintenance strategies, based on critical assessments,
the possibilities of finding damages and failures properties. Results of decision making
logic are as follows:
Eliminating failure causes;
Regular functional testing/control;
Maintenance based on time and state;
Working (acting) maintenance;
Corrective maintenance.
3.2.6
Application warnings and limitations
Methods of risk reduction should be chosen based on cost optimizations, so

that risks for health, security and environment satisfy criteria that correspond to
standards and regulations from subject filed.
4
CONCLUSION
Existing managing and maintenance methods of technical systems have a goal

to secure their desired reliability and availability, with a least cost possible. New
managing concepts, which are increasingly being applied in all fields of life, are based
on risk, complex category that directly links probability of unwanted event and its
consequences. This is all even more present in the field of theory and engineering of
maintaining technical systems. New methods of maintaining based on risks are being
developed for the area of preventive maintenance according to state, and they
represent an important support for the existing methods of maintenance based on
reliability. Its also important that in these new methods not only costs are being
observed, but also other important aspects such as security, influences on health and
environment, etc. This document has shown the basic document CWA 15740 which
represents a good and formalized support in risk based managing maintenance.
5
LITERATURE
[1]
CEN Workshop Agreement CWA 15740:2008 Risk-Based Inspection and

Maintenance Procedures for European Industry (RIMAP).
Todorovi J. (2003): Upravljanje odravanjem na bazi rizika, asopis Istraivanja i
projektovanja za privredu, broj 1, str.23-32.
Vasi B., Todorovi J., Curovi D., Popovi V., Stanojevi N., Curovi N. (2006):
Odravanje tehnikih sistema, IIPP - Institut za istraivanja i projektovanja u
privredi (ISBN 86-84231-13-9), Beograd.
[2]
[3]
996
PURPOSE AND METHOD EDUCATION OF INTERNATIONAL

RECOGNIZED PERSONNEL FOR WELDING
Dragoslav Dobra1, Sead Avdi2
Summary: To be welded product made in Bosnia was placed on the international
market, must meet the standard requirements for quality assurance in welding. The
standards require that the process of documentation and production related to
technology mergers carried out by welding personnel Welding holds a license issued
by the European Welding Federation (EWF). This paper describes a system of welding
personnel training programs to the European Welding Federation, as part of training
implemented in BiH.
Key words: training, welding personnel
1. INTRODUCTION
Welding is a complex technology, whose understanding of the knowledge in
the field of mechanical engineering, physics, metallurgy, chemistry, electrical
engineering, as well as other areas. The design and manufacture of welded products
of welding personnel are required and knowledge in the narrow field of expertise.
Specialist knowledge in the field of welding are acquired after completion of school
education, according to separate specialized programs.
In front of welded structures are placed increasing demands in terms of
production quality and efficiency. These requirements can be fulfilled only companies
that can meet the requirements of EN ISO 3834. To meet the requirements, companies
must have their own professional staff Welding, and to all levels, from welders to
engineers.
Since its establishment, the European Welding Federation (EWF), conforming
with the International Institute of Welding (IIW), intensive work on drafting guidelines
for the organization and training of internationally recognized welding personnel.
Guidelines for Training welding personnel are made on the basis of EN ISO
3834, which was completely replaced the withdrawn standard EN 729. Through
guidance and standard EN ISO 14731 are described in detail the tasks and
responsibilities of all levels of welding personnel, as well as the manner of their
1
M.Sc., Dragoslav Dobra, Maisnki fakultet Banjaluka, Vojvode Stepe Stepanovia 73, 78000 Banjaljuka,
(doras@urc.rs.ba)
Ph.D., Sead Avdi, Institut za zavarivanje d.o.o. Tuzla, Hasana Brkia 72, 75000 Tuzla,
(avdic.inz@gmail.com)
2
997
Dragoslav Dobra, Sead Avdi
education. Knowledge of the duties and responsibilities are essential elements of

quality assurance of welded structures.
To the staff of BiH Welding and welding products made in factories in BiH
could be placed on the international market, it is necessary to fulfill the conditions
stipulated in the general standard in the field of quality assurance in welding - EN ISO
3834. In the first place, it is necessary to provide qualified personnel Welding.
2. EDUCATION AND CERTIFICATION INSTITUTIONS
Guidelines EWF imply that each country has one authorized organization at the
national level organize and implement a system of education and certification welding
personnel. The work of this organization, "Authorized National Bodies - ANB" will be
evaluated and monitored by a series of rules, "Rules for the implementation of EWF
Guidelines for education, examination and qualification of welding personnel", are
responsible for the implementation of EWF application, implementation and
maintenance of established norms education, testing and qualification in their country,
and promotion of staff qualifications systems to the EWF.
The main task of the ANB in his own land, on behalf of the EWF, acting on the
qualifications of the staff and to ensure uniform conditions and implement:
granting organizations to carry out training under the guidelines of EWF,
supervision of exams (theoretical and practical),
qualifications of staff and keeping appropriate data.
ANB evaluates the work plan and a list of teachers approved training body
(ATB), which organize and conduct specialized training. ANB may authorize more ATB
in his country, as well as outside of their country.
Approved trainings body (ATB), authorized under the constant supervision of
the ANB, were responsible for the implementation of quality welding specialist training
staff.
3.
LEVELS OF EDUCATION
The system of education and certification welding personnel, which according

to the guidelines of EWF performed by authorized organization, mean education level
and more specialized to: Europan Welding Engineer - EWE, Europan Welding
Technologist - EWT, Europan Welding Specialist - EWS, Europan Welding Practitioner
- EWP as and specific levels of education and certification of welders (EW - Europan
Welder) and welding inspectors (EWIP - Europan Welding Inspection Personnel).
Schemes of certain levels of training specialists are shown in Figures 1.2 and 3.
To access a specific level of specialized training the students must meet the
prerequisites set by the ANB under the guidelines each EWF. Most require that
students have completed their educational background and some time working in the
profession.
After completing residency training and passed all the exams the student,
depending on the level of education, acquired the appropriate name for the EWF and
IIW specification and receives a degree in English. Diploma worth permanently.
However, according to the recommendations of EWF - they must constantly prove in
the profession, continuing training and work in the profession at a certain level. This
998
Potrebe i nain obrazovanja meunarodno priznatog osoblja za zavarivanje
application is particularly pronounced for EWE and EWT level. To acquire a name
(status) Coordinator European Welding Engineer Welding (EWE) has, according to
standard EN ISO 14731, after graduation to work in practice in the field of welding at
least two years. The status of the coordinator is renewed every 3 years, the scheme
shown in Figure 4.
EWE I
EWE II
EWE III
EWT I
EWT II
EWT III
EWS I
EWS II
EWS III
EWS 0
Intermediate Examination
EWP
Final Examination
EW
Fig. 1 The scheme of education specialist welding personnel to EWF

pipe welder
Special alloys:
plates welder
Aluminum
Alloy steel
welder corner joints

GAS
MMA
TIG
GMAW
GAS
MMA
TIG
GMAW
G7
MMA 8
TIG 5
GMAW 5
G6
MMA 7
G5
MMA 6
TIG 4
GMAW 4
G4
MMA 5
TIG 3
GMAW 3
Low alloy steel
pipe welder
plates welder
MMA 4
G3
MMA 3
G2
MMA 2
TIG 2
GMAW 2
G1
MMA 1
TIG 1
GMAW 1
welder corner
joints
Fig. 2 Simplified scheme of training of welders
999
1
2
Basic Techn.
Module
and Expirience
Acces condition Routine No
Interm.
Exam
Education Modules
Welding
Final
Exam
EWI
Diploma
Welding Tecnology
Module WT 1 / 2
Welding Inspection
Module WI 1 / 2
EWI - E
Welding Tecnology
Module WT 1 / 2
Welding Inspection
Module WI 1 / 2
EWI - T
Welding Tecnology
Module WT 1 / 2
Welding Inspection
Module WI 1 / 2
EWI - T
Welding Tecnology
Module WT 1 / 2
Welding Inspection
Module WI 1 / 2
EWI - P
EWE
1
2
Education Modules
Welding
EWT
1
3
Module
0
2
EWS
4
1
2
EWP
Fig. 3 The scheme of education inspector for welding (EWI), according to the
guidelines of WEF
1
Technical Knowledge Required

European Welding Engineer, (EWF Diploma)
2 years recent experience relevant job content and level
Demonstration of maintaining and developing tecnical knowledge
ANB Assessment
Issue of Certificate Certified Europen Welding Engineer

(Renewed ever 3 years by repeatig steps 2,3 and 4 above Surveillance)
Fig. 4 Levels of certification EWE
In addition to these specialist training, in recent decades, EWF and IIW

guidelines define the training techniques of welding related:
- ETS European Thermal Sprayer
- ETSS European Thermal Spraying Specialist
- EAE European Adhesive Engineer
- EAS European Adhesive Specialist
- EAB European Adhesive Bonder
- EWS-RW European Welding Specialist for Resistance Welding
- EW-UW European MMA Diver Welder
1000
Potrebe i nain obrazovanja meunarodno priznatog osoblja za zavarivanje
- Special Course in Laser Welding

- Special Course for Robot Welding
- Special Course for Welding Reinforcing Bars.
4.
STATUS OF EDUCATION IN BiH
Until 1991. year, in part Energoinvest in Sarajevo, there was the Institute of
Welding, which was a full member of the IIW and as such was entitled to training and
certification of welding personnel. Today, in Bosnia exists Society for welding, that is a
non-governmental organizations. What in the EWF has observer status and as such is
not a full member of the EWF. As an observer, BiH has the right to form ANB and
independently perform activities of education and certification of internationally
recognized welding personnel.
In BiH there are plenty of very good quality welding personnel (and often better
than individual staff in some Member States EWF) that due to the inability of
certification, is not able to place their services on the international market.
Training and attestation of welders in BiH, although not accredited, performed
a number of state and private enterprises. While most training companies based on the
guidelines of the EWF, some companies run by their own training program.
Certification of welders is based on EN ISO standards. Certificates issued by
welders in this way are valid in the area, for the domestic market, while the
international market has made recertification of welders in authorized institutions.
Likewise, when it comes to the qualification of welding procedures and
certification of welding operation, there are companies that are accredited and carrying
out tasks of testing and issuing certificates gossiping. These documents are not
internationally recognized and are subject to revision.
In recent times the company from its BIH Welding personnel sent to
specialized training in neighboring countries that have gained full membership in the
EWF, and thus entitled to have their ANB and ATB organizations and independently
carry out training and certification.
To companies from BiH closer to European market over the past two years
has been organized by the Society for the welding of BiH and the Institute of Welding
Ltd. Tuzla, together with the Society for the Advancement of Welding in Serbia (DUZSCertPers) as a responsible ANB and the Welding Institute in Belgrade, which has the
status of ATB's, organized training of two groups of participants for the EWT and EWE.
Participants of the course after four months, listened to presentations from four areas,
laid the theoretical intermediate examination and practical exercise and defend final
exam. So they have gained specialist titles that are internationally recognized by the
EWF and IIW Committee.
Planning of Society for the welding of BiH in the course of the year in BiH ATB
establish at least one center that would be part DUZS CertPers as a responsible ANB,
with the implementation of all activities for the establishment and accreditation of ANB
BiH.
1001
3. CONCLUSION
In Europe last year, in the EWF, many made on the harmonization of rules and
guidelines for development of education programs for welding personnel. Programs
EWF are almost entirely accepted by the IIW and through them runs education and
outside Europe. Internationally, the same programs and provides a way Trained
Welding personnel of equal knowledge, which enables better communication between
the flow of technical information.
Increasing requirements in the production and standardization of these
conditions requires that companies must continually educate Welding personnel in a
way that the acquired knowledge to be internationally recognized.
The race on the international market, and soon the home will be able to
include only the welding company (regardless of the capacity and quality of equipment)
that they meet the requirements provided for in the standard EN ISO 3834. That
means a licensed: equipment, welding personnel, procedures, techniques, methods of
inspection and testing. In a word, those enterprises that are able to meet the
requirements of conditions on the quality assurance of welded products.
LITERATURE
Journal Papers:
[1] K. Grubi, S. Kralj, . Andri, (2004) European system for education of welding
personnel, Zavarivanje i zavarene konstrukcije, vol. 3/2004, p. 133-138
Standards:
[1] BAS EN ISO 3834:2009. Quality requirements for fusion welding of metallic
materials, Institute for Standardization BAS TC 2, Istono Sarajevo
[2] BAS EN ISO 14731:2008. Welding coordination - Tasks and responsibilities,
Institute for Standardization BAS TC 2, Istono Sarajevo
[3] BAS EN 719:1999. Welding coordination - Tasks and responsibilities, Institute for
Standardization BAS TC 2, Istono Sarajevo
[4] BAS EN 729:1999. Quality requirements for welding - Fusion welding of metallic
materials - Part 1: Guidelines for selection and use, Institute for Standardization
BAS TC 2, Istono Sarajevo
[5] The documents, guidelines and rules EWF (Doc. EWF409-06, Doc. EWF 410-06,
Doc. EWF 451-02, Doc. EWF 411-05, Doc EWF 480-01, Doc EWF 452-94, Doc
EWF 481-01, Doc EWF 467-01, Doc. EWF 416r2-07, Doc EWF 483-01)
WWW pages:
[1] European federation for welding, from http://www.ewf.be, accessed on 2010-04-05.
[2] European federation for welding, What is EWF all about?, from
http://www.ewf.be/media/newsletterDocs/doc_23_1.pdf, accessed on 2010-04-05.
[2] European federation for welding, The International (IIW/EWF) Education, Training
& Qualification System, from
http://www.ewf.be/media/imagemDocs/doc_63_the_international_education_traini
ng_-_triptico.pdf, accessed on 2010-04-05.
1002
SAFETY PARTS OF BODY TO RESPIRATION WITH BREATHING

APPARATUS IN DANGEROUS ZONE
Aleksandar Majstorovi1
Summary: We are witnesses of life imperilling and working background too, as today, even
during the evolution. Our life interface is disturbed with different shapes of contamination:
natural disasters (floods, earthquakes, volcanic eruptions and the other); on purpose or
accidentally-different shapes like: wars, fires, chemical fates, using of dangerous materials in
production processes and every day life
In these mentioned specific conditions breathing is imperiled, like the vital function of mens life
and working interfaces. With the technical-technology progress, it is enclosed the development
equipment for protection and safety of the respiratory parts of body, faces and eyes, regard to
breathing apparatus. Applying the breathing apparatus, the total protection of the respiratory
parts of the body, faces and eyes to aggressive materials, smoke and warmth is enabled.
Today in the world there are many producers of breathing apparatus with similar technicaltechnological solutions, of which should be mentioned the main ones: MSA Auer, Drger
Interspiro. Using of the breathing apparatus in practice, continual performation of
different activities without consequences for the health and life of the user is enabled,
what firstly depends on the user`s education, regular using, controlling examination
and maintenance/servicing.
Key words: contamination; safety of respiratory parts of the body, faces and eyes;
breathing apparatus; education; maintenance/servicing.
INTRODUCTION
Since the formation of the first fire brigades in the period of the Roman empire
which governed Neron (1st century A.D.) people have been protecting from smoke and
fire with the improvised means which have been improved during the history.
The first breathing apparatus (regenerative apparatus on compressed oxygen)
produced in German factory Drager in 1910. and without particular changes was used
until 1924; when its way of carrying was changed. Until then, breathing apparatus was
carried on the chest. Of course this type of apparatus was modified introducing new
technical-technological solutions including electronics for controlling work and checking
those ones. One of these apparatus is Air Elite 4 h by MSA Auer the producer, with the
interval of four hours.
1
Aleksandar Majstorovi B. Sc. E. Mechanical, Banja Luka, City of Banja Luka, Department of Civil
Protection, Fire brigade, (aleksandar_majstorovic@banjaluka.rs.ba)
1003
Aleksandar Majstorovi
Apart from protective clothes and footwear in the fire brigades,the protection of
respiratory organs by the breathing apparatus has the most important role. Nowadays
we cannot imagine any intervention (fire, chemical accidents etc.) in the world and in
our country, without using the breathing apparatus. Considering irritated and poisonous
features of a smoke in the fire, and irritated and toxical features in the chemical
accidents, which at first affect respiratory organs, face and eyes, we realize their
importance and a role in the protection. The breathing apparates offer the protection
even from the different kinds of microbes, even antrax, by setting different types of
membranes into the lung automats. Such solution, gave Swedish producer
``Interspiro`` with its model 140/QS Spiromatis S, with building into the membranes for
SCBA models of apparates why it conquered the American market, after terrorist
attacks with biologic agent.[9] Of course, by this type of the intervention beside using
the breathing apparates, special rubber suits are used, and they are designed to offer
the preotection to a user in the most of dangerous surroundings, and after the using ,
it is necessary to carry out obligatory decontamination.
Uncontrolled burning (fires) cause manifold changes in the atmosphere of life
and work environment, polluting it with the smoke and its products-poisonous gases,
as well as the heating of air. With the complete (controlled) burning out of the organic
matters come to be carbon-dioxyde and water steam, and with uncontrolledcarbonmonoxide which is very poisonous and explosive. This gas is united with
hemoglobin in the blood pushing out oxygen, making the linking of carbohemoglobin,
which is more stabile than oxyhemoglobin. The affinity of hemoglobin to CO, is higher
about 300 times than its affinity to oxygen.[6,7] If natural matters are consisted of
nitrogen in the form of protein, then ammoniac, amino-acids and cyanichydrogen-acid
can be formed. Thermooxidative destructions produce thermodegenerative products of
the uncontrolled burning: CO, CO2, H2O, lower hydrocarbons, aldehyde, ordinary
lower acids, chlorohydrogen, chlorocarbon, chlor, acrylic nitrate, acrolein, ammonia,
nitrate compunds, hydrocyanic acid, phenol, formaldehyde, fornic acid, styrene,
amines, methyl alcohol, acetic acid, sulfur hydrogen etc. [2]
2.
IMPORTANCE AND USAGE OF BREATHING APPRATUS
The air is one of the most important elements in the working and life
environment, and its main components have approximately the constant rate merits,
which vary with the changing of the atmosphere pressure. As the average rate merits
in the air structure, are taken nitrogen 78,09%, oxygen 20,95%, carbondioxyde
0,03% and other gases 0,93% (precious metal He, Ne, Ar, Xe, Cr, Rn.)[4]
Disadvantegeous structure of the air negatively affects human organism, in other
words, his activity. Oxygen contents hardly ever goes under 19 %, and carbondioxyde
hardly ever goes over 0,3% in the working environments with the bad ventilation.[1] It
should be noted that if the concentration of the oxygen in the air goes down under
17%, problems with breathing appear, and with the concentration of 15 % burning of
the most matters stop.[6] If the concentration of carbondioxide in the air goes up more
than 3%, there are problems with breathing, with the concentration of 14 % suffocation
appears, while with the concentration of 25 % there is a momentarily death.[4] When
burning out with releasing of the heat, there are smoke and poisonous gases which
polluted the air. Inhalling the hot air into the lungs there is coagulation of proteins even
under the temperature of 42 Celsius degrees.
1004
Safety parts of body to respiration with breathing apparatus in dangerous zone
2.3 Means for protection of respiratory organs

Depending on the atmosphere surrounding and needs for using of means for
protection of respiratory organs, we can put them into two groups:
1.
Protective means on the basis of filtration dependent surrounding

(pic.1) [3]
-
2.
Protection made of serosol (mechanical protective filters with

halfmasks for protection of respiratory organs pic. 1a)
Combined protection from gases and aerosol (halfmask or mask for the
whole face and filters with the active filling with the specific intention for
protection of repiratory organs, face and eyes pic. 1b and 1c)
Protective means on the basis of isolation independent surrounding

for protection of respiratory organs and face (pic. 2). [3]
-
Tube devices (tube apparatus with air from unpolluted environment,

tube apparatus with the bellows, tube apparatus without bellows, tube
apparatus with air under the pressure of water).
Isolated apparatus with the open system (devices with the container of
compressed air, selfrescuer).
Regenerative apparatus with the closed system (breathing apparatus
with compressed oxygen, and apparatus with chemically bound oxygen)
Picture 2 Type of masks.

2.2.2 Breathing apparates
Using the breathing apparatus, a user is completely isolated from the
atmosphere environment, as well as from the affecting of contaminated atmosphere.
They are used in the environments with the lower concentration of oxygen in the air, or
where concentration of toxic gases are over permitted limits. Parts of the apparatus
(pict. 2) are:
Mask EN 136 Class III in which depending on lung automat prevails normal
or positive pressure. It was made of rubber or silicone with polycarbonate
visor. It covers all face and it has an additional protection for nose from the
heat. The version with spherical visor was tried so the width of visibly field
was increased and the inadequancy was manifested as a distorted picture.
1005
Lung automat EN 137 reduces the medium pressure to normal or higher

pressure about 1 bar.
Tube of high pressure (medium pressure) air is after the pressure
regulator reducing from 200 or 300 bars to about 5,5-8 bars depending on
the producer. At some producers` (optionally) the additional connection for a
mask for the other user or for the air system of rubber suits class III, is set.
Belts for carrying, anatomy shaped, made of nomex or kevlar.
Belt for fixation of the breathing apparatus around the waist, anatomy
shaped, made of nomex or kevlar.
Pressure regulator EN 137 - on some models, the adapter for fast filling
is built into (quick fill connection) which offers the filling of cylinder for 45
seconds.[10] It has an important role on the breathing apparatus, because it
reduces nominal pressure of the apparatus (which can be 200 or 300 bars)
to the medium pressure 5,5-8 bars.
Carrier of the apparatus EN 137 - they were metal earlier, while nowadays
they are made of syntetic fabric thermosets.
Manometer with a sound sensor and a tube of high pressure EN 137
which serves for controlling of high pressure in every moment, while
acoustic device serves to warn a user that in the bottle left more 555 bars,
and that the bottle should be changed. The acoustic device, depending on a
producer can be built into a manometer or a reductive valve (pressure
regulator) and it gives a signal to a user that in the bottle left the air for
retreating from a dangerous zone (approximately 5 minutes ).
Bottles for air with the valve EN 144 are made as: steel (4 l/200 brs ; 6
l/300 brs ; 6 l/300 brs-super lacquer), composite (6,8 l/300 bars ; 9 l/300
bars) and "full" composite EN 12245:2002 (3,4 l/300 bars ; 6,7 l/300 bars)
which are for about 50 % lighter than the steel ones. Composite bottles are
made of carbon and glass fibers, mutually connnected with epoxide resin,
and as a basis aluminium cylinder serves to them, while "full" composite of
the same structure are with thermoplastic basis.
The breathing apparatus by the rule can be used only by grown-ups (not older
than 50), healthy and a qualified person. By education, users are introduced to basic
parts, changes in new models, priniciples of work, and also that even the least
negligence and inadequance in the instruction, keeping and checking of apparates,
imperil his own life. 8 One of the most important factors is in using of air depending on
physical effort (T.1.) 5 It should be noted that depending on the period of using of the
breathing apparates with the longer work, two bottles can be used, which are with the
"T" parts connected to one reductive valve.
Table 1. Using the air depending on physical effort.
Cndition
Resting
Easy work (without physical effort)
Less hard physical work
Very hard physical work
xtremely hard physical work
1006
Example
Sitting, lying
Reading, writing
Less demanding jobs in workshop
Hard physical work
Going up the stairs,fast running,
stressed work
Air using
6
8-10
15-20
40-60
80>
Safety parts of body to respiration with breathing apparatus in dangerous zone
Once a year, examination and testing of the complete breathing apparates are
done by the authorized institution and for those the usable licence is isssued.
In the territory of the Republic of Sprska, actually Bosnia and Herzegovina
examination and testing of the braething apparates are done on the base of Command
about obligatory atesting of apparatus for protection of respiratory organs by "Official
Gazette SFRJ" nb. 49/87.
Picture 2. Parts of isolative apparatus Interspiro 140/QS Siromatis S.

1.mask, 2.lung automat, 3.tube of high pressure (medium pressure). 4. carrying belts, 5.belt for
fixing, 6.pressure regulator, 7. apparatus carrier, 8.manometer with sound sensor, 9.bottle for air
with a valve.
Because of a bigger safety and possibilities to monitor users, on the belt of the
braething apparatus, a motion sensor is set or it is already integrated into the
apparatus. After putting the apparatus on the user`s back, a motion sensor is activated,
taking out a key for blocking. If a user rests, after 20 seconds, a prior to signal is
activated, and if in after next 5 seconds, any movement is not realized, a sound signal
of 95 dB is activated, which means that a user is in a big danger, or in the
unconsciousness or he was injured because of the consequences of collapse, or falling
from some hight, explosion If a user who has not been injured or hurt, he can cancel
a sound alarm by putting the identification key into a motion sensor. On new devices
there is a possibility of following a user with the electronic way, what oferrs a great
advantage to a operator of the action so he can know in any moment where every
fireman is or if a sound alarm is activated, and his location and evacuation. The device
is charged with its own source battery.
3. CONCLUSION
Protection of the repiratory organs during the intervention of firemen is very
demanding because any protection is not 100 % safe, and there is always possibility
that this one let down. Understanding the principle of working and the intention of the
braething apparates, it can be guessed what kind of consequences might be caused
by their incorrect using and maintaining. To avoid these dangers, it is necessary to
1007
follow the producer`s instruction, to have everyday checks, and periodic checks ( after
using ) in the own workshops of firebrigades, as well as regular annual attesting at the
authorized institutions. It would be desirable that all bigger fire brigades are eqipped
with the control devices, and to test insulating apparates, but also to modernize all
these ones with new apparates in which move sensors and transmitters are already
intergarted, with which the safety of users would increase and their monitoring during
the intervention. Purchase of current equipment, education and trainings, reduce the
risks of injuring of a user and damaging the equipment.
Books:
[1]
[2]
[3]
[4]
[5]
[6]
[7]
Agovi . (1976). Protection at work. Economic review, Belgrade; page: 307.

Gai M. (2002). Protection from fires and explosion. University in Banjaluka,
Faculty of Mechanical Engineering Banja Luka; pages: 53, 55.
Devi M. (1983). Personal protective means. Library of professional edition,
Zagreb; pages: 57-83.
Delez G. Obuljen V. (1973). Chemical noxiousness-noxious and toxic matters in
the industry. High Technical school for safety at work and protection from fires at
university "Moa Pijade" Zagreb; pages: 5, 187, 188.
Dubravac M. (2009). Basics of respiratory system protection. Fire school, Ig,
Ljubljana; page: 20.
Kosti V. (1983). Basics of preventive protection from fires. Fire fighting
association of Yugoslavia, Belgrade; pages: 21, 22, 28, 29.
Zagorac M. (1978). Industrial toxicology. Faculty of protection at work, Ni; page:
96.
www:
[8] Grom P. Fire brigade Ljubljana http://www. upvh/ breathing apparates 04.08. 2010.
[9] http://www. interspiro. com/firefighting 15.10. 2010.
[10] http://www. msasaefty.rs 27.01.2011.
1008
STATUS OF MAINTENANCE SERVICE IN COMPANIES IN REGION

IN TIMES OF RECESSION
Stojan Simi1
Abstract: Organization of maintenance is a problem of leading managers in the
company both in industrialized countries and countries in transition. In industrialized
countries, maintenance is viewed from different points, with a view to its continual
improvement and development.
In times of economic crisis and recession, many companies have resorted to
unpopular measures, which are manifested in the reduction of the number of
employees and limit costs in all business processes. The economic crisis and
recession with a smaller or greater intensity affected the entire world economy and
could not pass by any countries in the region, as well as companies in those countries.
This state of the economy had a major effect on the organization of the maintenance
service in companies in most countries of the region.
The aim of this paper is to objectively examine and realistically show the position of
maintenance services in companies in the countries of the region during the period of
the economic crisis and recession, with proposals for the improvement of this position
in the next period.
Key words: maintenance, recession, countries in the region.
1.
INTRODUCTION
In the eighties of last century, began the process of restructuring the form of
international trade. This process is continued in the first decade of this century.
Classical forms of foreign trade flows, such as regular exports and imports of goods,
giving way to the international production of which the holders of transnational
companies. The consequence of these processes is the growth of the role of
transnational corporations in the economies of both home countries and the countries
they come to foreign investment. It is proven that the impact of transnational
corporations leading the process of internationalization and globalization, as well as
the creation of internal inter-company market.
The processes of globalization, international regional integration and
institutionalization of international trade have influenced the majority of countries in
transition adopt liberal laws on foreign investments and thus encourage foreign
1
Dr Stojan Simi, Modrica, Oil Refinery s.c., (stojans@modricaoil.com)
1009
Stojan Simi
investment in its own economy.

Bearing in mind the economic power of transnational companies that sell their
investments abroad, especially in countries in transition, in addition to the advantages
that these countries can bring in terms of additional employment of local people,
construction of new and renewal of existing infrastructure, expanding markets and
economic growth, the company can cause big problems in the economy of these
countries. Risk of negative impacts of transnational companies is particularly evident in
terms of general liberalization of society.
In the last twenty years, the countries in the region are faced with an
aggressive penetration of foreign transnational companies. The main objective of the
transnational companies that are strategic partners of companies in the region is to
ensure stable production and the greater financial gain. The first step was the
reduction of unnecessary costs that are primarily related to organizational units that are
the logistics of the production process. This is the more or less reflected the position
and role in the maintenance service to companies in these countries.
The economic crisis and recession that with smaller or greater intensity
affected the entire world economy and even led to the disappearance of many
companies with the economic map of the world, could not pass from either company in
the region. Of course, this situation in the last three years a greater or lesser extent
influenced the position of maintenance service to businesses of these countries.
The paper tried to look objectively and realistically show the position of
maintenance services companies in the countries of the region during the economic
crisis and recession with concrete proposals to that position in the period to improve.
2.
POSITION MAINTENANCE SERVICES IN INDUSTRIALIZED COUNTRIES IN

THE PERIOD OF ECONOMIC CRISIS AND RECESION
Previous two years will be remembered as the biggest world economic crisis
and recession in the last fifty years. The crisis caused by many factors, particularly felt
in the industry. Some companies have for years been leading the industry no longer
exist or have been before the shutdown. For example, in the automotive industry giants
such as General Motors and Opel have succeeded with a lot of effort to overcome at
least part of the problem in business caused by the economic crisis and recession.
In addition to reducing the volume of production, many companies that would
mitigate the impact of economic crisis and recession have resorted to the following
measures:
reduce the budget for advertising and advertising,
temporary stop development projects,
abandonment of the overhaul and repair performance only equipment
necessary for the performance of the production process,
temporary or permanent layoff, etc..
Like many times turned out to be the individual processes can not save, and
this certainly includes the maintenance.
In industrialized countries results from a large number of studies that were
conducted in the last ten years have shown that over 40% of companies that
manufacture, have no access to tasks formulated in vehicle maintenance. This data
suggests that a significant number of companies do not possess even the most basic
1010
Status of Maintenance Service in Companies in Region in Times of Recession
strategy, and that maintenance is applied indiscriminately or only in response. Even

when it was applied some form of monitoring of technical systems, only few companies
had a clear strategy to manage the maintenance associated with managing the
production. As a consequence, in many companies, part or full maintenance service
stayed temporarily or permanently lost their jobs. According to available data, the total
number of workers in the world who lost his job due to economic crisis and recession in
20 to 35% of the workers employed in maintenance. The number of workers holding
that remained out of work temporarily or permanently dependent on the development
of certain regions, type of industry and others.
3.
POSITION MAINENTANCE SERVICE IN THE COUNTRIES REGION DURING

THE ECONOMIC CRISIS AND RECESSION
Organization of maintenance is a problem of leading managers in the

company, on the one hand, and on the other and themselves on maintenance, both in
industrialized countries and countries in the region. Maintenance organization is a
complex problem that must always be approached from several angles.
Maintenance is very demanding, but also a very demanding activity in the
countries in the region. It is estimated that in these countries in maintenance of
technical systems, ranging from manufacturing organizations and across
manufacturing industries andorganizations to provide services, currently employs
about 140 thousand employees.
The current state of the economies of countries in relation to industrialized
countries is very low, which is inevitably reflected in the position of maintenance
service in those countries. When you look at the impact of the global economic crisis
on the economies of countries in faint, it is clear that the consequences of the
unforeseen and not to circumvent or service maintenance companies in these
countries.
In countries in the region, workers in charge of maintenance in an enterprise,
most remembered only when something bad happens. For example, when the bus
brakes cancel or damage occurs during the machines in the manufacturing process.
But before that, few ever question what is the quality of parts that the company provide
him, or how much money is annually allocated for the maintenance of technical
systems in the said company. Although the situation in these countries significantly
improved the arrival of transnational companies, there is a lot of companies in which no
one has the record of what was last maintained, and who maintains in any way.
The companies in the region where the maintenance department involved in
the organizational structure, as an independent unit or as a functional unit of the
production process, the rule is that the maintenance service scapegoats for all the
problems and delays that result in the production process. However, when top
managers tackled the problems of efficiency of the production process, come to learn
that the availability of working funds specifically responsible maintainers. It learns them
causes the maintenance department is a necessary evil in the company, which makes
the cost too high, but extremely important part of the production process. If top
managers want their company is competitive on both domestic and foreign markets,
they must fundamentally reconstruct the existing business or acquire new assets.
Thus, automated work tools requiring highly educated maintenance and higher level of
1011
Stojan Simi
service quality. It can carry a young and highly educated persons, who must in addition
have several years of practical experience. Today's structure of holding companies in
the countries in the region is mostly composed of personnel from the correspondence
school with an average age of over fifty years.
In order to improve efficiency and effectiveness of maintenance in industrial
countries for a long period in the application of appropriate information technology,
which is not the case with countries in transition. Information technology in the system
maintenance is a necessity, a lifeline enterprises, and today almost imperative. Only
few companies in the maintenance systems in the countries in the region benefit from
the information technology, something many uses obsolete, outdated and unreliable
information systems.
In industrialized countries point vehicle maintenance is a preventiv eaction,
which should be pursued in the less developed countries undergoing the transition
phase.
Technical secondary schools and technical colleges in the region at the turn of
this century have increasingly lost their popularity. This resulted in the production of a
decreasing number of employees in particular quality metal and electrical field. State of
the economy, has been fully completed implementation of the Bologna process in
higher education, inadequate laboratories and outdated laboratory equipment have
influenced the decrease of interest in engineering schools in the past. In addition, in
the course of study at technical universities, students are given very little information
about the importance of vehicle maintenance. Until recently, universities in the
Republic of Srpska during the regular study of any one subject was not fully committed
to maintaining technical systems, but the students with this problem very briefly and
superficially acquainted through other subjects. According to statistics in the region
from the total number of employed engineers in the economy of them from 40 to 60%
have a first jobin maintenance.
In recent years increased interest in vocational technical schools and technical
colleges. The relevant state institutions in the region need to harmonize the services in
the economy for the next period necessary qualification structure of employees in all
processes, including the maintenance process. States in coordination with enterprises
that are owned by transnational companies should be awarded scholarships
occupations in the economy.
In order to improve efficiency and effectiveness of maintenance needs in the
shortest period of time to rejuvenate these services with appropriate professional staff
profiles. Workers who start to finish left a couple of years, and workers whose health is
damaged companies should retire under a special social program or insome other way
to relieve the place for young professional staff.
Workers engaged in maintenance should provide professional development
and procurement of equipment and facilities to properly and efficiently able to perform
diagnostics in order to extend the service life of technical systems.
The organizational structure of maintenance service for most companies in the
region, with no significant changes, remained as before, more than twenty years.
Development and application of information technology to advance in this area in the
future is relatively much space. For example, one can use information systems to
establish direct contact with the warehouse and thus significantly affect the quality
monitoring of spare parts. By applying information systems to achieve faster movement
of technical documentation, better control of spare parts spending and in terms of
1012
Status of Maintenance Service in Companies in Region in Times of Recession
placement. Simpler and more to keep track of completed interventions on devices and
equipment. This would favor the concept of a planned preventative maintenance
system.
Transnational companies that are present in this region as majority owners of
domestic companies should in the future to bring one of the strategic decisions and this
is how they will perform maintenance of equipment. There are two different concepts,
namely:
have modern and efficient maintenance service within the company,
maintenance entrusted to specialized companies to perform these activities
with control and supervision.
In industrial countries today are almost equally represented both the concept
of maintenance. In countries in the region is relatively small number of companies with
technical and staff trained to provide maintenance services. Particularly those
companies with scarce professional personnel for the maintenance of rotating
equipment (turbines, compressors, blowers, pumps, etc.).. To maintain the
measurement and control equipment in these areas almost no one more specialized
company. Therefore, much more realistic option is that companies in the region in the
coming period in its organizational structure with modern and efficient maintenance
service.
It should be noted that management must be sensitive to the role of
maintaining the system of business. Key role in terms of maintenance should have
people in the profession, which, among other things, must be involved and consulted in
the selection of equipment, spare parts and raw materials.
4.
CONCLUSION
The global economic crisis with a smaller or greater intensity affected all
economic sectors in the region. Of course, this situation manifested itself on
manufacturing companies.
This paper reviews the current position of maintenance service and are given
concrete proposals to that position in the future on a higher level. In order to have
better quality and more efficient maintenance system needs to companies in the region
in future focus its activities in the following directions:
to make a strategic decision whether to run their own maintenance service
and will hire a specialized company for this purpose,
implementation of modern information system,
procurement of equipment and facilities for technical diagnostics,
rejuvenation maintenance service with qualified staff,
staff training workers,
scholarship for students needed profile,
implementation of quality management systems,
to use the experience of industrialized countries,
favoring the concept of planning and preventive vehicle maintenance, etc..
1013
Stojan Simi
5.
REFERENCES
[1]
Markovi, S., Tanasijevi, S., Jovii, S., Josifovi, D.: Improving the design
process to maintain quality, 34. National Conference on Quality, Festival quality
2007., Kragujevac, 08-11.05.2007.
*** http://www.pdffactory.com
*** http://www.poskok.info
Simi, S.: Position maintenance service to businesses in the Western Balkans,
Bulletin of the company on maintenance funds for the work of Republic of Srpska,
No 3, September 2008., pp. 25-28.
[2]
[3]
[4]
1014
IDENTIFY CHARACTERISTICS OF OBJECTS IN MACHINE VISION

Remigiusz LABUDZKI1
Summary: The paper presents the results of the analysis of sample images of
machine parts in the machine vision system. Visual inspection devices, including support quality control of mass production or mass, provide ample opportunities for practical applications and tangible benefits in terms of minimizing the number of defective
products. During production, we always deal with errors. They are formed, among others due to the imprecision of tools and machinery used for processing, or defects in the
material used for production, or damage to treated objects.
In order to carry out research designed to test stand to allow the registration of
the images of machine parts in motion, their processing and recognition. The position
will in the future on the following tasks: identify the object (by comparison with the
model or the analysis of the features), measurements of geometrical quantities, the
control texts, reading codes, and printing quality control, presence control (including
counting objects), or position control, control surface condition, the evaluation of color,
sort of.
Pre-registration and image analysis was carried out in the machine vision system consisting of: a white light source, high resolution cameras, I/O interface (GigE),
PC and the LabVIEW (2009) software package. Images of objects illuminated properly
recorded and sent a camera and an interface I/O (GI-GE) to PC. Processing, recognition and image analysis was carried out in the LabVIEW package.
1. INTRODUCTION
The introduction of the automation has revolutionized the manufacturing in
which complex operations have been broken down into simple step-by-step instruction
that can be repeated by a machine. In such a mechanism, the need for the systematic
assembly and inspection have been realized in different manufacturing processes.
These tasks have been usually done by the human workers, but these types of deficiencies have made a machine vision system more attractive. Our expectation from a
visual system is to perform the following operations: the image acquisition and analysis, the recognition of certain features or objects within that image, and the exploitation
and imposition of environmental constraints [1].
1
Ph. D., Remigiusz LABUDZKI, Poland, Poznan University of Technology, Institute of Mechanical Technology
(remigiusz.labudzki@put.poznan.pl)
1015
Remigiusz Labudzki
Scene constraint is the first consideration for the machine vision system. The
situation of the scene must be recognized by the machine vision designer and according to the required application such a machine should be developed. The hardware for
this sub-system consists of the light source for the active imaging, and required optical
systems. Different lighting techniques such as the structured lighting can be used for
such purpose. The process of vision system starts with the image acquiring in which
representation of the image data, image sensing and digitization is accomplished. Image sensing is the next step in order to obtain a proper image from the illuminated
scene. Digitization is the next process in which image capturing and image display are
accomplished. The last step in this process is the image processing in which a more
suitable image is prepared [2].
The first aim of this article is to describe a simple machine vision system that
can be employed in the industrial applications. Second goal is to describe typical examples of the visions systems in the automated manufacturing systems. Finally, it try
to present (selected features): identify the object (by comparison with the model or the
analysis of the features), measurements of geometrical quantities, the control texts,
reading codes, and printing quality control, presence control (including counting objects), or position control, control surface condition, the evaluation of color, sort of using LABView software.
2. SIMPLE MACHINE VISION SYSTEM
The basic structure of simple machine vision system is composed of a camera
and a computer. The camera faces to one of the planar surfaces of a detected object
and takes a picture, then the image is transferred to the computer where it is analyzed
and calculated to get the desired information (fig. 1). The principle it based on is the
homography of two planes in space. In other words a point in a plane of the detected
object and its projection point in image plane is one to one mapping as illustrated in
figure 2.
a)
Camera
b)
Surface of object
Plane of image
Surface of object
Fig. 1 Principle of machine vision system:

a) general view, b) mapping between the two planes [4]
1016
Identify Characteristric of Objects in Machine Vision
Point C in the figure is the center of projection. The mapping is linear when represent in homogeneous coordinates, which can be expressed as follows [3]:
u

s v
l
h1
h4
h7
h2
h5
h8
h3 X

h6 Y
h9 1
X

H Y
1
(1)
Where (u,v) is the coordinate of an image point in image reference coordinate

frames O1-UV, (X, Y) is the coordinates of a point in object plane referenced in object
plane coordinate frames O-XY, s is a scale factor. H is the homographic matrix of this
mapping, which is in fact a matrix composed of internal and external parameters of the
camera.
In most cases additional lighting source is always added to the system in order
to get clear image. The typical structure of the system is illustrated in figure 2.
LED light
Camera
Computer
Analyzed object
Fig. 2 Typical structure of the machine vision system [4]

3.
TYPICAL EXAMPLES OF THE VISIONS SYSTEMS IN THE AUTOMATED

MANUFACTURING SYSTEMS
In order to describe the applications of machine vision systems; four categories

of the visual inspection, process control, parts identification, and robotic guidance and
control mechanisms are considered. In this field, the most significant task of the
machine is for the Automated Visual Inspections (AVIs). The main concern of using
machine is to recognize that the part is well made according to the specified
qualifications. AVI and parts identification do not contribute a significant role in the
flexibility in manufacturing, however have considerable role in the automation task. On
the other hand, vision systems in the process control and robotic guidance can play
important role in achieving more flexibility in manufacturing [5].
3.1 Automated visual inspection
As described, the AVI does not enhance the flexibility of the manufacturing line
because the only use of this system is for inspection, but the utilization of this system
considerably enhances the automation capability of manufacturing process. The automated vision system can be used for the purpose of measurements, gauging, integrity
1017
Remigiusz Labudzki
checking, and quality control. In the area of measurements and gauging, the gauging
of small gaps, measurements of the object dimension, alignment of the components,
and the analysis of crack formation are common applications. Integrity checking in
automotive plants, food industry and other production lines is performed by using such
a vision system.
The medical and pharmacological products can be inspected by the machine
vision systems. Using such an inspection method in the production line has increased
the speed and reliability of the inspections. For example, during the automotive assembly, a vision guided robot recognizes the orientation of the engine heads and picks
and places them correctly on the engine blocks [6]. In another case, a system examines the fiber optics assembly line. A PC-based imaging system integrates hardware
and software to analyze the captured images for the possible fiber blemishes, chips,
and cracks [7]. As an another example, in aerospace industry, a vision-based robot
using the self-calibrating and self-teaching techniques has been reported that punches
rivets into the airplane metal sheets with the high accuracy [6]. A list of possible applications of AVI for measurements and gauging and quality controls are given in Table 1.
Table 1 Possible operations for automated vision inspection
Measurement and gauging
Gauging of spark gap
Measurement of belt width
Measurement of tool wear
Mark alignment and pick and place
component offset
Analysis of crack formation and
propadation
Calibration and inspection for

automation
Robot and vision system assemble
diagnostic devices, inspects and
grades rough wooden boards
Integrity checking and quality control

Automotive plants (enhances inspection of automobile parts,
safety inspection of brake assemblies, valve spring, piston rings
and rods, confirm switch integrity, Check auto axle assemblies,
asses oil-seal surfaces, inspect truck beds, check laser welds,
detects enginefuel leakage)
Food industry (classify fruits, grades moving products)
Checking correct printing pharmaceuticals labels
Paint finish assesment
Production lines (inspection of machined surfaces, roughness,
flatness edge damage and nicks, inspect bares acrylic-sheet
defects, inspects indexable insets, cutting tools, realtime webinspection system finds steel defect, steers package inspection,
automates glass-machine-tool inspection, network cameras check
components)
Electronic and electro-optic industry (inspection of printed circuit
boards, microcircuits, fiber optics connectors, semiconductor
inspection, elimination of surface defects in photographic film
webs)
Medical and pharmaceutical plants (validates medical containers,
inspect tubes for proper filling and sealing, ensures medical pill
qualities, ensures product and packaging quality)
3.2 Process control

Utilizing a vision system could help a better analysis, control, tracking, and the
issuing documents in different applications. Applying vision system also can help the
analysis of the nozzle-plates, and monitoring production process. It also provides online inspection and imaging options for the biomedical, pharmaceutical, metal finishing,
lumber production, and also in the automotive production assembly lines. Successful
operations of the vision system for the process control and tracking tasks are [2]:
speeds food processing and highspeed sorting,
1018
applying vision to nozzle-plate analysis,

fiber analysis in the wood panel industry,
event-capture system monitor paper-making process for breaks and defects,
control of flatness in float-glass manufacturing,
on-line inspection ensures needle quality,
checks display pixels
imaging options from biomedical, pharmaceutical to metal finishing and
lumber production,
shortens fiber-alignment time,
issuing documents (driver licenses, credit cards),
tracks pill bottles,
automates catheter manufacture CMOS and 3-D display offers crystal ball
viewing,
orients credit cards,
smart cameras check fast moving connector pins, smart sensors automatically perform multiple inspections of piston rings and rods,
robot/vision add flexibility to auto-part manufacturing (layered bin picking),
tracks targets on missile flights (measuring the accuracy),
inspection and alignment of wafers.
3.3 Parts identification
Parts identification and classification are one of the most important applications
of a vision system. Sorting of the automotive castings, parts, and identifying and unloading of parts from pallets are important applications. Sorting and grading of the food
and other products are another example of such identification applications. Some typical applicational examples for the identification and classifications are [2]:
sorting of automotive castings,
unloading materials from matrix packaging,
automobile well-to hob assembly,
identification of car bodies by outline,
automotive tyre-to-wheel assembly,
automatic decoration of chocolates,
precision application of car body seam-sealant,
sorting of fish by species, size determination and inspection,
unloading of automotive crankshafts from pallets,
machine vision and barcode inspection identify fast moving parcels,
identify license plate of moving vehicles.
3.4 Robotic guidance and control
The automation process is based on the different robots, which require guiding
systems. Alignment and adjustment processes also require smart guidance systems.
In general, in the automotive plants guidance of the robotic action is performed by using the vision system as smart sensors for position determination in the welding or
other processes. A list of the possible applications of the vision system for the robot
guidance and control [2]:
1019
Remigiusz Labudzki
4.
automotive windscreen alignment and placement,

high precession part-mating in aerospace applications,
seam location and following for welding car chassis members,
pattern-correct sewing in textile manufacture,
acquisition of cylindrical objects from the bin (bin picking),
smart vehicles (imaging a targetrich environment),
vision guided nuclear fuel subassembly dismantling.
IDENTIFY CHARACTERISTICS OF OBJECTS IN LABVIEW PACKAGE
Vision Assistant (part of LabVIEW package) is a tool for prototyping and testing
image processing applications. To prototype an image processing application, build
custom algorithms with the Vision Assistant scripting feature. The scripting feature
records every step of the processing algorithm. After completing the algorithm, you can
test it on other images to make sure it works. The algorithm is recorded in a script file,
which contains the processing functions and relevant parameters for an algorithm that
can be prototype in Vision Assistant. Using the LabVIEW VI Creation Wizard, we can
create a LabVIEW that performs the prototype that you created in Vision Assistant.
Vision Assistant offers the following features (fig. 3):
Script window records a series of image processing steps and the settings you use for each of those steps. You can run scripts on single images
or in a batch to analyze a collection of images. You also can modify and
save scripts,
Image Browser contains all of the images currently loaded in Vision Assistant. You can select an image to process by double-clicking it in the Image Browser,
Processing Window updates the image as you change parameters. Because this view immediately reflects the changes you have made in the
Setup window, you can continue modifying parameters until you get the result you want,
Processing Functions Window/Setup Window displays a list of image
processing functions you can use to develop an algorithm, or displays parameters that you can set for an image processing function. Each function
available through the Processing Functions window has a Setup window in
which you set the parameters for that function,
Reference Window displays the image source as you manipulate it in the
Processing window,
Solution Wizard displays a list of industries and corresponding qualityassurance tasks that those industries perform. The wizard loads an NI Vision-based solution for the task you select,
Performance Meter Estimates how long a script will take to complete on
a given image,
LabVIEW VI Creation creates a LabVIEW VI corresponding to the algorithm you prototype in Vision Assistant. Based on the options you select, the
LabVIEW VI Creation Wizard creates a new VI that implements the image
processing steps of the current script or of a saved script file,
1020
C Code Creation creates a C file corresponding to the algorithm you prototype in Vision Assistant. Based on the options you select, the C Code
Creation Wizard creates a C function that implements the image processing
steps of the current script,
Builder File ASCII text file that lists the Visual Basic functions and parameters for the algorithm you prototyped in Vision Assistant.
Reference Window
Image Size
Zoom Ratio
Processing Window
Fig. 3 Processing an image in NI Vision Assistant [8]

4.1 Processing functions: machine vision
This palette groups several image functions that perform the following common
machine vision inspection tasks:
detecting the presence or absence of parts in an image,
measuring the dimension of parts to determine if they meet specifications,
locating objects in an image.
Machine vision feature in NI Vision Assistant allows to analyzing the image
using the following functions (fig. 4):
Edge Detector, this function let to find edges along a line that you draw
with the Line Tool from the toolbar,
1021
Remigiusz Labudzki
Straight Edge (Rake), this function let to find points within the edge of
an object and fit a line to those points describing the edge,
Advanced Straight Edge, this function let to find edges by configuring
detection settings,
Circular Edge (Spoke), this function let to locate the intersection points
between a set of search lines within a circular area (annulus) and find
the best fit circle,
Fig. 4 Processing functions of Machine Vision in NI Vision Assistant

Clamp, this function let to find edges within a rectangular region drawn in
the image and measures the distance between the first and last edge,
Pattern Matching, this function let to locate regions of a grayscale image
that match a predetermined template. Pattern Matching can find template
matches regardless of poor lighting, blur, noise, shifting of the template, and
rotation of the template,
Geometric Matching, this function let to locate regions in a grayscale image that match a predetermined template. Geometric matching is designed
to locate templates characterized by distinct geometric or shape information. The function can find template matches regardless of lighting variation, blur, noise, occlusion, and geometric transformations such as shifting,
rotation, or scaling of the template,
Shape Detection, this function let to locate circles, ellipses, rectangles, or
lines that meet the specified geometric criteria,
Golden Template Comparison, this function let to compare areas of an
image to a learned template and return the differences found in the image,
1022
Caliper, this function let to compute measurementssuch as distances, areas, and anglesbased on results returned from other machine vision and
image processing functions.
5.
CONCLUSION
In the design and operation of a vision system, the image formation and visual
process, computational methods and algorithms, depth information, image representation, and modeling and matching must be considered. On the other hand, the systematic consideration is important in the efficiency and the performance of the selected
machine. The integration possibility, robustness, ease of operation, and adding intelligence into the system in order to make it a smart system are features of the advanced
machine vision systems. The universal capability, PC requirement (self contained), offthe-shelf hardware, connectivity and I/O control options are the key factors in this respect. For any production line and manufacture plant, there should be a good reasoning for utilizing such a machine vision system using NI Vision Assistant a part of LABView package.
LITERATURE
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Awcock G.J., Thomas R. (1995). Applied image processing. London. MacMillan

New Press Ltd.
Golnabia H, Asadpour A. (2007). Design and application of industrial machine
vision systems. Robotics and Computer-Integrated Manufacturing, vol 23, p. 630
637.
Criminisi A. (2001). Accurate Visual Metrology from Single and Multiple Uncalibrated Images. University of Oxford.
He Junji, Shi Li, Xiao Jianli, Cheng Jun, Zhu Ying. (2010). Size Detection of Firebricks Based on Machine Vision Technology. In materials of 2010 International
Conference on Measuring Technology and Mechatronics Automation, p. 394-397.
Edwards J. (1990) Machine vision and its integration with CIM systems in the
electronic manufacturing industry. Comput Aided Eng J. p. 1218.
Spotlight on robotics, Vision System Design, (2003).
Spotlight on robotics, Vision System Design, (2002).
Manual User NI Vision Assistant 2009.
1023

DEMI 2011 Proceedings

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

DEMI 2011 Proceedings

Caricato da

Copyright:

Formati disponibili

University of Banja Luka

Faculty of Mechanical Engineering

10th Anniversary International Conference

BANJA LUKA, May 2011.

University of Banja Luka

BANJA LUKA, May 2011.

PROCEEDINGS OF THE 10th ANNIVERSARY INTERNATIONAL CONFERENCE ON

A. MECHANICS AND DESIGN ................................................................................... 57

Michail Leparov, Georgi Dinev, Marieta Jancheva

Sran Bonjak, Zoran Petkovi, Milo orevi, Neboja Gnjatovi,

B. PRODUCTION TECHNOLOGIES AND ENGINEERING ...................................... 229

46. Aurel Prsti, Zagorka Acimovi-Pavlovi, Zvonko Guliija, Mirjana Stojanovi

63. Stevo Borojevi, Vid Jovievi, Gordana Globoki-Laki, ore ia,

76. Milica Grahovac

Mile Raievi, Miroslav Demi, Neboja Rako, Predrag Milenkovi

109. Branislav Aleksandrovi, Rajko Radonji, Marko apan, Aleksandra Jankovi

139. Milorad Panteli, Sran Bonjak

On behalf of the Organizational Committee of the DEMI 2011Conference, I would like

Banja Luka, 12 May 2011

President of the Organizational Committee

SUSTAINABLE INNOVATION AND INNOVATION FOR

Prof. Dr. Ing.Claudio R. Bor, Director ICIMSI-SUPSI, Manno-Lugano, Switzerland

COOPERATION MODELS BETWEEN UNIVERSITIES AND

TRANSPORT ECOLOGY AND GLOBAL CLIMATE CHANGE

Pei R., Veinovi S.

Fig. 1 Global temperatures

Ecology of transport and global change earths climate

Fig. 2 Climate disaster: Cause is the sun!

Fig. 3 Annual mean surface

Fig. 4 Is it Global Cooling or Global

Pei R., Veinovi S.

Fig. 5 Past and future Milankovic cycles [3,4]

Ecology of transport and global change earths climate

Pei R., Veinovi S.

Ecology of transport and global change earths climate

Pei R., Veinovi S.

Fig. 6 Biofuels are carbon neutral

Ecology of transport and global change earths climate

3.4 Environment and sustainability

Pei R., Veinovi S.

The term transport ecology or ecology in transport implies a complex balancing

Fuel and lubricants

Fig. 7 World companies offer other metal additives

Ecology of transport and global change earths climate

It is absurd to make partial and unrealistic regulations on toxic, PM and CO2

Fig. 8 Unrealized European Forecasts

Fig. 9 Regulations on reducing toxic

Pei R., Veinovi S.

Fig. 10a The high-speed, acceleration and

Fig. 11 New motor technology enables the use of a combination of fuel as a

Ecology of transport and global change earths climate

Pei R., Veinovi S.

Robinson A., et al., (2007.) Environmental Effects of Increased Atmospheric

DESIGN CONSTRAINTS AND ROBUST DESIGN AS THE MODERN

horizontal integration. Vertical integration considers development of product

Fig. 1 Horizontal integration of product development and design areas

Fig. 2 Power Transmission Systems Design PTSD model

solution and to every design component. Limitation Analytic Hierarchy Processing

Fig. 3 Relations in DPD module

Fig. 4 Example of DP minimization and variation

The axiom of independency is conditionally fulfilled. The members of matrix [C]

c11 V Hdes Tmax

Teeth wear (failure) probability PFi is presented by Weibulls functions, where

3. RELIABILITY AS DESIGN CONSTRAINT