Sei sulla pagina 1di 224

EFFECTS OF BOLT TIGHTNESS ON

THE BEHAVIOUR OF MERO-TYPE


DOUBLE LAYER GRIDS

MOHAMMAD REZA DA VOODI

THESIS SUBMITTED FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

CIVIL ENGINEERING

SCHOOL OF ENGINEERING

UNIVERSITY OF SURREY

NOVEMBER 2004
ABSTRACT

Double layer grids constitute an important family of space structures. Several industrialised

structural systems have been developed and applied for the construction of prefabricated
double layer grids and one of the most popular of these is the MERO-type system. A

double layer grid constructed by this system involves a large number of bolts and each bolt

also involves different degrees of tightness. The study of the effects of bolt tightness on the

behaviour of this type of double layer grid is the objective of the present research.

Due to the presence of a large number of bolts with various degrees of tightness and the

complexity in the load transfer mechanism of the bolted connectors, a purely analytical

approach for investigation is not suitable. Also, small-scale models cannot reflect the

influence of the bolt tightness on the behaviour of double layer grids in real-life situations.
Therefore, the present work is based on an experimental study which was carried out using

a full-scale structure.

The experimentalwork was carried out on a 10m by 10m double layer grid constructedby
that
components are normally used in practice. The grid was constructedin the University

of Mazandaranin Iran. The grid has 392 elements, 113 nodes and 12 possible support
positions. A number of different experiments were carried out involving two grid
configurationsand three different support arrangements.Since the main emphasiswas on
the study of the effects of bolt tightness,different levels of bolt tightnesswere testedfor
In
every experimentalcase. eachcase,all the bolts were tightened to a specific degreeof
tightness.Then, a concentratedload was applied at the central node of the upper layer and
the deflectionsof a numberof bottom layer nodeswere measured.

The load-deflectionresponsesof the test grids show that the degreeof bolt tightnesshas a
significant effect on the behaviour of the grids. The results also indicate that there are
major disagreementsbetweenthe actual and theoretical deflections of the grids. In fact, in
all the cases the experimental results are greater than those obtained from analysis.
Furthermore,experimental studies show that the load-deflection responsesof a MERO-
type double layer grid lie in a region whose upper and lower limits depend on the
configuration,loading and supportconditions of the grid.

2
ji

ýý ý..
.

:ý ýý

frijS JiA

1ý1-%,
4V* 4 9ý

I,, od.ilyLi"

jj
J1J
__
S..

3
ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to my supervisor Professor H. Nooshin for

all his help, advice and enthusiasm throughout my research studies.

Also, I would like to thank my supervisors Dr. P. Disney and Dr. M. Gorji for their

guidance and support.

I would also like to thank my colleaguesin the School of Engineeringof the University of
Mazandaran,Babol, for their assistancein performing the experiments.

I am grateful to the SepahIndustrial Factory in Iran for manufacturingthe test model of


this research.

I am deeply indebted to my parents and my wife for their support throughout the years.

The financial assistanceof the Ministry of Science, Researchand Technology of the


Islamic Republic of Iran and the help of the University of Mazandaranin supporting my

researchstudiesare gratefully acknowledged.

4
CONTENTS

ABSTRACT 2
.......................................................................................................................
ACKNOWLEDGEMENTS 4
.........................................................................................

CHAPTER 1-INTRODUCTION

1.1 Introduction 9
.....................................................................................

1.2 Double layer grids 10


............................................................................
1.3 The MERO system 12
...........................................................................

1.4 Bolts 14
...........................................................................................

1.5 Previouswork 17
.................................................................................

1.6 Organisationof the Thesis 21


...................................................................

CHAPTER 2-TEST SET-UP

2.1 Introduction 23
....................................................................................
2.2 The experimentaldouble layer grid 24
.........................................................
2.3 Grid geometry 24
.................................................................................
2.3.1 Model A 25
.......................................................................................

2.3.2 Model B 26
.......................................................................................

2.4 Manufactureof the components 27


............................................................

5
2.4.1 Material properties 30
...........................................................................

2.5 Assemblyof the grid 31


.........................................................................

2.5.1 Weight measurement 32


........................................................................
2.5.2 Assembly process 32
............................................................................

2.6 Foundation 33
.....................................................................................
2.6.1 Formwork of foundation 34
....................................................................
2.6.2 Steel reinforcements 35
.........................................................................

2.6.3 Concreting of foundation 38


....................................................................

2.6.4 Formwork for columns 39


......................................................................
2.6.5 Concreting of columns 39
......................................................................

2.7 Placementof the grid 40


.........................................................................
2.8 Roller supports 43
................................................................................

2.8.1 Hardening of support plates 44


................................................................

2.8.2 Hardnesstesting 48
..............................................................................

2.8.3 Installation of supportassemblies 51


.........................................................

2.9 Loading frame 52


.................................................................................

2.9.1 Cross-footing 53
.................................................................................
2.9.2 Vertical elements 54
.............................................................................
2.9.3 The top part of loading frame 54
...............................................................
2.10 Loading system 54
................................................................................
2.11 Measurementsupportframe 57
.................................................................
2.12 Data acquisition system 57
......................................................................
2.12.1 Displacementtransducers 59
...................................................................
2.12.2 Load cell 61
.......................................................................................

6
2.12.3 Data logger 62
....................................................................................

CHAPTER 3- THE TESTS

3.1 Introduction 67
....................................................................................

3.2 Test grid A 67


.....................................................................................
3.2.1 Setting up of grid A with normal tightness 69
...............................................

3.2.2 Testing of grid A with normal tightness 72


..................................................

3.2.3 Test results 73


....................................................................................
3.2.4 Settingup of grid A with specifieddegreesof tightness 78
................................
3.2.5 Testing of grid A with specifieddegreesof tightness 82
...................................
3.2.6 Test results 83
....................................................................................
3.3 Test grid B 90
.....................................................................................
3.3.1 Testing of grids B4, B8 and B12 93
...........................................................

3.3.2 Test resultsof grid B 95


........................................................................

CHAPTER 4-PRESENTATION OF TEST RESULTS

4.1 Introduction 133


..................................................................................
4.2 Analysis of the test grids with pin and rigid joints 133
......................................
4.3 Presentationof theoreticaland experimentalresultsof grid A 136
........................
4.4 Discussionof the results of grid B 141
........................................................
4.5 Analysis of grid B with considerationof the effects of the connectors 147
..............
4.5.1 Geometryof grid Band joint type 148
........................................................
4.5.2 Supportarrangementsand supporttypes 148
................................................

7
4.5.3 Modelling of elementswith considerationof the connectors....................... 149

4.5.4 Loading and analysis 150


.....................................................................

CHAPTER 5-CONCLUSIONS

5.1 Introduction 156


..................................................................................

5.2 Discussion 157


....................................................................................

5.3 Conclusions 161


..................................................................................

5.4 Recommendationfor further research 163


....................................................

APPENDIX- NUMERICAL RESULTS OF THE TESTS

A. 1 Introduction 164
.................................................................................

A. 2 Tables 164
........................................................................................

A. 3 Numerical results 165


...........................................................................

REFERENCES 222
.................................................................................

8
CHAPTER 1

INTRODUCTION

1.1 Introduction
Double layer grids are an important family of spacestructuresused to cover large spaces.
For easeof handling, flexibility in fabrication and speedyon-site erection,thesestructures

are often prefabricated.The componentsof prefabricateddouble layer grids are normally


mass produced in factory with a high standardof quality control. Several industrialised
systemshave been developedand applied in constructionof thesestructuresand the most
popular of these are the MERO-type systems. Such a system consists of tubular
componentsthat are connectedto sphericalnodesby meansof bolts.

In a double layer grid constructedby a MERO-type system,each bolt can have different
degreesof tightness.This degreeof tightness the behaviour the double layer
will affect of
grid. Therefore, the actual responseof such a double layer grid cannot be determined
precisely by using a purely theoretical approach.A searchof the literature revealsthat the
studiesrelated to the effects of bolt tightness on the behaviour of double layer grids are

9
rather limited. Nevertheless,a few experimental studies on the MERO-type connectors
show that the degree of bolt tightness plays an important role on the moment-rotation
relationshipof the connectors [3,7 and 19]. However, no study of the behaviourof double
layer grids influencedby the degreesof bolt tightnessis found in the literature.

1.2 Double layer grids

A double layer grid consists of two parallel layers of elements that are connected together

with web elements. From the configuration point of view, these two layers may or may not
have identical patterns. Also, the two layers may be placed exactly one above the other or

at an offset with respect to each other. Double layer grids where two identical parallel
layers of them are placed exactly above each other are referred to as "truss grids". Other
double layer grids with their two layers placed at an offset with respect to each other are

referred to as "offset grids".

A wide variety of configurations is used for truss as well as offset double layer grids. The

plans and perspective views of the three most commonly used offset double layer grids are
shown in Fig 1.1. In this figure the top layer elements, bottom layer elements and web
elements in the plans of the grids are drawn by thick, thin and dashed lines, respectively.
The double layer grid of Fig 1.1a consists of a binate (a simple two-way chequered pattern)

top layer and a binate bottom layer. A configuration of this type is referred to as "binate on
binate". The double layer grid of Fig I. lb is created by removing four nodes in the bottom

layer which results in the elimination of 16 web elements and 16 bottom layer elements. A

configuration of this type is referred to as "reduced binate on binate". In the case of the

grid of Fig 1.1c, both the top and bottom layers have an obnate pattern (a simple diagonal

pattern). A configuration of this type is referred to as "obnate on obnate".

There are also many double layer grids with a binate pattern for one of the layers and an
obnatepattern for the other layer. One of thesetypes of double layer grids is shown in Fig
1.2. In this figure the top layer elementsand the bottom layer elementsof the plan are

shown by thick and thin lines respectively.This double layer grid is "obnate on binate",
that is, it has an obnatetop layer and a binate bottom layer. Nooshin [16], Makowski [14]
and Hadker [9] presenteda wide variety of configurations for double layer grids that can
be used in practice. Also, some types of these grids are discussedin "Chinese Code of
Practicefor the Design and Constructionof SpaceTrusses"[5].

10
I',

(a) A binate on binate grid

/ f / f / \ / f I \ / \ / \

7 7

)V A T V
/ \ / \I \ / f / \/ \ / \

/\ /\ /f /\ /\ /\ /\

(b) A reducedbinate on binate grid

(c) An obnateon obnategrid

Fig 1.1plans and perspectiveviews of three offset double layer grids

11
An important characteristic of double layer grids is that the main internal forces in their

elements are axial forces. Bending moments, shear forces and torques are also present in

the elements in various proportions depending on the cross-sectional properties of the

elements and the jointing system. However, the non-axial force effects in these structures
are normally secondary. Because of this, the analysis of double layer grids is normally
simplified by using the assumption that the connectors behave as pinned joints.

Fig 1.2 plan and perspectiveview of an obnateon binate double layer grid

Most double layer grids are prefabricated. So, the componentsof these structures are

normally massproducedin factory with a high level of quality control. The connectoris
the most important part of a prefabricated system and the commercial successof any
systemrelies directly on its effectivenessand simplicity. One of the most popular types of
connectorsthat is widely usedin the constructionof double layer grids is the MERO-type
system.

1.3 The MERO system

The MERO jointing system is one of the


oldest proprietary spaceframe jointing systems
available,having beendevelopedin Germanyin 1940[1]. This systemis a multidirectional
systemallowing up to fourteen membersto be connectedtogether at various anglesto a
spherical node. The jointing system is suitable for both single and multilayer space
structuresof various geometriesincluding thosewith curved and irregular shapes.Consider
Fig 1.3awhere four tubular elementsare connectedtogetherby meansof a MERO jointing

12
system. The details of the connector are shown in Figs 1.3b and 1.3c. The constituent parts

of the system are:

9A forged steel ball.

0A forged steel conical end piece,

0A high tensile bolt.

0A sleeve with a window and

"A dowel pin.

Steel ball High tensile bolt


Tubular elements

(b )Section AA

Dowel pin Window

a) General view of four


connected elements
Sleeve Conical end piece

(C) Outside view

Fig 1.3 MERO jointing system

The ball is located at the intersection of the longitudinal axes of tubular elements. The

conical end pieces are welded to the end of the tube element. A high tensile bolt passes
through the conical end piece and is screwed into the ball by means of a sleeve. A dowel

pin is used to constrain the bolt to the sleeve in order to allow the turning of the bolt. There
is a window on the sleeve that allows the movement of the dowel pin and indicates the

penetration of the bolt in the ball.

13
The interaction between the tubular elementsand the connectorsin a structure usually
involves tensile, compressiveand bending effects. The tension in a memberis transferred
to the ball through the conical end piece and the bolt, as shown in Fig 1.4a.In this casethe
bolt is undertensionand the sleeveis inactive.

A compressiveforce in a tubular member is transferred to the ball by direct bearing


through the conical end piece and the sleeve,as shown in Fig 1.4b. In this case,the sleeve
is undercompressionand the bolt is inactive.

P4 I-- 0__I] III -p

(a) Connectorunder tension

P II II 41 P

(b) Connectorunder compression

Mý.. M

Gap

(c) Connectorin bending

Fig 1.4 Load transferin MERO jointing system

Under the action of a saggingbending moment the


connectorassumesthe deformedshape
shown in Fig 1.4c. Becauseof the bending deformation, the lower part of the sleeve is
separatedfrom the ball and the conical end piece. Thus, the lower part of the sleevedoes
not contribute to the transmissionof the bending moment due to the gaps.Therefore,the
bendingmomentis transmittedby the bolt and the upper
part of the sleeve.

14
The three casesof load transfer shown in Fig 1.4 are idealised forms. In practice, the loads

will be either a combination of a tensile force together with some bending or a combination
of a compressive force together with some bending. Furthermore, experimental studies
show that the degree of tightness of sleeves has a significant effect on the connection
behaviour [3 and 19].

1.4 Bolts

The main objective of the present research is the study of the effects of bolt tightness on
the behaviour of MERO-type double layer girds. These bolts are made of high tensile alloy

steel. The terminology for a bolt thread is illustrated in Fig 1.5. The pitch is the distance
between adjacent thread crests measured parallel to the thread axis. The major diameter d
is the largest diameter of the bolt thread. The minor diameter dr is the smallest diameter in

the bolt thread. The mean value of the major and minor diameter is the mean diameter, as

shown in the figure. The tension produced by tension load on the bolt may be found using
dividing the force by an area which is called as tensile area. The tensile area has a diameter

equal to the mean of the pitch diameter and minor diameter where according to ANSI
B1.1-1974, the minor diameter was found from the equation dr =d-1.226869p,and the
pitch diameterfrom dm=d-0.649519p.

Major diameter
Mean diameter
r-- Minor diameter

H H- Pitch p

45° Chamfer

il

1
Root --ý Crest ! dread angle 20t = 60°

Fig 1.5 Terminology for a bolt thread

15
In the Metric standardthe characterM followed by a number is being usedto call a bolt or

nut. For example: M12 indicates a bolt which has a major diameter of 12 mm. The

mechanicalproperty classesof the bolts are given in Table 1.1.

Table 1.1 Mechanicalproperty classesof bolts

Property Size range, Minimum Minimum Minimum material


classes inclusive proof tensile yield
Strength, Strength, Strength,

M pa M pa M pa
4.6 M5-M36 225 400 240 Low or

medium
carbon

4.8 M1.6-M16 310 420 340 Low or

medium
carbon

5.8 M5-M24 380 520 420 Low or

medium
carbon

8.8 M16-M36 600 830 660 Medium


carbon,Q&T

9.8 M1.6-M16 650 900 720 Medium

carbon,Q&T

10.9 M5-M36 830 1040 940 Low-carbon

marten site,
Q&T

12.9 M1.6-M36 970 1220 1100 Alloy, Q&T

The first column of Table 1.1 indicates the number of the bolt (class property) that is

marked on the headof the bolts. The secondcolumn indicatesthe rangeof major diameter
size in which the bolt is being made.The proof strengthof the bolt is indicated in column
three. According to the ASTM definitions, the proof strengthis the maximum stressthat a

16
bolt can withstand without acquiring a permanentset. The minimum tensile strengthand
the minimum yield strength are given in the fourth and fifth columns, respectively.The
material of the bolt is specifiedin the last column.

1.4.1 Relating bolt torque to bolt tension

Tightening the bolt by applying torque on it causesthe tension in the bolt. The question is:

what is the relation between the applied torque and the tension in the bolt? It is obvious
that the amount of tension produced in the bolt depends on the coefficient of friction
between the surfaces of the engaged threads. JE Shigley discussed the matter in his book
in detail (Shigley J. E. and Mischke C. R., "Mechanical Engineering Design", McGraw
Hill, 2003). He suggested an equation to determine the torque needed to be applied on the
bolt to produce a given tension in the bolt as follows:

T=0.2F; d (1.1)

Where T is the applied torque, F; is the tension in the bolt and d is the major diameterof
the bolt. Regardingthe data presentedby Shigley in his book, although the coefficient of
friction may vary widely, aboveequationcan be usedto obtain a good estimation.

1.5 Previouswork

Experimental and analytical investigations show that connectors in steel structures

normally behavein a semi-rigid manner.In contrastto steel rectangularframe structures,


limited studieshave been carried out on spacestructures.Some studieshave been carried

out to considerthe effects of joints on the stability of shallow single layer lattice domes
and the behaviour of double layer grids. Experimentalresults show that the behaviour of
the connectorscan have significant effects on the responseof the structures.

Murakami [15] studied analytically the elastic behaviour of a single layer latticed dome

with flexible joints under gravity loads. The flexibility of the connectorswas modelled by
a rigid part and rotational spring at the two ends of the elastic beam element. The
behaviourof the rotational springs
was consideredto be linear with respectto the flexural
stiffness of the beam element. The following points can be deduced from Murakami
studieson the elastic buckling load of the dome:

" Flexibility of the connectorhasa significant effect on the behaviourof the dome,

" The behaviourof the structurelies betweenthe pin and rigidly jointed casesand

17
" The variation of the behaviourwith respectto the connectorflexibility dependson
a number of parameterssuch as the dome configuration, slenderness ratio of the
beamelementsand the buckling modes.

Kato et al [10] studied the effects of joint rigidity on the buckling strength of single layer
lattice domes. A single layer dome, with regular three way grids, under vertical loading

was analysed. The jointing system was considered as screw-type ball joint. They found that

the member strength curve could be expressed in terms of a modified generalised

slenderness ratio, which reflects the reduction of elastic buckling loads due to the semi-

rigidity parameters of joints.

Ueki et al [22] studied the buckling behaviour of a number of single layer domes

constructedof memberswith flexible ends. They reported that the end conditions of the
elementshave an important role in buckling strengthof the model.

Yamada et al [24] studied the effect of joint flexibility and loading conditions on the
buckling characteristics of single-layer latticed domes. They reported that the joint

flexibility hasan important role in the behaviourof single layer lattice domes.

Kato et al [11] investigated the buckling behaviour of a semi-rigidly jointed latticed

sphericaldome of single layer. They reportedthat the structuralpropertiesof the connector


havelarge effects on the elastic-plasticbuckling load of single layer latticed domes.

Cheonget al [4] studied how connectorjoint type affects inelastic dynamic behaviour of

single layer latticed dome to


subjected earthquakemotions. Lattice domeswith pin, fixed

and semi-rigid connector joints were used as study models. They concluded that the
bearing capacity of members is reduced by the interaction effects of axial force and
bending moment. This interaction effect is large in casesthat the semi-rigidity of joint is

small.

Fathelbab and McConnel [7] developed an approximate tangent stiffness matrix for a

general spaceframe member. Using their method together with published experimental
results,they have shown the significant influence of joint characteristicson the behaviour
of spacestructures.

Lightfoot and LeMessurier [13] developed a computer program for the linear elastic

connectors. They used a physical model made from scaffolding tubes with flexible

18
couplers to show that the theory and the corresponding computer program correctly
the
predicted elastic instability strengthof such frames.

Swaddiwudhipong et al [21] investigated the behaviour of a MERO-type connector under


bending moment experimentally and analytically. They reported that the predicted
buckling loads of members based on pin-end assumption of members are normally too

conservative.

Rashed et al [17] studied the collapse behaviour of space trusses with a MERO-type
jointing
system. They stated that experimental investigation of joint behaviour and

memberjoint investigation have shown that bolt connectors are able to undergo the large

rotations,which may take placeafter buckling.

Shibata et al [19] investigatedthe effects of joint properties on the ultimate strength of

single layer reticular domes.The domes with the MERO-type jointing systemwere loaded
into the post-buckling region by controlling displacement at the central node. The

experimental results were compared with analytical ones obtained from the finite element
method, considering the semi-rigidity of the jointing system as an elastic-plastic springs.
According to their results, obtained experimentally and analytically, the conclusions drawn

are as follows:

" The joint systemplays an important role on the ultimate strengthof the domes so
that with increment in the bending rigidity of the connectors,the domes obtain a
large ultimate strength.

" The loosenessof connectorscausesa considerabledecreasein the ultimate strength

of shallow domes.

" Studiesare neededon both the buckling analysis and loading test of actual scale
to
specimens, establishing estimation for ultimate strength of single layer lattice
domes.

Kitipomchai et al [12] investigated theoretically the effects of bolt slippage on the


deflection and ultimate strength responseof lattice structures.The results of this study
illustrate that while the slippageof bolts may have someeffects on the deflections,it does

not significantly influence the ultimate strengthof the lattice structures.

19
A considerable discrepancy was found between the theoretical deflections and measured
deflections during the construction of a stadium in Split, Croatia. This stadium that was
built in 1979 has a shallow double layer roof spanning 180 m. It was constructed in MERO

system and when the temporary supports were removed the deflections were measured.
The actual deflections were found to exceed the theoretical predictions by about 70% [2].

Boris Androic studied the matter analytically and experimentally. He suggested that the

main reason for the disagreement was due to the greater softness of joints. He also

proposed an equivalent modulus of elasticity for the analysis that was 90% of the modulus

of elasticity that was employed in analysis of the roof structure. However, although the

analysis results with the assumption of this equivalent modulus of elasticity are closer to

the measureddeflectionsbut they still cannotpredict the actualresponseof the structure.

El-Sheikh [6] studiedthe behaviourof three double layer grids with a MERO-typejointing

system,both experimentallyand analytically. The overall behaviour and failure modesof


these experimental grids have considerable differences with those that conventional
analysissuggests.He suggestedthat the discrepanciesbetweenanalytical and experimental
behavioursof the grids are due to the flexural stiffness (semi-rigidity) of the connectors.
Also, he developeda nonlinearbeam-columnelementwith end springsto model the actual
behaviour of the grid components. Incorporation of these elements in the numerical

analysishas led to better prediction of actual behaviour and strength of the experimental
grids when comparedwith the conventional analysis. However, there were considerable
differences between the experimental results and the semi-rigid analyses for the limit

points of the load-deflectioncurves.

Fathelbab[7] investigatedexperimentally and analytically the behaviour of a single layer

grid dome with a MERO jointing He


system. obtainedthe moment-rotationcharacteristics
of the connectorexperimentally with different degreesof tightness.The moment-rotation
relationships of the connector were obtained three times with three different levels of
tightnessof loose,just tight and completely tight. The experimentalresults showed that,
the degree of tightness of the bolts has a significant effect on the behaviour of the
connector.

Chenaghlou[3] studiedthe effects of the degreeof tightnessof the bolts on the behaviour

of MERO connectorsexperimentally. Four tubular elementswere usedas cantilever beams

that were connectedto the supportsusing MERO connectors.Thesemodels were identical

20
in every respectexcept for the degreeof tightnessof the bolts that were tightenedat four
different levels, namely, very tight, tight, loose and very loose. The natural frequencyof

each model was measuredas the responseof that model using hammer test. The time-

signal responsesof the testedmodels showedthat, the degreeof tightnessof the bolts has a
significant effect on the behaviourof the model.

Saka and Heki [18] studied the effect of joints on the strength of double layer grids. They

estimated the elastic buckling loads of space structures using the slope-deflection equations
for members under axial forces, including both the effects of the rigidities and finite
dimensions of the joints. They found good agreement between the theoretical and

experiments results for octahedral units and a type of double layer grids. They stated that in

many casesthe buckling load of the elastically connected space truss becomes greater than
that of the corresponding pin jointed truss by their effects of the joints. But if the rotational

rigidities of the joints with finite dimensions are too small, the buckling loads become

smaller than the pin jointed case.

Schmidt et al [20] reviewed the results of teststo destructionof twelve double layer grids.
They reported that the characteristicsof the jointing system play a major role in the

responseof the structure.

1.6 Organisationof the Thesis

This Thesis has five chaptersand one appendix.It presentsan experimentalinvestigation

about the effects of bolt tightnessin double layer grids. The main parts of the presentwork
areasfollows:

" Test set-up

" Test procedureand data acquisition

" Data processingand discussionof the test results

The test set-upincluding the constructionof the double layer grids and the arrangementsof
the supportsas well as the loading systemis the subjectof Chapter2. Also, the equipment
usedfor the measurementsof the applied loads and displacementsof the nodesduring the
testsis describedat the end of Chapter2.

21
The test procedure,dataacquisitionprocessand dataprocessingare describedin Chapter3.
Also, the load deflection responsesof the grids are presentedin Chapter3.

Chapter4 coversthe data processingand discussionabout the results.The load deflection

responsesof the test grids with different degreesof bolt tightness are presentedand
discussedin this chapter. These responsesare also compared with numerical results
obtainedthrough linear elastic analysisof the grids.

Chapter 5 is concernedwith the conclusions of the researchand recommendationsfor


future research.

22
CHAPTER 2

TEST SET-UP

2.1 Introduction

Study of the effects of bolt tightnesson the behaviour of MERO-type double layer grids
be
may conveniently consideredunder two headings:

" The performance of the connector influenced by the effects of the degree of
tightnessof the bolts.

" The performance of the structure influenced by the effects of the degree of
tightnessof thebolts.

In the first case,the connectoritself is under study and by using experimentaltesting of an


isolated connectorits behaviour with respectto different degreesof tightnessof the bolts

can be studied. Subsequently,the behaviour of the structurewith the considerationof the


connectoreffects can be determined.

In the secondcase,the behaviourof the structurethat includesthe effects of the bolts with
different degreesof tightness is investigated.This approachis used in the presentwork.
Since small scale models cannot provide useful information regarding the effects of bolt
tightness on the structural behaviour, the tests were carried out on actual double layer

grids. Thesetests were carried out on two grid configurations and three different support
conditions.

This chapterdescribesthe processof setting-upthe tests including the constructionof the


double layer grids and arrangementsof supports and loading system. Also, the data

acquisition systemis describedat the end of this chapter.

23
2.1 The experimental double layer grid

The objective of the present work is to investigate experimentally the effects of the degree

of tightness of the bolts in a double layer grid that is constructed using the MERO-type

system. The experimental work is carried out on a 10 m by 10 m double layer grid. The

double layer grid is constructed in the University of Mazandaran in Iran. A general view of

the constructed double layer grid is shown in Fig 2. I


.

Fig 2.1. A general view of double layer grid

2.3 Grid geometry

In general, the influence of connectors on the behaviour of a structure can be investigated


by experimental and analytical approaches involving:

0 The connector itself or

0 The whole of the structure.

24
In the first approach the connector itself is under study and by using experiments on an
isolated connector its behaviour under load can be investigated. Subsequently, the
behaviour of the structure with the consideration of the connector effects can be
determined analytically [19 and 23]. In the second approach the behaviour of the structure

that includes the connector effects is investigated. It is this approach that has been adopted
in the present work where the behaviour of the experimental grid was investigated for
different levels of tightness of the bolts of the connectors.

The experimental studies were carried out involving two grid configurations. The overall
dimensions of the grids are the same and the grid model B is constructed by removing four
joints and thirty two elements from grid of model A. Some of the factors that have
influenced the choice of the grid geometries are as follows:

" The binate-on-binateoffset pattern has been chosendue to the fact that this type of
is
pattern simple as well as being frequently in
used practice.

" The geometryof the grid hasbeenchosensuchthat the elementsand the connectors

are identical.

" The overall size of the grid had to be limited to 10 m due to the limitation on the

availablespace.

" The geometry of the grid was chosento be symmetric. Also, the grid was tested

under symmetric loading and symmetric support conditions. Therefore, one was
able to check the symmetry of results, providing an indication of the accuracyof
the test results.

" The anticipated structural displacementshad to lie within the capacity of the

availableloading and measuringequipment.

2.3.1 Model A

The dimensions of the grid model A are shown in Fig 2.2. In this figure the top layer

elements are shown by thick lines and the bottom layer elements as well as the web
elementsare shownby thin lines. The centre-to-centredistancebetweenthe top and bottom
layersis 1000nun. Thus, the anglesof the web memberswith the horizontal plane are 45°

and all the membershad the samelength. The grid consistsof 84 top layer elements,196

25
web elementsand 112 bottom layer elements.Theseelementsare connectedtogetherusing
113connectors.

9898 mm (7 divisions@ 1414 mm)

Plan

ON

8484 mm (6 divisions@ 1414mm)

0
0 Elevation

Fig 2.2.Planandelevationof the grid modelA

2.3.2 Model B

The grid model B is shown in Fig 2.3. The overall dimensionsof model B are the sameas
those of model A. The new configuration is createdby removing four nodesin the bottom
layer which results in the removal of 16 web elements and 16 bottom layer
elements.
Model B consistsof 360 elements(84 in the top layer, 180 in the web 96 in the bottom
and
layer). The elementsare connectedtogetherby means 109
of connectors.

26
9898 mm (7 divisions @ 1414mm)

C
0

ýö

E
00

00
C71%

Fig 2.3. Plan of the grid model B

2.4 Manufacture of the components

The investigationof the behaviourof the connectorsis much more complicatedthan that of
the structuralelements.The complexity in the connectorbehaviour is mainly due to:

9 Geometriccomplexity,

" Presenceof parts with different material properties,

" Presenceof contact surfacesand discontinuitiesand

" Coupling effects betweenforces and moments.

A MERO-typejointing systemconsistsof a number of piecesincluding bolts. Thesepieces

are from different materials involving high tensile steel (that is, bolt) and different grades
of mild steel.In this system,discontinuitieswill be inevitable betweenthe connectingparts

27
due to the presence of bolts. The mechanism of load transfer at these discontinuities is

complex and the presence of gaps and prestressing will give rise to more complexity. The
degreesof tightness of bolts affect the width of gaps and prestressing, playing an important

role in the behaviour of the connector and the experimental studies show that the behaviour
of connectors affects the response of a double layer grid. The main objective of this
research is to investigate the effects of the degrees of tightness of connector bolts on the
behaviour of prefabricated double layer grids.

The influence of bolt tightness on the behaviour of the whole of a prefabricated grid can be

studied by experimental approaches involving:

9 Models that are accuratelyconstructedin a laboratory or

" Models whose componentsare manufacturedin a factory and then assembledon

site.

The first approachinvolves the testing of accuratelymanufacturedmodels in laboratory.In


this case,usually due to limitations of the laboratory equipmentas well as space,the model
dimensions are smaller than those in the actual structures but the imperfections are

minimised. Resultsobtainedfrom such experimentsusually provide sufficient information


either to evaluatea proposedmathematicalmodel or to assessthe suitability of different
techniques of analysis. However, laboratory conditions will not accurately reflect the
actualbehaviourof the connectorin real-life situations.

In the secondapproachthe componentsof the structure are manufacturedin a factory and


the assembledstructure will be tested in a suitable place. Since this researchis mainly
based on experimental studies of practical prefabricated double layer grids, the second

approachis more appropriateand this approachhasbeenadoptedin the presentwork.

The componentsof the experimental grid are constructedin a MERO-type system.This

system was manufactured by the Sepah Industrial Factory in Tehran, Iran. Due to the
simplicity of the pattern and the fact that all the grid elementsand the nodeswere identical,
all the componentswere producedin factory with a reasonablestandardof quality control.

28
40 mm

20 mm

E
ý
oc ck

'IT

20 mm

78 mm
21 mm

Ball
Sleeve with Window

20 mm
9Lni 21 mm 9 ºnun

O
N
E

0
M

E
00
N
t

18 mm 30 mm 18 mm
E
O
28 mm 4ýmýn 58 mm 4

Section of Conical End Piece


Bolt with a Hole for Dowel Pin

Fig 2.4. The details of jointing system

29
The dimensions of the components that are used in the experimental grid are shown in Fig
2.4. The diameter of the ball is 90 mm and the diameter of the holes on the ball is 20 mm.
Most connectors in the structure required eight member locating holes. However, the same
balls were used where fewer members were required. The dowel pin is a cylindrical piece

of 5 mm diameter and 35 mm length. The tubular elements for the experimental grid were
prepared from steel tubes of nominal external diameter 76.4 mm and wall thickness 3.5
mm. The tubular elements were prepared by first cutting 1200 mm lengths of tube and then
welding conical pieces to the ends of the tubes. The dimensions of a tubular element are
in
shown Fig 2.5.

68 mm 1200mm 68 mm

Fig 2.5. Tubular element

2.4.1 Material properties

The tubular elementsare from steel grade ST 52-3 in accordancewith DIN 17120.Three
tubeswere testedin tensionto failure and the test resultswere as follows:

Ultimate tensile strength 509-538Mpa

Yield stress 460-485Mpa

Elongation 22%-24%

The bolts are from 8.8 temperedsteel in accordancewith DIN 17200 of quality 8.8. The
first number 8 shows one eighth of the minimum ultimate tensile strengthin kgf/mm2and
the secondnumber 8 shows the percentageof one eighth of the ratio of the tensile yield
strengthand minimum ultimate tensile strength.Two bolts were testedin tension to failure
and the yield strength as well as the ultimate tensile strength of the bolts was obtainedas
529 and 689 Mpa respectively.

Conical end piecesare hot-stamped,forged and sandblastedmade from ST 52-3.

30
Sleevesare made from hexagonal tubes. The steel quality is 95 Mn 28K in accordance with
DIN 1651. The yield stress of this material is 380 Mpa and its ultimate strength is around
470 Mpa.

The dowel pins are made from tool steel hardened and tempered to the strength of 1400
Mpa.

The balls are forged steel spheres. The material is steel grade ST 52 in accordance with
DIN 17120. The steel is annealed and sand blasted [1].

2.5 Assembly of the grid

The components of the grid were transported to the experimental site near the Engineering
Faculty of the University of Mazandaran, Iran. All of the components were unloaded
It
carefully and under continuous supervision. was decided to assemble the grid on a flat

concrete surface. Attention was paid to ensure that the entire surface was level. The
assembly started from one side and carried on until the entire grid was completed. A
general view of the assembled grid is shown in Fig 2.6.

ýýiö
ý yk
3

Fig 2.6. A view of the grid

31
2.5.1 Weight measurement

For the analytical study of the grid, the cross-sectional properties of the elements had to be

obtained. In order to obtain information about the material characteristics, the weights of
the elements were measured individually. An element (Fig 2.5) consists of a tubular

element, two conical end pieces, two bolts, two sleeves and two dowel pins. The spread of
the weights of the elements that were used in the experimental structure are shown in Fig
2.7.

60

50
on
40

30
an

3 20

10

0
8.3-8.5 8.5-8.7 8.7-8.9 8.9-9.1

Percentage

Fig 2.7. Distribution of the weights of tubular elements

It was expected that elements would be of different weights. This is due to the inevitable

variations of the weights of different constituent parts such as conical end pieces, sleeves,
bolts, dowel pins and tubes. The weights of 50% of the elements were between 8.9 to 9.1

Kg and more than 81% of the elements weighted between 8.7 to 9.1 Kg.

2.5.2 Assembly process

The assembly process was started by a small portion of the grid at one of the corners. This

portion was a pyramid consisting of four bottom layer elements and four web elements that
were connected together by means of five connectors. This portion is shown in Fig 2.8a
with thick lines. The second step in the assembly of the first row was connecting three

32
bottom layer elements,four web elementsand one top layer element.This stepis shown in
Fig 2.8a with thin lines. The assemblyprocesswas continuedin this samemanneruntil the

completionof the first row, as shown in Fig 2.8a with dashedlines.

1% f\ 1 iý ff 1i ff Ii ff Ii ff

(a) The first assembledrow

(b) Two assembledrows

Fig 2.8. The assemblingprocess

Further continuation of the assemblyis shown in Fig 2.8b, where the assemblyof the first
is
row completeand the assemblyof the secondrow has started(it is shown by thin lines).
The assembly of the second row was continued in six steps, as shown in Fig 2.8b by
dashedlines. In eachstep two bottom layer elements,four web elementsand two top layer

elements were connected to the assembled part of the structure by means of two
connectors.The rest of the structurewas assembledusing the sameprocedure.

In the case of prefabricated structures the accumulation of the imperfections of the


componentsusually shows itself in the assemblyprocess.These imperfections that are
usually inevitable will change the theoretical configuration of the structure to certain
degree.In the initial steps of the assemblyprocessthe discrepancyis negligible and the

componentscan be located in their positions properly. In the following stepsspecially in


the caseof large structures,the discrepanciesincreasegradually and significant distortions

33
will appear. The main problem is when there are some constraints for the nodes or
elements of the structure such as supports. In the case of the experimental grid, due to the

nature of the chosen method and the freedom of movement at the supports, all of the nodes
and elements were fitted together easily. Actually, in this case the accumulation of the
imperfections only affected the overall dimensions of the grid. The overall dimensions of
the grid, when the assembly was completed, were measured to be 9908±2 mm by 9908±2

mm (the theoretical values were 9898 mm by 9898 mm).

In the assembly process, the connectors were tightened with an open-ended spanner or

monkey wrench. At this stage the main concern was the assembly of all the elements and
the bolts of the connectors were not fully tightened. Subsequently, all the bolts were

tightened after the completion of the whole assembly of the grid. The retightening process

was started from one corner and continued in the order of the assembly of the components.

2.6 Foundation

The foundation for the test set-up of the experimental grid was constructed at a site at the
University of Mazandaran, Iran. A general view of this foundation is shown in Fig 2.9. The
foundation consists of four pad footings that are connected by a rectangular grid of strip
footings, Fig 2.10. Twelve reinforced concrete columns were built on the foundation as

possible support positions for the grid. A plate was placed on the top of each column by

means of four anchor bolts. Furthermore, thirteen plates were placed on the foundation as
possible support positions for the loading frame and measurement equipment. The strip
and pad footings as well as columns are identical.

Fig 2.10 showsthe plan of the foundation and the cross-sectionsof pad and strip footings.
The building of the foundation involved the constructionof formwork, preparationof steel

reinforcements
andthepreparationandcastingof concrete.

2.6.1 Formwork of foundation

The experimentalsite was a flat concretesurface.The formwork for the foundation was a

wall of 600 mm height that was built using concreteblocks (Fig 2.10). The internal surface
of the formwork wall was coveredwith sandcementplaster (Fig 2.10). This formwork was
not removedwhen the foundation was completed.

34
In order to provide a clean and level surface for the steel reinforcement bars a layer of lean

concrete was laid on the bottom surface (Fig 2.10). The thickness of the lean concrete layer

was about 100 mm.

Fig 2.9. A general view of foundation

2.6.2 Steel reinforcements

The strip and pad footings are reinforced in both directions at both the upper and lower

sides, Fig 2.10. The concrete cover to the outermost reinforcement bars are 50 mm. The
steel reinforcement of the foundation comprised 20 mm diameter (type III) mild steel
ribbed bars. The bars were delivered in lengths of about 12 m. They were cut and bent and
then were fixed with mild steel tie wires.

35
m 4.26 m 2.84m

Fj__2.84
QQ
-L ý

Q
O-B QB of
Q
N
E
QQQQ QQQ . -r
X

D
Q
AA
=iA

1.80m 3.57m Loom 3.57m 1.80m

Plan
020@15 Cm O20@30 Cm

020@15 Cm

Section A-A
Plate 40 40x2 Cm

020@ 15 Cm 020@30 Cm U
o

25
\*\
8020 Anchor bolts
Block concrete wall Blinding concrete

Section B-B Base plate

Fig 2.10. Plan and details of foundation

36
Plate30x30x 1 Cm
if

A
4020
4020 Anchor bolts
Anchor bolts

60 cm
E
O

N
10@20Cm links
0
A

12020 Main bars

lip
E
0

Fig 2.11. A typical column configuration

Figure 2.11 showsthe layout of a typical column that is constructedas a possiblesupport

position. The longitudinal reinforcementsof columnswere 20 mm diameter(type III) mild


steel ribbed bars. Each stirrup was made of a sufficient length of 10 mm diameterribbed
bar. The reinforcementsof the columnswere carefully formed to the dimensionsindicated
in Fig 2.11. The vertical reinforcing bars
and stirrups were tied together with tie wires to
form the column cage. The column cages were accurately
placed and securedagainst
displacementby using tie wires.

The baseplatesthat are placedon the strip footings have eight anchorbolts the base
while
platesthat are locatedon the columnshave four anchorbolts. All the anchorbolts aremade
from 20 mm diameter ribbed bars.

37
2.6.3 Concreting of foundation

The concrete for the foundation was manufactured at the experimental site. The aggregate
including sand, gravel and crushed gravel were derived from natural deposits that were

available. Ordinary Portland type I cement was used in the concrete. The mix was designed
to achieve 28 days cylinder compressive strength (f c) of 20 MPa. The concrete mix had
350 kg/m3 cement content and 42% water/cement ratio. The following steps were carried

out under continuous supervision:

" The accurate weights of cement, gravel and sand as well as the amount of water

were usedfor eachbatch accordingto the mix design.

" The mixing time was 60 secondsor more after all the material was in the mixer.

" The concrete was sampled from mixer discharge point that was near to the
placementpoint on a randombasis.

A slump test was taken from most of the batches. The measured slumps were
between40 mm to 60 mm.

At the time of concreteplacement,all the steelreinforcementswere free from flaky

rust, mud, greaseor any other coating that might have reducethe bond with the
concrete.

" Every endeavourwas madeto avoid steel reinforcementbars from being displaced

or depressed.

" During concreting, adequatecompactionwas achievedusing a vibrator to ensure

good quality concrete.

" The surface of freshly placed concrete, specially between anchor bolts, was
levelled.

" The baseplates with slotted holes were bolted to the pre-installedanchor bolts on
the strip footings.

" The entire surfaceof the concretewas kept damp for a numberof days.

38
Minimum compressive strength obtained from cub tests after 7 days and 28 days moist-

cured concrete were 18 MPa and 28 MPa, respectively. The cubes dimensions were
15Ox15Ox150
mm.

2.6.4 Formwork for columns

Since all of the columns were identical, it was decidedto build them in three stages.In
eachstage four columnswere constructed.The timber formworks were preparedon site for
the construction of columns. Timber is generally used in Iran for in-situ concretework.
The formwork parts were substantialand sufficiently tight to prevent leakageof mortar.
The corners of the columns were bevelled by moulding placed in the formworks. The
formworks finished surfacewas coatedwith oil beforethe assemblingprocess.

The column reinforcementsas mentioned before were put in place together with the
footing reinforcements.Each one of the constructedcolumn formworks was placed in the

correct position and supportedby four screwjacks. The screwjacks helped to adjust the
alignment and the vertical position of the formwork. The height of the formworks was
about 2.20 m. In order to level the top surfacesof the columns a wooden guide strip was
provided at the proper location on each formwork so that the top surfaceof the columns
canbe levelled.

2.6.5 Concretingof columns

The concreteaggregateand cementtype for the columns' concretewere the sameas those
for the foundation. The concrete mix had 400 kg/m3 cement content and a 41%

water/cementratio. In order to have a better workability for concrete,the slump rangewas


changedto 60 mm- 100 mm. Particular attention was paid to ensurethat, at the time of
concrete placement, all the reinforcement and anchor bolts were clean and in correct
positions.

The concretewas sampledand a slump test was taken from eachbatch. Since the slump

amountin columns' concretingwas important,the batchingoperationwas slowed down to


allow completionof eachslump test beforethe next batch was made,so that the test results
could be communicatedto the operator and the necessarycorrections could be made.
During the concretingprocessthe main concernwas that the reinforcementbars were not
displaced.

39
The concretewas placed in uniform layers not more than 500 mm thick. Each layer of
concretewas fully compactedbefore placing the next one, and eachnew layer was placed
while the underlying layer was still responsive to vibration. The top surfacesof columns
were finished so that the baseplatescould be installed at the same level and in firm contact
with columns.Finally, exposedparts of all anchorbolts were cleanedusing a cloth.

Since the formworks for columns were not supporting the weight of the concrete, they

could be removed when the concrete has to


attained sufficient strength resist damage from

the removal operation. Since the formworks were to be reused, they were thoroughly
cleaned, repaired, and coated with oil again.

The column surfaceswere cured after removal of formworks. In order to obtain good

curing, wet hessian to


was used cover the top and the sides of the columns. The hessian

was kept wet by frequentwatering for four weeks.

The remaining eight columns were constructed in two further stages. In each stage
formworking, concreting, striping of formworks, and curing process were carried out as

explained for the first stage. After levelling of the top surfaces of the columns by concrete,
the base plates with slotted holes were bolted to the pre-installed anchor bolts.

All concretecube sampleswere placed in a pool and compressivestrengthtest carried out

after sevenand twenty eight days.Minimum compressivestrengthfor sevendaysmoisture


curedconcretewas20 MPaandfor twentyeightdayssampleswas30 MPa.

2.7 Placement of the grid

The processof creating a flat and level surface for the experimental work started with
filling the voids betweenthe strip footings. Thesevoids were filled with two 25 cm layers,

using the available sandand gravel at site. Each layer was compactedusing a compactor.A
10 cm thick layer of concretewas pouredon the top of the fill. The concretehad 250 kg/m3

cementcontent and was producedat site using the available material. The poured concrete
was levelled using a screedboard.

For the placementof the double layer grid a telescopiccrane was used. In spite of the fact
that the total weight of the double layer grid was about 3800 kg, the large dimensionsof
the grid and the presenceof the columns required quite a big crane.A 40 ton truck crane
with about 30 m boom was usedfor the purpose.

40
Fig 2.12. The grid about to be placed on the columns

Fig 2.13. Ensuring the accurate placement of the grid

41
Fig 2.14. The final inspection of the grid after placement

Fig 2.15. The grid in its correct position on the top of the columns

42
It was decided to hoist the grid from the four top layer nodes. Accordingly. the grid was

analysed for this loading and boundary condition to ensure that the grid could be safely
hoisted. Four cables were connected to the four top layer nodes in the central part of the

grid. Before moving the grid the tightness of the nodes was checked with inspection of the
dowel pins on the sleeves that show the amount of penetration of the bolts into balls. The

grid was picked up carefully and placed on the top of columns. The placement of the grid

on the top of the columns is shown in Figs 2.12 to 2.14. Fig 2.15 shows the grid in the

correct position after the placement.

2.8 Roller supports

The main objective of the present work is to investigate the effects of the degree of bolt

tightness of connectors on the behaviour of double layer grids. In the experimental process,
the displacements of the test grid under applied loads are considered as the response of the

structure. The support conditions have a significant effect on the response of the structure.
Roller supports in two directions were used for all the supports of the test grids.

Hardened Ball bearing Rigid PVC plate


steel plate 2.5 cm diameter 20x20x2.3 cm
20x20x2 cm

Fig 2.16. The cross-section of a support assembly

Twelve support assemblies were designed and constructed as in two


roller supports
directions. A support assembly consists of two hardened
steel plates, a rigid PVC plate
with seven holes and seven ball-bearings. A cross-section of a support assembly is shown
in Fig 2.16. The lower hardened steel plate was placed
on the base plate of the reinforced
concrete column. The PVC plate had seven cylindrical holes for the ball-bearings as shown
in Fig 2.17. This plate together with the seven ball-bearings the lower
was placed on
hardened plate. The upper hardened plate was placed on the ball-bearings, Fig 2.18.

43
Therefore, the upper hardened steel plate could move in any horizontal direction easily.

This arrangementplayed the role of a roller supportin two directionsfor the grid.

2.8.1 Hardening of support plates

The support assemblies transfer the vertical forces from the test grid to the reinforced

concrete columns. A compressive force in an upper steel plate is transferred to the lower

steel plate by direct bearing through the ball bearings. Since the contact surface between a
ball bearing and an upper or lower steel plate is very small, a significant pressure is
imposed on the plate locally. Furthermore, if the ball bearing penetrates into the plate then

the impression created on the plate prevents the free rolling of the ball bearing. To

minimize this effect, it is necessaryto ensure that the plates used are of sufficient hardness.

Based on information received from experts, twenty four 200x200x20 mm IASC 6582

alloy steel plates were purchased. This alloy steel was made in Isfahan Alloy Steel
Complex, Iran. Its composition is 0.34% Carbon, 0.3% Silicon, 0.5% Manganese, 1.5%
Chromium, 0.2% Molybdenum and 1.5% Nickel. This alloy steel is comparable with the

requirements given in the British Steel Standard Specification for construction steels (BS)
817 M 40 and the German Steel Standard Specification (DIN) 6582 as well as the

American ASTM 4340. The hardness of this alloy steel was tested using Rockwell
hardness testing method. The average hardness number obtained from tests was 12 HRC.
In order to improve the hardness of the support plates, hardening treatment was carried out
in accordance with the manufacturers' recommendations. The treatment was carried out at

the Department of Mechanical Engineering of the University of Mazandaran, Iran.

The hardening process of the steel plates was accomplished in three major stages:

1. Heating the steel slowly to ensure a uniform temperature,

2. Holding the steel at a given temperature for a given time and

3. Cooling the hot steel to room temperature.

In the heating stage, it was important to ensure that, as far as possible, the whole body of

steel had uniform temperature. Otherwise, one part of body could have expanded faster
than the other part, resulting in distortion or cracking.

44
Fig 2.17. Rigid PVC plate with holes together with seven ball-bearings

ý:
'
..::
a

..
__.

Fig 2.18. A roller support piece on a column

45
Uniform heating was attained by a slow rate of temperature change. An oven was used for

heating of the plates, as shown in Fig 2.19. Two plates at a time were processed, and each

plate was placed on four steel cubes. The space between the plates helped uniform

temperatures to be maintained throughout. The temperature to be reached was 850°C, as

recommended by the manufacturer.

After the steel plates were heated to 850°C, they were held at this temperature for 60

minutes. According to the manufacturers' instructions, this time was enough to allow the
desired internal structural changes in the steel to take place.

Subsequently, the plates were cooled to complete the hardening process. In order to cool

the steel plates rapidly, they were placed in a cooling medium. As manufacturers

recommended, (motor) oil was a suitable medium for IASC 6582 alloy steel. A suitable

amount of motor oil was poured into a tank. The oil temperature was around 17°C. The

plates were removed from oven and placed in the oil vertically (Figs 2.20 and 2.21). Once
the temperature reduced to around 55"C, plates were removed from the tank and left to

cool down to the room temperature.

Fig 2.19. The positioning of two plates in the oven

46
Fig 2.20. Removing plates from oven

Fig 2.21. Quenching the plates in oil

47
2.8.2 Hardness testing

In order to check the results of the hardening treatment, hardness testing was carried out on

the plates. There are different methods to measure the hardness of steel plates and one of
the common ones is the Rockwell C test. In Rockwell hardness testing, a conical diamond
is first pressed with a force FO, and then with a force Fo+Fi against a specimen of the

material whose hardness is to be determined. After unloading to Fo, the increase (e) of the
depth of the impression caused by F, is determined (e in mm). The depth of penetration (e)
is converted into a hardness number (HRC) which is read directly from a scale on the tester
dial (Fig 2.22). This number is obtained from the following equation:

HRC=500x (0.2-e)

The hardness number HRC is between 0 and 100, when 100 represents the hardness of
diamond.

F F=F+F, F,,

Surface of III
specimen
ý.. J........
0.2 mm v.
ýº.
..................... ... .. ...........

Fig 2.22. Principle of Rockwell hardness testing

An Avery Denison hardness testing machine was used to obtain the hardness number of the

plates. The machine was of DRWL type and made in the UK. Plates with thicknesses of up
to 20 cm and width of up to 30 cm can be tested by this machine. A view of the testing

machine and test process is shown in Fig 2.23. The average hardness (HRC) obtained for
the plates was around 42.

A further test was carried out to ensure the proper operation of a support assembly under
lateral forces. An assembly similar to the actual support assemblies was used, as shown in
Fig 2.24. This assembly consists of two hardened plates, a circular rigid PVC plate with
four holes and four ball bearings. The assembly was placed under a load of 300 kN using a
hydraulic jack. After unloading the plates were examined and there were no noticeable
impressions created by ball bearings on the plates.

48
i
ýC

Im

Fig 2.23. The hardening test


machine and the test process

49
Fig 2.24. Testing the sufficiency of the hardness of the hardened
plates of the support
assemblies

50
2.8.3 Installation of support assemblies

The roller supports could not provide resistance to lateral forces. As soon as a lateral load

of any kind pushed on the upper hardened steel plate, it rolled away in response to the
force. In order to maintain stability of the entire grid under accidental lateral loads, the
horizontal movements of the grid was controlled using some additional pieces.

Two pieces of steel angles were welded to the base plate on the top of support column at

the external sides of the grid. The support joint of the grid was picked up using a jack. The
lower hardened plate was placed on the base plate of the column. The PVC plate together

with the seven ball-bearings was placed on the lower hardened plate (Fig 2.17). The upper
hardened plate was placed carefully on the ball-bearings. This plate together with the PVC

plate and ball-bearings constituted the mobile part of a support assembly. The steel angles

prevented the mobile part of a support assembly from horizontal movement towards the
outside of the grid. Finally, the support joint was placed on the upper plate. The same
process was repeated for the other three supports. Fig 2.25 shows a view of a support

assembly together with a support joint on a support column. The four corner supports were

constrained by welded angles in order to prevent the lateral movement of the grids during

the tests.

Fig 2.25. A support assembly and connector on a corner column

51
2.9 Loading frame

Consider the loading frame shown in Fig 2.26. The loading frame allowed loads from a
hydraulic jack to be applied to the grid. The hydraulic jack was used to apply up to a 300
kN load to the central node of the top layer of the grid. The loading frame was made of

mild steel beam profiles and angles as well as plates. The beam profiles were delivered in
lengths of twelve metres. They were cut in appropriate lengths and welded and coated on-

site. The loading frame constituted of three main parts:

1. The lower cross-footing that supported the two vertical elements of the loading
frame.

?. Two vertical elements of the loading frame.

3. The top part of the loading frame that supported the hydraulic jack.

The loading frame was designed to transfer the reaction of the jack to the concrete base

through the cross-footing. All the components of the loading frame were designed for the
highest possible loads.

Fig 2.26. A view of loading frame

52
2.9.1 Cross-footing

Two 3.10 m long castellated beams (CPE 18) were welded to the three central base plates

a
on strip footing. Also four 1.45 m long castellated beams (CPE 18) were welded to the

three central base plates on the other strip footing (central base plate was involved in both

cases). As shown in Figs 2.26 and 2.27, these six beams made a cross that was fixed on
five base plates and transferred loads to the concrete base through the welds and forty

anchor bolts.

Four 60x'-)Ox1 cm plates were welded to the cross. Two beams (three IPE 16 profiles with

appropriate lengths that were already welded together) were welded to the plates so that
suitable supports were created for the vertical elements (Fig 2.27). The applied loads were
transferred from the vertical elements to the cross through these supports.

""

". "

Plate 60x 20x 1cm


Base plate 40 x40x2 an
Vertical element

"s "I \ Iof 1.1 lot \\\N. I Is "1

SS
a
ISS
a90

Castellatedbeam
(CPE 18)
Support of the
vertical element
(3IPE 16) Castellatedbeam
(CPE 18)
0

Fig 2.27. Plan of the cross frame

53
2.9.2 Vertical elements

The vertical elements transfer the reaction to the jack's load to the cross-footing. Each of

the vertical elements was made from two IPE 16 profiles that were connected together by

means of welding. The length of the vertical elements was 4 m. As shown in Fig 2.28 the
suitable supports for the top part of loading frame were designed and constructed.

Some protecting coating was burned during the welding process, therefore it was repaired

after the completion of the constructionof the cross-footingand vertical elements.

2.9.3 The top part of loading frame

A side view of the top part of the loading frame is shown in Fig 2.28. This part transferred

the reaction of the jack to the vertical elements. The top part is constituted three pair of
beams and a thick circular plate. The lower pair of beams was placed on the angle brackets

as shown in Fig 2.29. The thick circular plate with four slots was connected to the lower
pair of beams using four bolts and nuts. The thick circular plate could move together with
lower beams in longitudinal direction of them. This plate also could slide in other direction
due to presence of slots. Two other pairs of beams were placed on the lower beams as

shown in Fig 2.28.

2.10 Loading system

An Enerpac model RCS-1002 with 100 ton capacity hydraulic jack with 5.7 cm stroke was

used to apply up to 300 kN load on the central node of the top layer of the test grid. A

circular base plate together with four bolts was used to fix the hydraulic jack to the upper
thick circular plate (Fig 2.29). An Enerpac model P-80 steel hand pump was used to

provide the hydraulic pressure. A gauge adaptor was connected to the pump for pressure
gauge installation. The gauge was used to monitor the pressure of the hydraulic circuit
indicating the approximate amount of the applied load. A hose that was connected to the

gauge adaptor transmitted hydraulic fluid to the cylinder (Fig2.30). The hose was
connected to the cylinder by means a male coupler (Fig 2.29).

In order to install the load-cell to the hydraulic jack, an appropriate coupler was designed

and constructed. A sketch showing a side view of the arrangement of Fig 2.29 is shown in
Fig 2.31.

54
Fig 2.28. A side view of the top part of loading frame

Fig 2.29. Connection of the jack to the top part

55
Fig 2.30. Pump together with gauge adaptor and gauge

Thick circ ter of j ack

Circularb Bolt

Fig 2.31. A sketch of the loading system

56
2.11 Measurement support frame

The applied load to the test grid was transferred to the foundation through the loading
frame. Therefore, during the experimental process, the foundation had some small
deformations due to the applied load. In order to measure the displacements of the test grid

accurately, the measurements had to be independent of the foundation. To achieve this, two
13 m long beams were constructed from IPE 14 profiles. The beams were placed on two

steel basesoutside the foundation as shown in Fig 2.33.

2.12 Data acquisition system

In order to measure the displacements and loads a data-logger was used to collect process

and record the results. The data-logger was controlled using a Pentium II 1.33 MHz

computer. The arrangement of the instrumentation is shown schematically in Fig 2.32. As

indicated in this figure, the displacement transducers and load-cell were connected to the

data logger. The information received by the data-logger was pre-processed and then was

sent to the computer for processing. After the processing of the data by the computer, the

results were used for post-processing.

Load Cell Transducers

L
Data Logger

Computer

User

Fig 2.32. Schematic representation of the data acquisition system

57
Fig 2.33. Two views of the measurement frame

58
2.12.1 Displacement transducers

Two types of displacement transducers were used to measure the response of the test grid.
All of the transducers were manufactured by Tokyo Sokki Kenkyujo Co. The general

specifications of these two types of transducers are given in Table 2.1.

Table 2.1. Specifications of displacement transducers

Type Stroke Spring Force Allowable Temperature Range

CDP-25 25 mm 3.4 N to +60°C


-l0
CDP-25 50 mm 3 to +F(l"C
,4N -10

The transducers were connected to the measurement system incorporated in the data logger

using the four wire system shown in Fig 3.34. As shown in the wiring diagram of the full
bridge, the input pair of wires (red and black) was used for excitation and another pair of

wires (green and white) was used for output voltage acquisition.

Green
(Terminal B)

White
(Terminal D)

Red Black R=Resistance


(Terminal A) (Terminal C)

Fig 2.34. Displacement transducer circuit

I,

59
Twelve MB-B magnet stands made by Tokyo Sokki Kenkyujo Co., were used to hold

twelve displacement transducers during the test procedure. A transducer was mounted on a

magnet stand through a fitting holder. The transducers were adjusted and set vertically.
The plungers of the transducers were positioned on bolts connected to the nodes of the test

grid. The magnet stands were fixed to the measurement frame. A view of a displacement
transducer together with a magnet stand is shown in Fig 2.35. Before testing commenced.

each displacement transducer was calibrated separately. The transducers were calibrated
over their full working ranges using the data-logger and a set of measuring blocks.

Fig 2.35. A magnet stand and a transducer

60
2.12.2 Load cell

A load cell was used to convert the value of the load applied by the hydraulic jack into an

electrical signal. The load cell was connected to the hydraulic jack and transferred the
applied load to the central node of top layer of the test grid. The load cell also was
connected to the data logger using six wires. The load cell was manufactured by Tokyo
Sokki Kenkyujo Co. The specifications of the load cell are given in Table 2.2. The load

cell was calibrated before the test began using the data-logger and the universal test
machine. A view of the load-cell together with the hydraulic jack is shown in Fig 2.36.

Table 2.2. Specifications of load cell

Type Capacity Allowable Temperature Range Resistance

CLP-500KNB 500 kN to +60°C 350 ohm+1%


-10

Fig 2.36. Load cell between jack and node

61
2.12.3 Data logger

A CRIOX data logger was used in conjunction with a computer to collect process and

record all the data monitored during the test process. The CRIOX is a programmable data
logger made by Campbell Scientific, Inc. A summary specification of the data logger is

given in Table 2.3.

Table 2.3. Specifications of the data logger

Differential Analog Channels 6

Maximum Input Voltage f 2500 mV


Scan Rate 64 Hz

Data Storage Memory (Data Point) 62,000

Accuracy at 0°` to +40' ±2.5 mV

The data logger was programmed for collection, storage and analog to digital conversion of
data measurements. The program consisted of a group of instructions entered into a

program table. The program was given an execution interval equal to one secondwhich
determinedhow frequently the program is executed.When the program was executed,the
instructions were executed in sequence from beginning to the end.

The input instructions specified the conversion of the sensor signals received from the load

cell and transducersto the data values and stored them in the input storageof the data
logger. The programmable entries also specified the measurement type, the number of

channels to measure, the input voltage range and the input storage locations. Furthermore,
in order to convert the sensor output to engineering units, the sensor constants were
determined and entered into the program.

According to the information received from the manufacturers, the load cell and
transducers exhibit linear characteristics over their total range. Therefore, a linear equation

was considered as:

F=AxVoc+B

Where

"F is the engineering units (kN for load and mm for displacement),

62
"A and B are the sensor constants and

" V,,, is the output voltage of load cell or a transducer.

Each displacement transducer as well as the data logger was considered separately. They

were switched in turn through an amplifier. Transducers were calibrated using a set of
accurate measuring blocks. Fig 2.37. The accuracy of the measuring blocks was 0.005 mm.
The load cell was calibrated using a Universal Testing Machine manufactured by Denison
Mayes Group of the UK (DMG). Its standard capacity ranges from 0-600 kN and it is fully
A
computerised with modern windows software. view of this Universal Testing machine is

shown in Fig 2.38. The amount of applied load on the load cell was displayed on the
monitor of the Universal Testing Machine. Also, the voltage of the load cell was displayed
on the monitor of the data-logger at the same time, Fig 2.39. Several voltage readings were
taken at each displacement as well as load increment and constants A and B were

calculated using the method of least squares to obtain the best straight line through the data
points. The calibration factors obtained from different tests are shown in Table 2.4. This
calibration process that was programmed into the data logger was performed periodically
to confirm the reliability of the test results. A view of the data logger is shown in Fig 2.40.

Table 2.4. Calibration constantsfor transducersand load cell

Sensor Type A B Engineering Unit

Load cell 120.693 kN


-0.012
LVDT (CDP-25) 2.6814 -12.756 mm
LVDT (CDP-50) 6.6863 mm
-25.219

63
Fig 2.37. Set of measuring blocks

Fig 2.38. A view of Universal Testing Machine

64
Fig 39 Calibration

65
Fig 2.40. Two views of the data logger

66
CHAPTER 3

THE TESTS

3.1 Introduction

A double layer grid constructed by a MERO-type system consisting of a number of

components that are connected together by means of bolts. In this system due to the

presenceof the bolts, discontinuities betweenthe connectingparts will be inevitable. The

mechanismof load transfer at these discontinuities is complex and the presenceof gaps
to
and prestressingwill give rise more complexity. The degreesof tightness of the bolts

affect the width of gapsas well as the amountsof prestressing.Therefore,the bolt tightness
plays an important role in the behaviourof the connectorand the structure.

This chapter describesthe test procedure,data acquisition process and data processing.
Also, the loaddeflectionresponses in this chapter.
of the gridsarepresented

3.2 Test grid A

The plan of test grid A is shown in Fig 3.1. In this figure, top layer elementsare shown by
thick lines and bottom layer elementsas well as web elementsare shown by thin lines. The
test grid was placed on four two directional roller supportsthat are shown in this figure by

67
four little squaresat the four comers. The load was applied to the central node of the top
layer using a hydraulic jack that is shown by a little circle in the figure. The amount of

applied load was measuredusing a load cell that was placed between the hydraulic jack

andthe central nodeof the grid.

12

11

3 1

4 9 K
1 \
\8 /
1
\
\
^
1 /

6 7

Fig 3.1. Plan of test grid A

The displacementsof the bottom layer nodes 1 to 12 (that are shown by little solid circles
in Fig 3.1) were measured as response of the test grid using twelve displacement
transducers.Four CDP-25 transducerswere used for displacementmeasurementof the
nodes 1,6,7 and 12. Also eight CDP-50 transducerswere employed to measurethe
displacementsof the remaining nodes.The geometryof the test grid, its supportconditions

as well as the applied loads were symmetric. Also the components of the grid were
identical. Therefore, the symmetric nodes should have the same displacement. This

allowed the experimentalresultsto be checkedfor symmetry providing an indication of the


accuracyof the test results.

68
In order to investigate the effects of the degree of tightness of the bolts of the connectors

on the response of grid A, this grid was tested two times with two different levels of
tightness of the bolts. For each test all the bolts were loosened and then tightened using an

adjustable torque wrench. For each test the adjustable torque wrench was set for a specific
amount of torque and used for tightening all the bolts. So, the degree of tightness of all the
bolts was the same for each test. The amounts of the torques that were used for tightening

the bolts in two different tests were 60 and 120 Nm.

In addition, the response of grid A with normal tightness was obtained through a further

test. The term "normal tightness" is used to represent the degree of tightness that is

normally achieved in double layer grids in practice. Since ordinary spanners are used in

practice, the degree of tightness depends on the length of the spanner that is used for

tightening as well as the amount of the applied force by the worker. In practice, usually the

same spanners or monkey wrenches are used and the workers are trained to apply an
appropriate amount of force.

3.2.1 Setting up of grid A with normal tightness

Since the test had to be carried out outdoors and since each round of the setting up and
testing of the grid required around four hours, the test had to be startedin early morning.
The settingup of grid A was carried out as follows:

1. The componentsof the grid were assembledon a flat concretesurface.All the bolts

of the connectorswere tightened and then the grid was placed on the supports.The
tightness of all the connector bolts was controlled again with the grid on the
supports.Tightening processwas carried out using an open ended spannerand a
monkey wrench of length around30 cm in three stages.In the first stagethe bottom
layer connectorswere tightened. The tightening processstartedfrom one side and

carried on until all the connectorsof this layer were tightened. In the next stagethe
top layer and finally web layer connectorswere tightened in a manner similar to
that of the bottom layer.

2. As shown in Fig 3.2, a cylindrical steel piece was designed, manufacturedand

connectedto the central node of the top layer of the grid. The top surfaceof this
steelpiece had a concavesurfaceso that the entire convex surfaceof the free end of
the load cell would be in touch with the cylindrical steelpiece.

69
3. The hydraulic jack and load cell were mounted on the loading frame. Since the

thick circular plate could move in two perpendicular directions. particular attention

was paid to fix the hydraulic jack in the correct position so that the load applied at
the centre of the node (Fig 3.2).

4. Twelve magnet stands were positioned on the measurement support frame.

5. The displacement transducers were numbered as shown in Fig 3.1 and mounted on

the magnet stands. Also, they were adjusted and set vertically. The arrangement of
the displacement transducers are shown in Fig 3.3.

6. Both ends of the cables of the displacement transducers were numbered. These

cables connected the transducers to the data logger. A further cable connected the
load cell to the data logger.

7. The computer and data logger were switched on and the given program was carried

out.

8. The whole of the grid A, supports, hydraulic jack, transducers and wiring were
inspected carefully.

Fig 3.2. Load application

70
Fig 3.3. Two views of the transducer arrangement

71
3 2 Testing of grid A with normal tightness
.?
Testing of grid A with normal tightness was carried out by running the test computer

program. In order to condition the test grid and check the operation of the test equipment.

around 50 kN load \Nas applied on the central node of the top layer of the grid and after a
few seconds unloading as carried out. This process repeated a number of times and
during this process the displacements of transducers were checked to ensure that the grid

as detorming symmetrically.

"B+

"r

Adjustable holt for I'lulluer


holding transducer's plunger

Displacement transducer
f=itting holder

Magnet stand

Fig 3.4. A sketch of adjusting displacement transducer

The test process was continued as follows:

1. Each transducer was adjusted using an adjustable bolt that was connected to the

node of the grid for holding the transducer"s plunger. Also the fitting holder was
adjusted so that the plunger was at its maximum stroke (Fig 3.4). Particular

attentionwas paid that the transducerswere set vertically.

?. The displacements of the nodes under zero load were read and the transducers were

checked so that their strokes were close to their maximum range.

3. In order to carry out the main stage of the test, the computer program was run

again.

4. Since the linear behaviour of the grid was under investigation, using a hydraulic
jack a small load around 2 kN was applied on the central
node of the top layer of

72
the grid. The state of the grid under this load was considered as the starting
referencepoint.

5. The load on the grid was incrementedby around50 kN.

6. Sixteen channels of the data logger were used for receiving input signals. One of

these channels was used for the load cell, twelve channels were used for
displacement transducers and three remaining were not connected to any sensors.
The execution interval for collection, processing and storage of a data measurement

was set equal to one second. So, the required time for acquisition of two
consecutive pieces of data for the same channel was equal to sixteen seconds. The
whole system was left to collect around three different measurements from each
channel and to reach a stable situation.

7. Steps5 and 6 were repeatedas long as the limitations of the test set-up allowed.
The test limitations were the maximum load of 300 kN (that was obtainedfrom the
theoreticalanalysisof grid A) and the maximum strokeof the hydraulic jack.

8. All the obtaineddata were storedin a file for processing.

9. The grid was gradually unloaded.

In order to increasethe accuracyof the obtainedtest results,the test procedurewas carried

out in the samemannertwice. So, two different set of data were obtainedthat represented
the responseof grid A with normal tightness.

3.2.3 Test results

During the tests, displacementsof twelve bottom layer nodes of grid A together with
different applied loads were measured.Theseresults togetherwith somemore information
including date and time of test, temperatureand battery voltage were stored in the output
files. An output file consistsof twenty three columnsas follows:

" The first left-hand column indicatesthe order.

" The next two columns denotethe year and the number of the day of the year when
the test hasbeencarried out.

" The next two columnsdenotethe time when the test hasbeencarried out.

73
" The next column showsthe degreeof the temperatureof the environment.

" The next column denotesthe battery voltage.

" The next twelve columns denotethe displacementsof nodes 1 to 12 of the bottom
layer of the grid.

" The three remaining columns relate to channels that are not connected to the

sensors.

9 The last column showsthe applied load.

In order to obtain the load-deflection responsesof nodes 1 to 12 of grid A with normal


tightness,the following procedurewas employed.

"A set of data from output file was chosen.This set consistedof rows of the output
file for which the displacementsand load measurementswere in a stable situation.
A row was consideredto correspondto a stable situation where there were no

significant difference between the values of displacementsand load measurements


in that row and the respectivevaluesof the previous row.

" It was decidedthat the first row of the above set of data has to be used as the base

point for linear load-deflectionresponsesof the nodesof the grid. So, all values of
displacementsand load measurementsof this row had to be equal to zero. Thus, the
first elementof eachcolumn of the set of data was subtractedfrom all the elements

of that column. The resulting set of data was used for data analysis in the next
stages.

" Since due to symmetry of the grid, loading and support conditions, displacements

of nodes 1,6,7 and 12 had to be the same,the averageof the displacementsof


thesenodesin eachrow were consideredas the displacementof node 1. The same
processwas carried out for the data obtained from the secondtest. The results are
given in Table 3.1. Finally, the displacementsof node 1 versusapplied loads in two
different tests were plotted as shown in Fig 3.5. In this figure the load-deflection

responseshown by thin line relatesto the first test and responseof the secondtest is
shownby thick line.

74
" Also, displacements of nodes 5,8 and 11 were the same as node 2. So, the average
displacements of these four nodes was considered as the response of node 2 and

given in Table 3.2. The load-deflection response is shown in Fig 3.6. Furthermore,
Table 3.3. and Fig 3.7 show the displacements of nodes 3,4,9 and 10.

250
200

iso Ts1
bm
li
T__a o%

.i loo
so
0
0 10 20 30 40
Displacement
mm

Fig 3.5. Experimentalload-deflectionresponseof node 1of grid A with normal tightness

Table 3.1. Test results for node 1 of grid A with normal tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

36.01 5.46 36.01 5.56

75.30 10.88 76.56 11.24

114.21 15.80 121.29 16.91

156.77 21.07 162.02 21.58

199.14 24.38 195.66 25.11

75
250

200

150 ýTest
b
cl Test 2
100

50

0
0 10 20 30 40
Displacement
mm

Fig 3.6. Experimentalload-deflectionresponseof node2

of grid A with normal tightness

Table 3.2. Test results for node 2 of grid A with normal tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

36.01 6.34 36.01 6.47

75.30 12.74 76.56 13.16

114.21 18.48 121.29 19.76

156.77 24.51 162.02 25.39

199.14 31.69 195.66 29.98

76
250

200

F50- Tes
1

100-+-

50-

0 10 20 30 40
Displacement
mm

Fig 3.7. Experimentalload-deflectionresponseof node 3

of grid A with normal tightness

Table 3.3. Test results for node 3 of grid A with normal tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

36.01 7.22 36.01 7.43

75.30 14.63 76.56 15.18

114.21 21.18 121.29 22.66

156.77 27.99 162.02 29.02

199.14 36.12 195.66 34.23

77
3.2.4 Setting up of grid A with specified degrees of tightness

The tightening of the connector bolts of grid A to a specified degree was carried out as
follow:

" All the nodes of the bottom layer of the grid were levelled using adjustable steel

props as shown in Fig 3.9. An adjustable steel prop, Fig 3.8, was manufactured
from an outer tube that constituted the bottom part of prop and an inner tube with

some holes that constituted the top part of the prop. A threaded portion with a slot
was welded to the outer tube. A high tensile steel pin that was located through the
slot in the outer tube and two holes of the inner tube was used for coarse
adjustment. A threaded collar located below the pin was used for tine adjustment

and accurate levelling of nodes (Figs 3.8 and 3.9). Particular attention was paid to

ensure that all props were in firm contact with the concrete base and the balls of the

connectors.

I. "

-a

Bottum layer i
of the grid Inner tube

High tensile
in a hole
ales in inner tube
Collar
Threadedpart
Slot in oute

Outer tut

Base plate

Fig 3.8. A schematic view of adjustable prop

78
Fig 3.9. Levelling of bottom layer
nodes using adjustable steel props

79
" When all bottom layer nodes were levelled, all the bolts of the connectorsof the

grid were loosened using a spanner. The loosening operation was started from
bottom layer and continuedto top layer and completedwith web layer.

" After loosening of all the bolts of the grid, the bottom layer nodes were levelled

againusing the adjustablecollars of the props.

" All the bolts were tightened using an adjustabletorque wrench (with a spanner)as
shown in Fig 3.10. The torque wrench was model EVT2000A and made by Britool
Limited Co, UK. Its adjusting torque range was 50 to 225 Nm and its length was
597 mm. The amount of the torque of the torque wrench for tightening the bolts in

the first stage was set at 50 Nm. The torque wrench indicated with touch, sound and

sight when the required torque was reached. The square peg at the end of the torque
wrench was not suitable for gripping and fastening of the sleeves of the connectors.
So, as shown in Fig 3.10, a spanner was attached to the torque wrench for

tightening of the bolts. In order to connect the spanner to the torque wrench a

square hole was cut on the handle of the spanner. The position of the hole was
determined so that the overall length of the torque wrench and the attached spanner

was 730 mm. So, the applied torque to the bolts was 20% more than the torque that

was set on the torque wrench and was 60 Nm in the first stage. The tightening
process by adjustable torque wrench-spanner started from one side of the bottom
layer and carried on until all the connectors of this layer were tightened. Then the

top layer and finally web layer connectors were tightened in a manner similar to
that of the bottom layer.

" All adjustableprops were removedafter the completion of the tightening process.

The remaining steps for the setting up of grid A with 60 Nm degree of tightness was

continuedin an early morning as follows:

" The load cell was mounted on the hydraulic jack. The hydraulic jack was fixed in
the correct position so that the load applied at the centre of the central node of the
top layer of the grid.

" Twelve magnet standswere positioned on the measurementsupport as shown in


Fig 3.3.

80
Fig 3.10. Tightening of bolts using adjustable torque wrench (with a spanner)

81
" The transducersthat had alreadybeennumberedin the order shownin Fig 3.1 were

mountedon the magnetstands.

" The cables of transducers and load cell, whose ends had already been numbered,

were connectedto the relevant channelsof the data logger.

" The displacementtransducerswere adjustedand set vertically.

" The computerand data logger were switchedon and the test programwas run.

3.2.5 Testing of grid A with specified degreesof tightness

Testing of grid A with 60 Nm degree of tightness was started by the conditioning


To
operation. elaborate,around 50 kN load was applied on the grid and after a few seconds
unloadingwas carried out. This processrepeateda few times.

The test process was continued as follows:

1. Each transducerwas adjusted so that its stroke was close to its maximum range.
Particularattentionwas paid so that all transducerswere set vertically.

2. The main stageof the test was carried out by re-running of the test program.

3. A small load around2 kN was applied on the grid. Sincethe linear behaviourof the

grid was under investigation, during the calculation of the results, the stateof the

grid underthis load was consideredas the starting referencepoint.

4. The load on the grid was incrementedby around40 kN.

5. In order to reach a stable situation and record the correct measurementsafter each
load increment, at least three different measurementswere collected from each

channelbefore the application of the next load increment.

6. Steps4 and 5 were repeatedas long as the limitations of the test set-up allowed.
The test limitations were the maximum load of 300 kN (that was obtainedfrom the
theoreticalanalysisof grid A) and the maximum strokeof the hydraulic jack.

7. All the obtaineddata were storedin a file for processingand the grid was gradually

unloaded.

82
In order to increasethe accuracyof the obtainedtest results,the test procedurewas carried

out in the samemanner twice. So, two different setsof data were obtainedrepresentingthe

responseof grid A with 60 Nm degreeof tightness.

3.2.6 Test results

The load-deflection responses of nodes 1 to 12 of grid A with 60 Nm degree of tightness


from output files were calculated as mentioned in Section 3.2.3. The output files containing

the results are presented in Appendix A. Due to symmetry of the grid, loading and support
conditions, the displacements of the symmetrically situated nodes had to be the same. In

order to achieve more reliable results, average values of measurements of symmetrically


situated node displacements were considered as the response of each one of the symmetric
nodes. The load-deflection responses of nodes 1,6,7 and 12 are shown in Fig 3.11. Fig

3.12 shows the response of nodes 2,5,8 and 11. Also, Fig 3.13 shows the response of

nodes 3,4,9 and 10. The numerical results of nodes 1,2 3


and are given in Tables 3.4,3.5

and 3.6, respectively.

Two further tests were carried out to obtain load-deflection responsesof grid A with 120
Nm degreesof bolt tightness.The following stepswere carried out for eachof the tests.

" All the nodesof the bottom layer of the grid were supportedusing the props. Also,

all of thesenodeswere levelled using the adjustableparts of the props (Fig 3.9).

9 All the bolts of the connectorswere loosenedusing a spannerso that all the sleeves
were loose.

" All the bolts of the connectorswere tightened using the torque wrench-spanneras

shows in Fig 3.10. The tightening processwas carried out in three stages.In the
first stage all the bolts of the bottom layer connectors were tightened. The
tightening process started from one side and carried on until all the bolts of this
layer were tightened. In the next stagethe top layer and finally the web layer bolts

of the connectorswere tightenedin a mannersimilar to that for the bottom layer.

" All the props were removed.So, the grid was supportedon the four corner nodesof
the bottom layer. Testing of grid A with 120 Nm degreesof tightness was carried
out in a manner similar to that mentioned in Section 3.2.5. The obtained results
in
were stored output files.

83
" The displacementsof nodes 1 to 12 of the bottom layer of the grid versusapplied
loads were calculatedfrom output files in the samemanneras mentionedin Section
3.2.3. The numerical results are given in Tables 3.7 to 3.9 and the load-deflection

responsesof the nodes are shown in Figs 3.14 to 3.16.

200

150

Test 1
100
Tect 2
0
50

0
0 10 20 30 40
Displacement
mm

Fig 3.11. Experimentalload-deflectionresponseof node 1

of grid A with tightnessdegreeof 60 Nm

Table 3.4. Test results for node 1 of grid A with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

20.58 2.85 27.58 4.90

32.67 5.00 59.20 9.62

63.93 10.51 94.46 14.39

89.93 14.78 120.81 17.81

120.64 19.58 145.53 21.43

160.25 23.48

166.61 24.30

84
200

150
Test 1
0
100 - Test 2

50

0 10 20 30 40
Displacement
mm

Fig 3.12. Experimentalload-deflectionresponseof node2


A
of grid with tightnessdegreeof 60 Nm

Table 3.5. Test results for node 2 of grid A with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

20.58 3.60 27.58 5.97

32.67 6.10 59.20 11.63

63.93 12.62 94.46 17.31

89.93 17.68 120.81 21.26

120.64 23.20 145.53 25.45

160.25 27.81

166.61 28.82

85
200

150

z Tes 1
g 100 Test 2
0
50

o-ý
0 10 20 30 40
Displacement
mm

Fig 3.13. Experimental load-deflection response of node 3

of grid A with tightness degree of 60 Nm

Table 3.6. Test results for node 3 of grid A with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

20.58 4.23 27.58 7.04

32.67 7.15 59.20 13.65

63.93 14.74 94.46 20.28

89.93 20.65 120.81 24.76

120.64 26.95 145.53 29.55

160.25 32.26

166.61 33.44

86
300 T

250 -

200 -
Tl

0
150- Test 2
100 -

50-
0,
0 10 20 30 40

Displacementmm

Fig 3.14. Experimentalload-deflectionresponseof node 1

of grid A with tightnessdegreeof 120Nm

Table 3.7. Test results for node 1 of grid A with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

22.52 2.28 28.99 2.97

37.54 3.81 48.93 4.94

54.11 5.43 84.39 8.35

73.26 7.23 117.78 11.41

98.38 9.56 138.49 13.24

124.26 11.88 163.34 15.39

159.47 14.95 198.57 18.38

190.79 17.77 224.17 20.42

214.34 19.97 239.19 21.61

241.27 22.50

87
300

250

200
Test
150
0 llr--ý It
100

50

0
0 10 20 30 40
Displacementmm

Fig 3.15. Experimentalload-deflectionresponseof node2

of grid A with tightnessdegreeof 120Nm

Table 3.8. Test results for node 2 of grid A with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

22.52 2.77 28.99 3.61

37.54 4.64 48.93 5.97

54.11 6.56 84.39 10.00

73.26 8.71 117.78 13.61

98.38 11.49 138.49 15.75

124.26 14.21 163.34 18.23

159.47 17.78 198.57 21.68

190.79 21.05 224.17 24.02

214.34 23.64 239.19 25.39

241.27 26.51

88
300

250

200
Test 1ý
150
0 Test 2
100

50

0
0 10 20 30 40
Displacementnun

Fig 3.16. Experimentalload-deflectionresponseof node3

of grid A with tightnessdegreeof 120Nm

Table 3.9. Test results for node 3 of grid A with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

22.52 3.21 28.99 4.22

37.54 5.38 48.93 6.98

54.11 7.64 84.39 11.64

73.26 10.11 117.78 15.78

98.38 13.32 138.49 18.18

124.26 16.47 163.34 20.96

159.47 20.50 198.57 24.81

190.79 24.17 224.17 27.45

214.34 27.04 239.19 28.99

241.27 30.25

89
3.3 Test grid B

Grid B is obtained by removing four bottom layer nodes of gird A which results in the

removal of 16 web elements as well as 16 bottom layer elements. So, due to the reduction
in the number of elements of the grid, grid B is more flexible in comparison with grid A. A

general view of grid B is shown in Fig 3.17. Also, a plan view of grid B is shown in Fig

3.18. In this figure, the top layer elements are shown by thick lines and the bottom layer

elements as well as the web elements are shown by thin lines.

Fig. 3.17. A general view of grid B

Grid B was tested with three different support arrangements.

" The first support arrangement consists of 4 two directional roller supports at the
four corners. The support positions are indicated by small squares in Fig 3.19. Grid
B with this support arrangement is referred to as grid B4. It should be mentioned

that the prevention of the lateral movement of grid B4 during the tests is achieved
in the manner described in section 2.8.3.

90
" The secondsupport arrangementis the sameas that of grid B4 with 4 additional
two directional roller supportsat the positions shown in Fig 3.20. Grid B with this
is to
support arrangement referred as grid B8.

" The third support arrangement is the same as that of grid B4 with 8 additional two
directional roller supports at the positions shown in Fig 3.21. grid B with this
is
supportarrangement referredto as grid B 12.

Fig 3.18. Plan view of grid B where the top layer elementsare shown by thick lines and the
bottom layer elementsas well as the web elementsare shownby thin lines.

When testing the grids, the load was applied to the central node of the top layer using a
hydraulicjack whoseposition is shown by a little circle in Fig 3.22. The amountof applied
load was measuredusing a load cell that was placed between the hydraulic jack and the

central node of the grid. The details regarding the arrangementof the hydraulic jack and
the load cell are discussedin section3.2.1.

91
Fig 3.19. Plan of grid B4

Fig 3.20. Plan of grid B8

Fig 3.21. Plan of grid B 12

92
In each one of the tests, the displacements of twelve bottom layer nodes were measured as

the response of the test grids using twelve displacement transducers. These nodes are

by little in Figs 3.22. The details concerning the displacement


shown solid circles
transducers and their support arrangement are discussed in section 3.2.1. In order to

determine the manner of the behaviour of the test grids during the test process,
displacementsof the nodesin the caseof unloadingwere measured.

1 12

4
11 <
/ N
/
V
\I
\ /3
\ V \/ \ /
1

4 9

/ T\8 1
5

6 7

Fig 3.22.Positionsof loadcell anddisplacement


transducers
on the grid B

3.3.1 Testing of grids B4, B8 and B 12


The main purposeof the experimental work in the present researchis to investigatethe

effects of the degreeof tightnessof the bolts of the connectorson the behaviour of double
layer grids. To this end, eachone of the test grids B4, B8 and B12 were testedfour times

with four different levels of tightness of the bolts. For each test all the bolts were initially

93
loosenedand then tightenedto a specific degreeof tightness.This was carried out using an

adjustable torque wrench. So, all the bolts of the connectors had the same degree of
tightnessfor eachtest. The actualtest procedurewas carried out as follows:

1. All the nodes of the grid were loosened and then tightened to 60 Nm degree of
tightness. Loosening and tightening process was carried out in the manner
described in Section 3.2.4.

2. The grid was hoisted using a crane. Four support assemblies were placed on the
base plates of the four corner columns. Attention was paid to placing the support

assemblies in the correct positions so that the welded steel angles could control the
horizontal movements of the test grid (see Section 2.8.3). Then, the grid was placed

on the support assemblies.

3. The hydraulic jack, load cell and displacement transducers were mounted in correct

positions. The load cell and transducers were connected to the data-logger using
thirteen cables. The setting up of the loading system and measurement equipments

were carried out as discussed in Section 3.2.1.

4. Testing of grid B4 with 60 Nm degreeof tightnesswas startedby the conditioning

operation.To elaborate,around 50 kN load was applied on the grid and after a few
secondsunloadingwas carried out. This operationwas repeatedfor a few times.

5. The test processwas continued as discussedin Section 3.2.2. Displacementsof

nodes 1 to 12 of grid B4 during loading in


and unloading were stored a file for

processing.In order to increasethe accuracyof the test results, the test procedure
was carried out twice.

6. When testing of grid B4 completed,the grid was hoisted using a crane.Four more

support assemblies were placed on the four columns to create the support
for
arrangement grid B8 (Fig 3.20). Then, the grid was placed on the eight support
assemblies.The test of grid B8 with 60 Nm degreeof tightnesswas carried out in
the mannerdescribedin steps3 to 5 above.

7. After the completion of testing of grid B8, the grid was hoisted using a crane.
Placing four more support assemblieson four remaining columns, the support

arrangementfor grid B12 (Fig 3.21) was created.Then, the grid was placedon the

94
twelve support assemblies.Grid B12 with 60Nm degreeof tightnesswas testedin
the samemanneras describedin steps3 to 5 above

8. When the testing of grids B4, B8 and B12 with 60 Nm degree of tightness was

completed, all the bolts of the nodes of grid B were loosened and then tightened to
120 Nm degree of tightness. Then, steps 2 to 7 were repeated to obtain the

responses of these grids under applied loads. Subsequently, this procedure was
repeated to test grids B4, B8 and B12 with degrees of tightness 180 Nm and 240
Nm.

3.3.2 Test results of grid B

In testing grid B with different support conditions and different degreesof tightness,
displacementsof nodes 1 to 12 of the grid under applied loads were measured.The load-
deflection responseof these nodes from output files were calculated as described in
Section3.2.3.

The geometryof test grid B and the applied load in the all caseswere symmetric.Also, the

componentsof grid B were identical. Furthermore,the support conditions of grids B4, B8


and B 12 were symmetric. Therefore, nodes 1,6,7 and 12 had to have the same
displacementsand the sameapplies to nodes2,5,8 and 11 and the nodes3,4,9 and 10.
So, in order to achieve more accurateresults, the averagevalue of the measurementsfor

eachone of the abovegroupsof four nodeswas consideredas the responsefor eachgroup


of symmetrically situatednodes.Therefore,displacementsof nodes 1,2 and 3 representthe
responsesof the abovementionedthree groupsof nodes.

The load-deflectionresponsesof nodes 1,2 and 3 of grids B4, B8 and B12 with different
degreesof tightnessare given in Tables 3.10 to 3.45. Also, the graphical versionsof these

results are shown in Figs 3.23 to 3.58. In each one of these figures two different load-
deflection responsesof a node under applied loads are represented(see step 5 in Section
3.3.1). The load-deflection responses shown by thin lines relate to the first tests

representingboth loading and unloading. The responsesof the secondtests are shown by
thick lines for both loading and unloading. So, eachfigure containsfour responsecurves:

" Loading for test 1 (thin line),

" Unloading for test 1 (thin line),

95
" Loading for test 2 (thick line) and

" Unloading for test 2 (thick line).

Generally, all the unloading branches of the load-deflection responses of the grids, which

are shown in Figs 3.23 to 3.58, are in good agreement with the corresponding loading

branch. This indicates that test grid B during all the tests remained in the elastic range.
Furthermore, all the four curves in most of the figures are very close and have good
conformity. This confirms the reliability of the test results.

96
180
160
140
120
100 door, Test
ö 80 Test 2
a
60
40
20 -0 0
0
os 10 is 20 25
Displacementmm

Fig 3.23. Experimentalload-deflectionresponseof node 1 of

grid B4 with 60 Nm degreeof bolt tightness

Table 3.10. Test resultsfor node 1 of grid B4 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

42.13 6.10 39.71 5.80

81.62 10.53 81.84 10.67

123.97 14.92 124.85 15.06

157.98 18.57 162.15 18.82

163.25 19.16 156.23 18.28

111.24 14.18 107.73 13.59

78.77 10.93 65.82 9.10

44.98 6.86 -1.54 -0.28

97
180
160
140
120
-z 100 Test1
0 80 Test 7
W-2 60

40
20

0
os 10 15 20 25
Displacement
nun

Fig 3.24. Experimental load-deflection response of node 2 of

grid B4 with 60 Nm degree of bolt tightness

Table 3.11. Test resultsfor node2 of grid B4 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

42.13 7.70 39.71 7.35

81.62 13.16 81.84 13.28

123.97 18.57 124.85 18.66

157.98 22.97 162.15 23.24

163.25 23.65 156.23 22.57

111.24 17.56 107.73 16.84

78.77 13.43 65.82 11.37

44.98 8.66
-1.54 -0.34

98
180
160
140
120
100 Test 1
b
ö 80 1 Z/- Test 2
a
60
40
20ý
0
05 10 15 20 25 30

Displacementmm

Fig 3.25. Experimental load-deflection responseof node 3 of

grid B4 with 60 Nm degreeof bolt tightness

Table 3.12. Test resultsfor node3 of grid B4 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

42.13 9.01 39.71 8.58

81.62 15.30 81.84 15.39

123.97 21.45 124.85 21.50

157.98 26.41 162.15 26.69

163.25 27.19 156.23 25.93

111.24 20.26 107.73 19.42

78.77 15.57 65.82 13.17

44.98 10.10 -1.54 -0.47

99
250

200

Y 150
Test 1
bcd
Test 2
100

50

05 10 15 20
Displacementmm

Fig 3.26. Experimental load-deflection responseof node 1 of

grid B8 with 60 Nm degreeof bolt tightness

Table 3.13. Test results for node 1 of grid B8 with 60 Nm tightness

Test 1 Test2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.29 4.43 45.65 4.10

96.35 8.03 103.80 8.40

138.49 10.98 145.50 11.15

189.41 14.37 187.42 13.80

224.31 16.73 224.95 16.11

170.75 13.57 188.74 14.01

115.89 10.13 133.65 10.63

64.75 6.55 92.83 7.90

-3.51 0.54 57.50 5.34

-2.64 -0.13

100
250

200

150
Test 1
-o
Test 2
100

50

0
05 10 15 20 25
Displacementmm

Fig 3.27. Experimental load-deflection responseof node 2 of

grid B8 with 60 Nm degree of bolt tightness

Table 3.14. Test resultsfor node2 of grid B8 with 60 Nm tightness

Test 1 Test2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.29 6.12 45.65 5.74

96.35 10.77 103.80 11.34

138.49 14.61 145.50 14.98

189.41 19.00 187.42 18.47

224.31 21.96 224.95 21.39

170.75 17.91 188.74 18.70

115.89 13.39 133.65 14.30

64.75 8.69 92.83 10.71

-3.51 0.32 57.50 7.35

-2.64 -0.24

101
250

200

ýt
3Z150
100 Test 2

50

0
05 10 15 20 25 30
Displacementmm

Fig 3.28. Experimentalload-deflectionresponseof node3 of

grid B8 with 60 Nm degreeof bolt tightness

Table 3.15. Test results for node 3 of grid B8 with 60 Nm tightness

Test 1 Test2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.29 7.84 45.65 7.06

96.35 13.03 103.80 13.73

138.49 17.51 145.50 18.00

189.41 22.61 187.42 22.05

224.31 26.05 224.95 25.53

170.75 21.32 188.74 22.30

115.89 15.98 133.65 17.14

64.75 10.43 92.83 12.91

-3.51 0.20 57.50 8.94

-2.64 -0.34

102
300

250

200
Test 1
150
Test 2
100

50

0
05 10 15

Displacementmm

Fig 3.29. Experimental load-deflection responseof node 1 of

grid B 12 with 60 Nm degree of bolt tightness

Table 3.16. Test results for node 1 of grid B12 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

43.45 2.64 45.00 2.74

97.01 5.37 93.06 5.16

146.61 7.65 131.25 6.91

185.90 9.28 172.73 8.64

238.36 11.41 226.94 10.78

281.82 13.09 281.59 12.84

276.33 12.93 240.77 11.42

209.38 10.56 182.39 9.22

160.22 8.39 130.59 7.07

113.69 6.63 91.74 5.05

68.25 4.45 54.43 3.42

103
300

250

200
Ts1
150 Tacr 7

100

50

0
05 10 15 20 25
Displacement mm

Fig 3.30. Experimentalload-deflectionresponseof node2 of

grid B12 with 60 Nm degreeof bolt tightness

Table 3.17. Test results for node 2 of grid B 12 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

43.45 4.25 45.00 4.47

97.01 8.66 93.06 8.36

146.61 12.29 131.25 11.16

185.90 14.96 172.73 13.94

238.36 18.36 226.94 17.40

281.82 21.14 281.59 20.71

276.33 20.87 240.77 18.39

209.38 16.97 182.39 14.82

160.22 13.88 130.59 11.42

113.69 10.74 91.74 8.61

68.25 7.29 54.43 5.63

104
300

250

200

150 T__a o%

100

50

05 10 15 20 25 30
Displacement mm

Fig 3.31. Experimentalload-deflectionresponseof node3 of

grid B 12 with 60 Nm degreeof bolt tightness

Table 3.18. Test resultsfor node3 of grid B 12 with 60 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

43.45 5.34 45.00 5.74

97.01 10.89 93.06 10.66

146.61 15.48 131.25 14.16

185.90 18.84 172.73 17.68

238.36 23.20 226.94 22.04

281.82 26.77 281.59 26.24

276.33 26.42 240.77 23.31

209.38 21.49 182.39 18.77

160.22 17.58 130.59 14.46

113.69 13.56 91.74 10.91

68.25 9.21 54.43 7.15

105
200 , ----
180
160
140
120 ---
Test 1
100
Teat 2
80
60
40
20ý-- -
0-
05 10 15 20 25
Displacement
mm

Fig 3.32. Experimental load-deflection response of node 1 of

grid B4 with 120 Nm degree of bolt tightness

Table 3.19. Test resultsfor node 1 of grid B4 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

55.29 7.21 51.78 6.97

97.85 11.89 98.52 12.04

135.37 15.84 138.01 15.93

170.69 19.55 174.65 19.46

182.54 20.81 182.33 20.18

125.06 15.52 159.73 18.21

80.74 10.91 125.29 14.88

-1.32 0.21 78.33 10.00

106
200
180
160
140
120
Test I'
100
° -Test 2
80
,.
60
40
20
0
05 10 15 20 25 30

Displacementmm

Fig 3.33. Experimental load-deflection responseof node 2 of

grid B4 with 120 Nm degree of bolt tightness

Table 3.20. Test resultsfor node2 of grid B4 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

55.29 8.98 51.78 8.63

97.85 14.68 98.52 14.56

135.37 19.44 138.01 19.57

170.69 23.86 174.65 23.86

182.54 25.32 182.33 24.74

125.06 18.92 159.73 22.32

80.74 13.28 125.29 18.28

-1.32 0.10 78.33 12.31

107
200
180
160
140
120
Test I
100
°ý 80 Test 2
.
60
40
20
0
0 10 20 30 40
Displacementmm

Fig 3.34. Experimentalload-deflectionresponseof node3 of

grid B4 with 120 Nm degree of bolt tightness

Table 3.21. Test resultsfor node3 of grid B4 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

55.29 10.55 51.78 10.07

97.85 17.06 98.52 17.13

135.37 22.46 138.01 22.56

170.69 27.43 174.65 27.41

182.54 29.08 182.33 28.42

125.06 21.79 159.73 25.65

80.74 15.47 125.29 21.05

-1.32 0.08 78.33 14.34

108
300

250

200
z Test 1
150 Test 2
0
100
50
0
05 10 15 20
Displacement
mm

Fig 3.35. Experimentalload-deflectionresponseof node 1 of

grid B8 with 120Nm degreeof bolt tightness

Table 3.22. Test resultsfor node 1 of grid B8 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.37 4.11 53.75 4.59

102.46 8.09 103.78 8.19

144.81 10.95 146.34 10.99

184.96 13.52 184.30 13.31

220.07 15.81 223.79 15.68

247.93 17.62 254.51 17.57

204.49 15.15 207.12 14.89

156.88 12.32 154.02 11.73

115.41 9.71 111.02 8.99

63.63 6.05 73.50 6.39

109
300

250

200
Test 1
0
150 2
-Test
a
100
50
0 00-0ý
o5 10 15 20 25
Displacement
mm

Fig 3.36. Experimentalload-deflectionresponseof node2 of

grid B8 with 120Nm degreeof bolt tightness

Table 3.23. Test results for node 2 of grid B8 with 120 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.37 5.61 53.75 6.17

102.46 10.79 103.78 10.83

144.81 14.56 146.34 14.49

184.96 18.02 184.30 17.59

220.07 21.01 223.79 20.67

247.93 23.37 254.51 23.08

204.49 20.19 207.12 19.64

156.88 16.49 154.02 15.53

115.41 13.00 111.02 11.93

63.63 8.25 73.50 8.59

110
300

250

200
z Test 1
0
150 T- 1%

100
50

05 10 15 20 25 30
Displacement
mm

Fig 3.37. Experimentalload-deflectionresponseof node3 of

grid B8 with 120Nm degreeof bolt tightness

Table 3.24. Test resultsfor node3 of grid B8 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.37 6.81 53.75 7.51

102.46 12.98 103.78 13.03

144.81 17.49 146.34 17.35

184.96 21.57 184.30 20.96

220.07 25.06 223.79 24.54

247.93 27.86 254.51 27.41

204.49 24.15 207.12 23.36

156.88 19.83 154.02 18.58

115.41 15.78 111.02 14.40

63.63 10.18 73.50 10.43

111
350

300

250

200 Test
-
Tr--. 7
150
100
50
0
o5 10 is
Displacement mm

Fig 3.38. Experimentalload-deflectionresponseof node 1 of

grid B12 with 120Nm degreeof bolt tightness

Table 3.25. Test results for node 1 of grid B 12 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

47.85 2.65 55.09 2.95

109.53 5.54 105.35 5.34

139.82 6.84 149.03 7.21

197.77 9.18 185.68 8.70

230.03 10.46 228.04 10.33

262.52 11.83 259.86 11.49

299.61 13.48 299.59 13.00

254.39 11.91 245.60 11.14

205.45 10.10 204.11 9.62

153.87 8.06 160.44 7.92

114.36 6.38 112.37 5.89

70.90 4.35 72.65 4.05

112
350

300

250

200 i
Test
150 Test 2'

100

50

0
05 10 15 20 25
Displacement mm

Fig 3.39. Experimentalload-deflectionresponseof node2 of

grid B 12 with 120Nm degreeof bolt tightness

Table 3.26. Test results for node2 of grid B 12 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

47.85 4.45 55.09 4.90

109.53 9.08 105.35 8.70

139.82 11.17 149.03 11.68

197.77 14.92 185.68 14.05

230.03 16.99 228.04 16.67

262.52 19.18 259.86 18.58

299.61 21.87 299.59 21.04

254.39 19.34 245.60 18.05

205.45 16.46 204.11 15.59

153.87 13.21 160.44 12.88

114.36 10.55 112.37 9.66

70.90 7.35 72.65 6.77

113
350

300

250

200 T1

150 Test 2,

100

50

0
05 10 15 20 25 30
Displacement mm

Fig 3.40. Experimentalload-deflectionresponseof node 3 of

grid B 12 with 120Nm degreeof bolt tightness

Table 3.27. Test results for node 3 of grid B12 with 120Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

47.85 5.64 55.09 6.20

109.53 11.52 105.35 10.99

139.82 14.16 149.03 14.77

197.77 18.91 185.68 17.76

230.03 21.51 228.04 21.10

262.52 24.32 259.86 23.52

299.61 27.71 299.59 26.64

254.39 24.51 245.60 22.84

205.45 20.82 204.11 19.71

153.87 16.71 160.44 16.28

114.36 13.41 112.37 12.22

70.90 9.28 72.65 8.55

114
200
180
160
140
120
Test l
Iti 100
80 Test2

60
40
20
0
05 10 15 20 25
Displacement
mm

Fig 3.41. Experimentalload-deflectionresponseof node 1 of

grid B4 with 180Nm degreeof tightness

Table 3.28. Test results for node 1 of grid B4 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.92 5.87 41.90 5.13

101.35 11.77 92.79 10.72

147.86 16.88 145.22 16.11

171.99 19.51 167.16 18.35

102.89 12.90 143.69 16.26

-1.96 0.42 110.56 13.12

68.88 8.74

115
200
180
160
140
120 Test
100
80 Test2

60
40
20
0
05 10 15 20 25
Displacement
mm

Fig 3.42. Experimentalload-deflectionresponseof node2 of

grid B4 with 180Nm degreeof bolt tightness

Table 3.29. Test results for node 2 of grid B4 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.92 7.19 41.90 6.23

101.35 14.34 92.79 13.01

147.86 20.42 145.22 19.49

171.99 23.48 167.16 22.16

102.89 15.59 143.69 19.68

-1.96 0.44 110.56 15.91

68.88 10.66

116
200
180
160
140
120
Test 1'
100
° Test2
80
60
40
20
0
0 10 20 30
Displacement
mm

Fig 3.43. Experimentalload-deflectionresponseof node 3 of

grid B4 with 180Nm degreeof bolt tightness

Table 3.30. Test results for node 3 of grid B4 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

48.92 8.44 41.90 7.19

101.35 16.48 92.79 14.94

147.86 23.34 145.22 22.29

171.99 26.80 167.16 25.30

102.89 17.76 143.69 22.19

-1.96 0.38 110.56 18.14

68.88 12.20

117
300

250

200
Test
150
0 Test 2
100

50

0
05 10 15 20
Displacement
mm

Fig 3.44. Experimentalload-deflectionresponseof node 1 of

grid B8 with 180Nm degreeof bolt tightness

Table 3.31. Test results for node 1 of grid B8 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

59.01 4.30 53.97 3.93

101.78 7.18 101.35 7.20

147.63 10.25 148.52 10.32

202.03 13.73 195.46 13.18

241.52 16.29 247.45 16.29

277.27 18.90 277.50 18.07

223.53 15.93 156.85 11.17

165.40 12.27 108.15 8.05

118.24 9.47

77.44 6.66

118
300

250

200
Test 1'
150
0 Test 2'
100

50

0
05 10 15 20 25 30
Displacement
mm

Fig 3.45. Experimentalload-deflectionresponseof node2 of

grid B8 with 180Nm degreeof bolt tightness

Table 3.32. Test results for node 2 of grid B8 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

59.01 5.81 53.97 5.32

101.78 9.61 101.35 9.51

147.63 13.49 148.52 13.45

202.03 17.93 195.46 17.19

241.52 21.13 247.45 21.18

277.27 24.40 277.50 23.43

223.53 20.63 156.85 14.55

165.40 16.15 108.15 10.57

118.24 12.34

77.44 8.74 : 71

119
300

250

200
Tst
150
Test2,
0
100

50

0
0 10 20 30 40
Displacementmm

Fig 3.46. Experimentalload-deflectionresponseof node3 of

grid B8 with 180Nm degreeof bolt tightness

Table 3.33. Test results for node 3 of grid B8 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

59.01 7.06 53.97 6.47

101.78 11.59 101.35 11.48

147.63 16.15 148.52 16.14

202.03 21.37 195.46 20.48

241.52 25.09 247.45 25.18

277.27 28.97 277.50 27.88

223.53 24.44 156.85 17.37

165.40 19.17 108.15 12.66

118.24 14.67

77.44 10.44

120
300

250

200
Test
C 150 Test 2
100
50
0
os 10 15
Displacement
mm

Fig 3.47. Experimentalload-deflectionresponseof node 1 of

grid B 12 with 180Nm degreeof bolt tightness

Table 3.34. Test results for node 1 of grid B12 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

50.91 2.34 54.40 2.41

101.82 4.45 102.88 4.37

152.28 6.58 154.88 6.35

192.97 8.37 200.95 8.03

236.31 10.11 245.70 9.59

282.39 12.07 281.01 10.78

235.64 10.52 244.82 9.48

165.21 7.88 193.49 7.49

111.68 5.71 141.49 5.41

63.41 3.62 92.35 3.32

121
300

250

200
Test
150
0 Test 2
a 100

50

0
os 10 15 20 25
Displacement
mm

Fig 3.48. Experimentalload-deflectionresponseof node2 of

grid B 12 with 180Nm degreeof bolt tightness

Table 3.35. Test results for node2 of grid B 12 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

50.91 3.96 54.40 4.10

101.82 7.54 102.88 7.37

152.28 11.09 154.88 10.64

192.97 13.96 200.95 13.35

236.31 16.81 245.70 15.94

282.39 20.02 281.01 17.90

235.64 17.44 244.82 15.78

165.21 13.37 193.49 12.57

111.68 9.62 141.49 9.20

63.41 6.21 92.35 5.80

122
300

250

200

150 Test1
m
ä 100
Te, + ')

50 4

0-0
0 5 10 1S 20 25 3
-50
Displacement
mm

Fig 3.49. Experimentalload-deflectionresponseof node 3 of

grid B12 with 180Nm degreeof bolt tightness

Table 3.36. Test resultsfor node 3 of grid B12 with 180Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

50.91 5.20 54.40 5.45

101.82 8.80 102.88 9.65

152.28 14.30 154.88 13.85

192.97 17.98 200.95 17.35

236.31 21.63 245.70 20.65

282.39 25.69 281.01 23.09

235.64 22.36 244.82 20.43

165.21 16.90 193.49 16.37

111.68 12.41 141.49 12.10

63.41 8.14 92.35 7.83

123
200
180
160
140
120 Tes
a 1
100
Test 2
80
60
40
20
0

05 10 15 20 25
Displacement
mm

Fig 3.50. Experimentalload-deflectionresponseof node 1 of

grid B4 with 240 Nm degreeof bolt tightness

Table 3.37. Test results for node 1 of grid B4 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.36 6.06 50.89 6.11

98.93 11.61 96.08 10.99

137.32 15.65 137.76 15.15

171.54 19.16 168.70 18.12

114.94 13.83 165.18 17.81

72.39 9.27 154.43 16.83

20.62 3.06 105.29 11.98

-1.96 0.02 61.86 7.14

11.19 1.03

124
200
180
160 1Oosý
140
z 120 Tes
1
14
"v
«s
100
0 80 Test2
60
40
20
0
05 10 15 20 25
Displacement
mm

Fig 3.51. Experimentalload-deflectionresponseof node2 of

grid B4 with 240 Nm degreeof bolt tightness

Table 3.38. Test results for node 2 of grid B4 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.36 7.39 50.89 7.38

98.93 14.17 96.08 13.32

137.32 19.99 137.76 18.31

171.54 23.17 168.70 21.87

114.94 16.72 165.18 21.49

72.39 11.23 154.43 20.31

20.62 3.70 105.29 14.45

-1.96 -0.07 61.86 8.63

11.19 1.18

125
200
180
160
140
120
Test 1
100
Test 2
80
60
40
20
0
0 10 20 30
Displacement
mm

Fig 3.52. Experimentalload-deflectionresponseof node 3 of

grid B4 with 240 Nm degreeof bolt tightness

Table 3.39. Test results for node 3 of grid B4 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.36 8.57 50.89 8.52

98.93 16.33 96.08 15.30

137.32 21.79 137.76 20.97

171.54 26.53 168.70 24.99

114.94 19.19 165.18 24.56

72.39 12.93 154.43 23.20

20.62 4.31 105.29 16.56

-1.96 -0.07 61.86 9.89

11.19 1.40

126
300

250

200
Test 1
150
-Test 2
a0
100

50

0
0s 10 15 20
Displacement
mm

Fig 3.53. Experimentalload-deflectionresponseof node 1 of

grid B8 with 240 Nm degreeof bolt tightness

Table 3.40. Test resultsfor node 1 of grid B8 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.80 3.78 52.65 3.93

97.18 6.96 100.03 7.18

152.90 10.55 152.46 10.61

198.52 13.37 191.50 13.02

256.21 16.92 245.24 16.27

275.08 18.08 274.42 18.04

216.07 14.82 247.87 16.62

142.59 10.33 188.43 13.16

78.09 6.01 125.69 9.23

-1.52 0.05 75.02 5.82

127
300 T

250 --

200
Tes

G 150 Test 2'
100
50--
0--
0 5 10 15 20 25
Displacement
mm

Fig 3.54. Experimentalload-deflectionresponseof node2 of

grid B8 with 240 Nm degreeof bolt tightness

Table 3.41. Test results for node 2 of grid B8 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.80 4.98 52.65 5.25

97.18 9.13 100.03 9.54

152.90 13.74 152.46 13.96

198.52 17.53 191.50 17.06

256.21 21.95 245.24 21.25

275.08 23.42 274.42 23.48

216.07 19.21 247.87 21.64

142.59 13.44 188.43 17.20

78.09 7.92 125.69 12.22

-1.52 -0.11 75.02 7.74

128
300

250

200
Tes
1
150
- Test2
0
100

50

0
05 10 15 20 25 30

Displacement
mm

Fig 3.55. Experimentalload-deflectionresponseof node 3 of

grid B8 with 240 Nm degreeof bolt tightness

Table 3.42. Test results for node 3 of grid B8 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

49.80 6.09 52.65 6.42

97.18 11.09 100.03 11.54

152.90 16.54 152.46 16.73

198.52 20.82 191.50 20.38

256.21 26.08 245.24 25.24

275.08 27.83 274.42 27.92

216.07 22.83 247.87 25.70

142.59 16.03 188.43 20.50

78.09 9.51 125.69 14.55

-1.52 -0.17 75.02 9.33

129
350

300

250

200 Test 1

0 150 Test 2'
a
100

50

0
os 10 15
Displacementnun

Fig 3.56. Experimentalload-deflectionresponseof node 1 of

grid B 12 with 240 Nm degreeof bolt tightness

Table 3.43. Test results for node 1 of grid B12 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

60.76 2.77 54.40 2.66

107.93 4.73 101.13 4.40

148.96 6.46 142.15 6.35

192.61 8.25 186.24 8.11

225.74 9.64 230.33 9.80

264.35 11.21 261.04 10.93

318.10 13.35 295.04 12.12

244.20 10.76 241.30 10.29

189.32 8.78 182.07 8.10

121.10 6.03 123.94 5.79

70.42 3.80 65.59 3.24

130
350

300

250

200 Test

150 Test 2

100

50

0
05 10 15 20 25
Displacementmm

Fig 3.57. Experimental load-deflection response of node 2 of

grid B12 with 240 Nm degree of bolt tightness

Table 3.44. Test results for node2 of grid B 12 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

60.76 4.59 54.40 4.34

107.93 7.85 101.13 7.63

148.96 10.70 142.15 10.38

192.61 13.53 186.24 13.19

225.74 15.68 230.33 15.94

264.35 18.14 261.04 17.75

318.10 21.66 295.04 19.77

244.20 17.43 241.30 16.78

189.32 14.20 182.07 13.19

121.10 9.82 123.94 9.48

70.42 6.26 65.59 5.40

131
350

300

250
200 Test 1
150 Test 2

100

50
0
05 10 15 20 25 30

Displacement mm

Fig 3.58. Experimentalload-deflectionresponseof node 3 of

grid B12 with 240 Nm degreeof bolt tightness

Table 3.45. Test results for node 3 of grid B12 with 240 Nm tightness

Test 1 Test 2

Load (kN) Deflection (mm) Load (kN) Deflection (mm)

0.00 0.00 0.00 0.00

60.76 6.00 54.40 5.62

107.93 10.11 101.13 9.84

148.96 13.71 142.15 13.33

192.61 17.37 186.24 16.95

225.74 20.10 230.33 20.41

264.35 23.23 261.04 22.75

318.10 27.65 295.04 25.24

244.20 22.35 241.30 21.43

189.32 18.22 182.07 16.89

121.10 12.66 123.94 12.15

70.42 8.17 65.59 7.00

132
CHAPTER 4

PRESENTATION OF TEST
RESULTS

4.1 Introduction

The behaviour of a connector in a spacestructure is not always bending dominated.In a

numberof spacestructurefamilies such as double layer grids, domeswith high rise to span
ratio and towers, the behaviour is dominatedby axial force. This differs from conventional
framed structureswhere behaviouris mainly governedby bending.Therefore,an important

role of the connectorsin spacestructuresis to transfer axial forces.

The experimentalload-deflection responsesof the test grids with different degreesof bolt
tightnessare presentedand discussedin this chapter.In order to comparethe experimental
and analytical responses,the numerical results obtained through linear elastic analysis of
the grids for relevant nodesare also presentedin the presentchapter.

4.2 Analysis of the test grids with pin and rigid joints

The analysis and design of lattice spacestructures are normally simplified by using the

assumption that the connectors behave as pinned or rigid joints. However, the actual
behaviourof the joints doesnot conform to either of theseextremes.The actualresponseof

most practical connectorswill fall somewherebetweenthe two cases,resulting in what is

133
generally termed semi-rigid behaviour. Also, the behaviour of a semi-rigidly jointed space
is
structure betweenthe extremesof absoluterigidity and absoluteflexibility provided.

In order to compare the experimental load-deflection responses of the test grids with

numerical results obtained through analysis of the grids, the test grids A, B4, B8 and B12

were analysed using the ETEBS version 8 package. The analysis of each grid was carried
out twice. The first analysis was performed using the assumption that all the joints of the
grid behave as pin joints. In the second analysis all the joints were modelled as rigid joints.

In order to create the test grids data in ETABS, the geometric particulars of grids A and B

were consideredas mentioned in Sections 2.3.1 and 2.3.2. The cross-sectionof all the
elementswas considered to be a tube with outer diameter of 76.4 mm and thickness of 3.5

mm. The material was considered to be steelwith the modulus of elasticity of 206 kN/mm2
for all the elements.

X, YandZ
degreesof of
freedomare i are
constrained ned

All the
supports
a single
raint in
direction

Z upward X

Fig 4.1. General support condition of the grids

The supportcondition for the analytical model was chosenas follows:

" Three translational degrees of freedom of one of the


corner supports were
constrained, as shown in Fig 4.1.

134
" Two translational degreesof freedom of another corner supportwere constrained,

as shown in Fig 4.1.

" All the other supportshad their vertical degreeof freedomconstrained.

In the analysis, the self-weight of the grids was not taken into account. This is justified on

the basis that the joint deflection measurementsduring the testing of the grids did not
include the effect of the self-weight of the grids. Therefore,to be comparablewith the test

results,the analytical deflection calculationsshould not considerthe self-weight.

The testing of the grids was carried out in the elastic range for all the cases.As far as the
theoretical results are concerned,the deflections for nodes 1,2 and 3 were obtained for a
central load of 100 kN. Thesetheoretical deflections are shown in Table 4.1.

Table 4.1. Numerical resultsobtainedfrom analysisof the grids for

pin and rigid joint conditions

Deflection for Deflection for Percentage


pinjoint condition rigid joint condition difference
mm mm

Node 1 6.8145 6.7316 1.2%

Grid A Node 2 7.7934 7.7164 1.0 %

Node 3 8.6742 8.5957 0.9 %

Node 1 7.1018 7.0278 1.1 %

Grid B4 Node 2 8.3965 8.3225 0.9 %

Node 3 9.4280 9.3493 0.8 %

Node 1 2.8665 2.8148 1.8%

Grid B8 Node 2 4.8516 4.7839 1.4%

Node 3 6.2075 6.1295 1.3 %

Node 1 2.3259 2.2715 2.4 %

Grid B12 Node 2 3.9891 3.9117 2.0%

Node 3 5.1788 5.0869 1.8%

135
In Table 4.1, deflections of nodes 1,2 and 3 of the test grids obtained from analysis of the

grids with the assumption of pin joint are shown in the third column. Also, the deflections

of the nodes when the joints of the grids are assumed to be rigidly jointed are given in the
fourth column. The comparison of the corresponding values in these columns shows that

the deflections of the nodes for pin and rigid cases are very close, as may be seen from the
fifth column of the table.

The fact that the analytical responses of the test grids for both pin and rigid joint conditions

are very close is predictable. It is a fact that the manner of modelling of connectors for

structureshaving the following three conditions has no significant effect on the responseof
the structure.The conditions are:

" The structurewith the assumptionthat all the connectorsbehaveas pin joints to be

statically stable.

" The structureto be loadedat joints only.

" The proportions of the cross-sectionsof the members are such that the bending

rigidities are not abnormally high.

In practice,the analysisof double layer grids that satisfy the aboveconditions is simplified
by using the assumptionthat the connectionsbehaveas pin joints. In the presentwork, the

results obtained from the pin jointed analysis of the test grids are considered as the
theoreticalresponses.

4.3 Presentationof theoretical and experimentalresults of grid A

The experimentalload-deflection responsesof nodes 1,2 and 3 of grid A with different


degreesof tightnessare shown in Figs 4.2,4.3 and 4.4. Each figure correspondsto a node

and includes the load-deflection responsesobtained for that node. In these figures,
responsesof the nodes of grid A with 60 Nm degree of tightness are drawn by dashed
lines. Also, the responsesof the nodes with normal and 120 Nm degree of tightness are
drawn by dotted and dash-dot lines, respectively. The letter T is an abbreviation for
"tightness" of the bolts. The numerical results obtained through linear elastic analysis of
the test grid A for the relevant nodesare also shown as a straight line in the Figs 4.2 to 4.4.
The analysisresultsare shown by thick lines and are labelled "Analysis".

136
300 i
OOOF,
250
i.
200 f"F normal

IT 60
150
0 Ir"ý j
Iý'%
100 ,
I ý`
,
III
"i

50
.yi

0
05 10 15 20 25 30 35 40

Displacementmm

Fig 4.2. Load-deflectionresponsesof node 1 of grid A with different degreesof tightness

of bolts togetherwith the analysisresults

300

250
T 120
,4
T
I normal,
200 .
i wo 60 i
M
C
150
100

50

0
05 10 15 20 25 30 35 40
Displacement
mm

Fig 4.3. Load-deflectionresponsesof node 2 of grid A with different degreesof tightness

of bolts togetherwith the analysisresults

137
300

250 - T120

.'T normal
200 -

, "' T60
150 - ,."t'"" ,-
0

1004 r
i
-' i
-.. ,.
50 V
Wit' ' -'

o;
5 10 15 20 25 30 35 40
0
Displacement
mm

Fig 4.4 Load-deflectionresponsesof node 3 of grid A with different degreesof tightnessof


bolts togetherwith the analysisresults

A general observation of the graphs of Figs 4.2 to 4.4 reveals that unlike the analysis

results, all of the experimentalresponseshave a nonlinear behaviour, in particular, in the


initial stagesof the loading. Furthermore, it may be observedthat all the graphs of the

experimental results are on the right-hand side of the graph of analysis results. These

observationslead to four interestingconclusions:

The first conclusion is that the actual behaviour of the test grid is rather different from that

suggestedby the linear analysis.The analytical load-deflection relationshipsfor the nodes


of the grid are linear whereasthe experimentalresults reveal that the actual responsesare
nonlinear. This does not imply that a nonlinear analysis would have given rise to results
similar to the experimental responses. Actually, the difference between the linear and
(geometrically)nonlinear numerical results in this casewould be negligible.

The second conclusion concerns the fact that the experimental results in all the cases
indicate that the grid is more flexible than the numerical results suggest.This can be

explained by the fact that the analytical model cannot accurately representthe actual test
grid. In order to produce data for the analysis of grid A, it was assumedthat the grid

138
consists of tubular steel members with a uniform cross-section that are connected together
at the intersection points of the members. These intersection points were considered as the
nodes of the grid in Fig 2.2. However, the actual test grid is constructed using a MERO-

type system. A typical component of the MERO system consists of a steel tube element
together with the end connectors, as shown in Fig 4.5. This component is connected to the

rest of the structure by means of the end balls. In the analytical model, the centre points of
the balls are considered as the nodes of the grid and the geometrical as well as mechanical

properties of the tubular elements are considered for the whole of the member. In reality,
the main part of the members of the grid is tubular, but the presence of the connectors

whose actual behaviour is not taken into account in the analysis has a major influence on
the response of the grid.

Tubular Element

Fig 4.5. A typical componentof the test grid

In order to discussthe matter in greaterdetail, four tubular elementsof the test grid that are

connectedtogether by means of a MERO-type connector are shown in Fig 4.6a. (The


constituentpartsof the MERO jointing systemas well as the mannerof transmissionof the
tensile, compressive and bending effects in the MERO-type system are described in
Section 1.3).

In the analysis process,the arrangementof Fig 4.6a is representedby four lines that are

connectedtogether at a node, as shown in Fig 4.6b. The length of each one of theseline
elementsis 1414mm, that is, the distancebetweentwo adjacentnodes(see Section2.3.1).
This length consistsof two parts, 1200 mm (85% of the line element) for the tube and 214
mm (15% of the line element) for the end connectors.In reality 15% of the length of each
line member that relates to the end connectorshas different geometrical
and mechanical
properties in comparison with the tubular element. Furthermore, the manners of the
connectivity of the tubular elementsto the connectorsas well as the connectorsthemselves
are not consideredin the analytical model. This explains why the analytical model cannot

139
represent the test grid accurately and, therefore, a disagreement between analytical results
and measured responses of the test grid is inevitable.

Tubular elements
connected by means of Analytical model of
tubular elements

Analyt
of co

k-I (b)

Fig 4.6. A view of four members of the test grid together with its analytical model

The above problem also exists for the space structures constructed using a MFRO-type

system when analysed using the standard analysis method. In the standard analysis method
it is assumed that the whole of the structure consists of elements with uniform cross-

section and material. Also, it is assumed that the elements are connected together at the
intersection points either rigidly or by pinned joints. For instance, a considerable
discrepancy was discovered between the analytical deflections and the measured
deflections during the construction of the Split stadium in former Yugoslavia 121. This

stadium was built for the VIII Mediterranean Games in 1979. It has a shallow double layer
barrel vault roof spanning 180 m. the stadium was constructed in MERO system and when

the temporary supports were removed the deflections were measured. These deflections

were found to exceed the analytical predictions by 70%.

Some researchers have claimed that the discrepancy between the


actual and theoretical
responses of the space structures is due to semi-rigidity of the connectors. They also
mentioned that the actual behaviour of the structure lies somewhere between pin and rigid
joints conditions [31. The comparison of the
corresponding values in the third and fourth

140
columns of Table 4.1 shows that the maximum percentage difference of deflections of the
nodes for pin and rigid cases are 2.4%. However, the experimental results indicate that the
percentage of difference between deflections of each one of the nodes with 60 Nm and 120
Nm degrees of bolt tightness is much more 2.4%. This shows that the actual responses of

grid A could not be justified by semi-rigid behaviour of their connectors.

The third conclusion is that the behaviour of the test grid is indeed affected by the degree

of tightness of the bolts of the connectors. This may be clearly seen from the response
curves in Figs 4.2 to 4.4. During the tests of grid A, the grid configuration, load and
support conditions as well as the member properties for all the caseswere the same and the
sole variable was the degree of tightness of the bolts. Therefore, it can be concluded that
the degree of tightness of bolts will affect the behaviour of the grid.

In Figs 4.2 to 4.4 each graph is a load-deflection response of a node of grid A

correspondingto a degreeof tightness of the bolts. In order to obtain the responseof the
grid with 60 Nm and 120 Nm degreesof tightness, all the bolts of the connectorswere
loosenedand then tightened to 60 Nm and 120 Nm, respectively,using a torque wrench
(see Section 3.2.4). However, in practice the bolts of the connectorsare usually tightened

using ordinary spanners.In this casethe degreeof tightnessdependson the length of the
spanneras well as the amount of applied force by the worker. Since the amount of the
force dependson the position as well as the physical conditions of the worker and cannot
be accuratelycontrolled, the connectorsare not tightened to a specific degreeof tightness.
The term "normal tightness" is usedto refer to this stateof tightnessof the bolts in a grid.
The fourth conclusion is that the load-deflection responsesof the nodes of grid A with

normal tightness, as may be observed in Figs 4.2 to 4.4, lie between the graphs
correspondingto 60 Nm and 120Nm degreesof tightness.

4.4 Discussion of the results of grid B

Figs 4.7 to 4.15 show the experimental load-deflection responsesof nodes 1,2 and 3 of

grid B with different support conditions (B4, B8 and B12) and different degrees of
tightness(60,120,180 and 240 Nm). Each figure correspondsto a node of grid B with a
specific support condition and includes the load-deflection responsesobtained for that
node. In Figs 4.7 to 4.15, responsesof the nodeswith 60 Nm degreeof tightnessare drawn
by dashedlines. Also, the responsesof the nodes with 120 Nm, 180 Nm and 240 Nm
degreeof tightnessare drawn by dotted, dash-dot and full lines, The degrees
respectively.

141
of tightness are indicated by T60, T120, T180 and T240. Here, the letter T is an
abbreviation for "tightness". The numerical results obtained through linear elastic analysis
of the test grid B for the relevant nodes are also shown as a straight line in these figures,
labelled "Analysis".

A general observation of the graphs of Figs 4.7 to 4.15 reveals that unlike the analysis

results, all of the experimental responses have a nonlinear behaviour, in particular, in the
initial stages of the loading. Furthermore, it may be observed that all the graphs of the

experimental results are on the right-hand side of the graph of the analysis results. These
observations lead to the following conclusions:

" In spite of the fact that the grid was tested in the elastic range and that even

geometrically nonlinear analysesof the grid suggestlinear for


responses the nodes,
the experimentalresults reveal that the actual load-deflectionrelationshipsof these
nodesare nonlinear. This nonlinear responseis thought to be due to the nonlinear
behaviourof the connectors.

" The experimental results indicate that in all the casesthere are a considerable
disagreementbetweenactual deflectionsof the nodesof the grid and the analytical

results. This is due to the differencesbetweenthe analytical model of the grid and
the actual grid, as explainedin Section4.3.

" The degreeof tightness of the bolts of the connectorsaffects the behaviour of the

grid. This may be clearly seenfrom the load-deflectionresponsesshown in Figs 4.7


to 4.15.

" The experimentalload-deflection responsesof the nodesreveal that in all the cases
the flexibility of the grid will decreaseclearly when the degreeof tightnessof the
bolts increase.This trend continuesuntil 180 Nm degreeof tightness of the bolts.
After that increasing the tightness of the bolts has no significant effect on the
behaviour of some of the cases.However, this increasein some casesresults in a

more flexible behaviour.

142
300

250

200 -
2
150 -
0
100 -

50

0-
0 5 10 15 20 25 30
Displacement
mm

Fig 4.7. Load-deflectionresponsesof node 1 of grid B4 with different degreesof tightness

of bolts togetherwith the analysisresults

300

250

200
z
150
0

100

50


0 5 10 15 20 25 30
Displacement
mm

Fig 4.8. Load-deflectionresponsesof node 2 of grid B4 with different degreesof tightness

of bolts togetherwith the analysisresults

143
300-r

250 --

200

150
0
100-

50-

0--
0 5 10 15 20 25 30
Displacement
mm

Fig 4.9. Load-deflectionresponsesof node 3 of grid B4 with different degreesof tightness

of bolts togetherwith the analysisresults

300
T 180 1
250
I T 240
T 120
200 ,

z
150
0 T60

100

50

OH'
0 5 10 15 20 25 30
Displacementmm

Fig 4.10. Load-deflectionresponsesof node 1 of grid B8 with different degreesof


tightnessof bolts togetherwith the analysisresults

144
300-r

250
180 '1*. 0, T 120,
200 + JA,

150-
0
a T60
loo

50

0-'
0 5 10 15 20 25 30

Displacement
mm

Fig 4.11. Load-deflectionresponsesof node 2 of grid B8 with different degreesof


tightnessof bolts togetherwith the analysisresults

300
T 180
250

T 120
200
1 240 j
.;I
z ;;
150
0 T 60
100

50

0
05 10 15 20 25 30
Displacement
mm

Fig 4.12. Load-deflectionresponsesof node 3 of grid B8 with different degreesof


tightnessof bolts togetherwith the analysisresults

145
300--

250 -

200 -

150
0
04
100
50

o -!
0 5 10 15 20 25 30
Displacement
mm

Fig 4.13. Load-deflection responsesof node 1 of grid B 12 with different degreesof


tightnessof bolts togetherwith the analysisresults

300
T 0,
250
4 t T240,
l
'T 120
200 i

.
Z /' ."; ý/ T 60

G
150
100

50 ii

0
05 10 15 20 25 30
Displacementmm

Fig 4.14. Load-deflectionresponsesof node 2 of grid B 12 with different degreesof


tightnessof bolts togetherwith the analysisresults

146
300

250
, ýý T 120
200 4
1 18U
T
' ''ý 6'0
150
C

100

50


05 10 15 20 25 30

Displacementmm

Fig 4.15. Load-deflectionresponsesof node 3 of grid B12 with different degreesof


tightnessof bolts togetherwith the analysisresults

4.5 Analysis of grid B with considerationof

the effects of the connectors

As can be seenin Figs 4.7 to 4.15, there are considerabledisagreementsbetweenthe actual


deflections of the nodes of grid B and the analytical results obtained from standard

analysis. The in
experimental results all the cases indicate that the is
grid more flexible
than the numerical results suggest.The main reason for the disagreementis due to the
relative softnessof the connectorsthat was not taken into account in the analysis.In order
to obtain a better approximation of the responseof the grid the actual stiffness of the
connectorsshould be consideredin the analysis. The aim of this section is to propose a
simple method to consider the effects of the connectorsin the analysis of the grid. This
method is based on the use of the SAP2000 version 8.08 packagethat is easy to use in
practical works and grid B with three different supportconditions was analysed.

In order to analysegrid B using SAP2000 packagea complete set of data was produced.
This set of datawas included information about:

147
" Member connectivity, nodal coordinatesand the mannerof connectionof members

at nodes,

" Support arrangements and support types,

" Geometricaland mechanicalpropertiesof the membersand

" Loading information.

The mannerin which the effects of the connectorswere consideredin the analysisof grid
B is described below.

4.5.1 Geometryof grid B andjoint type

The plan of grid B is shown in Fig 2.3. In this figure the top layer elementsare shown by
thick lines and the bottom layer elementsas well as the web elementsare shown by thin
lines. The centre-to-centredistancebetweenthe top and bottom layers is 1000 mm and all
the elementshave the samelength of 1414mm. The grid consistsof 84 top layer elements,
180 web elementsand 96 bottom layer elements.These elementsare connectedtogether

using 109 connectors.

As discussedin section 4.2, the analytical responsesof the double layer grids for both pin

and rigid joints conditions are very close. So, all the joints of grid B were consideredas
pinnedjoints.

4.5.2 Support arrangementsand support types

Grid B was analysed with three different support arrangementsthat are mentioned in
Section3.3. The supportarrangementof grid B4, that is grid B with 4 supports,is shown in
Fig 3.19. Also, the support arrangementsof grid B8 and grid B12, which are grid B with 8

and 12 supportsrespectively, are shown in Figs 3.20 and 3.21. The support positions are
indicatedby small squaresin thesefigures.

The supportconditions for grids B4, B8 and B12 were consideredas follow:

" Three translational degrees of freedom of one of the corner supports were

constrained,as shown in Fig4.1.

" Two translational degrees of freedom of another corner support


were constrained,
as shown in Fig 4.1.

148
" All the othersupportshadonly their verticaldegreeof freedomconstrained.

4.5.3 Modelling of elements with consideration of the connectors

A typical element of grid B built in the MERO system consists of a steel tubular element
together with end connectors. This element is connected to the rest of the structure by

means of the end balls, Fig 4.16a. To model space frame members which include the

effects of joint characteristics, the joints at the ends of the member could ideally be

simulated by typical structural components. Fig 4.16b shows a possible simulation of the
typical element that was used in analysis of grid B. In this figure each end connector
including conical end piece, bolt, sleeve and a part of ball was simulated by using one

variation at the ends of the element that is referred to as "end element".

When the individual componentsof a connector are flexible, the axial deformation of
theses components can be considerable. The connector deformation may then have
significant effects on the responseof the grid. The constitutive relationship curve for axial
force-displacementof a connectorhastwo important characteristics,namely,

" Initial stiffness and

" Ultimate strength.

The magnitudeof these two characteristicsgoverns the behaviour of the structure in two
different limit states,that is serviceability and ultimate limit states.In the ultimate limit

state,thesetwo characteristicscan be important and should be consideredin the modelling


of the grid.

In the MERO jointing systemthe axial rigidity of a connectoris usually lower than that of
its attached tubular elements. The modelling of the behaviour of a connector requires
information about the axial force-displacement characteristics of the connector. This

relationship should be found experimentally. In the proposedmodel each "end connector"


was consideredas a pipe with small cross-section.

It is assumedthat the tubular elementshave a uniform cross-sectionand material. In the

caseof grid B the tubular elementis modelled using a pipe with outer diameterof 76.4 mm
and thicknessof 3.5 mm. The length of the pipe is 1200 mm, Fig 4.16b, that is the length
of tubular element.The material of the pipe was consideredto be steel with the modulusof
elasticity of 206 kN/mm2.

149
Ball Tubular element

11
(a)

Pipe representingtubular element


1

"Small pipe" representingend connectors-


07mm 1200mm J 07mm

Fig 4.16. (a) A typical component.(b) An analytical model

The length of the "end element''is 107 mm and its material was consideredto be steelwith
the modulus of elasticity of 206 kN/mm2. Each one of these"end elements"was connected
to the tubular elementrigidly. It is assumedthat the "end elements"have a uniform cross-
section in such a way that their axial stiffness is equal to stiffness of the whole of the end
connectors.The cross-sectionof these elementsthat is an equivalent cross-sectionfor the
connector must be found from force-deformation relationship of the connector
experimentally.

4.5.4 Loading and analysis

A concentratedload of 300 kN was applied at the central node of the upper layer of grid B.
The position of the applied load is shown by a little circle in Fig 3.22. Since the joint
deflection measurementsduring the testing of grid B did not include the effect of self-

weight of the grid, it was not taken into account. The deflections of nodes 1,2 and 3 of
grids B4, B8 and B12 whosepositions are shown in Fig 3.22 were obtained.

In the proposedmodel when the cross-sectionof the "end elements" is the same as the
tubular elementsthe load-deflection responsesobtained for the nodesof grid B are shown

150
in Figs 4.7 to 4.15 by straight thick lines. Also, smaller value for the cross-sectionof the
"end elements" will result in greater deflection of the nodes. The author has examineda

number of different cross-sectionsfor the "end elements". When the cross-sectionof the
"end elements" is considered as 200 mm2 (about 1/4 of the cross-sectionalarea of the
tubular elements)the analytical results indicate a good agreementwith the experimental
results. The analytical results from the proposedmethod are shown in Figs 4.17 to 4.25.
Also, the experimental load-deflection responsesof nodes 1,2 and 3 of grid B with
different supportconditions and different degreesof tightnessare shownin thesefigures.

300

250

200

150
0
100

50

0
05 10 15 20 25 30
Displacement
mm

Fig 4.17. Experimentalload-deflectionresponsesof node 1 of grid B4 with different


degreesof tightnessof bolts togetherwith the analysisresults

151
300

250

200

0 150
100

50
OJ
0 5 10 15 20 25 30 35 40

Displacementmm

Fig 4.18. Experimentalload-deflectionresponsesof node2 of grid B4 with different


degreesof tightnessof bolts togetherwith the analysisresults

300

250

200

C
150

100
50

0 5 10 15 20 25 30 35 40 45

Displacementmm

Fig 4.19. Experimentalload-deflectionresponsesof node 3 of grid B4 with different


degreesof tightnessof bolts togetherwith the analysisresults

152
300 7-

250 -

200
z
150 -
G

100

50 -

0
0 5 10 15 20 25 30

Displacementmm

Fig 4.20. Experimentalload-deflection responsesof node 1 of grid B8 with different


degreesof tightnessof bolts togetherwith the analysisresults

300

250

200

150
G

100

50 t

0-'
0 5 10 15 20 25 30
Displacement
mm

Fig 4.21. Experimentalload-deflectionresponsesof node 2 of grid B8 with different


degreesof tightnessof bolts togetherwith the analysis
results

153
300

250

200

150
0
100

50

0 5 10 15 20 25 30

Displacementmm

Fig 4.22. Experimentalload-deflectionresponsesof node 3 of grid B8 with different


degreesof tightnessof bolts togetherwith the analysisresults

300
T240
120
250
T 180
ýý
ý
"ý T 60
200
z
150
C

100

50

0-'
0 5 10 15 20 25 30
Displacement
mm

Fig 4.23. Experimentalload-deflectionresponsesof node 1 of grid B 12 with different


degreesof tightnessof bolts togetherwith the
analysisresults

154
300

250--

200 --
z
c 150
a 0

100 -ý-

50

i
0
0 5 10 15 20 25 30

Displacementmm

Fig 4.24. Experimentalload-deflectionresponsesof node 2 of grid B12 with different


degreesof tightnessof bolts togetherwith the analysisresults

300

250

200
z
0g 150
0
a
100

50

o-
0 5 10 15 20 25 30
Displacement
mm

Fig 4.25. Experimentalload-deflectionresponsesof node 3 of grid B12 different


with
degreesof tightnessof bolts togetherwith the analysis
results

155
CHAPTER 5

CONCLUSIONS

5.1 Introduction
In the construction of a prefabricated double layer grid the accumulation of the
imperfectionsusually shows itself in the assemblyprocess.Theseimperfectionsare mainly
due to:

" Lack of fitness of componentsduring the manufacturingin the factory,

" Deformation of componentsduring loading, transportationand unloading,

" Imperfectionsin supportpositionsandgeometryof assembled


partsand

" Human error during the assemblyprocess.

Theseimperfectionsthat areusuallyinevitablewill changethe theoreticalconfigurationof


the double layer grid to a certain degree.In the initial steps of the assemblyprocessthe
discrepancyis negligible. During the assembly,adjusting the amountsof the penetrationof
the bolts into the spherical nodes is often used to compensatefor the accumulation of
discrepancies.Variations in the amounts of bolt penetration, lack of sufficient spanner

spacefor tightening of some bolts, effects of the tightening of one bolt on the degreeof
tightness of other bolts and human error will result in different degreesof tightness for
different bolts in a double layer grid.

156
In a MERO-type system, due to the presence of bolts and sleeves, discontinuities between

the connected parts are inevitable. Due to the presence of gaps and prestressing as a result

of the presence of bolts, the mechanism of load transfer at these discontinuities is complex.
Also, in this type of connectors, the transfer of compression through the sleeve and the
transfer of tension through the bolt will give rise to more complexity. Furthermore, each
bolt is subjected to different degrees of tightness and the degree of bolt tightness affects the

width of the gaps as well as the amount of prestressing.

The above considerations lead to the following conclusions:

" The degreesof bolt tightness in a double layer grid that consistsof a large number

of elementswill lie in a rangerather than being a fixed amount.

The bolt tightnesswill affect the load transfer mechanismof the connectorswhich,
in turn, affect the behaviourof the whole double layer grid.

" The study of the effects of the degreeof tightnessof the bolts on the behaviour of
double layer grids is complex and cannotbe done in a purely theoretical approach.

" Small scale models of double layer grids cannot representthe behaviour of such

structuresin real life situation.

In the presentwork, in order to achieve a good understandingabout the influence of bolt


tightness on the behaviour of double layer grids, a programme of experimentalwork on
two actual double layer grids was carried out. The grids were 10m by 10m in plan and
constructed from actual size components.The grids were tested with different support
conditions and different levels of bolt tightness.The deflections of a number of nodes of
the grids were measuredas responsesof the test grids versusapplied loads. The results are
presentedin Chapter four of the Thesis. The collection of these test results lead to the
discussionsand conclusionsthat are presentedin the sequel.

5.2 Discussion

Since the 1980's extensiveresearchhas been conductedin relation to semi-rigidity of the

connectorsof the rectangularsteel frames.But the studiesrelatedto the effects of the semi-
rigidity of the connectorsin spacestructuresare rather limited. The following comments
be
can concluded based on experimentaland analytical studies of researchersabout semi-
rigidity of the connectors:

157
" Flexibility of the connectorshas a significant effect on the behaviour of the space
structures.

" The behaviour of a semi-rigidly jointed space structure is between the extremes of

absolute rigidity and absolute flexibility provided that the semi-rigidity of the
connectorsis due to flexural deformability.

" The laws of variation of internal forces and displacementsdependon the sensitivity
of the structureto the semi-rigidity of the connectors.

" Shallow domesare more sensitiveto semi-rigidity of connectors[3].

The behaviour of a connector in a space structure is not always bending dominated. In


double layer grids the behaviour is dominatedby axial force. Therefore, flexibility of the

connectorscannothave considerableeffect on the responseof double layer grids.

In Section4.2, a test grid was analysedtwice. The first analysiswas performed using the

assumptionthat all the joints of the grid behaveas pin joints and in the secondanalysisall
the joints were consideredas rigid joints. The comparisonof the numerical resultsobtained
from analysis of the grids for pin and rigid joint conditions shows that the deflections of
the nodes for two extremescasesare very close. This is in spite of the fact that the test
resultsgiven in Chapter4 indicate that the deflection of a node of the test grids will change
considerablywhen the degreeof bolt tightnesschanges.Therefore,the actual responsesof
the test grids do not lay betweenthe extremesof pin and rigid joint cases.

The double layer grids are less sensitive to the semi-rigidity of the connectorsthan other
types of spacestructures[3]. This is due to the fact that the main internal forces in double
layer's elementsare axial forces and non-axial force effects (such as bending) in these

structuresare normally secondary.

In view of the above considerationsit can be concluded that semi-rigid behaviour of the

connector cannot justify the actual responsesof the test grids with different degreesof
tightness of the bolts. Therefore, bolt tightness is an independentphenomenonand the
studiesrelatedto the effects of this phenomenonon the behaviour of double layer grids has
not beenreportedin the literature.

In Chapter3, a test procedureis presentedthat enablesrealistic testing the effects of bolt


of
tightnesson the actual responseof a double layer grid. The load-deflection responsesof the

158
grid are presented in Figs 3.23 to 3.58. These figures show that unloading branches of the
load-deflection relationships of the grid are in good agreement with the corresponding
loading branch. This indicates that the test grid during all the tests remained in the elastic

range. However, a general observation of the graphs reveals that all the experimental
responses have a nonlinear behaviour, in particular, in the initial stagesof the loading. This

nonlinear response is thought to be due to the nonlinear behaviour of the connectors. In

what follows the internal force-displacement relationship of a tubular element and

connected bolt under tensile force is discussed. It is assumed that tubular elements and
bolts have elastic-perfect plastic behaviour for the corresponding force-displacement

response.

Since stability is only of secondary concern in tension members, a tubular element

achieves its ultimate strength when all fibres of the cross-section have yielded. The

strengthmay be expressedas:

Tu= FyAg

Where Fy is the yield stress,Ag is the gross cross-sectionalarea and T is the ultimate

strength. As given in Section 2.4.1 the minimum yield stress of a tubular element is

measured460 Mpa and the gross cross-sectionalarea of the tube is 8.4 cm2. Therefore,
assuminguniform distribution of tensile stress,a tubular elementbehavesin linear elastic
rangeup to 386.4 kN internal force.

For tension membershaving holes, as in the MERO-system, the reducedcross-sectionis

referred to as the net area. Holes in a member cause stress concentrations(no uniform
stresses).Theory of elasticity showsthat tensile stressadjacentto a hole will be about three
times the averagestress on the net area. However, as each fibre reachesyield strain its
stressthen becomesa constantFy, with deformation continuing with increasingload until
finally all fibres have achievedor exceededyield strain. The strengthof a tension member
having hole may be expressedas:

Tý= Fy An

Where A is the net areaof the cross-section.Furthermore,severalmethodsare usedto cut


holes. During the cutting operation the metal of the edge of the hole is damagedand this

will affect the net cross-sectionalarea.As shown in Fig 2.4, the net cross-sectionalareaof
a bolt can be estimated2.14 cm2 (3.14-2x0.5=2.14). Also, the yield stressof the bolts is

159
given in Section 2.4.1 as 529 Mpa. Therefore, the ultimate strength of a bolt is 113.2 kN
and after that the bolt will continue to deform without inducing additional resistance.

In view of the above consideration, a bolt behaves in a nonlinear range before 113.2 kN

internal force. This nonlinear behaviour is due to stress concentration in cross-section as

well as uncertainty of net area of the cross-section. After 113.2 kN internal force, the bolt
behaves in plastic range. This is in spite of the fact that the tubular elements that are the

main components of the test grids behave in elastic linear range up to 386.4 kN internal
force. Fig 5.1 shows a failure that occurred in a bolt of the test grid while the overall

behaviour of the grid was in the linear range. This bolt was located in the middle of the

bottom layer of the grid and the internal force was maximum.

The same situation happen when a tubular element was under compressive force. In this

case, a sleeve is a medium through which compressive forces are transmitted from the
conical end piece to the ball (Section 1.3). In spite of the fact that all the tubular elements
were in the linear elastic range, some related sleeves were deformed considerably. A sleeve
that deformed during the test is shown in Fig 5.2.

Fig 5.1 Failure of a bolt during the test

160
Fig 5.2 A considerable deformation of a sleeve

5.3 Conclusions
A study of the load-deflection responses of the test grids leads to the following

conclusions:

" Since the unloading branches of the load-deflection responses of the grids, which

are presented in Chapter 3, are in good agreement with the corresponding loading

branches, it can be concluded that the test grids during the experiments remained in

the elastic range.

0 Although the grids were tested in the elastic range, the results indicate that the

actual load-deflection responses, in particular in the initial stages of loading, are


nonlinear. This nonlinear response is thought to be due to the nonlinear behaviour
of the connectors.

161
" The comparison of the experimental load-deflection responseswith the analysis

resultsof the grids show that in all the casesthere are major disagreementsbetween
the actual and analytical deflections of the grids (seeChapter4).

" The experimentalresults indicate that in all the casesthe actual deflections of the
grids are greaterthan those obtainedfrom the analysisof the grids.

" The load-deflection responsesof the grids in all the casesshow that the degreeof
bolt tightnessof the connectorshas a considerableeffect on the deflections of the
grids.

" The studies show that the load-deflection responsesof each grid with a specific

support condition lie between two limits, which are the upper and lower limits of
the responses.Each one of these limits is related to a specific degree of bolt
tightness.The lower limit is related to the lowest degreeof bolt tightness.This, in
practice,is thought to be around 60 Nm. The degreeof bolt tightnessrelated to the
upper limit dependson the configuration, loading and supportcondition of the grid.
The upper limit of bolt tightnessfor grid B seemsto be around 180Nm.

" In practice the bolts in a MERO-type double layer grid, unlike the bolts in the test

grids, can have various degreesof tightness.However, evenin this case,it is logical
to expect that the load-deflection responseof the grid will lie in a region whose
upper and lower limits correspondsto the loading and support conditions of the
grid. The load-deflection responseof grid A with normal bolt tightness confirms
this fact.

" The experimentalresults show that the width of the responseregion dependson the

stiffness of the grid. This width will increase when the stiffness of the grid
increases.

" The experimentalload-deflectionresponsesshow that the stiffness of the grids will


increasewhen the degreeof bolt tightness increases.This trend continuesuntil the

upper limit of the load-deflection responseis reached. After that, increasing the
tightnessof the bolts does not have any major effect on the behaviour of the grid.
However,this increasein the stiffer grids give raisesa more flexible behaviour.

162
5.4 Recommendations for further research
In terms of the effects of bolt tightness on the load-deflection responsesof double layer

grids, some topics for further researchare listed below:

" Studying the actual responses of double layer grids with respect to the
configuration, span, loading and support conditions using experiments;

" Although it is thought that the bolt tightness has no major effects on the internal
forces in the elementsof a double layer grid, but further investigationof this matter
is a worthwhile researchproject.

" Developing a three-dimensionalfinite element modelling schemefor the MERO-


type jointing system so that the effects of bolt tightness can be modelled
analytically.

163
APPENDIX

NUMERICAL RESULTS OF

THE TESTS

A. 1 Introduction

The difficulties and high cost of full-scale tests are the reasonsfor the experimentalstudies
into behaviourof structuresare usually carried out on small-scalemodels. So, information

on the real behaviour of a structure,in contrast with that of a small-scalemodel, is scarce


and valuablefor researchers.

The effects of the degreeof bolt tightnessin double layer grids are complex and the results

obtained from the tests conductedon small-scalemodels cannot reliably reflect the actual
behaviour.Therefore, in order to investigatethe effects the degreeof bolt tightnesson the
behaviour of double layer grids the use of full-scale test grids is unavoidable. So, the

numerical results presentedin this appendix together with the information given in the
Thesiscan be useful for further researchin this field.

A. 2 Tables

The numerical results given in this appendix are related to results from fifteen
obtained
different tests on two double layer grids. Grid A with four supports
was testedwith three
different degreesof bolt tightness(three tests). Also, grid B was tested three different
with
support conditions (four, eight and twelve supports) and four different degreesof bolt
tightness(twelve tests).

164
In order to increase the accuracy of the results, each test was carried out twice. The letter
"a" in the title of each Table shows that the results are related to the first test. Also, the
letter "b" indicates that the results are obtained from the second test. For instance Table
"Ala" is related to load-deflection responses obtained from the first test on grid A with 60
Nm degree of tightness, while Table "Alb" contains the load-deflection responses of the
second test.

A. 3 Numerical results

During the tests,displacementsof twelve bottom layer nodesof the test grids togetherwith
different applied loads were measuredand stored in output files. The positions of these

nodesare shown by little circles in Figs 3.1 and 3.22. Also, the loads were applied to the
central node of the top layer of the grids. The amounts of the applied loads and the
displacementsof the nodesare given in the Tables Ala to A15b. Each of thesetables has
thirteen columnsas follows:

" The first to twelfth columns indicate the displacementsof nodes 1 to 12 of the
bottom layer of the grids.

" The last column showsthe applied load.

Due to symmetry of the grids as well as the loading and the support conditions,
displacementsof the following groupsof nodesare the same:

" Group 1 consistsof nodes 1,6,7 and 12

" Group 2 consistsof nodes2,5,8 and 11

" Group 3 consistsof nodes3,4,9 and 10

Therefore, in the calculation process of the load-deflection responsesof the grids the

averagedeflections of eachgroup was consideredas the deflection of eachof the nodesin


that group.

165
pN 1ý0 oo 'T et "-+ tn ON M \O It O \O MOM \O 'tN (D M "'T IRT
C) 1.4 N W) 00 -4 ýp 'D OC 00 ON OM [- -4 Vl MMMM
,nN 00 00 An N \O 'C -* -t MN Q\ kn - ',O ON C\ ON 00 cV ÖÖ
N pý Ö -4 NN', C 4MMMMO OÖ Cý Cý Cý OON
S-NN (V MMMM 'Rt vn ýD \O 10 [- t- 00 00 00 (7\ ^- --4 -

mmwlkoW)wl -N ltc+MON%OI'00Nwn\. D Nr. io 11, IT


oO tra C14 IT
-4 t- Wi Ch C> Wn 1ý0 \.0 MNOý I- NN-- 'IT 4n MM
kn C7\ Wk s--ý-ýtWnv)"Yl-O\'ttO\ONv1ONNNNo0en --
0 MO4ý MM CýIt ýPo0ell r-. 4.
_i. -4ti et ,O Oß_4 -i

c) rýrý%0Cý 00t-ýCv1Mý0--4%0 - KA r, 1týNd -, +"-+ o\N


irr ýCcýrý 00 ýOvi " Vlb Mrs. --4 "OM000N[-ntleMOM -14 -4
Opt oON 00t-ý-+lýN0: [-%0 NNooN'0--4-4-0---4O: 1,O%DIC
ooO\vi ivlt-Naý000 Men Cod-i-i_. _. N ýpýp
GQ
QK oý Oý --- "-+ -- .-+ .-. NNNNNNNN

O
e) 00 h_ NÖ Ö 00 QNNNýn-
6
a)
it 66 kn o o
NMMMM
OM. 00 vom)00
tl-
%D IF)
00ýý ÖÖÖÖN4ril-:
M_
bQ NN
NNNN N
10
E
MM wl nMetIi'Ii' C,4 Ch ON
tr-v'1ý-M t-NI! 'NOý00-ýO00 t'-
Oh vý *o 4N O\ (7\ p ýt --4 O -4 in 't VO --+ Re) v) et '-t in M o0 o0
b W., kn V?
OOM'ýq
ýO ýO ýO tý [ý O\ ýj et Viw OÖ
NNNNN
tei

W-1 qt NN C\ W) - MW) 110


O4n 1ý0'rNN00N '¬ 'r mN
00 ý,D MO NO M IN V) IT Oý vt ýt M O,ý 00 O- [N %,O to v') 00 O V's 'T
NOM ^'"Ch -4It en 000'ý ['- 00[-rý-4OCýmmm'ITM110NC
0000C;
,.. ý... ýýyMen cV a: 'e; 4 oö C CC; C; Ci '. 6 6
fV Nhýýýo? C 0C --4 -4 ý ". "^ CN C" NCVCV N
I

w0
y) M q-t N O\ NNM 00 v1 ý,O ýo nN v1 C% tN NNN wl m 10 %0
-RY N "r ON V1 00 qtt --4 M "O NN 00 V1 - 110 I'D 1O ýO 't 00 MM
t7-, O-4 10 Ih 1.0 Ný-+OMOýott)et%0Nl-l-[-t-O: tt%Dýo
l., in ýo ýo "D %0 ON as ON C* tt ON 6
_4 -i
4M v1 Vi vý \O oÖ 0 C)
I" .ý . -r . -, . -, . -" NN
-o
ca 00 00 00 v1 M "O [N ON ON 00 '0 00 NM%.
W) D rý en 00
ý-r. ret"''tNv1 -"yN0000'0MO [ý ýO
00 r. Neý t) [ý tý
NN-+NO
!t ON 00 in " l- l- 00 "t
Gý ON 0
OO
OO. wi t/'ivi oÖ01"-ý"--ýr". qq: qqheter-nOýOK
--ifVNNe -,

10 ý000M 00 00 00 00 00 CC (t
NN N-N O\ N v'1v1 ý nCh
v1 QN Nm O
ýt -4 .--O
MM ON 00 lif mO C% %M CD

OO

10
o 0O V 0
td N-. O0,0M-NMNo Vý1 Izt 00O V))N0 000OQ>O, OOO
0 Ö. -ýoOýOvý 00 VýCOýOýOýOýOýDýOýOv10000
OOýOýýt Ocýrý ÖÖÖÖ-+MýdD
OM-tetýt
0.4 OO QMM V'i t
.
tý ýD o0 r-+ . -ý ý--ý ýý
"-» . -r NNNNNNNN

Q
(1) r- r- 'n v, e r9 l- r- ý,C ýD M V) r- 't M 0% 00 00 00 00 eh P-+ V) r-
pp pp 00 Nt M Q> 'O VI NN OO M" CD O\ '27 M 'Z MMM vl MM 't
OOýOONrt00le --4 - le O\MOGýOýOOt-r- I'- [-o0 oý -- -,
%,O ýO '! OON '10 ýo 0 r+ O',
F-+ }
N ýer MM vi OZ CZ OZ oý ON O %d

00 O\ 0O et tý [N ýG NM ýt ýO vý NM OW N 'n S \O
fir ý--ý --ý
M4tNNv1vlMcho0Ol''- NsONN-4 C\['- Ob%Dv1
. 00 %D lt N lý lý IN v1 N le M Oý Q% 00 et OOOO.
fis\o 00 e '40 \O
v) OýQ% 0% 0% MMMýON
.r -4
OONNNM v'i kr1 ke1 \G OÖ NN

00lß Q>- e[- aýe-"-+Mý, 000 v1 an \D


00ý"--ýMN V10ýQ>"-+M enr4rq e0000
N >0vlNý-+Ný--4OooýOOo000[-v-bMMM M N01 ^"'"'ý
e: OZoOO0
OMMMMMN

166
NM - \CO MM (- IZ '0N --000 mýFN V) V) -4 -4
MýoINNo0oo O ý7v v)NVlb 0000O- v1v)MknNN%0%D
, _,
M
NM N-NNM QZtzt-c4 Noýo1etÖet +Or-+NNMýt
Ze- e-A5'z ZO ZO
In kn v) \O O\ O: O\ -r -4 ,--i ,--- .. r . -r -4 4 . -+ -4 . -e rr . -4 -4

VI 00 [- MN l- v'1 NN l- %0 %,O O% G\ Oý
--r --4-4
N V'1 [- N r- V1 t- MMMO t-- M -et e %M %D vb 1,0 V7
v)'it e Ode --ýONOOMOkn v) tt -+NNNOOOOO
O, O>-t-tc>riMei
M MrzC. 62QK--i.
-i. --icV vAw; i 'i Z6 c; ic; r;
Iº 0? II I+ r I+ I I+ ý+ cV N cV NN cV NN cV NNN cV

N
a) tMt/1W-) t---t't 1-00-4 -, c MNN%O - l-tl- ll- knitW)
1.0 c00c0No0ýbýG "tNNýt. "tetN000OOýoOO-+-+
-4O*, vl n--ý--ýr... --ýýOo00000N-4týtýtýWnO-, ON OýMNNN
öA N
^ý' ý+ -ý+ ý+ tV cV NNNNNNMMMMC? C? MM

O
a) [- l-
a) er ONI- ON - N[- ýt "TI-t Wl NM ONTO V-1 MM v)v10000
p ýn 1, p yr M ON ON p ýO Qs I'O ý,O 't - r-- oo O-, 00 v1 et 'Rt O\ O\ O\ 00 00
oA mll- pOýW*i N W)10ON 00ýý110001ý0ý01. OO0wlwl W) --q"-r"-rMM
MpOýC; 6 t. GNC
-o ,fiý1,6 N. -4"iCri 16 [-ý t-ý o6ÖNNN4ktivi1.
NNNNNNNNMMMMMMMMM
e

00 M O> 00 v1 0% NM l- NM v1 tt '0 t- mM- ým l- U1 mt


O\ 1.0 MNN-M r-+ v) Ne %C MN O\ - r- [S 00 MM Ilet O> 00
1t''t 00l-NNd'hO\ O: N et -e NhNNNOOQ%OýO
. --i Mv1
b MM tý c; --ý .cý
ý+ c; -ý kei
l-z l-ý
ý--ý . --ý '-+
NMM V'i
ýNNNNNNNNMMMMMMMMM
OÖ 06 OK .4MMM k/1 v1

M en m C14 10 MMM 00'N 110 10 -14 "


tIT
- d' 'RT qt t'- W) 11' l'- 'ti' Ct 00 [ý 0000000nN NMt
"D It ýo ýo oO O 'Rt 'Rt W) 00 NNMNNN 00 00 IT t:t [ý 00 00 oO 00
[- t-Z n l- 0ý "--i .r ri NMM44
C? C? MM

v1 eh e-- NN'r 6nCýOýtetýf


vyv1 \O - MN -+ýo ýo o0-4 el 'ný}. ý}
OQIA ýýIT etýM NIYItmmlZ1O O'. ON OOOO
3 týl-knC'ý Oýt-l-ON00
~~N^;
00 Zt1%D'so4T - OO'R: r - 1,0ýDýOI000OO
"'"~ NNN l-: l-: 06 Ö. 4r4-- M444vi. 11C .16
d MM'ýýýý? ý"-I ýýýNtVNNNNNNNNfVr CV
Ici IIII

an
d' ýO ON v) 'r1 It h -1+ \1O -N v1 N wA M I* M ON 00 00 MN 00
NMOooetgt O"--týW'i o0[-0000t-tl- NO-N- r etOO\
vl IR:
NON OOO oo[ý\1O--q qtT et -4oo0000 NO\etqt 'd. MIt IT en N
CD en OOOV1nl, l- Met 4 wi606
Op N 0ý--ýý--ýcVMM4et
O wi vi vi 06 Oý Qý Cý ý+ IIIII. ýý cV fV tV CV fV fV CV
ý"
0 I I


0
0
I'D %D M G\ "-r 00 - 'O NO m 00 -NNNC t- 'O
M eF 't [l- n %D MMNN 00 00 V1 00 00 "--ý O Cý O', Cý 0ý MMMM
v Mr"týrýcli44 i oÖGýCýC; etýl-ý l-z tooC; C;
aý cV Nvioooo .-+NNNNNNNNNNMMM
S

"Ci
N If "IT IT W) e}
00 (2N oMON"
.- In .-Nm .o rl. 00 N -4 as r-
-4
cu IttWl0ottlVtet'Rt 14)v1O-+-4ooOvnv'1[-- ONNNMOONN
t, 'IT 000000 MMM1ý6 ht-: Iý: ÖCVCV cVMvi NCý6
Nv'iOýOýOý --ý-+-+NNNNNNNNMMMMMml MMM

NMm 00 t- 00 00 ON 00 00 M 00 N 00 00 NN
eF1D%0N -N "-4
aý N%0 00l-[-v1ehqtMOqtqtt 01Mqýten C-0000[ýýC
N Oh - ON 00 C\ N 00 00 00 N It IT ON M QN oN ON IN [-
C, O1 M tN tC er d; ýo ýo
't 'T 'It N444 Cý -4 .-4 .44 MM4
06od06 v'; v1 1ý6 oÖ OOO -+ eh
E-+ ''''^' ''ý^"ý+ýNNNNNNNNMMMMMMMM

ýt I 11, %0 10 tl- t- Vn N O
%. M Oh ý,O Q.,
ON ON It 'T R I- <t V1 in V1 MMM 'IT
vn v1 110 00 l-l- 00 qtt NNNO%ýG "G ý,O c7N mmONýo [-
NN 00 ON V1 00
00 V1 Vý W) NMM'4OlýNNN Cl C%
"--ý "r 00 00 00 00
,. MMM Oý Oý Oý lý t. 0\ O\ OO
fV fV M[ n. 06
"I .I,.. i ,. .r ,"NN tV NNNN tV NN fV MM
I I

[- Qh - t- [-l-[-l-t-t-t-NNN-NNNv'lct000041 o'c'
IZ 00 etlet `ýtnvi vlv)ttNNqt N[-[-
OON hc+1-t e"CD /1v)ýt`t
00 00 0o tý
OOÖM vi try vi t-z 00 od pd c; rV cV rV m n* -Y v%
'i i" ri wi vi wi

167
-v
0 'O\
[, N -'t' NýNN'000000 Q\O W>výjM-4"-+0000 0%nß--
1-ý ti N .r .--- -e- "-r . -r .--4 Q'. NNNNM O'. C'. OÖÖ-
ýO ýD pp V1 ¢f "-i --i
ýt ý7 eF d rN cý oý cý cm; wi v' er; er; 7 ýt rz 25 c> NNM kn \O iD

0E 00 N- 01 M 0, M 00 00 00 00 00 00 wý vi M- \O N tN N [N \O fT 00
et et
bE V) 2r N`d'a'. - a'. - NNNM VI"-r'. D-'O e O'. MN---4ee MM0% e
p N00in0CD 00Q'. 000en Ct, lýr-M00 N00NN 0 '. 0
ni -+o0%0
ZNNO C) 000 l0 v) v'1 v) v) N000 I- oo oO ýNNM "6 rZ:
FNN h eF ýt v tz
Vý V1 V1 ýG N OÖ OÖ OÖ 00 Oý Qý
111111111111111111111 "-r . -i ý--i . -ý . -r .r ý--i

Gn r-r
a) NMM"'e 00 00 00 00 'C '. 0 N O 00
in %0 'eh MMM'ld- V1 NM
ýO N 00 NN 00 00 1ý v'1 et ýP C 00 Q'. p'. -N
0M 00 in
'i- V') M--N in in to
in in - "--- NNM ON 00 00 00 N O'. O
in OOOO . -, N-N 00 00 00 '. O in vn ý--ý .--1 NN
ZNN 00 00 00 00 00 M 00 00 00 V-1 O' O'
MMMMi oÖ oÖ
"+ f+l ýO %C ý6 %6 [ý 00 06
11 1111111111111111111111 . -ý . -"ý . -r . -ý . --ý r"r . -. r 1..

0
C
a)
MOONN'. oNNMhkn v) 00MMMMen ecx%O\e! '\Ov)%0
00 1 n". l1- c>CDc> O"-+O'. O'. '. 0MM000'. 'e 0000 0N
bA tý oo et r-+ O "--ý "-r N 00 00 00 C'. \O ON tr 00 in in in '. C %C in 00 00 M
-'.
%V 00 00
O\ 1,0 NN Ninoooo*0oo ei c;
0 -+OOr,
b ýý "-+ MMMM eý hh$ 1f N
c; c; c;
00 OÖ OÖ 00 . -"r .--i . -"y .r ý--i ý--i . --i ý--ý . --i -NN
11111111111111111111111111
E
O\ 0000 MOO
000' 00\ONNN&q NIýI- nMOO Q\ OýN0ý0h OýNMM
0MN C'. O O'. ý--4
'-r "--ý MNM 'O NNNO et in o'. ('. --4 e Q'. 00 00 M r-+ -9
b E lT leM'f Min le '0'. -M-+"-rip. -". --4 -r0 CM O\o0.
10 0 vý to v1 00 00 00 00 liO "i ""r ýp et Ilt `fit -rNO'MMN0etef
"-"r
r_ Z .r.. r N v1 ý1 v1 oÖ oÖ 00 fý! fV
iý ri fi7[IC OÖ Qý Qý Qý . -r .--ý . -r .-r ýr ý--i . -r .-r . -ý '-"+ NNN
cd
y
rZ
O 00
cit MMin"-r--+N-Q\%0'. C%D%0Mvi '0\O N NNet N- eMv)M--
0E et et a,. c 0c> 00 00 oo oo Nlm, Me '0
0E CD OO000- 4t - '0 V) kn t*- 0 ('. N (D in in
NN N'. rqe-"et00in VnNMM00-CD MM\ýOý, O
o 0000 le ýoliO110Ne -eeet[', 00ppeV NM
ZMM` vi vi Vio40ö0öÖ-- --
\O %6 ý-O tz 00 OÖ 00 OÖ Oý .
11111111111111111111111111 .--1 -- . -1 . -.. --i -. .--1 . . -/ ý-- .NNN

40.
S.ýi
. E 00 00 " let S'O'D'C OC A--- - --i \O MM 00 M 00 i rN M (> (eo
me O'. O'. vn NNN 'C in ON O-, o'.
C". \O 00 00 00 00 NN et N- in NNN gt NN
o MO'O\O'0'0mot'MMMM 000000"-40000 N-- e v) N`t e
Nt-l- t- eeh
en MMMNOýOýOýNNNN,..,,..,,..,

fy "-I "I "-"+ MMMM V1
MMM
!r Ifl V1 ýO ýG ýO [ý 0I 0I "-I .r.. ý .I
111111111111111111111111111 .-I . -I .I . -I . -r
I

00 M V1 in N wn v-b v) in M
'D "-r ýh ýt ýh ýO m 00 0'. 00 00 Q'. M 00 in V) nt
00 00 00 00 NOOO \C NNN 00 01 M r! "
0 0
oO

et

OMMMMMNNN
--+
"r
-4 00 00 00 N cý O
t*- in
00 ýD ýO 00
'et N
0% 0% 0% %0 ON O'
00 00 00
Z MNNNN ýC 00 00 00 00 tý `ýt ýt ýt ýC NM ýO
cV N cV ýt In h
1111111111111111111111

Ir N
E
i{r 00 00 O, '.0 'C in vy e -4 m o'. O\ 0% MeN 0% \O \O Q'. 00 . -+
ýy 1ýin00000,00000000,00NNNN0'. 0N
.. r MM OO NNO (> ýt oO 00 O\ 00 NMMM - 00 V100 '.0 in00if 1
r-+ r"+ et ýt 00 NNM O0 00 00 NMM
Z vi wýlqtOýMMMen MMMen
NNMer; 1r; 1r; er; er; l[lt ei d; odÖÖ
060'O'01O;. _. "4. -r , . -v. -"rP-i. "rNN
Gi

'L3 00 Ni f'. ate %t v)Mvi ýi ýtetN00000'. 0'.


eh M Q'. 00 00 00 C'. Q-, O\ 00 00 00 00 et "-+NN'0in' MUMM
cd ö ýt h in -NO ýO ýO O ýt K1 N ý--ý NN
N 0^'"^+-Nýt MNNNN00-
0 C? CO -+OOMNC, "MRt ne 'l v1e it
Z ""4 000OetNNNNNv1v, V1i'1NriNhe.
wio00000ÖeV N
" "-ii N er; Vi vi Vr1; hln( oÖ Oý Oý Cý Qý .. r . -i . -"ý . -r .--i . --i . --ý "--ý NNN
Cd

NN 0%C P%000000
U ,0F in dMMONNNN
NNr, 4r, -4 l, r, t, 00 O ý--/ V1 vý vbO 0000NN-
o0
tu 4 ýýt C)O N NNNv'b
r4ý r4 ""4 MMMMM ýO Qý"--ý. -+. -ýrr NN NNNMM C'
Z c+ºMMMN V1 v1 v1tnM ÖMMM
"ý -r OÖ ÖÖ
E- 1 1" 1
MMMMMeA1; 1r; ei ý[ý ýniý 01Qýý"ý+. -"v. -ý. -ý.. r. -r. -+. -ýtV N

N
0E O'. Q'. %0'ýtýt'ýteTý--ýN000000
t- t- ^' vn`t Me etNeN0000NM ON NE
17 ýCM"--/NNNON
z
.--IO'P""4-gMNen NtZ, %D'Det ,. ONM
^" No000ooooCD 00 t' " t- NN l- [*- le ee 4t NNN O'. CG 00
\O'O'O'ONin Inv).
r. -r. tz[z06Ö .
MMM v1 hh 'O rz tjtrzC. 6 Qý 0' cý O' -retetetl-:
"--ý "-"i . °ý . -r 2 . -r . -+ . -ý .--ý "--ý NN

N ". CS Ch 0'. N ýntý'ýtlýl NNtv'. N X0'0


b 00 00 00 ýt V ýt et "- -,. -rNNN'0 --ß""r'0 ýt
ýý0- et M t+1 MMN 00 N 00 00 NNNN 00 00 -N "-"ý mt "-r "-"
--4 .. 4 O' OOOO
000000000000
FO NNN000OO . -"1. --i . -r in Vt V100 ýp ýp
F-. vi v' vi ý 7, OÖÖÖ iet
l\ lý lý -""-r 1-4 er 116%ý6 %dl-z cýQK
n 00 00 00 00 00
I 111111111111111111111111 . -. ý . ""ý ý--1 . -ý z, z . --4 -4 -4 . -4 '-r 1--4 . --4 . -4 -ti -4

168
b
03 NNNOkn ON
0 %n I'rt'T erpNýON Qý.
ON 0% \0--iNNN
1.0 ON C*,
''ý
"-+ "-ý vý M cV cV cV cV cV vi tv V'ý v'1 v'1 ý "-ý fV cV
ýOýOýOo0Oý01OýOýCýOý-"4"--ý"--ý. -. Nýt et
NNNNNNNN
N

yE 'Tr qt C' an - --ýN-aN00NGG(2s(3N n


b 'mot et O\ t+1 00 00 t- N '0 N
-4 -4 00 00 . -r '. O ON O'
z týNNC'ÖOÖÖÖÖCV N NN 4
r; 'n
-4NNNNNNNNNNNNNN

rA
r= MMN
o --ý ON O\ 0\ 00 tN "-+ N an an NN
NN1'. O N eh et M-O- O' IN 'n an an
NNMN an an an an an an er NN O\ an -4 .. r
Ö --4
öq oÖ 06 00 -i -; -; - --; O NN
ý4Mr444'.
cVNCVNCV NNNNNCV NNNN
O
O
- '. O I'O %D an O\ Q'. Q\ 00 00 00 MMMd
r-+ \0 an
NNN '0 00 00 00 NN tN O [N N tN 00 O '. O V7
bA "C3 00 00 00 O\ W) W) N V1 Vl kn Vl IT IT .ONN
a) ÖÖÖcV 444
N:t IýTeh10 N NN06000
FO
Fý NNNNNNNNNNNNNNNMMM
E

MMVN ON N \.O 1ý0 kn \.D ý} n v1


NE- ^' NO I-'O'O'0'0 N'. 0 '0 an ON '. O M-
'Ly d' et -t 000000 00 O' ON O' an mNN
10 0NNN eh \O 'O '. G 'O ,O 'D N 00 00 00 O' -NN
NNNNNNNNNNNNNNNMM

00
Oý 00NNN-+"-+ v1MMNt-MC14
N' etONNN"-+-+ N\C'Cv1--ON --ý
"O \-O 00
0E M O\ Oh Q1 Q\ Q\ Q\ ct M:t Ct d: ON kn MM
ti --i .--i . -i t+1 d 'ýt d' IT IT e}' '.O [N [N [N N O'. ÖO
F-r NNNNNNNNNNNNNNNNMM

N Q\ Cl 10 0NNC'M
l NN- ON Ch mm nNN .
It) 00 tý tý D n
ON %. ý0
"T 00C'4 . NNNNV1ýtetet[ý--ýpýQý
0

l
10
on
Pl. 4)
5 't'tt -t Oh MMON -, ß-+v1 'R7 an NN
00 00 00 N 00 C\ 01 00 00 00 fit 10 \D \O ON'. O 'O
Q\C\O\NIr- Iýnl-l-l-C\CNC\O\
w0 0
kti N
Nwl MM
'n 'n 00 00 00 00 00 00 O'. OÖO .--I NMM
r-+
NNNNNN

Lr N
'O 00 00 00 NN [N M SC \O %0 "--- "--
it W) v)
v1 IN 00 00 00 N Ný N 0\ .4
öMM 00 00 00 v1 N 00 00
K1 'G in vn an an an an "OOON 00 00 00
ý4 OOOMMMMMM4 ýC \O \O \C N 00 00
CV CV CV CV N CV CV N CV CV N rV NN CV CV CV CV

10 It
NNN '0 00 00 0'. N 00 IN \D CNN '0 V) 't '! t
tu -4 N ý7 ýt an M et M C' O\ 00 00 N an 0O
?? ß; 00000000N 0000000 - 00
J 0
NNNM '0'ý6 '0 N
'C oö 06 00 oö ONN
CV N CV CV rV CV rV NNN CV CV CV N CV MMM
cCi
MN 00 an an an an an v1 'G 't NN
MNMNNNNNN O\ an an an an eý C\ ON
MMMý000000c)
ÖOO -ý00000000ýpp)p\
-mot 4 444v1\d'.
cri ýZo NNNNNNNNNNNNNNNNNN 'O %6

NNNN
b 00 00 0 an-00
00
an an 100000 NN 00000 -
00 00 00 00 . "ý . "ý -OOOO t- \O '.G NN
OE an an
z OÖO--+'e! 4wiý6
et etet4
NNNNNNNNNNNNNNNNN \6 \.6 \ý6 od0. C.

NE ON C\ an C'
O'. ON 00 NN
't ei' - ~-ß -
bOOON an an an vn an an 00 [N N C'
'IR \-O \O "t 'T et 't et et "T - \O \O \.O \.O 00 C: ON
Qý 0'. C' ÖNNNNNNM4444
"ý'^ý an an an
ý+tV (V (VNNNNNNNtVNNNN

169
- "--ý , Cen en en NN0000"O'Oc 00 DQ\ONON\0m QýONO-, 00'fl'1
IT
G In
C, 00 I'D O\
ýýtÖÖÖýÖÖÖr-i.
C\
MMMM ON ON -o
M oO cO 00 00 V1 0\
ON p1 ON et p, % I'D \p \p
NNN
NMMM M-
C; c;
NN
c; wi
--iMMýCýDIýcýI4--C;
"--+ "--ý MMMM kn kn W) wl tn \C 00 00 00 00 Q1 '" -4 -4 . --4

ON vy "-+ NMN 10 - %D NNM


tC4It0ý --+ 'tt N
V')NONv1Ov)Ot-ONo0WI)ONv-) kntn %D -4 ýý00 MMMV)MN w-4 00
OIt'MNN 1Ol-l-NNN\0N W)
m -4NNNkn-4 et 'Tr %n O ýC\OýOýOtý0000
". 4-. 4.4. r-4--r1Oýo%D 0*1
Neq NC-44: 4v: W; 6
\. 6

t
rA
ON MMNN ýD 01 [- M Nt N-t to mMM tf mM t-
oo I'D ON t- tl- v1MMNNNetý. 0ý0--+.
It MO 00 00 00 %D 10 W") W) vl MMOOON --ýoo0O VyýG'i'It WnI'O V) V)M"--4 ". "
00 tl- 10 v) O 10 %0 ýO 00 00 00 00 O mt et
%.
1 VII 00 ON 01% ON ON G% O
4- C; 06 Cý Cý

O
I)
0)
{-w ITeq%0 ro "-+ 00 ON t- W) It 5 [SS
M Nkn QýO[S[- l- tl% [- et d'
q1tIf'%. O%000%0%D4n4
bQ ['. ON O O\ O\ N I* I'O et
kn 00%O W) 00tn0000I-- tl- OOCD O1. O"--4"-4
It m 01 ON 00 00 00 01 k/ 1 ON m ON QN 14T ýT qt:f' N
b N00 W 'Rt'etI- NNNNNN,.., IT tn r4
NNNcVvi N
OaÖcV
E . -w .. r h tri w; wA 06 06 06 06 OÖ Oý ý-r .--ý ý--ý ý--ý "--r . -ý .ý.. r . --ý '-r . -y . -"r "--ý "-r NNN

Z
-1 -"r 0% 0% 00 M l- l-- l- n %D N %D [- l- \O Oh %D '¬ W1 10 v1 it)
m10tl- *- `0 1.0 Il- k 10I'Dq'1k N -r --1
[-l-e! 'IT t-10% "Tmm%DW)4n NýCV1et
- %0 t- [- %D I'O -OOOM 'T r-+ NNNN
MI-"D-o'ot-W)wl 00 1* et Rt et 00 00 00 00 00 ýo %1C
w wl't C"; ; 444.000OÖOÖO6C; 66
(V N Cý Gý Cý Q; CI --i MM
"-r .. r . -r .r . -r .r . -r .--i . --i .r .-r N (V cV N fV N$

O
4 N qt en NNN 't et NM_ 00 [ý [*- tl-
Ov)l-%D"OIONNO. ý,p ýp N__ -4 --1 MMmN %D -
01 v1 MNNN -. ON C%ONýFNT ")m-4 [l- lý[ýMCh M00'GON O
,tt ýt "t MOM et 'rt t tt \o qOOO ON ýO t-
ti [-00 0 Oa66 0-4 "ý - ý-r -
cli 4'i 44
06 CM606 06 6666 6 NN
06 oÖ oÖ oÖ .-r . -r .....
ii .-y .. -y .-y . -y .... cV CV CV NNN fV
42
MN% 1000") -14-14ýo 0000mt-MMC%ooaav1
tn00000G) 4t11Oý, O"-rpNvlýD- -4 Nt- MIRttn%C
r- kn %D1Ovl l-[-N1O"-r. -ýý--ý. -i'i'MNNON00t'-
W) [-l- n\Gýoet'00%D
MO 00 00 oO 00 00 pp pp pp pp MNNNNN6N wl, W) 4n ON Men MM-4"-r"--4-4"ONN
aMMMM cll NN44444 ý6
wi wi w; ýC OÖ Q; Cý Oý
kli tr;
b
on 10 'n- MtýM NI-in t-on 00Nin it Ntm 00 ONNt
Vm 00 ON m CN N l- l- M . ply(-[-n
M 00 . -+ NMM ON NMNN v1 <7' MM 110 t1-
CD t'- [- %Dl- 10 N OOOMM-4 OMOsetMONMONl- t- [ýlý-0000
-14 'Rt It It ýt Vl Vy 111 V1 Vl 00 00 ON ON 00 00
CF MMMM
ýA ý1 Vý VI l 00 06 00 00
ý" ý" ý"I . --i . -"ti . "ý .ý .ý . -r . -ý . "-i
II

aLI mommm 00t-ýo%D'0 "I r- NNNNWnlfl


MON NOOM It V kt)"Ften 00v1en
00 I-N VlNT00 V 1,0%DIOýGONN- Cg r4 - OI-
MmOmm [*- 00 [ý kn v'j et ON C% MMMM- "- rtN
MMO ON QN m ON ON ON ON OOOOM It et
en MMMM ýO
IM .MMMMw; ý6 ý6 ,O ý6 01 Oý Oý Qý Oý .+-;
ý6 \6 \G CI CI GI CI Oý ý-r .--i .. r .I ý--ý .-w .
t -ý . -r . -ý . -ý -I .-r ý-I .-r '"r fV fV

lL MNýDýD V1v1NN- - MýGýC'Rt


o tnen 'Rt'Rt14D 00 t- %D %D ýo It IT m 00 00
ns m -0ON 00M%0etNNNIt [l- %O VI [l- t- W) kt) -ANN
O tO 00 V1 0o[o .O MO,
'C G
ý0 00 00 00 00ýO MMMMN V1 "t qt 't 00 00 00 00 "" ýD 10
'% o ONNýt 4 ýo0o0o0006 ÖÖ6
r! 1+MM

00 M t- %0 %. D V') ON 00 00 00 V
NNOOV) 00 00 00 00 00 ýO 01 m ON ON 00 ON 00
WI)ITkn NOOOM[NýOýo ýoýDcn
N~"" 0000 t- aNt-t-I-OMNNNNOMMMM0000nt-ONW) ant- lt ' _00M NN V1
CO', O\ Oý[, "O'C%O'Ooo V1
lrl Cý lll
cVcVcVcVof "6 "6 "6 ý,6 06 06 0006 00-i-+
ý' . -r .I -I -I -I . -i ---. .I--- . -y
I IN

"--ý 'r "-r 00 00 N iý [l- 00 ON 0% 00 tý


00 [ý et IT t- as t} ý-+ . --4 . --+ 00 NN 00 ON
IT IT t- Ir- ýGlOýCoo
"T M oo 00 o0 wlýten NNr-+lýv1v)inC-4 ONIt
oO MMNNN 01 t- t- -
oo 00 MMMM[- t- ý-+ M Men en 4n AWlh 00 0
t- t- Iýt-
Mý. Mi. MrM. ýýr. ýi
QýC1c1CýQý.
--ý. ýýiýýi. NN

l*-OhOSet99
'ItW _nýýv)O"--4-14 "-r[-00 00 00 00 00Nt+1 v1l-l-OsO, NNNNNN-
-, .- t- NýMNr+OO ý
00t-pp C\ mm ehwl W) W) W-1v)etv1O
OO 00 ppOOC\ MýDhgi't NNNNNOOOOOO"N
0000 týt, vl . -4r0NNNei 4
Vl Vi W1 tei V1 et et v1ýC
lý Qý Cý Cý Oý "-r ý-r .
-r . -ý . --i . -i . -ý ý-+ .r . -r

170
M %0 00 00 00 00 l- O-1 00 00 00 00 00 O% I7
N10 00 Ittd; IT d
ON I'OOOOOOC1400-
-4 -. +
t1i %6 1-Z 0000Ö ý%. 6 \6 %6 16 C C; 4-;
000-+NNNNNM. -4
NNNNNNNNNNNNNN ZtetRt et

00 kn W) tl- 00 M 01 V1 W) 00 Nt [ý
NhooI*- "-+ - Os [- Iffi ON I'D
ý010WI) WI)00ýGýo V1 v)Ml-00t-%0in
00 W) M 00 00 00 00 00 l- ON Cý ON C\ Cý n -4 . -1 . -r . -4

NNNNNNNNNNN

rA
Metrt0 00 00 00 0r -

M ýO M vý et 'ýt MN ýO V1 V1
O1 "-+ V'1 Vý M ýO ý V'1 et
b0 N
ýO N "-r . --ý (D Qý M ýO kn v1 vi v) l*- et Mt mN
00 r-A - --ý --ý -OO l- [- [- N I- V1 MMMM
10 N
rV V1 %6 1.6 \j6 \c; \G l-: GO 00 06 CO 00 Olý ÖÖÖÖ
9NNNNNNN9NNNNNNMMMM

E
0
N 01 O\ M
N W) -
Vl
\C v) 9NMý
ON 00 l- in \O N
'fit 'tT 'IT M
l-
-4
ON 00 00 MO
vA et ei' NO
\O vý vl N
O*, ON
.-r -4 . -4 .. r 00 l- \.O S%C\O
r. NNNNNNNNNMMMMMMMMM

O QýNen t-1CýthýPkn
C]. Nr-+ - ON ON t1wItM
nM 'IT MNO O O ON Q\ -4
I" et ýo \1p vn ON 00 v'l \p
\O O "-y ---- t- tn
. --i et ýt 't "f O\ 00 00 00 00
cV M V'i ý6 %6 %6 \z; %6 06 oÖ oÖ
NNNNNNNNNNNNNNNNNNN oÖ oÖ 00 0ý C\ Cý 0ý

to.
OOtw- . --I pN V1 l- t- NMMetv)vnl- QNtn.
%D S'C O v1 vn et mMO - M
Qh ON 00 [- l- O 00 00 N %0
N ýt MNNNNN l- NNNN ýo Itt 'it It efi

^'^"^+1-+NNNNNNNNNN
10
on a% 00 OO\. -D N W) N000\I1- NN
v1a-nr-+. ýOQ\t-\ONr-+OO
oo Qý N 00 00 00 00 [N _ t- 00 00 00 W1
00 OOO O\ O\
v'1 - 06 06 06 06 06 O- ÖÖÖpO NN
-4 ""r -4
NNNNNN
CA
NNmkn W)mN"-+G\M ch \ O\ et C\ ON [l- 10
"--ý "-r 00 - "-r ON 00 (- \D O 00 \O \ýO \D NM K1 -O
IT ? It oO 00 t- [- l- O O\ OOOON et q1t It et
NNNNNNNNNNNNNNNN 06 NNN

"ri
výNýOýDýDýC'a Ný1l-o0IZ%Mle Nvi NNP-+
N et mM p--4 aý Gh [- n
cu v? h 00 MMMMMh I- OOOOO
%0
0 MM [-: t-: [z[-: rincýOOOÖÖ
vý vý vy v1
NNNNNN9NNNMMMMMMMMM

W) ON Cý t- t- v1 "O v1 t- h v)
v1 vy r_> m r1 - "t - `t tt wt r-+
-? ' ??? :; Oý OO "t NN -4 00 Oý Cý ON %0
etMv)ýOOOOOOhv)v)v1v)
" -; c+1 v1 V1 t6 ýj Vi V7 OÖ Cod oÖ [-:
NNNNNNNNNNNNNNNNNNN oÖ Ci Qý 0ý 0ý

M
rn (""W-ý 00 00 M
lý [ý4 %O v) -4 'IT -+ Iý n 00 I'- Iý
QR Q\ [N'f; 't It IT QN O% l- N 0ý OM O\
IT lFt It It M 00 NNNN-N
"-ý "--ý M V; W, Vi Vi .- 1-4
V wn;
NNNNNNNNNNNNNNNNNNN

M [ý MMMM
.r . -r ýp"-ý. -r. -.. r. r... lýN r-+ lý MMN --ý
NNE--+et et ýt ýt
O-Ö"-i. -i. PlN
"II "I "i' -i. -ir-i N P1
III '+cVNNCVNCV Nr N

171
,s
cd r1 00 - 00 00 00 O\ M "O MMN ON Oh ON cM ON IO '. O 10
OON ON NNN In C\ ---ONN In O ON ON OONO ItY et IT
a t- I- "C MMMa r- %o NC .q r- M-i'4
6 t.,: .4 V11f)1ý0 Cý 00
.M
-r -4 ýD l- O, 0
^+ ,6'. M v1 v1 00 "-r -. 4 -4 -,t ý v'1 ý1 ýD t-ý Oý
ONONQNO',
ON Q1 0ý
eq M en on M . 1,4n t- tl- t- 00

NE et 00 Nqtt'Rtýt N%01. DRt'-t0%Rt


10 EO In 00 N- 'OIt N v1"---+%D
0tN -1N 09ýheh
-M 0N N nCC) In (In 0000 ON N10Ch Q NINCC 00
N C) N
aý ___ NN
ý"''a 'ý^ý. ýýNNNNNNNN
aaIa a

N
rA
0E- 00 In =--N . -r en NM 00 n In et OC In In 'O et 00 O\ NN
-E OO\(fl -. . -r 10 C) W) NMMNet
IT et 00 [N etN'. 0 IRt 't
In In In \O -4 N GQ Oý In \0 %0 '. O '0 In r-+ '.0 '.O O\ [N G' In In In
d4 z : WiCý\O\. R\IRC MMMktitl: Mw; w;
.; OýOýO; Ili %6006 -; NNN
ro NNNNNNMMMM

0 O

4) 00mN00tN\Orn-4r-. 0 -* -. +O'Q0Nen V '.0 In n0Oeh'O\O'G


oA In "--ý M . -+ 00 NNN O\ M C) O\ N IO \4O q! C) O et MON'. 0 --
ttI-4 OQ\NNNMen OInO\0\0' -+
fl- NNNNC NNN
10 NIn\ýD. %gWi 4t-:
-+M\Dhl-: Oý-4 MMM00000cV 060606
N W) 00 C\ O\ C\ -4 -4 ^4 -4 -. + r+ NNNNNMMMMMMMMM
I

NI'CMNNt- t- In O\ In InMO\'0\G\Ot-en t- t- 00 . rN'p'O


N NO In 00 00 00 .-+ 00 N et et Q\ '0 et et TP In C) N tN N
'L% E 1.0 M 'O "f eh qt:t et NN 00 00 CM 00 MMM
Itt \O In Mt Rh
O \C In NN IO N \O IT et et
et M It W) W) In (14 4 v'i vi [-: Oý cV eV tV 16 06 Oý Oý ÖN '0 t-: lz N
"ý N "'ý+ ý+ -ý+ NNNNNNNMMMMMM

00
NE e- '.0 0\ 0\0It [N\OC-, 00/0 - t-\D%DN'. 0-NN1O
cs. .0 -v1ý000%0"-. 00 M--4'. 0'0'0 dN0 0'-M In wItetM
O "T00NNNNN\1000000"-+en ýO\C\ m\ ýO\000 '0NO\O\O\
N It d v'ý v1 vl \O C) N M4
(fl cli In Ný C'. 0'. C' In In .16 06 NN CN N
06 ^"''a ''"^'~r+r+~r+NNNN NMMcan M

en Cý -+ N (14 00 m \0 N In tl- W) NNNNN \O \G \0 \0


.C n In m In t+l "-+ N 00 00 In In ON'. D In \O - In mN 00 In I'0 %0 \D \O
bM In O\ 0' N 00 O -4 ON O'. Q'. 0\ N 00 00 00 O\ M --1 -4 ON In OOOO
O' \0 W, N
`Z tV M vi r-: l-: nÖ CV MMM V1 ý \-6 \C \6
NerN6 N6 NF 00^'^'^"""'ýr+. -ý, -. ý-+NNNNNNNNNN
ii
10
ao \O
ENNM
Q\N00et-'.
'1 - 'p
ON1C
00 C) C\
N0C'.
C) C\
C\O\CNN00InMen
ON M 00 00 M O\
00 N NNN
MMMNN In In In In
M In tN O C\ 00 -N et 1.0 In MNMMMNN In In Cl 00 OOOO
NNE-+MNNO4 NNMýt\
o ý \, 6
z MOr-N
rA asIIIIIaaiaaISa
-00Qýý. -+ý. +NNNNNNNNNN
a" aaaa a$ a

kn
00 N
00 N- -4 00 M C\ \C In \O In 'C 00 M M 'O 'O N 'O
NNM ý! M q! MM
00 O\ N tN 00 "0 - In vi' M 00 ý--ý M 00 'RT - 00 00 NO In 0'. 00 00
'd'-NC'NNN 00 C) IRT IT eq t:r - CD C)o0N
O --ýO't it -4--+-
'G O I'0 OOO 00 OM44 .t1i .OOO
z' c+i v'1 .16 .6 .16 06 NMMM
ý' M 0? 0? 06 ^' "ý ^ý ^+ ý+ N cV fV cV N cV N fV cV MMMM
U
'L3
en --+MV1IM In It NV'1\0 hd C\'[h-400N00-+en
NMNN Q\ N 00 C\ t- -4N. -"r
-+ "-. 00 N \p ON 00 O O' et t- C) NMN
cl E o0 et N .-r C) [N N 00 M -- O In [N NNMNI et NM 0% tN N l-
O InNC'0 oot- In t-: OýeV NCV\ 06CQ 0ýÖN"6 t. 4 [-z P:
14 ", Ný0000000'. -- + ý+ -NNNNNNNMMMMMM
Ct
M
N ý+ NN rP "Ft N et 00 00 [N M O' N 'O '. O 'O '.C %0 et
'.0 In 'l 000MNN'. 0 In ýt ýt v1 vn
'd ý+" l- N-+\O'C 'et' 'O 'O %0 Itt ON 00 '-T 'RT N In N C(0
O -O\MNNO'ttNNNN-Invi
1: 100 ýOOO\OOM -4N00NNN
cd ý'' VnI'ei ice.
Vytý: 0dC'dC .
H '; ý? ý? ý? -r --ivj%O%6
ý' ^' ^a ý+ NNN (V tV fV NNMMC? MM

N
E ýt MN In "t' ýt M "--ý IT 00 %0 ýt N
ýOOýnv1h-In000)0000In'0 In ff "-r NMM 00 00 - eý ;s It
ö Oý NM NýTMOOtN'OO'. 1 \D
C) 00 00 (fi N \0 0'. 00
-4 (a) '" MNN '0 In 00 'c %0 '0
N 00 00 [ý ýO ýO Op
C'j 4i "i ý6 ÖÖÖ ri et 6 ý6 G OÖ NMMM
ý+NNNNNNtVCVCV MMMM

ý7 M\
N %D '4D 000'. '0\000 O'ß-4
CK t, - C\ In - t- 00 00
'.0InNNM M0'ý
Vl NI --4 O\ 00 '. O 00 MM Ný N 00 00 00 00
O "-r In In M C) 'C MMNN In In In N ON -+
OIT ON \,O%. O\, COOIýpNtV -4 -i} -4 00 00 00 00
Netlýt%: lýÖ"-+MMr or; or; vivlvi
NNNNNNNNNN

172
as ý 01 -+ 00 00 00 O\ ON -t' O'. Q'. 0% 00 N '.O 'e 'r kn en 4n N '.D Q'. C\ 'O 00 IC MM
O 1.0 ýD %.
O %O 00 v) ---N ON O "-r "-ý -r N ý+ 00 00 - ON N In In
--ý
v) 00 00 00 00 '0 O M
FQ ýÖ4
"t "t ,ntNNNM 06 NN cV Ö_M
G0 rNNÖO -' NNNNM ch '.O '.O '.O [N 00 O*1 O'. O'.
O ---1 MMM It in '. O NN tN O'. "-r .--i '" ý- .r -4 ""
-4 --- -4 rr -" -, . -" -"

Ne 'o v1-NNON ýt 4n4nW)-4m


'Cy E -. 4 000'. 0'. O'. et O'. NIIW) N- MM
Oon 00ON ON M-00'i Ii'e! ''tI'OO'. O'. O's ON 0%00N%DON ON It 00It NN
O %0 -4 '0 v1 In ;z .-, v'. ON O'. O' MNN et Rt 'T O' 'T O'. NN O'. N In OO
'z' ON[, MMM. 'o 6 Mý6
.; .i-; lýlýtýlýoo6-; CiCV Ciýýý\6
"N\O\O \O 0d ^"-4^r^4., r+-4 -4-4-4,. -+NNN NN NN NN

. -r
ti r--r
VE '0 l v1 In v1 - 00 00 00 00MNN -' -4 00 vl - 00 M MM
C\ - .r . -r -+
-0 E \0 \D 'O NNO 00 00 00-+ kt-I OMMMmNMO ON 00 00 NNNN
w O ON O \0 '. C 00 OMOOO \0 00 00 \0 'G '. O \0 N ON N "-+ ' 00 on G'. N tz
.O OQýI*: : tl:
%6 1ý6 l-: Qý-4 fV fVN(V 4wit-ý 0600006 1-NC4
IT TTTT ^ý ýý r+ ý --ý NNNNNNNNNNNMMMM
ii$
0 0
~MMNNNN
UA IA N '. C NN 00 00 00 00 00 M
MONNNm00-. 4Q'. oo00 -- 'I' n000OQ'. M-4'r't 0CMCD 0
bA "L7 E [NNNNNO'. CInINNN 1 Oen MMMMOOO'. OI'Gen "-+000000
y0 tlý MMM4
"o Z "-y ""r _; N 'rt lý CýCýCý "-+ ýO t: tý l-: [`-: Cý ý-+ %6 r OÖ 00
1.4 "NNNNNNNNMMMMMMMMM
cat

Oý C'. IT N0 In -+ 00 NN N Y1 00 [N 00 00 G% In m-- v1 - In N
OEm N -4
--+ %D NO -14 0N %0 '0 -14 00 ;; N-NN0 IO ON 'O I'O N [N G'. 0 ý+
bE r'+ N '0 \O %D 00 mONNNNN 00 NNNNm: t en N 10 N
b O6 06 0Mý6
M -- 'l, Itt
r_ Z 060606Oýfli41; v triw; t-zC --ý NclJNef %C[-: n
cd ------ --ý 1- 1NNNNNNNNMMMMMMMM
y
e
O 00
[s. '7en IT IT m'It '00\et. M NO\ oON-trt'MOO'R7--4N
4) O'. vn00000000MNN--Wt NýtO'. Q'. OOM-'i Q'. ONOM%0o0ON
ö '0 \G In In In an an 00 OOO- e' NV1 v1 %D 10 G\ \O ýP -- \C MNN t"
Oý [, ý MMMM N CN
Z -4 CV cli cV
NNNNNNNNNNNMMMM M vi l-: oö 06 06 cV

r-
E Nt
4) -N 0\ \X) 'O 'CNtNet'tW)0000m0'. en 000 NN
It Mv. v1'ttMr-+OON ON ON 00 t-t-kn Inif'v1\10-\DetIT \O --ß`d"00
p-i--r. N-4 [ öÖ
-, -+cV CV CV of viýý
""" "aaaa ^" 1"' ^'1"i ^+ ^"-+ NNNN
r"+1" 1""r. -+"NNNNN

ýO -4 I It It -N 00 It N 00 N V1 In -NaN 00 ýt 0\ d' 00 ON 00 'O cy,ý ON


O "-+N00NNNIRtIT M1 00 --ý-+CD 0'. 00'. 0'x-+ O\NN[N1Cn NN
M t- N W) in ON c+1 t- NN In NMNN N
00 r""--ý"-ý"-r"-,
et et en 00 --ý -+ NN 00 OO
t- t--4.4-4c-l4\-CtýtýtýnoÖCý
O" cV cVcV M4 \6 .16
F1 ýýýýýý"U ^'+~~~~r+~~NNNNNNN
a U

E et N"-"ý. -r.. nNOýýOýONCýNN "-ýý--ý"--ýel, 00- v1NN tfl '' ''


N In N 00 Ný 0 '.0 OMNNM O\ 00 e' 00 C\ ON C\ \G 00 CD C\ a--N ON ON
C\OOMN-4 -ý 00 MMM-mt-000000000\ \0 It M'000
O OlN\lC1O\0\D- IR1In\0
"iýD\C; \ r-: Cý. 4riCVC4C4Mvit-Z Ö_; MM
Z+ NMQýQýOýOýý o60006
ý+ý. ý+ý, 4NNNNNNNNNNNMMMM
1
et'
N ýONInNNNNNvnv1'ONNN NNNNN O\00%n 0 000'. N 00
td ID C\ %D It
%D t- -ren '.ONNNNInC'. N-N'Ocn - N00
O ýD In In
In In In N r'. mMNN R7 00 00 00 00 00 00 00 NNNNN e' zt
Z wiwiwiw;
1-ý OOOÖNvio00606Cý-+e ý6000fiticq V wit-: t-:
NNNNNNNNMMMMMMMM

-- cc NN'O\Oen NNNN 1ONNNh'O


NNMNNO NNtNNN
in In en MMN In [N 00 00 00 -- In 00 00 00 N ON 10 1ýO
ON O\ W) e* N In N
fl O 'i' - 'D K1 00 00 00 00 00 00 N- "--ý OVMM
..ai eI "tfi 00 00 00 00
z [-z 06 O
NNOýOýOýQýý. ý. -ý .ý .r. -iNNNNNNNNNMMMMMKIM

N
OE v1 "t en N r-+ -NM NN 'O IN C\ Q\ N Ný
-4 00 en 'O M 'T tr 00 \O In
't MO 0\ \O 'O N O'. N '. C N '. O O
N- . -I M oN 00 00 00 eý 00 NMM tt en \G In
Qh tN e' ýt N %D O'.
-4 -. 4 '. 4 -4 M O\ 00 00 00 qD !t 'O %D \O N NN In
z 000O"I'n C; QMM
N cV O'. Oý O\ 0\ ý6 o6-; NNNNM V1[ý: 0006 O6 c;
.-. w ,.., ..., ...., ..., ,.., _NNNNNNN CV CV N CV cM M tý1 M

N 00 '0O' lf'vn`'00M
ý- NtetNNNIt NG\ -tO'V)qT M O'. O\
ö N e1' \O Q\ N% DM t+1 N
V7 In 00 NNNN --4 OON ON Oý C'. O 00 N tN
00 \O In NN O'. 00 00 00 O
O "--ý M 00 00 00 00 e' O In O O\ O\ 'd- 00 00 00
00 Q, NNMv;
oÖ oÖ oÖ oÖ 0ý --ý fýI ý! MM vi v1 r; Wi

173

0M 0, . -r . -r r- . -r NM [- [- [- 00 00 00 00 hM 00 'fit e ýt mM- `e V'1 MM-
O> O% O\ Oý VI 1 Wn NOO o0 0o 00 00 'l M ýO 140 \O IN O' O'N Ný
o0 00 In -
r-i N In An mIOM 10 IMM (D OOO
Nit .O QN OMMM V) In \0
ý+ 1.6 kei kr; kei
-t": ý+ete+ItcV'ý oÖMMMMMvi - NNNN
"' "--ý "-ý "--ý "r --ý NM le le lY 'e mh lt le In \D N 00 00 00 00 00 Ch

In N00 1ýONo00000oQ --4\ONoo00 , -4%D \Dýor-(: [-MMON M


lý 00 ýf Oý O Oý h r4 OMM et 1,0 "-" tN (D m \m
'Li er In v1 In et mO et et m rt
r- CD Ch %(Z mt en OýNNýOOO%MMN OOCD CD OOOetet
0 C) m gt et
r14 N vlb 140 N le 10 %D \O \O \O V) e 00 N
z -; fV fV fV cV M vi %6 \O \c; \;
. --i . -i . --i "-i (ýj ýD lý [ý tý tý lý [ý lý OÖ Oý ... . -r ..... r .... .. r

rA 1-4 V1 t7
MN ýC hM
Oý -+ 00 N ýC ýt dMNM ýf V1 v1 V1 V1 W) MM kn v1 V1 IT
It Iýt "-+ %D "-r O plý ON 00 I- NO %O ilD ýp 110 ýO ONMMMN
ýO tý NO 00 h vl OMMNNNNN 0% NO [N tN IN NN F- ON MNNNN
,o X00 6M
0 Vi týl-z [l: l-: [-z t-: t-z 06 cl cV NcV NNM%6 6
060606
.caq Z 1"''I
I
'I 1'5 ^"1-4''''^+ß-i ^+. -ýNNNNNNNNNNNN
c
.4

O 0 1.0 ON ON ON M-
000 ON ON-ONMN00000 Vi "rmM'7t-%D MMen d'1ON
00 00 --MO It --0C
RT ýO 10 00 00 00 00 00 00 00 N \.O "-r "--I - "--- .--4 N ON MMM
Nn1, ["
oU ýýý NNN44N44NNrC;
-o
N 06 06 *6 06 C;
E
t- Nwl W11d'v')Mý00 00 0t*- W* rC\O\00 NN C\
M v1 NN- -4 -- MN'rY
vl t- '" 14D Tr NM- C\ Q\ 00 ýo [- 00 t- O, ý Q\ 0 00 0 OO -
O Itt I"' IRf MNM \1O O\ ON Q\ O\ 00 N .G v1 "--- - . 14 - . -" . -r 00 NMNM
ooM0000000\M 4 4'tqttNi
V'il-t-: t-: 0000
b 'z noooo0006Cý-4-. 4-4 ''^"""^"^+r+r-+NNNNNNNNNMMMM
066C vi l-:

00 N00 as ON MN[-1Clwtwl eF't-4 MNMMM1.0.0t-tN"r'r


4 00 O\vlMMOO1.0ON NMN0--+00-4 n"
"a e Nv)MNNQ%0000 000000N - 000 NtNONON' 1,0-4k eneý
"0 V'1N Vn'n'nr No 000
11' t" Cl "N 00 M1ý ý
Z -- \C ýC ý ýD ý OO Ö "--i . "r .--i --i . -. ý Ndý tý tý n
1-06060600NNNNNNNNNNN
1
w0 r- ON I", ON ON lFt Nt in NN et Oh I'O NNq! t - wl ý10 %D v) ý.G ON -NNM
E Oo00CD %ONN0OO*%%D%D -NCh ht-It"r-%Dn%Ot-t- W)00%D%Dkf)
110 O 00 . --4 . 14 O O) t'ý N ON Cý 00 00 "D %D v1 OM
--I kn tn kn v) kn 0 04ý Vý C1 ON Cý
0E-. 4 W) MMM 1ý0 00 t7 00 -" --4 . -" --4 - .. r 10 O0..
r ý--ý . --i ý--ý "-ý NM wi Vl vl h

-b
ao ýo %O--+en OIt %O 00Ch d' W) t, - t- NIRt M NýD[-NcnaN-4-4 co
'ttNOOMO*, 1ý0 N["rl-4 t-N")MON"M4If1ONN"'4 Mt'%IT
W" eh 'T en O as in mt MMN --40 W) NNCA
.a -t! tt t- r-+ V1 V1 W) et 00 l- 00 00 00
N NMh N 00 M ýO "-r tI1 00
. -r . -i -4 r 6
ý--; -4 -i . -4 ý--; -; M wi w; w; V1
y .r ýI _; . -r M wi ý16 t-: t- l-Z t-: l-Z t-: r.: OÖ -4
.. r
.I .r . -v .r .4- .r . -r '" "I".

vi r1 121 NN1, OetMg! NN"-+ O, %0Nv)h"--ý-QýMNN


irr
NNNOOI,
ÖEO -, 4 O O\
121 NC
Ch oNNNNNN
00 NN '0 fl V1 O 0% 0 '0 NN
'D M 00 \O %c eO
0% '(1 0NNN
V'1 O> N fl 0\
tof
0% G\
oO l- [, - Cl,
rV M ýG oÖ 0; "-i .-i .-i . -i .-i . --ý ýf ýG ýC

Gi

10
1 'IT
5 G% 10MMM0000f-%0%D%D V'IttMmf NMM 00 00 VyN Itt
4) ýO wl NNNM N- O qt 'It MN
"-+ NNNN et - ON \ \O 't''l"
NtNt\O\D\GCCNMM"-r"--- O to nMM
Q\CD 000^4M ." . rvlIttMet t1 tfl '0 NC\ 10) 4n W)
ýO 06 06 06 00 Cd oÖO6C; Ci4444t jI, ý OýÖÖO
Fý t- °° 0ýOK? `'4^'"ý . ý+CVNCVCVNNNNcVcN MMm
O;

M ON M1ýDv1 V*i hV1%. OIRto0hh%-OvnMOONen


00 t NNv1V-, MNd'M, CNNN
Qi wb It º000000'b V*j of''MNC 10 c) a
D
%. ON 00 hN 00 00 00 00 00 00 00 00 en oN't CCCC
00 . -r O ON "--ý .r O
O, Cl O1 O1 C1 O . -r ,.. ý . ""ý N 00 %, MMM
Net%Ohht-: lý[ýtýoÖO"-+etý! C; C;
E-

N
OOCIA vlv'1vl, 4411441vy C
0%oooOnvl00
nNNNN%Ovi Oý0 ý,0 V1Nvi NNN
0%ýNr -+"- 000 O%, O\D
b t-o0000000t'wä NNNN . W Mr. i' 000000
'10 1"e NNN VýO\O\C\o\OvoON(: C2N
etýtet-mai
ýDtýnlýnýýtýCýÖNCV fV i
CK fVNc-,
f c06 c06 0; 0;
84'NNNN "S I "8 "8 "I
4
"6 "I "8 `1 cV N cV NN tV N cV

00 r -4 P--1 .-r 00 %0 v) . -+ ,CN


,yE r40l 00 et 00 m 1,0 N l- -e M v1 et aý 0% aT O>
et e -eCO ViONoOMOhýtOO\%0x-00.
p f4 ýO NNNNN C -4 o 00 00 t-. t- t- '.p O et -rOOýIýenr-4le
y CýMMfý1MM00M00%Oppe -4 NN--- O% O\ DSO
fr lý 00 1 10oo +ONNNNNNM
. -i . -ý ý--i "-i M V; ý6\; \C
Vl \C

174
bN1.0
ý ehNt et MM--0%--- ON ON '1tIt . r-4-00
o -4 l- %- Q'. C'. GC et IT NN ON N O'. 00 N v1 NNNMMMn 't q1t W.1
t, N000ýCýCýwiwi W; wi t"g Mýg ÖMMM[ýt-ý l-- CN
MV 10*%okn on kn 'OI'O1C C'. O'C '. Ocntn 000Ný6\0
-r-
00 00 00 N Ch Itf

N
1'. - -ý. "rýr. -, Nv1' v) If7v1 W) MMMkn Ol -NM
co 'PtNNNNN'0 -+. -+N"" N- MMNItý"T
NOý-+. NN NN 00 ON 00
bE -. -+. -r%pNNNt- NNkp0O[-[- I- MMMMe1'MN
t-C7%6 (D d00000000 dodoo
vi', iNNNNN 't? et
FO
F+ ^i ^i NNNNNNNNNNNNN
1111111111111111111111 r+ --i ... . -. , . -i . 06 00 -i
11 11I

CA W) v1 'n v1 10 Q'. Q'. O'. (7'. '. O M "--ý '. O tN [N


'-' %00041' MMMM ON ON e1'
E-N n4e)%D '. (fl Itt IT NNNN 000CCON 0010 '. 010'. 00000N
I'D ND "o 1.10 NMMMMMM
b ý+ 0 ON -NN0000NN O'.
Oý -- ýNNNNNMMMMMMMMO MMMM
yNMMMMMMMMMMMMMMMNNNNNNNN 0ý l-: l-: l-ý 00 OÖ
.L z 11111111111111111111111111 "-+ --ý Qý
aQ
W
0
O
rr10M MMN00'O'. 0 nan Vn'n NN 'n 00N1- 0'.
V"4 MON W*, 00 -+-4
W1hIRTC14 r-+0000 W)It11, %DCN000NCNONINOMm00
4r Ci in Cj 4n 0000000000NNO00MMMMON
GE egNNN ON -4
M Vi vi vi In W)1010ý'0ý'. o' a'. ''et't
QMMMMMMMMMMMMMMMMNN N O'a'. et0606O
b M N1 NNN 00
'ý "ý

z 00 MNM tM 00 N '. O N
It %0 N . -+ tea c+1 M an MN O'. t-
140 oo _N RTýý OOO
z M vi 1*
^+ l/y vý v1 v "D ýC ýp v1 cV cV Oý ýt Oý Oý Cl
MMMMMMMMMMMMTMMMMN `ýt eF et
1V NNN

00
It ccan0'. 0'. ON NN 'IT 1RtITM "t mtNit NNNt
NN NN%0 "ItMON
-+ %COOOo0001mm00CNON kn tt t- t- r- ""--0lnrt
Ö ýO NM et -I- et MNOOOOOO o% MN C'. O'. 12N 00 00 00 00 Q% p'
[- Qý "-i "ý ""-ý ""-ý "--i "-ý cV tV CiN C4 C4 4 Cý oo OR
z NNMMMMMMMMMMM1 V) h v1 --+ -- --i _; ý6 ýp
N1N1 N1 N1 NNNN-14 x8 n

to. N
r
00 -00NNN
00 an 00 '-t v) an In NNMMNNMN
00NN0000NN'. 0 Ift (OW)egt- ra nv)00
MOrlIA vl c+1 'D '.O 'D MN N- 'O '.D O
V'1hVlN-r-r--i-4V-4 000v7--
X00 'Q; -00000000Q'. O'N
pE CýoNC; 00000oöt-w; ke; W1--ý. -r .. r. --i 00001n
,z .--ý ý--i 1" . -. i --i .--ý . -r w" NNNNNNN "--4

Ici
ao tD
E
0000 tea an an VnC''t MMýt Rt0000"t m
N00NNNNtN1. GaAv1a1N V't
Nen tn ami'T'NIt tea tea tea N an 0'. tea
10 a '-+N- MMC+MetQNoNa%cNp%as ppen N000NN NNNNO'.
vi wi wi "--ý r-.. -i .--ý OR 00 N
111111111111111111111111 l1 ý1

Vl M t7 1.0 G'. O'. .-r MNNNN


NNr N0'. 0, N01'. -+ 00 N 00 et eh MN ý} et M
p0MM tqttIttIftta'. NM' '.0141-4 O411 OOan
-" MMMM e7 OOOOOO 00 NN Ch Oý ON . oo oQ 00 N N, _
N Oý "-i . -i .-i -i . -i . --i NNNN cV CV
NNMMMMMMMMMMMMMMNNNNNNNC11, oo an vi vl C'i "-i -+ NN NC
z
"T -1-1
1

10
-o p tNan 10 anNN'OýCtNt- vanNNan00no00ON0In 0N
-t NN
NN-
NN'.-T en qtt
0N
tu
0
E NC> InI n'i
ÖMInvlvlvivitriýýýCýýDýD\6
'ýtNNNNNN -- IT NNN 10 10 10
r4 ,0 MMMMMMMMMMMMMMMM \CC-i c Qß\6 et etOýQýCý
1M' M1 N. NNNNN
Cd
IRT
d "4 W1t No0NNNNNNNN00 t-
a) ye "Itt- an (2'0\00'. 0\0*, o\ C., Q\ONýMNNanananýtN 0
b ."
"It ýýNNMNNOOOOOOQNCýM"-i. -- C)C)a\ 00
zO -+. -+N, }ý}etoooooo
cV -t vi V1 Vn Iii vn '0 \O 'p 'p \C \C
M IT MMM vi to N Oý cý cý lý et oÖ o6 06
(I MMMMMMMC? MMMc CV CV NNNI

N
E -i' an oo .-r . -r '- N
ON 00 O\ 00 00 \G \O 'O
O"4t+1O'. O\O\000000 OO\O0\Q'. r-+ "aN '0 M --I M 'O 'O \0
COC\ MMMMMO" 'D'0. -. MMIf %0 g7M'T'17 ýT`ef
Oý-+000\ Q\ OOOOOnQ\
z odC; NNCV NMMMMMN(IiC 0\ ON - -. 4 "-+
[t, , N\pNNN000OOÖ
NMMMMMMMMMMMMMMMNNNNNNNN
--ý "-+ ý+

NE0M -" an ON V 00 00 00
ö an -r an an an an Vy M %D ý.C !t NN-- [N
E", -a 00 co O\ Q\ O'. 00 OOO -4
. -r O 0\ ON v'1 M d' tea Nt O 0'. Q'. 00 00 M
N O\ MMMM tea O\ OOOOO
Z 00 00 Ny Q, Q'. Q'. ON an 'p %D \p
06 Q\ OÖÖÖ 'O Ö ;- et V et
NNNNNNNNNNNNNý -i -i -i OÖ 00 1! ý h IA Vl NNN
. +ý. ý+. -rß-4 060606
I

175
bcd
Oý Oý et Oý Sn %D NNMM IO t. m v1 M kn tt) NMNN ON [-
0h lý "-.. -r .ý .-.. r
o0 00 MM -4 -4 OO --1 %1p et -RT 10 OMONNNOýO 0% tl-
C14CS VII (14 MMý-4NON v) "It NN'tt MOýýýýrý:
OON
i-pON tz ý
NIýcV +, ýt
NNNNNM cV cVcV cVýGÖetetý
et tt -t et .4Z 00 00 00 o0 00 ý6-NNNNNMon
O\ "--4 -4 -4 -4 -4 -4 -4

N ['- 00 Cl 00 -4 . --4 (7% et -Rt 00 00 W) l. % [, N 00 %D 'tt 0% M


NE It - l- NnN 'OT m 00 00 00 "NMMMN
N ý--ý W) M 00
-0 E OO 00 Oh 00 ýo %, O 00 00 00 00 M 00 -I- mt ýt
v)
et
v1 00
ON in
'"
00
. --r
00
.. r --r
00 00
. -,
ý,O

N
. --4 .
-,O ý,O
0 Met v1 V1 V1MlýV1vb V1Y1w10,.. ýtVNNNM
lý: ccc

H
00 00 00 --4 --1 C\ ['- t'- I'O 00 --+ --4 O\ C\
är O C: o
O %Crl
N 00 00 ON
[ýtCN 4f, V1 If) No000QNO%l*- OQN"-+QN-4 ONO-, NOý
NNNNetin MNN--4ý MN--4ý, o W-1 %D 4r., V
NMt'- t- t- V'1 NN

z IIIII

-1 -4
INNNNNNNNNNNNNNMMM
[. [.:
1.4 --4
OÖ ONMMMM4O
1-4
OÖ OÖ OÖ OÖ OÖ 0ý --ý
MM

C
0 NN 'O 'O kn r-1 -N 'O 0% oo NNNN0
"-4 Co 00 00 N \O IN
00 M <T MMV, 4 -N 'O \O \O Q1 O '0
00 N "O N 00 C% 00 [ý l*- 1.O O r-+ v1
lit oo NN v1 MN tN NN Q> vt v1 'D
%ei v1 vi 00 oo O \O '0 \O eý 00 00 00

ö 1448.......
NNNNNNNMMMMMMMMM
tz c; c; c; c; cs

Z 9991899IIIi

0\ 00 en O\ C\
lýrtNNýt'ctN00
O ON MNNN Oý M 00 M
00 00 00 00Mýf ýOýDýDýOýCIýOýýOýOýOýOýtý0ý00
n er

10 ö
MýDNNN
[-: t-: l-: noö, r, jmay;
eh44vit-Cý000Ö---; Mv1h
06
Z Ia1 ""' ^' -S -+ NNNNNNNNNMMMMMMMM

0 00
Ci.
l"

fn
V1
M
W) h
Wb Vl
an
IT
V'1 V'1 Ittl-
mNm
C, % Vy
IT
10 Rt MM
00 00
0\01Q\ONt-
ON Ni
ON t'- . -r . -. r
\O\00\O;
It
O
MM w1 Qý W1 %D ýO
ON OK V1 00 M 00 00
00l-\CMMMMON00
ýO V1
'I,
00
N
\C
n0ý
00
N
0ý --ýv1 V1 v1pýr NNE--
V'1 vi W) v1%000CD r\ t-
0 b 06 06 06 ý''
tI
4'' 4'' 4'' 4'' 4-4 -4
-cl
NNNNNNNNNNNNNNMM
: r- r- t- lýC-4 --;

N ^' ýo -+ýD
oO Q\ N tl- N
MO0%mQ\00.
C14 \0 %00000M
et M l- 00 00 O\ N
1440Nm
it O
gTON Cý\G\, OMt-ýo\,
l- 110 ýo in mN-NNNN
o +- m ch -N
00 00 Q\ N
öE "+N rt rt et
-iOO\O\00m[-OQ\ON
00 00 "-ý OOO sC et
Q\Mtt)--ý\OMM\.
hC
ON
\D OÖ
O
Ö
Z . -i -; -i N V'1 NNNNN Q\ 1-4 . --ý - .-r . -y - -4 -N

tA "w-4 ^' n\0-4


N4 C) C) -+NýýtIl- W)vý1NNNMýnnÖ
ýýýDDD -4 -4 vy 00 Q\ en \ `T C14 eq hO v1 ON
NN V'1 In N "--ý "-r
w° I .-ý .-y vn to 0
. -. ý -.
4_; NM V'i \6 %6 N6 \6 oÖ O\ Qý
r-+ 0 Oý------------- ---

1r
IT It v'1 v1M\. OM9! M9! I7' \Oý} %O V1 t/1 v1MONv1\ inW)I0 -N n
4 N O\ N 00 00 -4 MO O\ 00 00 00 00 M 00 V1
NON-. wt IRT wl "" 4 "--4 --4 00
NM v) NNNN l- 00 nOOOO It v1 N n ýa V)
%01.0 er It et It Q\
-4 C; 4\D%6
0 -14-4
O0 O0Q`? ý? "'ý "ý'^""ý
116 \6 ý606C; V CVNCV
ý"ýýýtVfVCVNNCVNNNNNNCVCVMM
cV ef%6 [-l-[%ntý0ý"-r. -r
z
ID

%D ýD NNN '-+ M "-, .--i -- 00 00 m (N '0 00 %D M


[-%0- 0MON 00 00 t- 00 NNN 00 00
"-4-4 ONt-OýOoot-CNt-ýoIn NON ooON 00 er 41)-4It
O ON NNNý. p ON OOOO
0
-0 Ei ON oo et "It et qt O- 00 00 00 00 V-+ V) ýO M ON 0% ON .4M W) N

z 0001
III
O tri oö 06 06 06 06 -4 M4444
I'"'I II
-S SIIII
vi [-: Cý OÖOO
NNNNNNMMMMMMM ý-+
M v1 116

S
117
ON 00 NNN V) ko00t-t-W)s1n%Oý.
ON 00 ll- 00 00
CýDv)"f 0000 wt et
IT 00 N It MMN 4n [ý 10 10 %0 V1 !hOO9 'At V -T t- n l- t-
R7 Ei oo00MMpv'it-z
riÖCV
O lý[-: l-: 444
4e0C;
0000 000ÖCVvi
ý+NNNNNNNNNMMMMMMMM
z
-4 --1 h2-iMMMIý nMMMN
etýtNMN'"D It --+s en NMNNen t'--4M
ON MON V Ov1v1V1'Rtm '1*000N
OON Rfc)ý0-
-0 E "I M ['ý oo t"*
O0%O
p
MMOOOOO.
-+ Oh et er ete*erN-+Oý

z pip 0ýC ý et l-Z


yrýýýýýýý
SSIIISSSINNNNNNN
t-: t-z t-: l%z Cý . -i KIMMMMhýp00
fV
00 00 00 00ÖNN
NNNNNMMM

'0(> V'1výv1[*- V) Chow)-4 V)rýlfiNNNC+letet


ch -r V% V'l vi 00 0 ON ýt 0 \OýDep-%ONOOM
\0 0 [ý M rý rý t- 140 O\"
V) qt NNN t- 00 MMMNM rn -9 \O \O \O \O '. O
0 mmYD \D\C\C\O-e\Dý, 0ýD%mm-;
ýc O v1 vý
NNCV
vi v)
cV
Vi
r4
r4 t-
+v rl:
OOOO
tý:
(D
ý: týZ t-j
v)
06ÖO
p-, Q>

141460888888188110188

176
V1 NMMMM V1 ýY M lt Vl> lt IO IO
vlýC\OMD 'ýD"-4-MNONCD NOCD
00o0000N0 NO NNN
ýtoÖO6
ýO V1ý! 1 ý! 1 "-ý . -r OÖ 00 OÖ 00 00 "0 ýO
N,0 00 00

0000Nd M'TTttONON M. -rt-%0l-. -+00. --1


GG" C) pOO l- l- it- oo t- oo t- 00 N in
MMM 00 00 00 00 OO 4n m
0006000 vi V'i vi 6OO6OM

y
ON ON %0 %D ýo h %0 %0 %D NNNNNN
NNN 4n vl tel h lý n
[l- NNNN v1
r-+ MMMNNNN --ý
.Z
\O"--ý -- _;. --i_;
(1Q lit MM t+1 r1 M C? NNNNNNNN 91 O\

O
00 00 [- O\ 00 [ý lý M le ce 'fit m
N v1 M ON (: 00 oý NNN l- N
0000000000 V) V1 V'lMen MMMOOM
ýn ýn ýn vi vi vi ýn oö oo oo N tV N cV rV Wi n
I cý1MCý MMMfý NNrVCV CV NNN000?
E

OK 00 . -+ v) V1 v'1 N 'Rt %n m- t-
N et et oO ý-+ MMMONM- 4n t-
00 00 00 O--- ýD ýO %D ýt `qt qt IRt Ii' 110 00
hV1 v1 V1 V1 h kei 06 06 oo fV f-1 t": Iq
CV cV P1

b y
MMMMMMMNNNNNNNNo000

O
00 %D "e -t 00 oý 00 et V-1 ,0 -ýO w-, vi
-+ NN oo v1 vl v) O r-+ OO --+ NN "-+ ,O ýO
0% Q% O\ vlb NNN vl v1 v1 OOOOO et "
N -- ---; --rr V1 V1wi 000ÖON"r
---;
Ir MMMMMMM (V NNN9NNN 00 0?

r-+ 1O(14 NN- Mv)C, 0CN -+ý, GW)


.. r NNNN 00 00 .D
O-
00 et %O \O \O
OOOO of IT 00 00 00 .b- .b
IT MMMMN "O
ÖÖ6Ö0ýOýCýetýt`ýt ÖÖÖOýTN
N CV NN

b
co vt
hMN I et --ý V1
00 O 'O V1 N
ON ON ON 0% MMM [- [- t-
wo 00 MMMM 00 M
Cý Cý Cý Oý Cý Cý et et ýt cV 6ÖÖÖ 'ý? M

nn 10 n vý let Rt 0% nNNNN C\ o
00 00 n oO en NN 0% "-+ G et OOOOm ý--ý
OOOOMMMOCONNNNv1O
vi h vi OOOO ýt 00
MMMMMMMNNNNNNNNe 0?

'L3 00
,t 00
'T
%0 %0
MM-O as 00 0% W)
01% 00
00
(7ý
ON 00
00 00 00 00 ON NM
"--ý 00 00 00 ON ti' --ý
NNNN qlt '": et O0 00 00 IýO IO %R 10 00
"G%'DýG%0 viv)An cN000000 V NNCV
MMMMM ß-4O
Mime N Co' NNNNNNIGS

-4 -4 00 00 ON 00 t- t- r- t- eq ")
[- t- lz [z M MM 00 01 00 %0 %0 I'O %D
0 CIA ON ON NNN Nt -
oo 1.0 10 I'O MMMM t- v)
vl v1 v1 vi wi tei --ý oÖ oÖ oÖ Ci c'l cal cV -1:
MMMMMMMMNNNNNNN... 0Ö
H

cý rn \o %,
0 1,0 h-t V1
nItt ;t Vy 1O 110 M It
OýO%OOMen A ý00ON ON etN
MN\G O SC -. 4 -4 -4 -4 r-ý 00
%,
tV cV MM cV cV cV ý-+ ýC
MMMMMMMMN CV CV CV cý1 CV CV ý+ pý

M 00 MM W) .-14 . "-i Cy 00 00 M O% Oh ON M ä-
, ý! It t-
IT IT 00 00 00 Oh OOO 0% ON ON OON
0% QN ON ON MMMN %D %,G %0 ON C;N ON OO . -r
00000000 v'i Vi 'fl OOÖ
NNNNNNN

177
00 00 M-N %0 %D r- r- r- 00 t- %0 V, 1 MMeý, G t- r- V) V) VI Q1 r-
%Drýt'- --tOC>0000le \ON%COONOOMOI-t'- vlVI V) M O00
N t- n ýe wl -t %0 %ýC et N et O -t m
OOC7ÖCVtýlýlýtz [ýMNMN
..,,.., vitol-[-: tI[ell) NOOOr-+MMMMMM ntý[ý
N N V'1 h V'> V') VI V1 V'1 ýA
WAN 1--4

Mv1v 00 NN oo et Oý Oý vý ýD N V1 ý+ NM lý "--ý "-r . -r . -r .--i of 00 V1 v1


NýýNNv1ooMOýetetOýervýerMN000OOý! MOO
n Vý Vl 00 OO 00 00 t- [ý [, - lý "0 et m t %0 OO
-t -e --+ MMMMM \C
MýýOMýMIýn[ýlýtýtý lý^+, yNfV NMýýýýGýýýOýÖÖ
Nh v1 N r-: ri tz tý rý lý OK "-a . -i . -a .. -I -I -I . -I . -r .rr . -4 NN
a I

00 00 1-M N l- mh In M ýO -4 01 O\ It
rA M tt1 et 10 n W*i 10 lý
It
%.O
"i' et 'tt
%o %0 %0 WS v11 tn
-
N O, ý Q., 00
t- et et MM
00 00 t- [-
-+
N
-MM
ýO vn a., 0ý
O% MN "f IT ^i* 00 00 00 00 00 00 00 M- ýO V1 wA Cl MMMMM 00 N t- tý
-4
Q, O,. wi160ýCýQý
ý ýtý ýtýteF eFýt etw Cýc; C; keiwiw;
'n ^" "I ^"^" "-S ^+"-+NNNNNNNNNNNNNN
I S I

c4
0
MONON10 W) N n- nN 00 -N00 O V'n--1 -14 -4 ON 000\IRt W)mcq
mt l- [- N[l- 00et4nMv11 'gTen Ot-00M O0ý V)M0 NON0 M000ON
a) l-ý l-- en O"-r. -rr. r. -4-4. -r-4-4kD00rýrý1, N V1'"-4""-4O-4 CM-4
%, O
O V1 h t"' r-i
%6 t- l- [- 00 ýMMM er tz cý Cý c1 Cý Cý OM .t4
a) 6%6 ýý *6*6--"'ý^+ ^+ý-4 -+ý- NNNNNNNNNN NMMMM
10
E

Z ON v$ %D "
TON -
qt N Cý ON ON C ON 00 Ir- -"
MNýoo0 00 00 00 00lý1.0
-4
tl- MNOs
00 r- r% %D MMN
ooOvn --
r-+
(D M00
r-+ Nt 0N
W)NON
00 V1000NW) sn *0 %0 %0 10 10 ýo %0 0% -4 _0 QQ W) %D it '-T rlMMt-V1en en
IT OOO%NNýýýý%6
6t %6 .16 -4 cri eliMM%6oÖoÖoö0öCod GýfVMM
w; ý, 'I -1 -+NNNN NNNNNN NNMMM
.I
cd
1..
O M INT M (l- N WA W) V'1 h v1 V1 k- ON N ON M tý MMM
l- --4 r! "-ý "-ý "-+
-OMM Oh V1 W) wl V1 V1 N %D iN - Ch r- t- Cý vl V1 NM-
ON 10 4n O ý--ý a Vl
mt %D 00 00 00 00 00 00 00 00 ýD %D Vl V1 00 ýD MMMMM It Cý [l- [l-
NI-t-Qý... ýMet'ýT imivi1ýG0Ö0ý0ý
IIII of ýtýttý7 Wi000ÖÖMw; kr;
NNNNNNNNNNN
i
°
w
et%O\O o0Q>O>M0000INv'1vi O\l-c+1V1-et V'1NNNv1vy000\ ON
VI 'et " I/) NO Q> N %D ýG %D MM V) MO 00 00 M %D 00 00 00 %O ýO vy 00 O
MMO lý M 00 MMMMNNO cý ýo vä wl %D O\ V1 V1 V1 v) v) met mM
ýýýýttýQýOýQýOýQýOýQýOýýOOtV NcV
cV dÖ

b
on I-(N OýNýOý--ýý-+M[ýMMmt C% "t Dt- MMNnNQ\ M NNN
N %D ý-+ v1 'ýt ýO [ý . -r rlý 00 of l- [- %D v1 eh m oý 00 [- 00 10 YO M "--ý ýf 'lt
vb [- M ýD N oo c- '40 l- ýO ',O vl kn O 00 \O \O \O ýD 00 v) w1 v1 h v1 NMNN
w° pp ""r ""-ýýOn0%O%G: Q>O: 0%(x%V) pfý NN r4rf %ý6 %6ý6lýZ oýÖÖ
.. --i .. r .. r . .. r .-r .. . -r .. 99

-N -+\O N ON0000m 000000N MN - 00"tMMNZt'ýttet


4 ItMMnNNOv's vlý0wl, invnNýt0O[-ýOýG%t-MNNýýýOC W) tn
eq r- %D 00 00 IRt ON CNON ON C1ONCN"tNv) VII l4r) V'). -4 MMMmmm'Itnh
0 ýGOOMOM
ýtýf eheýetetýtIti06000ÖVIVivi CQ
vitoi Vlo6C

110
1O 10 NNIýW) MMM'TM - NM %0 W) W1 MNN O\ ON MM
'e! ' 00 00 -MNO -
00 00 00 ON 00 %, p ON l- 00 et MMN 00 N [- %O M 00 OMM
b MOO Rt l- I- I- ýO 10 ýo \.O \O \1C OMNNNNO \O \D ýo ýo ýD N- \O ID
ce etOONvll-: h n[-z [ýnt-: 06. +: 4 414
0 t'I-z cl Ocr 0ýOýÖM.
a aIII ^'''" ''ý''ý''ý'S
SSI ý+NNNNNNNNNNNMMM
S
ed
00 .-r .-r . -r 'ep tý MM -4 N "-r "-r 1-4 wl IRt VII et 0% IT Vy V1 V1 MNN
1Otl- NO IýNO[ýll- 00 -" "14
4n ý10 k1M VyOW') -+Ot-wiwt O\O\Q\t- o00o000
%0 ON 00 ON NMMMen MMM\0000., 0\0000
ON O, O\ V'1MMMen Mt- 10 'TItt
0ý Oý Cý Cý Cý Oý N V1 v1 v1 vl OÖ
'ýr+ NNNNNNMMMMMMMMM
I at aI II"
H

\O 0o V1 kn IttMMMMN-o0N[*- oo0000%tlf'[re1
0%-r-+O-e Ml-\C\O\C%D 12r 't i --,
h eer-+CD 00oýQ%O\NOo1cýQhQ>Ch -4P-r00=
Y1000\v1NNNNNNN NMN000%(>C
oO MMNOM -4rß ýrý t-r-0eNN
VI wi vi vi m V1 vi vi OÖ ÖÖppM u1 Vi V1 kei wi %j6 OÖ ÖÖ
ýýýý---4--'"-4--4 -44--4 .- ýýýýNtV NNNNfý NNNNNMM

\OO\v \ rg00"t-[-t-t- ON%ID\OI-, O-, 000MMMN V'1MM


00 " %D O Oh M ýt O %A O v% 1/'1 KA 00 00 OO
v)1Iý` ON 00 00 tt mt -t ýt M 00 NN
VlbMNID l- %%Vvý %0v1 V'1 v'1Nl-0CD Ch %00%OCD CD O"-+N0000
MOOON `e e4 NNNNNNNOtVMM
rl d' \d l-: [z n tz lz cý ÖÖ
. -i ý-U . -i et ýO OÖ OÖ OÖ OÖ OÖ OÖ OÖ 00 . '
11111111111111088I111U1UI . -r . .- . -1 "-ý "-r NN .r. .-. . -U . -I
11111111

178
n tl- v) M oo 1ý0 00 'V Nt NN lý% l%% [- Rt 'tt wt 00 00 00 00 m Cý m 0ý
MO000010 MM- MON ON nl-000000NN MM'-+0%'tOO
MM Ö 4444_; ""r ~ Qý ýC --+ NN
cV N cV vl vi Ui "4 _-; [-: [.:
tý nnn- 00 00 00 00 00 00 00 00 00 00 00 NNNMMN cV
00 00 00 00 00 pip

in N Vn V1 -4 N 00 00 00 00 eh --4 %0 Ii' 44-1 l- NM ef' - %0 00 MM ON et It'


OOOOO I-T %D %D%0 1.0 *10 4nýtON00%000[-I'C c7N 00týtýtýNNN
OOOOONNN NNNNNCh ONC%O1% ONON Men MMM'1t NN
ÖÖÖÖÖÖ ÖOÖ O ei' et
"-"i . --i . -i . -i ri . --i V1 V1 "-i .--i . --i . --i
NNNNNNNNNNNNNNNN -+ --ý -+ --ý r-+ r+

H MNMN-, O-tv)v1%Avi NN-+MO%OýNMl-'z 0,0ý00>


%, oýc>
oý000>00vl00ONQ>cýON -,O%emot' Ir) OMnNO0%O>Q-, O>MMM
MN NNNNNN-+ OýMOOOO r-+- -4 - MNN
it

NNNNNMMMMMMMMMMMNNN
ý1Q -I -I -I r-+

O
Nnn MM M le NM MMN le 00 00
Qi
U
Ö o0 00 00 %D MD \D VN1 vv
4r O- 000-0[-l-nl-l-I-[: l-MMe -hZt OOOOO oONN
ýfetet'ýi V etýý Výv1ývýv1v1ýv1000O6 NNNNN"""C
MMMMMMMMM Cý1 MMMM ýj N [r 9
b MMNNNNN fV
E

ON OIN t- QKtl- 00 lýlr)I1) It ON F- F-.


1ýr 00 -v1M_M_
MMMM Q M t- t- ON 1- 00 t'- kn IN
In Öns - C) CD CD C:)
ON Q\ Cý O*, G\ CC1 O\ VW'1 itqT
) If) vl byT 00 00
"C1 MMMMMMMMMC? MMMM, MMNN tV fV fV CV N [7
ai

-. 4 1-4 O\ONav1t- kil t- V"l wlmM mv1\CIONM"--4"--4"-r


C1. kill gtr4 1.000\000'0'0 eh W, N* i/' knm -tNNN - Oh
Cl, [ý lý lý NN Qý "--ý . -. ý .-ý .-ý rr . -r .--ý "-y 00 NNN----- ON fl M
Qý O; CýO; Cý "-i ý Cl O 00 eh Oý Oý Oý Qý Qý M lý tý
. -i .-ýi . -i .--i .. .. -i
NNNNNNMMMMMMMMNNN oÖ v1 v1
ý-+

v')vnvi 41100 l- týNNM0h0% t-"en 0%%0ýo%oN --4 v)l"


.r OOO Om v1[NN NnýOv1kn NO\NND V) lt`ltMz! ' 00vr
MMMMM -tvivý V'b V) V'1wý krý v-b MNe C,4rq ýD%0e1,0I-Q%v1
3 C
OOOOOO --i . -+ . -ý .--ý . -ý .--ý .-ý .. r .. -i ""-i ýG ýC ... i ,.. i ... i ,.. i .. yOMM
NNNNNNNNNNNNNNNNN
. -+411

ao
00 N 00 V1 yr 00 NNN 00 It qt v1 N [- t- N Wt W) W) %D 00 N
0 M of M OO OO MM ttl q'i V1 e}' MMM Q\ 00 00 00 Ch v1 V1 er eý N ý--ý M
w NNN
ÖOOOOO
- -ýNetd --t- -, - oC o%CýO-N
ýG ýD v1
r-+ "-i . -r .-i . -i .-i r-i . -i .-i . -i . -i O ýO ýO "-i .-i .--i . --ý .--ý
VNNNNNfV NNNNNý 1S ýý
IISSI IIIIII

00t-10%0 00\ 0 ON 0'0In00'0 00 10 mm


v1 00 t-I'O %D ýo 10 00 00 I'O %0 IT et M eh N 'C /O 'O 'O G,ý mCmmNN
[- SO ý,G NO %0 NO -- -4 -4 . -r -4 -4 - 00 00 1, tý rý Cl . -r -4 - M NO ýD
-4
OýOKCýOýCýCý"-ir-i. -i. -i.. rý-i. -i. -iOOO440ý0ýCýCýCý "-r'ý"

10 NNN ON ON C\ ON ON CNcc lýhN 00 NN00. -ý


NNN co 00ýo00ýoýOýOv, etN00ýoMMMn"-1-4 ONCN
4-40
ed vl NNNNNNNN 00 00 00 ON ON et 1.0 %D %D %0 "NN
4 .
16 116 ý6 %6 ý6 ý6 .16 ý6 tei 4r; V1000ÖeYNtVNCV Ci --"--
MMMMMMMMc ý1 tM r1 M Me MMMMN CV NN cV cV N fV r
tti

"-r"-ý"-ý"r OýOýNNNNýO efiN oON-+ýO\OýOýOe%DýDv1


OOO t- v'ä O> ON O\ Oh mN- rý v1 NMN"
le le el emMOOOOOO0CDele\Or-r-mMMMMM v1 vi vl vlb u1 --4 ON
fl v) r, 1
. cd ýOýýDýCýýpoÖOOOO0ÖO0o0oÖ00[ýn1ý00oÖetýi'etýi'
MMMMMMMMMMMMMMMMMMMNNNNNN ýi'"-ý"--ý
E--ý 111111111111111111111111111 Oý Qý

00 00 \O -4 - ". r .-r r CN 00 [-
00 00 00 00 v) v'i --i %O I'O .\O
-4
\O \0 VII ""
00 00 !t "r .r r+ MM
NNNNNNn l'- l- l- l- n -- C) W-4 -- to \O o kt) t.0 tt1 NN
l- t- 'T 'RT MM 00 \0 t oý p\0 \.D NN
ÖÖÖÖÖÖ_;
MMMMMMMMMMMMMMMMMNNNIII14.4-- 14-i. te4 _;. --ý-4. -r kti wi 4QýcýCýcýQýk? vl
P-4 -+ %D %0

MMMM aý " v) r- r- r- M (> 0% %0 M V) "-4 NN


NNNN rý -rN 00 00 N V)
x-14 -- r, - -r --4 --4 -4 (> O\ 00 MNý oo oo [ý . -r OO 0% a% 00 O
00 00 00 00 00 00 0% "-4 "-r "-r .--4 OOO 00 00 00 [- rý iý "-r .
--ý . -r OO [- %M
cc; c; c; c; c; c; -; cV NCVCV Nr4rV dp MM
NNNNNNNNNNNNNNNNN. "4- -4 ldýdNNClicVtV
. --i - rI . -I -+. -i . -+
I

179
b
Cd Mit It
NT 't - MItIt'MtI- -. --4%0MON -Rt'Rteto0 NNNNIO
IttIt qt VT MT ON INN\OOOO0010 Oin%1on O\-i'-4--+M00000000\O
ON ON MON NI-NN'R7 MN Nc--; 6 0 1 M
MO 'o "t et et "t on Z N r4-4 CV pi od C, Vi c!
MO in t0 OONM -t et -4 l4 [" l- 00
eP 0
00 00 00 00 \O
"

vl vl rr 00 V1 V1 in \O
OO -4 qýt in tt It in an M in N in on in - 00 . -+
't q7' O\ in vn %D IO NO 00 ON Nt NMO GO O'. O in N
'L7 %0 t- NNNlNNM 00 00 . --- - . --ý ON IG
Vnv1NNON ON 000 O. 000000'O'V'qýtON N(14CI4
Q OOOOWAON O1a', N
(71 Ö6 c;
ZM C4C4M %6
V1 lý: N 06 -6[ýcy
11111111111t1111tt11111111111 . -r .--4 NNNNNNN "--ý

H . -r
ti
'0 -1NNNv1v1 NNNN'O%0Nr- %D 1.0 1.0 0'. i100
ooN'R7ýr NMN -- C'in
IRtN%0I'D.. O - in00. -1 00MMM-ON1N%. 0 11 ID in III NN
E ýMin'nin00 'n -- NQNMMMet
Sr v)vlv)hI n 0000NMMCG NNNN00
.
ýi cal kri 1! i 00 . -4 . -4 . --iMvi\0 10 \pl.: Cý000-4-44
NNNNNNNNNNMM1M1M1 -; 06
fr IT IT IT M1 MMN
IT
11111I1I111 111
ýIr 11111 11
O
O
Q% N Ný N
it 00 in '.0 0'. ON ßt '0 -N Rt 9'- '0 00 00 N ON in
in rN
qtt N 00 00 ON 'Rt NMMM v1 rf qtt et NO
ON C'. ON O' M tN NN
v1 NNNpe7 ý-. 00 NNNO
b 0 . '"hq M
ý q! %0
sQ, Q, Oý OM vikA V1 v11r
E ý["ý 400-1 NNNNNNNNMMMMMMM
F+ 1111111I11 111111111111tt11t11

Z ON ? in 00'0 n in vn'D0%on ehN-N-inn


N0000M t (:D 'RT N00-4C> -1-- MM
Ei M 00 00 ON NN -+ý C)-4gr m'i N- MNN'1' NM(' 00
t- ON O'. NM Q'. p'. ON OMNNNOhO0MOOOOM
"ti o ON tý ýtýIIt-ZÖMM
Q z ýo6OýCýO C; 44lä
6t Z6 M cli VlvlviN
Fý NNNNNNNNN
1t1111t11t111111t111t11111 MtMMiiM MM MMM

00
L1. [r of q VII in N' 1 [N N'D in ON
Ll. M O- NM mmN 10 'D on Itt NN'. 0 W) M
in ONN "-+ N W) W) oo %D O p'. M p\ - in
00 00 C\ O-M- MNN 00 00 N I'D N
p ^" --4 W) MMM r-+ NOOO tý 00 MM W1 --ý .--i -+ . -ý 00
ýC NNMMM
N
z0 f`l 46 V'1 k1i OÖ _; . --i .--i M V1 \G .16 116 %6 OÖ ppp
fir"-^' -N NNNNNNNNNMMMMMMMN pp
111t11111t1111tt11111111111

M O'. N '0 %0 -4 . -r MNN 00 Q\ MN NNN


yE O'. N 'C in v1 W) oo OO9! "-r . -r - an 00 00 mot M
O- -NNO of C\ NNN in 'O [N IN v1 OOO C' 00
.pC- NN I1' NOOO rt M qt NMNNN
MMMM [-MM%O,.,.; 'O M 10 %0 N ýt of qt M er
Q NO M
z 000? - MM 4\C; [-ý l-[-t.: 01 0 00"-ý"--ýý--ý,
rr -4 N V1 oÖ oÖ oÖ "ýQý
tt11111t11111ttt1tt11111t11tt NNNNNNN-

an
0EN In \O \G N in in N ýt M WI %D \D ý
4.0 .aN "i' MN- 00 N "t e! ' '-O N- "-r mot' N --4 MNN
00 MM O' 'ýt Mt -- 4m
in '0No0MIRT0oanWe VT 1.0 N 00 00 NN tN %D Q'.
w0 Q NMMMMMt-
[NO V1vy1N". 7MMen ýOmt-r-r-"t-t Rt Rt\0
-t-t\O. ý MMMt}"6 6 4-4
r. + I.: tý[zrý: 0ý0(=;
or; 06 06 00 .r . -; _ipi
1t1/11t11111 -1 ýrIt -4 ýi111i1J. ... --i ___NNNNNNN_
Is 1tt11111

E -. 4 ON 00 in NNNN 00 0\ 00 tm 'Rt ;tO\ C\ NNN t- ch


(U 00 vl 'O N Mt vl "-+ .rO [N 00 \O [* "--- -4
N Itt - MM -4 -4 . -w OO O\ on .4
O ý-+NM ýpOMM't NMM't000'. --- yn.
oE tt MMNNO
ýi N4 \6 %6%6 %6 06 000""ý. ip
cV fV N tV NN -i. -i.
1,11t11111NNN cV M tý1 MM Cý1 MMM

10
QN NNR: YMC'. 'sOv1M 0000 C'. 00Nen
'.D NNNv100 NN nNNmN 00 in
ö 10MNOetOO. -+00't MNNetQ'.
ON ONO\Q\Q\. 'sON\p C\ Q\ p'. '. p N
Oo00000t- --1-4. --ýMNNN\OO\0 %DýD00. retet`tT tt et'ý-4
r--1 06 0606. --i vi Iry1ej%6Cý606 O
111t1111111111111111111i11111 _NNNNNNMMMMMMMMMMMMM

on ON 00 00 '. p .-N N'- 00 tNN


o\OOOet 00 . -. MNM et i v1 '. 0 et in IRT N r-+
Wlt- in t qtt0 00 0.0 CI4(14 ýN
öEN oo 00 00 .--ý M \C 00 00 00 00 00 00 V) 00000'. NMN0ON
00 N t- m' IRT fit' V't 00 NNNMMMMN
ý+o000oOCýC;
ýJ'ý' V; cl; C; Q; N\, 6 \6 ý6 oööfV cV N
cd Oý Oý Oý Qý Oý fýl eT l[ tý oÖ oÖ oÖ oÖ oÖ
. "".. r r. r . -r .-r
E- ^"11t1t11t1NNNNNMMMC? MMMMMMMMMM

N
-00000CMMNNNane'tetON tNNNNMMMMM
ON OinI'0'00ý NNNNONOOOO\NNNMMvlv1v1'CNNM --ý- NN
Q 00 N 00 ON ON Q\ 00 '. O \0 \G - O'. O\ OONNN
Itr O\ OOO 00 t- tN OOOC.. O\ G\ ON O\
z
1t11111111INNNN tV cV NNNNMMMMc ý1 MMM

0EM- 0\ %0'O \.G -4 ".. - Ný c+1 00 an O\ 00


'ti NO It NNN v1 NNM O\ -N in MÖÖ0NNNN en
r-+ O' p'. O "-ý MO Q'. O
QO 00 "4MMM OO 'C '.
-4 "-+ NN 00 \D (+1 00 NN 00
ýi Mit N"t. NýT t- t- t- 00 M Rt R "RT -4 -4 -. 4 -4 .r
C; M MM'it NCV
il 00 00 00 .. r . -" . ,6 may cV C4
t1111 No 1t11111t11tt11NNNNNNNN --i . -.
4 . -4 -r 1.4 1.4 -4 NNN
. -4

180
Cid NNNNNNNN
00 00 00 00 NNN C7% C00
O- N'f N
1.4 vi MNNo0oOOR 00 00-N
ýD %4 ýG ýO NN ' H N
"-r
r . -.. ý . --ý "--4 00
"-"4 "-"; 0 0O O
.r.. 00 00

N
ýf N tN W's IO N_M 00 M
N O'
b --+-- 000
0tl- OÖNON
i.,
z o%oioýoývivi_; _; _vOýo

cm
H
- MN et ý--ý QNN o0 N o0
v1 00 V'1 NO OO Rr C*. as
00 NNN 00 N 00 00 N --ý vl N

o 06060606-1: 060606 "R


O
O
4)
P-4 s -it t'- M ON Nt O %O M- 4 vn 0 -,f
00
O
b 0 ONCý00IT
MNNN
oÖ .'C
oÖ 'T --'
"O"D'.T
MNN t- M
ºý MMMMNNN NNE

/fi r

"D - 00 cf v'1 00 'O et N NN '. O


M 00 Rt 0% V) Oý [l- NMMO
b MNNNN t-
10 . -. "'" ON p\'. p
o
r. Z N NN NV Ir-: ;
-4
CI R" N
Co en en en Cn NN NN Vi v ý"Nv r

IT --I a% 00 l'- M Oh . "r W's 'If


O 00 MO [N N
Ei IO M- OO kn m -4
a V 00 [ N N- '0
Ö
NN N 00 00 N O-, N V1
O E 00 00
NNNNNNý-+ý-+.
00 O 'ý 00 00 00 '
.
ti o -ývýV1I

N 00 - 00 t- Itt N ON ""r NW *1 0%
E tN -4 0 00 *ft 'D W.,
00 en -m
'Ly 'q'"qT M""-14NNNMN O
O ON Q'. Q'. Oi %.6 'p O o0 '.p
'
ca
Ici
I'O
00 NO'M '. O et 00. --rMp\-0'.
1-4 OE It 00 I IT c+1 '--ý 00 t- V'º v1 O O'.
0.
C.

O
4n 'T ý' t NMen en ON VIen
O'. Q'. O'. O'. %6 '0 -M
901 -4 .4 -4

ci'ýDv1
"'+MOoON - c lw cYMNO
GN 00 00 l- wwwwwm W)
0
'O
00 06 0606440606 06 M 00 00
F+ NNNNNN Vj yý
.-ý . -ý+t

10 IRt
lL OE Nvfv1N
ai .0 %D .--- O'. C% %0 Itt NO- %0 .-.
ND% W)C)C)MMMOOOOM
O O
0'. N V
Z MMMC+iC'. CV ý-- 09 00
cI1 MMM CV tV NN CV ý ýD S

M ýO 00 NMNMN O'. 00 . --1 N 'Rt


0N - iN 00ina V'. -4ý0-4

cd
0 MMMMMMNNN
knte;oo44.4 CP6
ON
ON -00 00

5 '1' 'C W*A NI*


4) N' T en _" MN
O O" RC 00 "tO 00 N "! 1 Mh
. C3
woo) v 1 týtýtýtýtýRtýOýt
O
Z vi Sf C'. 0'. Qý 06 MN
NNNNNNý. 1pýp

N -lam 'ýt . NN Nei ý0


öE N'O , v1 - t- v1 "" "-+ N
N ""4 . "r .-ý 00 00 00 00 00 00 It
et' ON

z ooooýýýý_ý
NN NN"-ý-rr ý--i.
. -1o
-ý.. ý,... ý-. ý

181
N
Mý 10 \10 ONqe ON 'T \. O\O-ý-+-[-RhI'DNNvlr- Nv1M-4"--1
ll. ll- l*- [- t- 00 [- 00 00 00 l- 00 V'b MMM it kil 00 00 N ON 00 00
"-+
tý tý wl kn F41 00 00 00 00 00 00 m ll-
aocV Ö- N4VNNNNN lZ
00 ~~ -+
ÖÖÖÖOPÖÖ
cal N --- r+ N wl v') v1 -n V'1 N O\ . .. r ... . .. r . -r rr 1" 1.4

N0000qýtý-+v1Mýo-4aýQ*, %lDNNo0eto0as v'l%'Ot--+00%0 %0


Wý 00 00 t- [- MN O ý,O -+--4v'1"0
00 %D %. M %0 00 O !t QN OO Oý, Oý, (ON O,% ýt .t
v1CD OQ% v-, Nl- OONOONl- l-- GNCýC, erM NOýoo00000en M
Cý NN (14 N- V') O, 00 O, ON t, 'o
_4 NNNMv; lz oÖ oÖ Cý ÖÖÖ66Ö

N
N lý %D %0 V*j v1 -t 00 V'º NN x-14 N tn O\ -- 00 ON W) NM It 00 00 00 00 [-
N-NOO IT N 00 00 00 00 . --4 n" I", %D \O ýO 00 %D ON O hO00 tl- 10
M NN N -. +
It 00 ýC 1.0 %0 ýO er NO %lO %D IO N ýt l- 00 ON tý l- OOOMM
- . -r . -. --WlM en en MN 4
r . -, "i vl wry \6 Cod Ö -4 fV 4 kli v'i tdi 4+
.: M 00 06 00 00 C\ -----
dQ . --i -NNNNNNNNNN

O
[- NN[- Wl NNMM[- kell intnr%IT MIT v10%v1Men IT Men
M %O %, C V1 W') 'O -4 M --4 -4 C) C) MO Cý ON ON MM V1 M TT -4 ON C) (D 4 '4D ýNO
1..
O%11Dý-OVII V) ON Mkn ýt ei'MMONNO Vlv14n -- 110-4 v1 VII MV C)C>0"N
ýp"-y"r"ý"-ý"Oen N-tit'T 0Nt.,: 4w;
týz [, oÖÖM: [ýoÖoÖoÖ lý:
10 ''" ^''''Mtý: clý cyý cyý CýNNNNNNNNN
I I
E

Z
0 10týt'-
ON 411 V'1It
t'- t- 00 t'-
'ttNw)"Tm
Wl 00 %D W*)
wt NOl-
MI- NýONON
ý.pýOMMN
W)
NVII
V1 v1ON
kt) ")
IfhqttMMM
-I - Ot--t-

00
P-q V1 1%O %0 h
MMen
n"'"'''ý^'M
V1 V) [l-
en ý.O'ýt
%D t%
"*t-
00 VO ttl
t- t-
"-+
O
00 ON ON O% M
w; t-: l-: 1-: 000"iI":
00
vi[-060606t-:
M 00 00 Cf 00 MMM 41 V1
I-:
b ''"^S ^I ~r"r"ý-I NNNNNNNNNNN

l-[-l-00%0-#Alt--4-+N%Z-1 rýnl-ýOv1%, Co000-4 e V1gtM


VI0000000'N %00CD 0Q>Ovi "r"
wnm00000000 (2%-4CD CD %,DtýNNM0>00
0
a
t- t- t- oomv)ee
00 O000
Ill-
mt-emO-tMMMvi
\iC -+ oo r4 NNn..
M V1 v1 V1 V1
[-OV'1V'iO:
[ý Ö
Oýo , 0ýc
""r "-ý ""y 'ýT Q ýf ýt MM
, äff

'""'""Ný0ý000ý0000^'- -+ý--ý. -rr+r+NNNNN NNNNN

-N ýO M c+1 v1 et mM" IZ eP v1 er ýt "-+ 00 Oý pý


OO(D -tt[-NV)NMMMMGýMMeý NnýtlýNtýýp MMMMM
O lt bn OO [- Nm G\ C\ 0% ONMOOO "-+ 00 Iý OOMO ýO ýO ýO ý-+ --ý
'et`e MMM%G 00NNN%0 et NNNNMv)
f4 -,C \"o %m 1 V'1 ZO \O t-Z l-: l-Z oýoiOKcTaýci
-4 %d OÖ .-r .--4 . -+ .-4 .- -i . -9 . -1 .r .-U . -a .-I . -S -I . -i

b
on M 00 I'O NN C\ tn M [- M t7 v1 00 00 00 ý0 00 --1 ý.p . -r v1 v1 M 00 M \O N
ooý%OW'l ---+ et O\httwl W*l hvl\ 110 V1v1v1v1%O'rtgh'Nt"14-
VII TtwlV1 W) ýo[-00it en MhOQ\t-t-t-l-gtMt-l-O\\1OMen en 0000
w° N0000 00 Q1 M Q\ G\ O\ -O0NCiNC
'i : l.: t': C; ÖÖÖ Cý Qý
,Dl,:

U
it -+ v1 v1kn ei'en -+MNN%0 00ý1'-4 ON ON MC\ O\ Mio-"-r-Nr-
-4-4 "4^4 -4 M00%ONý". 4 wl GýrýýoMen MM%Oen V100. -+-ý--ý. -ýen en
ýMd R7 et ON Mt ND 14) v1%. 0o0Ntllvly) WjIRtýo Rt It W) tn tet 0000
O 00 NMMM Cý O
%6 oÖ oÖ 06 oÖ ON444 ý6 116 %6
ý''ý''+ý--ý"-ýNNNNNNNNNNN

et ON NN v1 et ON ON ON Md en 1.0 %0 v1 00
--+ 00 1 ei 00 v) V1 W) v1 00 ,f
14 x14-14 --+ONm N NN[ý--4 Ittn -+. - (=en 1.0en t-1ONNNen tl-
Oh NNNN 4t v1 mot' kO ýD NO %-O N ON "O [*- tý tý 00 Mh 4n v) 00 OOO lý N
O ~~~ÖNMMMMý6
00 Mý6 0000006 NcliC
a 'ý 'n 'n 'n 'n "--ý --ý ý--ý ý--ý -+ ý --ý -+ NNNNNNNNNMMMMMM

Cd
\o
a) t
vi wl W) W*l It tl- 10 Ch 00 00 00 Irr ON 00 00 00 t- IT t- VI An -. 4 Cý ON ON (D %-D
r 00 00 00 00 00 v1 wl en MMM- 00 Vy [- t-
ce 10 00 %0 4t) tn t- t- tl- tl-
t- t- N V1 v) v1 t- ONO N
l-ý [.: t-: r-: Oý Net' ýi' d'ýOlýlýnoÖtý
E- ý' ISaaaaaiiia ^' " a' ^' ^I ýi a+ NNNNNNNNNI
a

nntýCh ýOON v1NNNI'O koON00t-M00NN-4


n lý n tý tl- -NOOO I'ONNNNM
Cl %D M -4 W) of 00 It Oh ON 00 N v1 N vl vl
00 00 -4 -4 .-4 "D 00 [ý cf 00 "--r . -r . -r nO
NcV OR 00 00 00 0 0ý00000c., -4 ý4 r+ ONNNN
it-: c; 4
cVd or; wike;
TnTTCýQý"'^'"" NNNNNNNNN

NOo - -r. -oo IT et It IT 000N-


It Nqt et lt ýtýNl%. l-ý ONn'4C\ 'D et
'ttwt et"ten ýr-4, -r-il*ý M v1 v1 v1 W) kDOv1`etvt"tON0(D 0t*ý
vy et et et et "T N- 00 00 00 00 N 00 00 NNN
C> MMMMil- v) vl et Rt et O\ O "-r "-r -
Ii'NNN N0o ýýý-4
ONMMMttýoÖppppQý
iitS ti t,: [I OÖ "-I . -r .-r . --ý . -r . -ý ý-r . -ý
.-ý . -, NNNNN

182
11c) wl W) Itt
C43 4n 411W)
ö ööövv
t- t-
10 10 00

N
5 r- n t- n- M 00
4)
OOOOn0

H 00 o0 00 00 N wl
ON O\ O\ O\ O\ <t 0
N i'r 00 00 00 00 M o0 V7

z ------- SS S

0
O ooOýOýCýýtOýM
00 O\ O\ O\ O M
00 %,
br ý ". 00 00 00 00 "--4 ON 10
-0 E 00 00 00 00 'n Wl
b p . - ..... -r . -r . -. ý
z 1.
", II SS

Z
0
E' tNr- NOrf en
'o
00
Iti p
NN NNQN
OýOýOýOýOt-:
C\QNýo
c":
z
9 " SSSS

00 Wn 1ý0 110 10 O
O v1
4) 5 MONO "O"Ot ýov l
-0 Ep - . -ý. -. r... r.. r. -,
z ""SS S

w0
tONeý
N NN C14 N
C4 oO
E
p ell; MMMIq 00

p0
10
z -4------
-0
p
E"r
MMmM
qtý01%
qtImt 00t-
00000000l,
MMMMNMM
a% %D t-
Q'l00

VII 00NN 0 0
o Ol-Ö oO
oont- [- Ov)M
E
p Ca Cý 0ý Cý -i tl t-.

z -----
10
v1et«TetýOr- m
'L7 [ N ýv1v1"-r'ýtN
cd '10 E 0 C. ON aN c7N - W.,
O 0 ?
Z to cV cV cV l-
NNNNý
cd
IO

e
M 0o en '!en
V1 -'r
1 V*j W)
(M --r
.10
p lý OOO ON 00 \O
cd Z Oý G; Qý \G a? 00

N M 00
"a E IOMen en NC, 4-4
p c"ýtýriririNN
Z N cV
I+

0E V ýA . -ý "--ý V1 M O,
"b \O "-r - -4 . -r O \C

z c>-,f 444*ý llý


N "-r "--- V--- "-"- -4 ý-r

183
N '0 'D 00 m 00 tý l- M 'It NM ýt N vn 00 MMN-- [- V1 (7) 00
Iýt [-1ýDkO"-+'tOOO4: rooMdooýc Ot--*- r- "RT W)v1Oo0NMkO
CJN 00 OOOM 00 00 00 -4 N l-- l- *N [- 4 M
O, (21 MÖ -4 _; -. MMMM Oý Q;
pp 'O l-Z r-: t-: t-: 00 06 oÖ oÖ ON4 v1 i1 ZO r, t- [, [N [- R:t It
I. V) hN 'T 'T eY l F- 00 O\ ON O\ ON -- -4 . -r 14 -4 -4 1-4 . -- .. r --1

kn N W*j ^4 N 1. M O% tý M et oo d' 00 110 00 00 N %D %0 - [- IO


OIttch0ON kn IT ON Itt00 00 00 00 00 ) mtct Tqt %1ONNNNMM-t
kn ý -4"-4tOw-+00ON "--ýN'ýOOOOOo0ý4OMýo"Oý, OQNQýCýON C;N rý[--
MNNN Oý NMM 00 MM p666 od od
"i od 06
(V f`1 (V M [-: C-ý [-: 06 CI "--i -r . -, . --i . -.. -, ---- "-r NNNNN.. ý . -i

y V'iNNN'!! O-4W*i
%.
"-+0000. --4Nt W*b v1 V1 ON%. O1000000IN NNNNNvnvn
a) C) "-ý -4 "-r 00 ONNMM -4 00 -4 . --i -4 . -4 00 [, t- [- ll- M kn V1 V1 V1 V1 'T tt
. -r .-r -4 [- It ýt '' tt 00 00 01% 00 00 00 00 v1 [- l- MMOOOOO 'It qt
[- t- t- wl ct ON ON ON "O
oÖ Ö
O MN v1 vi N
NNN'f -; w; CV cV M W; v1 wi v1 "1 (i
..oA lz 0000000; "I - .--ir-+. -r..,. - -NNNNNNN NNNN

0
MOýQý00NMMMlS, t"-+v)O\ 'eh d' M'v1et vlä v1NN
a) Mý-+OOýMe %Ov1NNo0Nn} et -t00etv) NN[--+N. NNet.
O--4 --40r-Ch-4 -r0ý0CD ýýOýC%O%, pM00(D 00000000 -e -+
vi 000000ý. OOO, Q\O,
a) OMvilzlýCz[ioöOMv1riOoö0ö0ö0ö0övi r;
-o -U+NCU1fVe oÖ U+ . -I NNNNNCV NfVfVN
IUUSU UIIU IU
E

t-[-t-etO%oc00 hqtt11000I'D %Dk1ýON


In Nv1 W) qt V1ýt 00
NMMM%0mMMNNtý- .--4O M%0-ten etN00CA "-4N"--HMO
tel MMMM W) NNN00N[l- 000O%DONMen NNMMMMeFIT
OOOW, NOOOOM
Ici 06 OÖ 06 OÖ 06 kt1 ý6 OÖ oÖ oÖ o0 oÖ tr1 kt i
--ý
Cfi cli kri .4M
tti

iy
Fr O\ [- [- %ýDMMM 10 V1 " et %D kel \o f4 M lt mm 00 lt 00 00 l- 00 N (%l
- v) eNrMQhl- \o vi IOleýnNo0l-00C-000%0000%DN"O"Ov1%Moý00
Qr MNý--ýOýv1rývletMONM O\CD O\G\Ch
e gtxt 'O'. GDýoOýoMO
Qr h v' tt vl 00 1,1,1, - M N
-. V1 V1 OÖ 01 ""r -4 'ýt ý! ýt et ýt c'V N
NNNM CS CS GI Cý"-+. -I . r. -n... i,. rr. -iNNNNNNNNNN
8 I
4r

e --, 14DN 10 [- --, \O N-M - \O IO ID 'V' Q%-


vy 00 M-Ch
NOvb. -+OO\t--. -ýNOt- . -ýOOc> 0- tehQ>Ml-[-r-[ý-4(14
le 00t-l-N"Olit KA bi00t*- ge - MMV) N min v1Oheeeee e
00wwle Nw=mCD t-0CV NNN_Ii'výtýrz cO c06
N V1 ýC \O t-czcýcýO -i
r
. -ý .. .--ý "--ý rC N 00 - .-ý . -+ .--i ----- .-r ----- .--i --

b Nvl V1\00000 le kn v1NN"- - - gyp - V-4<2%Nýtl-


00W0% -r V1 M
. v1ýoOOmt O\vlt-000Mv)p--4OOootZO- - N[- N
%M MMNO 00 M et et c) --+ (D I- OO (D N o1 Mm ýO M v) v) v1 vi P--4 et
MNNNýv%0ee ýe-t'Om-;
Nv1Nl rVMMMMt000ÖoÖÖÖÖÖÖaýoÖ
wo CVNCV l Oý---------. -ýý--ifVcV tV NN"-r "--ý

I IkrbMONviiO00 MMNýC
s000ON'tt etÖ-Ö000ON 00 0000 OK
Oý-O1GvIItlO% 000NON-. 4 -+NMl'ý llý tý0000[ýON(2N(n MV W)knk W) ýq
t- ncVmo't' ÖcV 44viP-: nt-z
06 00060606 l-z l-ý lei
IIIIII '''''''''^'^I ýý+ý-+NNNNNNNNNNNN
III

'C3 NO\O\C 0000cs0000 n N'0-N'000mt- '0 \OM-W)%D%DNON 00


N -4 OOM ýY M ON v1 \.D It Ný4M CN It Q\ M Cý ['ý 00 M et maß'111 n
ed -N OMNOOOetoONo0000000Nen 00NNNNNen
ý000r-4 4\6 NNNNNCý
ýýý -4"4-i_; _;
'n ý+ IIII Cýl CV cV N CV CV N. NNNMMMMMMCI
IIII

ON v1 I1' "? ' vl eF NNMM


d 00 00 \O [ý 00 Itt V1 ON ON O

l- l-
v*J N
00
--I
l-
NM
00 O\
qRt -4
G\
vý \,p 00 v1 of I'D 00 00 \0
0ý [- "-r -4 00 Q\ Rt mMM ýt ý,O ý,O O
M oO ON t- 0- l-
0 000000M00ýýpcV
. --4 et -4 00 00 00 00 . -r N "-r .rM t- NNNNN et
ýflýtýtý[ýtýÖCV V'ivi V1týoö0ö0ö0ö0ö Vh
ý+criMMMl Qý ýr-I .ý -r. cV NNNNcVNtV
cii I
. ""i, --i ra-I NNN
F

C, ý 00 00 00 00 0, ý NM %O nV [- nn as rý
Oen hIttgTo%<D ý4en ON Oý NNM
r- qt t'-
nWn10 N1ý0 Oeh'T V'1t-t-MNMt-o D'o[, - 0000
ýt l- kO ýo CN - oo oo ON MM v) NNNN__
ON cV N 00 O 00 00 00 O v1 v1 v'1 v1 O
-ýh d; "t 6 Ö
NMMMM6 .; cri Vi6D ý6 N, oö cV tV fV vi v; vi vi vi vi M
Qý Cý Ca ". ,. ,. ,. .s... ýy fV N fV NNNN CV NN
s a s i

lt 0% O\ . -r 00 l", O\ cl 0% ON v1 l- l- "-4 Q\
.-r . -r 000" r- %00
v1 00 -4 mot N v1 M l- -N
v1 0lý h 00 10 %ID -oo Z0 oý -e e [- O O n 00 00 O Q\
0%(D cýcýcýOqt NMMMVN t-rtle
r- M\O%. O\Gvi ýe eON v)Q%O%O>I7- NNNNM""'
M000000NO __r-i-. cK
NNNN iiI.. NMMMMý7 ýD odooodc;
VI lI [I ,. ,. ,. ý,. NNNNNCI
I .I,. .I,.. r" I
I S I I

184
lý Oý Cý lý tý NN ý--+ OMi M

NR 0ý

%0 10 [- t- N v, ) v1 ON O\ 00
Ißt IT 1.0 %D l- OO ON MMM- ON
NNetqT qT 11, ItON ON ON v1N
00 00 vi vi V) Vi NÖÖC;

GM %M l- r- 00 c --9 NMM v1 v)
Vli 1.0% 0r -0- wi %D %M 00 l-
,Kt m'v1erl v1%0 %IONNNýYM
t+
rr" NN oÖ o0 00 00 00 MMMOO
.
NNIIIIIiIINN

O
-mmItt\O%. OCS 000\ It C14
en l)M<tI'O1. O00 [- 00 -%ýO
{. n
dQ "-1-1l-l-nt-t, l-l-00ýOO
W) OOOC; (6 44 0ý (0ý
et
NNNNN N- r-+ r- 4
E
1

Z 00 Ir- t- 00 Ch t-- 00 00 mt
0
00
oMNN00
o00
rn-t-
0----
00 00 NN
N
V1 kei .i -i-i -; .4 Vi
10 NNNNNNI -I "-+ ý-+ Cý CS

00[- [-00-4N8 `ßt-4


0
clq
NN
i -iriNNMM(D
OÖ OÖ OÖ OÖ OÖ MM
009

N c+1 v1 v'l "G \iO t, - NNM 00 00


"O vl vý vý vý MMM V) 0
l-z1.1: ýtýtOÖO'NO
,. 4 .r- . --4 44 - 44 -4 Z- -4 . --i

10
on NN"t Mkn M[- NON -+
Itt It -N ct "t Nt wl MO
,it It MMMMM 000 - 1.0

y
OOOÖw; vi vi 41 kii -r 14. -iM
tlý

r. +

N
br It NM `ý 00 00 M qt 110 tt Iei
NNtMV I-t-[-o0C)
NN'n'n V1tnW) ltd W) en
0
vi
CV CV CV CV NNN ý+ l vi

N- M kn 1ý0 1O kn MM O1%kt)
n7 IT %0 M til tl- 00 00 00 h %n O kn
tu MMOOOOOet -- C14 -+
O Cý Cý v1 V1 V V1 Cý Cý Cý M~
NN (V NNNNýýý O
fl ýý
.

ON (14 M W) \O l- t- \O %.
O n-
M VI Itt in t- 00 C\ - 00 00 Ö
ItT ? ----- GNNÖ
cd
E-

-. 4 " --+ 1C O\ [- 00 ty
M IT et W) wl W) N V) Vn
ý"yýMMMMMý--ý0000 et
mM Cý cN Cý Oý aý 00 't 't M 00

Q> M "--4 --4 0% tt tý e c*, 00 oc


ONOO - NMtýN -oo
NN000OOOýlO, O\o-

185
b
ce P" -a -M \O %n v'1 00 \D M QK O\ It N NO 00 00 v)
QN0\-0ON 00 00 00 N
0 ý[ý1'- ý"-- t-\0\0M00o00ý': rýtNN'tt tývlVlvlt \ýo-4
ýýW) tn MýfO MMOO V) N-,
a ýý"'4
vi 06 ON ýt
00
Gý OOOONMMM
W) 0 ChOý[. NMMtý:
et ND l- t- 'O
t.:
\O "-+
týý0
--
§-- . --4 MV In N Vl V1 tý 00 -q _4 . -. - -i .-,.. i . -- _. q rr . -- _- _I ."- -r

--+ 00 00 0 Qý N %D %V vi vli lt et 'KT [- nt [- 00 0% et m ýC "-+ .ýNV, 1 IO N


O ^ý -+ N ýO %N OOONM 140 &A
'L7 E --4 of "-- MMM lý 1,0 4,1 et M
^y^+--4NMtýV)v)v1N%0 vlý 000M00 MOOI- NKA Vl OONNN
O 'ýt eQm 00 -e ýt N OO
0 . --ý . -i . -ý fvj Vj \O l-:
11848481
[z l-Z CÖ
$eit
,.. ý
II
MMMMe
. -I r-+ -I .-I
d r'i
.Ir. r
tz.:
I
Oý 000ÖV;
r+ . -4 NNNN "--i
V1 ei
. -ý

rA
ý,cr- r- I'-
a) -rNNIOt-
GNW'bk1[-C%l10R'v1tnNan
V1o00 "N
1ý0 C>
; -4ONýt t Ol GNl- [*- NNnv1 V'1
QN[- NN o0IýI'Oýt--+-4 v1en l-kn vn
ý--ýOOO - et[-0000NNNNr-+.
kn OO l- [- M fn M
--4erM 1-4 OO[-', OMMOl- týtý
4
oÖ Ö 406
an
0 cV NM. 1CoÖ
I
QiOI. -I
fV
. -I r.,...,,
I
.I1.4-
"-i . -i . -r M
fV fVNNNNN
V1 Vi
NN-4
06
I.

-I+
0
a) 0
a) q1t Ch
MM
00 ""^' - lr(2% N 00NON t-1O- (14 qtt mN00mN
W) 00 \O W) It 'fit M 00 Ih 'tt --+ ýt M \O M [, ý 10 \O 00 [-
öA "r v1 \O 00 O ON [-
V1M tt N V1 \C \0 \0 4t t- OMMM V1 Vl \O l- [- of "RT It IT IF N
a)
'v -a
0 E N V;
C;
-i --i. -itVýt[-: 01QKc"i44wit%: 0ý0ý0ý06c1i
-"
NC4
-14
ý; °O ' ;' "I' "; ý+ cV NN
E
z III
^I ^I cV cV cV cV cV cV cV NN cV

Z "-+ t- l- C t- "-+ ýO ýO ýO o0 0o cV 00 M Oý o0 00 0o v1 C14 N Oh Th ON 00


0 C'
00
[-
et eF

O, QNMN1,6C;
oO v1
00 v1
OMMM
r-+ 1O %D 1O
C;
oo O
MOM
4"iý6
[- mT O
v)
[-
et
ýO
vn
N 10
1ý0 I'D
MM 00
ON ON It
MNN
vl [- [*-
oo ONO
1ý0 OMMN
ON

"C1 -0
0 E 1; '; t; "I'-4-. 4-4-
06 06 06 06 6NMM41ý606O60606
I I+ VIII NNNN NCV
-i. -i-i
NNtI
III
9
ti
z
\O 'O [l- 00 - ON 00 00 00
C1.
00 r-+ 110
týON
oo ýt rh
00 NNN
N --+
--
-
M Vy 1-h If ý. -
O VI 1Cý-Go00000\0kn
O\ Tt 00 ON \C V1 er 00 t- vl
O\ as N I'D ON 00 00 [ý et t'- N 00 00 %D v1
N NOQ\ON
eh vl en -4
0\O\M'ttIt et
bE NNO \.Dvn0ON "-4"-r-. p,;
4 %. M VittiIti
V'it.: OýÖÖ_4 CV .444 4odo oo

0
Oi CI ... -r .-r . ... r NNNNN (V NN .-I ý-I . -+

N n lý
NW
C)
[-
-b
-
Mm
ONlZ
-
00
r-
-
MMM
-+"-NNeiONNooo0ONMI'-
1.0 I'00ý0-ýOW) etetmegNN-i'ooet")en
N00'O
1
%D \O: 3
ýý
'T

0 0 v'1 \O so \D C\
wl el oq -1
M-ý MMr
.! -1 :O-M
4. ý-r ý-r
oo %O ýp
OO
-MMM

lu
Z NNNN C\ . , i .
V1 V1 V1

on t*-
N--ýO., M(14
0 ýONo0O\O"--+.
t- W) 0O11DvlvnNrl-00
0\
-rCD
v)Itt't
NtV100et"--ý-4.
ChW14"\
eFvi
-+en
%D1100ýet
[l-
ýtQ\ON
W')
MIýtOht-
en Vlv1
At) 1-N000
MIt
kt*i v1t-\DV'tet
V1o00000
--ýM

w0
r. +
bE
0
W)
N fV NM
O\ NM
\ýO
O\ 000Mp
lI Gý Cý Cý ..
cV of 44J:
I
ý-I .- . -r ..
vi [ý: oö
ý .I
oö 06 Ö CV CV CV ýýý
.-I "-r .NNN
ýD
(V ""r .--i . -r
I...

O
V)
OE
ON tf-4
Moo.
N oo NN
--ý
ßv1"-rNNoOMm NýOýOv10ovýetet-+.
ýtýCO-+--ýtýMMMOýOýooOýv1NNOýoONNNNOOnv1
\C -+ ýo \D ýD [ý 00 00 00 MMM
-+mm N --i Qý

V
0 ON'Itc+jVi
o000cod
00 ýD "--ý "-r .r
0ý"-ien en M t%z
pý NNN
n[ýl-ý
ýo ON O\ ON

z VT ýCOOO

r-ki11OON t-10- - ett-k0 0%0%oOen


C\tn . -, W) 0000 ON ON ýten N0N %a
%0-MQ\[-
O -0 E 00Mtn - \0M000\0
r NNNNM NN"-rW) [N O\ C\0 -- CD C\00-a\Q\Ir
4 0 tr-Oý-+\00\[-l-I-lam
too
--q ýT 00 W) tn v1--4
fý cf"d [,: ntýQýf4
-- CNC\CD
NMVll.:
- 00nt'- -! ýt`d'ýt
ntýnÖCýÖ
N cV N. fV (N cV cV NNN
cd

00 00 It vl V1 00 MMMNMMMM
"t C\N\Oer ý10 IT NN ON N 00 If to M
ON 0\ 0-1 ON qt:lO\eho000000\1-Mkn
'n 'n NOOOOM tol 110 'd x/100 kn M
v1 0\ NNNNNNNN
O ONýClDpfV '-+ C\ O\ O\ NNNN
fV CV N4\6060ý0ýCý
E2
Z MMMM
00 .-I .- -4 .a. ,. '. - ,. NNN
ttiwiwit-:
(V NNN
Gcrý0ý0ýcV
CV NN fV
fV cV
NN
a"I,. I a i,. a

00 G\ -- O% %MO> O\ 'et et-tMNOýN"-r. \Cvi


-0 E MNN%MV)r-00
M V) %C1DcýooO: kn -4NCD
ONNNNMNN--+-tNOOO\O--
C)c vkn0MNN--4
-4NtA
r`
00Nh
00 le 00rýv'1
-, O 000000
Z "-I
NNI [ý
I IIII "-ý .r . -r I . -.. r . -S -. cV NNN fV NNN "-I "-I .-S

NE %D 'n \O if Q\ Q> oý oý \o m
'Cy E 00 00 yr e 00 W" n 00 et -t G% M 00 ei" M
OO
\p . -r o0 00 00 00 Vr e 00 O\ 17" "-+ \O Q\
.--1 N 00 10 #A vl vl vn oo vi ANNO-4 OOd'N -+Oh
NO 0 00
0 OO 'ýt et r4 Oh d' 't -t et 00
MMýi
výý--+NNvn00o000Met-t
06 p6ÖC; W; W; vi

186
10
Ö4 ý let 00
Cýý výÖnnn M ý r
v iv i
a "o cl cl cl - 1 In V1 Va t- W) ch

-NNN fn 4
C,
t) NNN
eq IC) IT 00

N
-4
ý+ W) N to tý NC --q m NO % rn W)
NS . "T N0 10 wl [- 01 et ON
'b e N 10 %0 '.D It' N W", M IT O 00
0 ItetýtýtetM
ºý vi000[l:
. -r . -. 4 1-4 -4 t? 1" 1.4 . -i
ii

ro)
U Ey -N'. O N [N --+ O\ M 00 \0
-4 t- IT en 't Ch Ohen [": r
w+ "a N ""r "-" - 00 NNNN 00
. -r

aQ
.. r z 00MMM0ý'0'0ýQ\O00
0
U
U O lý\Ovy'7Nlýl- v)en Q\00
it MO O\ 00 e! IT `: t' O " -+ M-
^4 O\0000 Cl; W) vnr+ONCD
Qi -0 E
F NI "--i .-ý .. r .-ý tI to to MMM
E
+ SIa

z
0
IT
N b
N
\C
tý-Q\'.
N00
N 00
00c'
N

0\(: >00
00 00 00 - \C W)

M
o h NNNNO.. 00 OO
"0 r-+ et ýt
O
H
00
" qt N 00 N 10
C]. -4 00 MN
E t-W*l OO\Q\0oO1ýOv1000
0 M' f0 00 l0N '( V '10N
Z OO N-+MNNIýDvler
-4 . -1 -4

wO N an NMNNýt'. 0 NN
týý' -
et M O Oý0N0 ýM
ýy
z viNÖÖÖv ýhhoo r yr
ý}

10
ao N'0'Oan
-0 '0ÖI'OtON N
o 00-MM M O' 0 V1N\p V1

N '.D 00 tN ' W) N ýY as
MN \0 V lM 0'4 '. O MON N O\
an Vy If) -, t '0 rn \0 M0
Z pp V1 MM
,tl vj ,*000
NN "-r 7, OÖ OÖ h1
1

10
'O --4 00 00 W-1 0\ Nv ß "t
ý N 00 NMO .i ')
Cd 1Pt IT
O 0 ý--ý "-ý . -r .--ý MO 00 et M
C; C;

0.4 1
Cd
n
ýr MNQýanN'o V1"-r
N E"
aý 0E NN "rOýOMM--ýGýtýd
V'nan'c7N00'GrnO'. tG
cV" NCB Q! Cý\G\ýC
(V N7 "-ý " ýr 91 lý
E- [ý nMM

E
V IT
T JE 0\ 00
O
"'ten
00
anMO'Oen N0000
00 MMM Ir- 1.0 Vy et to Cy
z MMM ýi. 00 00
00 00
NN

0E an NMW! 1 00 O, N an N
00 tN
E 00 NNvlMONO\-'%DV'1
0 MM [- [- t- "-r O'. 00 00
. -i an
z hVi000oOv? d'???

187
Cý kn 1-4 C, N Cq C4 C4 tn 47N C7N ON M
--4 Vn 10 t- t- W) %D 1.4 -4 M P4 Cq
oo 00 IT -4 00 C) CD C) M 00 t- 00 00 (3N ki-I W) en W) Q
C> ON 00 N -4 Nt tn tn ,t C4 eq
(D t- r- t- C'4 "D 'Rt
OR*ý Cý 06 ("i wi kli wi \6 Cý 06 (D 6 6
C, C"i (=; C;
Ci Ci -ý 06 Ci Ci V; -; c; c; 6
IT -4 cf) kn tn (5
kn (0 N4
kn \0 00 CN .4 'T in Zo t- t- r- r- t- t- t-
I It -4 -4 -4 -. 4 -. 4 -. 4 --4

fj
%D m-- r- aN mm C-4 C4 Ch '4D 00 oN
E 14) 00 p CD 00 t- t, 00 m 00 C-4 r- m --4 (7ý r- 00
00 -4 f, -4 00 \0 in "D ;s en -, o CP,
0 00 C7\ -. 4 m 00 00 . 1, rn -4 C>
Rl* 00 t- fn \0 tri en M t- t- M t- t- r- \0 V. I W) w

t--: 06 C-iC-i C-i C-ivi 4 \,6 \.6 %6 Cý


en
ti
00 tn ýt \C) as t- %ýo m 'o en cn 00 t-
G CP% W'l I! t t- wl Rt G '. 0 ('4 cp\ (=ý a-,
C) 'D fn \0 mt t- tr c) 00 - ý- C14 ON W) qt
.0e-- Itt t- (ON 00 C\ 0\ t- ý,b \.o C-4 -. 4 -4 t- t-
1: ýo "tt qT q* ýt Iýt eq C> Q C)
. 00 (0 IC ý.o ý.o Ci Ci j:
aQ . Wi 'r; ti \ý6 06 6 C; C; .4........
en Iýt ýt IT Iýt 'Tr mt, IRt
Ci C-icti \6 Cý(:)ý(:)ýclý- --4-.1-.4'. -0--4-. CýCýCýCýCýCilCýCýC'ýCýC;
l
O
Qi
Q% tl-
-4 kt) \0 m t- r- lko m" W) 00 en 00 W11 IRt kn t- tl- kn 00 ON kn
- --4 00 C14
G Ei C\ W)
v" --4C) m \10 r- (D 00 0\r- 47\
-4 (7\ C) 00
ýt 4n 00 00 rý RT\0 00 %4D
-4 V*j 0a W) 'Rt
00 m 'Rt 0 tý t-- eq cn Ft 'tM (7ý RTkn
(7\ m tý r4 -4 CNtý%
ýo CII
10 w, in oo \b C-i %eit%:06 06 06 - 00 00 00 00 00
zIIIIII- Cý Cý Cý 1ý Cý
t- 00 00
Cý Cý Cý Cý Cý Cý 1ý1 Cý
-1
E 11
C\ "t -t -t 0\ -t p t- zr - \0 en m 00 \ýo :r- V-b \.D t- \p r- CYNen (N -4 --4 r- 'T "
m r4 C) ýo C> CM C) ON CD C'4 " C> 00 t- C-4 (7\ C4 ýo t- ý,o kn :sm
oo t"- -4 tl- 11, Q0Q 00 W-1(7ý Ch Cý 00 C-4 qt \0 -IT
N W.) (D ") \.o \0 \o "" C14
cn -4: -1: Cý "6 r.: rýz t.,.: 6 6 ý
0
zIIIIIIIII.
C4 M t- C\ ti
-. 4 -4 -4 -4 -4
06
-. 4 IT r4
Cý M
Cý cc;
1ý1 J:
'ý \.
Cý t.:
Cý rý:
Cý t-.
Cý r-Z
Cý r.:
1ý1 rý:
Cý r.:

CA
O 00 I'D -14 Mtl- t- "\0 W-1 " M-4 W-j g
! i» '-D C\ t"- - \0 kn M- W') C4 "t Mt ýo
C1400 qt
1-4
en
C) (2\
t- C\
00 W) a ") C'4
C\ 00 00
qt ýi
1-4
.0E "t -" C\ o
ý, <D 0\ \0 Iýr C'4 C> M "T 41. q:t -tt "D V) tf, W) W) t- 00 Q : C\ t-
oo C'4 rn It 1,
c; 6 4
-4 t- t-
Cý wi wi t-Z cx-. C; (6 C'i
zIIaIII191 -4 .1 C'i C;
C7\ G\ -4 --4 -4 -4 .4 -4 Cý Cý Cý Cý Cý Cý Cý Cý Cý Cý Cý

en \. O C\ 0\
E -14 C14 00 \10 W) W) t- t- en -t
99 \0 r- M 00 en \0 Ch %10 en
-tl(L) E t-
Cý4 00 C> t-rn (14 C) en
eq -0 %0 t- C., 00 t-ý
r- I'D t- r- 0\ - en 'tt
C) (3\ tn cn
C;N c7N ýo
(: ) \, D W)
C'I -4 -4
IT
N W) t- t-
13N (=) tl- V)
(D (=) 00 W)
t-ý m eq
t-ý t-ý
tn 6 C-i C-iC4C-icc;4 %6r-: t-ý t-ý 06 C; C; C; Cý 0ý 0ý
cli ke; t-:
4111aIIaIIIý
06 06 06 06
V-4 W-4 -4 -4 ccc -64 Cý

ýo kn 00 \-D e4 00 m C-4 0\ t- -. 4 C-4 t-% C\ fn - C14 ON 00 8 ýt t, g t,


wl
m Nom-It 00
C-4
C> WI)
C4 oo
m-(14 C>
00 IRT t- C) (5 ON
"t C4 S- ON 00
C\ C\
00 .ON C\
IT 00 C>C\
Q C> 00 QC,4 t-p C\
G\ W) "t m C)
ON Go t- ýo %0 ýo
wO 0 00.00."ll. 171.Q. ID. ID. "I. c;
W) \-D 00 00 00 00 eq cli cli cli li \6 \6 -4týz-406 0ý
-. 4 Cý
-4 Cý
-4 Cý 42ý
-4 -4 0ý4 Cý

aU
- 0\ vt C7\ "
El
F
ID m C) fn
en -Ft
t- kn
C-4
cn
00
C7\ -t -4
00
el
"t
W) "
-4 t-
-4 tl-
00 \0
00 wl
eq
00 eq
00 \0 \0
" 10 t- Q\ V)
ON 0 \0 t-
00 00
1
r4
O o
ON \0
17, "o "D
\0 0 (ý C\ C14 %I) %D 0\ C\ G\ 00 \0 xn r% t- t- t- '. ". -. 00 00 00
C; Cý 4
C,6
C'i -ý Cý \6 t%: t-z tý: tý: 0ý -; C4 C4 C'i '": \C; , C; \,6 \ý6 wi W; W;
....... 4 -4 V-4 -4 -4 C4 C4
166111 Bill

b \0 00 t- \0 vt t- - %,o V) t-. 00 eq W) eq \0 -4 ýý t-ý t-ý fn ON Oý 0\ W) Cý


\0 -
4 43 E
, I* t- en C) 00 %ýo-4
C04N kn r-4 I'D \0 eq 00 00 00 en Ch \o
cd -4 -4 C4 00 CZ4 C)N C7ý 00 00 0 W) tn 4n kn ýo W*i kn C,4 - "-4
O zo \C! r4-1wl wl 00 -,T 6 c; C; ci Cý ý6 ý6
15tIIIIIIIII '.., -4 V-4 -4 W; .4 V-4 P-4
1.4 "" ý6 \6
".. 4 06
ý4 C'll C'i Cý
-; eý C'j (i ý,6 %6ý6 ý6 \,6 \,
6 ý6
Cý C14

ev) 't 00 C\ -4 t, - %D ('4 ý4 'T tn (4 r- " t-


In 00 t, tý (7" "ýt " t" Ir, eq
wl "" C\ C14 %D " 0%
00 "t 00 M C\ \0 MM Q) W) tý
(ýl m \0 -4 en -4 -14 V-4
C) Rt Rr 40 rn -4 -4 -4 Cl C\ Q t- t- t- ýo t- t- t'. t- cn en C, 4

-6 CiC4C-iC-4
W;tý:0ýcýCý
0ýc; Ci 44
F-+ zII. 06- -1:ý6060606060606
eq cqCýCýCýCýCýCý06
CIA
0E" " ýo
ýo
W,
IT
't
-4
0\
-4
eq
lrý M
00
cq \. o 00 t- en "T C'4 (: \ 00 %D o
ý, mr %o eq
'Tr ýt -4 ý,o t. ý cq eq C> kn
C) t- W) el tn (b -. 4 tý.
o oo 00 V) in tn qn C> cs tn -I- I- c7NC) 00r, t-. %D
z Ci 'ei "ri 06 c; fq ('4 C.1:
0? Cý
eq C-i
cq 4 eq cq cq Cý

ON ON m%0
%D
eq ýo
00
eq
C4
t-
10
IRt
kn
t-
t-
t-
-4 fn rý ,m -4 t- c;,, "I en
t- 00
Z"4
m
\-D W)
\. O t-
It 'tM aN
cr\
t-
t- kn
wl en S C> I'D \0
C> ID \o \. D N-0 r- m 00 00 00 t- 00
eq r- r, r- t, 00 00 wl W) kn
'T
Cj Cj Cj
C'ý"i 6 \,6 "6 06 CýCý 0ý :3ý0ý 0ý (:)ý
Cj
ý
t-: ti ti t-z C:

188
Oý ýO'tt ýt lý 00 r- N "--ý "-r "--ý l- l- NN
ON eNN\Oo0ýo"-4"-r-rM l'- l'- t-[-0%O\%0v1
cý OK aý oý OK cý GK ÖÖON %0 110 110 t- t- 0% M v1
ýOvýV)vb 000 .--ý ýO v) v; V1N. t

V1 MMM Nt -1+ N l, - ýo 00 kn . -4 M V1 V1 00 t-
oMMMO ON ON 00 [- t el I'D MO- e* ON et et et "t
It 4n v1 in t- \-O 10 1%G %D %-O vla 4: t mN ý10 %0 %D %0 MN
Cý 06 OÖ 06 MMMMM clq oq oq t- t- t- t- CW)\
N W'l
"", - "-4 --4 ""- -"1 --4 r", oÖ 0? op 00 oÖ fV CV cý1

Q% V'1 an t7 O t*- MMM N- N


a) NNNO\ON\C'O
r+ . --q
VINN-4 ON r- V11OV'1Itt1'- W)
O 00 00 00 00 00 00 00 00 OOO 0% 0% 00 ON 00 [- et O

oU 4 cNi(-i cli 116\6 \6 16 %6 C; C; MMMM Oý Oý


NNNN... ý . --i ý--ý . -ý . -ý . --ý . --i . -r .--ý "-r
*42
0
a) 0% NN 00 t-
a) 4n NN
-4 00
"*MN
1D W) 00 00
OOON
[- ýo et %0 C% 0% 00 [- M
--+et 00%0NMM-ýO00
b0
a) 06 1ý6 %6 .16 OOOOOMMMM Mý OOOO ON
b cV NcVNCV NNNfVýýý-ý+ý V1 Vi V1 ViNý
E

Z
0 I'D Wlbon --ý"-ý"--I
No00000N(4C14
[-: w; vi vi Gý Ci Cý
ll- Il- OOl% O*A 00v1Nq!
-- VII 'tIRter
Cý NNNN
Wt I7tl-
fV
tIt It t'ý
[-Iýen
o0 oo 00 -4
ON
1C
M
10 NNNNMIM t":

9
y
t2
O Qý M d'
N t- Vi \O MN rq 2 -- C% 00 -+
Q O"NM OQh(D C% Ov)NNMM"Olt
C14 l- v1 vý VI \O ýo \O ýo v) 00 00 l- [- l- v1 -NN \O O
rVN N\ OÖÖÖÖONNNMIO
NNNN . -+ .S r-+ -+ . -i . -r .-+ .-I .I . -S V1MMM

Cý 00 C\ 1O M Qý IND NNM sn G\ as ON It
et ýt M ON N--O 00 00 ON O ýO N v1 v1 v) -14 O\
äßn ['- OON CýON ON MMNNr+Oet -it qt ON kn

Ici
ao
00týM000N 0r- eýO - P-- mt v)NN
MMNO[- t- eln ef Met OON '0'0-- 0
w0 vi \C e IZ h 00 00 00 00
CM9OOOOM
O\ 00 00 00 1M 00 00 00 ýO O
cz pp pp pp eti fl; 00 00 00 t'

M 'ýtN ON . -+ 00 -4 kn MN nn r- Ch
N ON O 00 00 00 t- Rt 00 t- d' Tm
'kr)
O o0 %D n ýo 00 ON ON C% ON NNNNNM \O \C ýo N ON
N ýf [1 'tt N 00 06 cli cli
C-i
NNNNN ý--ý "-ý ý--ý - r-r ý--ý . -ý .-r . -ý - V1 v1 V7 V1 M

'C3
00. -r. -4 00%D 00 00 NO Wt%nv1v). -+-4 vl"-4N"-"
lC to NN 00 IT et 'IT NO l- n l- MM %lG 00 (- 00 [l-
cd
O tt 1t4.4 00060606.4. --i. -i-4 14 -4
00 00O000
NNNNNý-14"14114 .-+. --i . -r . +NNNNO

. cý
00 C1,00 00 nt-knmNQNOOttIT
NO -+men en MOý
ON ON vl tn en - %0 MN 00 00 4n n e- t- r-
O1% Q% 00 et
00 MM to en t- t- t- \O %0 %0 00 00
tu N6 \.6 ý.6 \6 ÖÖÖÖMM Mri NNNNN
00 00 O
(Ii cl;
E-+ NcVCVNNNNCV 1V 8+ -14 -14 -1 . 11 1? 1? 1? 1? ýj

NOO ON N 00 "-y - 00 eh M "-+ 00 ON


\O w) V1 t-
,NN ON ýO OO Il C14 e4 00 ON
et tt [- "O I'O ON C\ 00 00 00 00 et en NN \C
CV cV ci cV .16 \C %6 ÖÖÖÖC; Ö?? ý; ? `ýt
NNNN ý "-i. -+. -+. -+. -+. -ý.. r. ýý-I MMM M

NC NMMMN O\ lt O l1-
%. MN l/) v1 9 [- t- IV
,0 te-b vl \0 et NN 00 N [, Q% qý ON Q> O
ý V1 v1VA\C\D\Q%Ovint(n Oý1.0
-+O: \o""N
Q; 06 OÖ OÖ 00 MMMM llý llý n [l IlO \O Vl Wi hh ei
-r .--4 -r . -r .I "-1 -. -r o0 00 00 oÖ CÖ C týl cNK?

189
-n [- - [- N 00 ON M +n IO Mt et I! ' ON l4t 00 00 00 00 00 v'1 \C
m ZtIR et- NO\t- t- 'F- O\OOONs- O, ý MMCo MOOOOMON \O
OO ON ON W's NO 00 00 ON Vl yOOOO MMMMMMMM
06 [' Qý
'r 00 06 vi vi CVfV 00 (- 000000N't"td'-i'ý: r .4 1- \D[-
^+ Nt it IT N IRf V'1 V'1 W1 %C00 "-_ _-. 4 -4 -4 _- -"4 - . --ý -4 -4 -4 . --r - ." -4 . --ý

^4 NN 00 le IO [- t- M -et Q: %0 ýo ýO %, O --4 M cý em 00 et m ýe mM t- O% t'n


["010N't-"4\OCD CD "CCD MMMMC>[---4MetmmntM'tle (2%
O: MM O> 00 00 lý [, - ý--ý M 'e le eT -t mNý, O 00 00 00 00 00 00 00 00 00 O-
VInt*- Ch MMMMet000: O\O>O\oýN0_j_- 4
N [ý lý _- - "4. -i-i. N eh V'1
NV vj ýC [ý 00 00 00 00 00 I
.-... -. --4 . -r -- --4 -
I1I88118808

In
In Q NW11 W*i v %D MMM tl NMNNN\OMNNNNMNNOOwl N
NMMM -4 00 00 00 tý 00 ON 00 00 00 00 N l-- t.0 %ýO ýo %D I'O 00 M
W) Olý O1% ON M ý10 v1 OOOOOMN
15 v) NNNN
ýt ýt et
i
ON ON ON ON Q% ON G% ON -14 ON M
^" k! 1 kei
oA O Cý _4 MN ke -'i
t, vi [-: oÖ 00 06 06 oÖ 00 06 06 ÖM
M -------------rr. -ý.. r. -r. --ý -. -ý NNN

0
a) %0 -4 --4 00 Nkn qt It fn MNMNen
Rt fn ýokowl%0%Okn %0 ýD01% mNl-
l--40% %0000(14 -OOIkOI*- W) !t W)"tWl%0 q1t aý0%000NO., 00ON Cs t-(14 M
00 ON 0% 00 "t OOOOM V1 W) W) W) vl k/1 [- Q., 0% ON ON ON (7) ON ON 0% NM 0ý
bq MMM1CN-- r_; N4\D 6 ÖÖcV4vi
MMM--- . \6 %606000Ö6
.-I .-+ .-+ .I . -I . -r - .I . -r NNNNNNNNNNN
5
z %1DON C\O\in N 000Od'-+ 0%00(7% 00110%D
0 t-NNNýOCý%DMmv1in
rnmMMO
"C v1 Vl h vl
W1 000N
N4
in
NOmOcN
%O1Ov1Ovn
r- ob
%D 00
-+v1V1n%0it)
Mt- l- CNoot-
chOOOOOOONN
V', hen
t- l-- OO n
N
00
N6 %6 ýD \O ýO 00 O "--i . -i .--i .--i r-i i N
-4 . -i ... 'eh V1
r.
cqs IIIII

IIIII ilr
I
oil
ý--ý .-r N CV NNNNNN fV NI

W
r.
0 Vn 00 00 00 \O
%DI'O1OooONN-
IT %-O %0 ýO l- M Iýt v1 IT ýt `Rt - O\ 00 00 00 00 00 ON O\ - -4P
[-MNen NNNNo0zt'ttetet'tkn OON
10"tv1ON ON ON oo(D C)ON MC\MCO 4N[-kelt- - t- t- t- t- tl- 10 "T t-
MOOOOtt --l -14 ^46 NMMMMMM V1l-zt-lz 06 C;
NMMM wi N Qý Qý Oý . -r ý-r -4 . -r . -r

00'-400 'f '0NNNen N00 v) M" -4ý4 \0 'ONlq! 'i! '


-r--4
ý--ý O v1 O 'O 00 V'1 wl V'1 O 00 to 00 N 00 00 00 N 0\ NNNNNMNO O\ 00
to v1 Q\ h 00 NOOO ON ,N oo N t- t- NV1 In Vl V'1 W) W) v)
MNOoNOoN0\ [, 000vl
"-ý ý--ý ,
'It W) W1WAv111)ON
MhV; tf1 V1 OÖ 06 OÖ 06 OÖ OÖ Cý --4 .
I
"-4 "ý . -r .- "w . .r .r .ý.
I

b
wl, t- %O W) nON ON n'RT1ýOON ON ON Cs en NIT 0%en ON en MMen
an Ch OV kl*it-ONhýArtýN 010ýOýOýeten O--4- NNNN--4N
-14W)
N
I'OOONNen ON 10"--ý. -ýOýýOOIýIýlýtý00 wt IRT'T ý:td' It ItM
O\ x-14 NN ll- vn 00 O\ ON M 00 M eh [-
"t `t 'T -t
N Iý N tý
MI Vy \O OÖ 06 06 06 06 06 Qý ý--ý "-ý . --ý . -ý . -i . -ý . -ý . --ý . -r . --i . -ý

00 .O MD 'O Oý tt d' 00 -e ý ;Zr2le


[- O V1ýf NNOh v1Neh`ýtNnýV
etN --ýN Cý\Ov1vli " et1D\O%D\O-4ON tteT e -t -t -t e0%G>M
ý "-r "--ý -N "-+ 00 00 vi
.yMMMMMM V) ýO lý lý [ý [ý P [ý lý n Cý ý

00 ---t d' 00 qt tt -- qt mm ON Oh N t- C\ [t ON n l- lF
00 MMM 00 NNOO - cý
ON OMOOONN oo N O% ON r C% ON pN
V1't IT It v1O\C\MI-QN IT It IT tetON000\ooC .DON
oo 0000 Mn M
O It It It v1 ýt OO ÖÖOÖÖOOO
.4M vy NC ý6 ý6 \O \6 oC ý+ M v'i
NM cIi cIi of oÖ r .I,. NNNNNNNNN
IIIII.. .. .. CV N
Cd I I I I
00
M'$r t- - V1 N V1 MQ: C\ - aý
0% O\ 00 0% 00 00 -- r-
140 r-. 00 V, ß}' ONMNNONNOOC%--4
M 00 t- .-4 -+--t OC>v IIZ M
00 N O: N l*- i- O Mt l- MMMMM O\ Q1 00 00 00 00
vý MMM1--1-4 00 00 00 - et O
OOO-4M vi \6 %6\6 %iC t-: cýCÖÖÖO OO Mvi
SfIMM1 -;
ni oS . -I .--. . -r .".r . "1 , "I CV CV NNNNN tV NN
,. ".. --.
E-4 88 S S 44 S

10 -ON v)1-00o0Mn% I'D, 0 W1 in W) 4cnt-00nt-r-00oo, et 00


tn vlv1mot'O4nl-OO%0ONn000hmC%ýýý+aýOONOsQNOOlý%
W) 00 00 ll- \.C tl- 1O
Il- NNMMN Q\ ON 00 00 m
N O\ ON O\ M 00 -4 "o l-ý \O \-O \O n l- 00 00 N
O\ \O 606 4444l.:
MMMM (Ii et 'f' oÖ oÖ oÖ oÖ 04 06 06 oÖ Ö f-1
'er t Oý . .... ..... ....... -w

Q%NNN"- %o")t-I-NNen
NNNN 00ooooo000-t t- MMQýoooa 0000 W)t- N
0000 -14 v) oo o0 Rt M 00 [- t- t- t- t- N It et .t d et et O v1 QN
-+ I} r1 th
- -+ h Cl er en MMMMM t- I". qqh et et III* V1 0
OOOOCNNt- ntl- eto0IT r-+
gtetNtet
11111111111111111111111111111
,.. NNNNNNNNettri

190
e' i. 00 00 Oý -ý"N't 00 00 Oh O\ - "-+
NN 00 +NOý0%oýýDýoO>01 r+C>M--+M e r'- "t MMMM
et %V l- 0% O\ ON 00 00 00 oÖ 00 c; O'MM o0 00
00 oÖ M et vi Vl vi 9O
GOOO CD NNNNNNNNNNNNl [-l-[-Ncl NOýO
--ý--4--4 NNNCIA NNNNNNNNN "-+"-+ -4-- --4--1%, 0 ý,0 -q-q

Wn qtY qtt 00 N %D MNN oO NNN 00 00 N- Oh "t ON et 1ý0 l- [- 00 I'D


ooCýON NMOrttol v11gTkn nv 1-trIT v1W-1(o OMNNN00a',
. r. r. --ýooNvývývývývývlýnvývlvlNýtetrtetOOO I- enq. I
W) W) in W1 Vl [- l- t- l- l- l- [- l- l- [- [, r qýh qT ch '[h "-i .--i "--ý "-ý
. -r ý"-ý . -ý .. r .ý .--i .-r .ý .-ý . -ý ,.. y .. r .--r . -+ rr .-ý . --i . -r .. ý --4 1-4 --

a) 00 C% O%CN en MMMMMen Men M eTNNNN00 Cý0ý v»d'


MI'Det ! t[-"-ýNNNNN NNNN--ýNN"(14 [-ý+C)"
.O
00 MMMM`ýt V1ýOýpýpýpýp\pýpýCýpýpýD NNNNIýtýIýMMd
NNNNN NNNNNNNNNNNNNNNNý--ý--+--ý"-S

0
V) v1 d' le v% - O\ Oý O\ V% %IC V1 v1 %A t- v1 v'1
a) NN-- v1 -MM-MMMM --+ M %D %0 N V1 w1 NOO t- V1
I- -i - -. -! o0 o0 V1 vl Vl V1 %n N Vn Wi ei hO [-ý [- tý l-: d: e -t O\ c. 00
bA 'n Vl
a) O %0
%, %0 %D 00 ý> Qý O\ O\ Cý O\ O\ Oh Qý O> Cý tt ch et O\ Oý Cý MM
10 NNNNNNNNNNNNNNNNNNNNNý-+Iý-I I
ý-I ý-+M
E

z l- l-
MMMen
t- [l- 00
I'OCh
t- 00 ll-
O\OO\Oo0G\C\
00 \O [- IN IN 00 00
ON OOI*
'et mN V1 ki) 00 [N NN fV NMN

0
0
rr ""ý "--ý tý [ý kn \O \O v1 v1 kq v) \0 10 r+
M V1
00 00
V1v1MNN-
00 00 %n kn to OOO
-47N10
t-

b9 NNNNNNN NNNNNNNNre Nre re re re ýý-+-+MM

Gn W) M It MN
NO-O %D -
IN It "* %0 1ý0 Nt
v1 wl "D %0 vl
%D '[t %0 INO tN
ýp v1 ýp %0 t-
N ON O\ ON O\ -- 00 00
.-n NOON "-+ ON ON OO Ul N
^+-+-+hM00000000001,0000v -I- :r 000ov'1N
a cV eq cV cV cV ke) vi vi vi wi vi w; kl i V'i kej 4 ýp Np N6 et
NNNNNNNNNNNNNNNNNNNC "--i .--i .--i .-i
NC -+ýMM
II

ao
I tr I'O wl et 00 IRt MM OO MNMMM It as V1 N V1
aý OýOCD 110[-000000 0O\GNo0mQ ýOýOýGV1"-+ý-+"-rFt
-4 -4
It 0N
ooON ON ON hNNNNNNNNNM -- --ýý-+nnnN-C\
3 .--ý .--ý .. r .--i . --ý - -4 --- .-"ý .--i . ý-i . -a ý--ý -4 -4 .-ý . --ý r+ ý--ý --- t- t . -i .-i

b N l" l- -4 " 4t1 h et et "It M 00 er MMM GN Oh NNMM ON M Q*, v1 00 M


Oh OKON OO%en -- NNNMNNMen MrtM"-+'" O%Ovn O -t OOet
an [- l- t- 000ýM"" ""r-4. -4. --4"-4 . -4-4-. 4000OOO W) V)v. 1 ON ONNOM

Gn
qýt efi 1G eF qt ON 00 k0 00 l- oo ON 00 00 00 00 t7' oo "-+ NN t- %D 00 It Rih -
[-tl- 0OnnOIT "IRTfn IT V1'TtRtIt ctON O01OOMNet0000v1ý--+
00 00 00 00 CN 00 00 00 00 00 00 00 00 00 00 00 R:f oo i- 00 00 NNN kn it) N 00
4444ÖÖÖ
0 -4 . -i _;
CV NN(V
. -i
NNNNNNNNNNfV NNNNN(V
--; . -i
\O v1
MM
- .4

10
ON 00 00 00 NM IT It M et IT et et Itt It t- 00 O\
v1 It ITttl-ON %0 hC-ýDt-t-l-nC-t-00. -r .r -4 . -- lý n -" N
'C7 OOOOO-41tIttIt -4ON . -4 Nn [-. lý %ýO%ONen
cd IRtIýt'"Te! ' it't'RT OI1'1Ol-nMMen 0000v1wl
NNNCVCV N(VNCV CV eV CV N(V NNNCV 1VNN--0
º-a "14 1+MM
ai
00
"tN'trfgmgv1'etggmgg"tt7 'CO\C\Oý.
OOýOOOGht-00t-t-l-ýol-t-[ýN%0ONONQNllý i 0000ýY'tT
CL) ON 00 ON m ON 00 MMMMMMMMMMN ýoI-l-tý VlM
ce t- NO ýo ýo mMM ýo ýo Oo V)

E-4 cV tV NNtVtV NNNfI


ialIS NNE-ý+ý-ý+ýýýMM
ISII

P-4 -4 O\ C\"--ými
O% O\týMmNN lit 0 VI
%,
Mv1Y1VN1tt\O\D%0P-\Dt-\JO%0Ntift-000,
OOOOO O00 oMO>O> -- NO
IV O> O\ oý O\ Q> cý ON Q: C> O> O> Kn Qb O\ O> h
MMMMCrIet et tT 00 00 00 V1
ýiý1 Vyvlvlvl Vi vi vi nl-i cV N'17tV
NNNNNNNNNNNNNNNNNNNNNr-+P.
-ýr-+ý
Iq

e! et NOSet 00 eh 00 Cs 00eF"t00NM
l- l [ý t. - O ýn ýp ýO ýO ýO Vý ýD ýO ýD ýO nV1 %0 ýON M fl N
oo °°°° 00 00 NNNNNNNNNNNCh et ýt ýt 00 MN Oý et O OH
Cl CD _ 00M
výhýllvi 060606o6060606ododoÖ:, 6t : oo"OM
'eiCi

191
--ý 00 00 m
r- t- t- n
00 (14 ý ýO %1or- 11, ýO tý NN W*bW, kO 10 ýt NNN
W) Mt qt ýt %D V1 (- MMMONN "14 00 00 MM
fl- t*- C\ 00m "-y"-4
NN -rMr t 1, r nknQýC\Ooo--i.
\O It'ýtt%:
cV 0ý 0ýt1ivi0000N`ItMt -it-i000
0ý MMMMN It NtIT 1ý0 t-CNO\O\CýC\01%
ýf 44\. G 00 "--- -4 --- "--- "" 1-4 ." . -+ .-" . -r -4 "- . -i -4 -4 . -y

10 tt t1' t7 [l- v') t- . -y 00 t- %D


r-~~~~ W) a, W) "" . -r .-y kn t- m ll- tý 'ct tl- ON O\ ON
Mn
--M-00000o00'OON NNNNN
ON'RT'ti'et'J70%NvnelO%Dt- - OOONtntnk1v).
OMMMM cCh "Dý, p[, -+MN N NNNN
"-+l-CýCNC%-;
wi l! 1 -,O 00 O\ C\ Q\ O\ .... calNcVcalNMd "; vjvivi vjvj
. --r .. . -.

Gn
N ON Olý Olý v1 (*- rý ON 00 tr [l- . -r
'n ON Cý ONO1% ONChm"IRt -4 O\ 00 __ ." .--_ %p mf' Ch
Ir-" - N%OMCD 000t-m%DýOI'O%O%, Oqt
+ l'- tý rý 1ý0 --, CN N'C I'D ON 00 O's r4 NN %o M O*, QN ON ON vn ."
N It 11' et e* ýT It
OO -+M VCC pt 0ý0;
a4 Qý c ONMMMMMM
I' I" a; I". NN CV NNNN CV

0
00 O\ ON '0 0\ '0 tý \D 00 NMN N-
ONet '0 NNwl N- V'l MMM et et er
- M1-Ot-t-M(>"-rpp-OO\OO"lt
N ýe t- v1 NO [- md'Nlt'(D CD Or-+. --+. --+
v)
oA V) ON O'Nc Oc l-t-Nd:
00 M 00 00 00 MMOOOO rý V) -4 -q . -1 -q -0 -4
rýZ rz r,: 2; 4flicVCV
cV ýýC%Dýd
et'ý %C
10 M MMMM lý Oý"-ý. -r.. r. -ý, ""ý.. r.. r,. r.. ý(V CV N NN N NcV NN NN fV
c-
Z
0 --4- nON 00 C\nMM V1 C% 00MetNMei'MN1OMIh't'U'MM
N --ý NM 00 e! ' N v', v1 ON [l-
Wlvyct ýNWAý000000NOý. v'1 et l- l- tT O-O %D Rt Oh OOO ON Ch
V1en
--1 . 4%0 W) Q
b M""'""'""'1-4-4
-i --ýNNNNNNNN
--1 ß. -r00
NNNNN

N
0 [-
ýO
O0 0
o0
-+ 00 M
00 "--1 ON ON O
ON 'If
ON 43N0000[-vi
Oý 'tt
M
N
ýO MM
ooet
'tf M
-tt
V1
00 00 00
--4-r. -4Q4n
-4 . --4 . -r N
oo kn IRt on
ci1%0c
CN t., [ý% n
v1
lý,
N -4
WIC-4M
-4 00
%D -4 NNN
000NCNt-QN
-4 -4 -4
10 MOep'Rt6C '" , .r
4
MMMMM W;W; W; ý6 n06 06*6Odcý6 NN NNNN
.16 oÖ 0ý . ,. .y,..
p ,.. ,, _, ,, ,p,, , _, .p,., ,. ,.,. VV NNN fV N
p p p 1"

N ýt ON m O*s mn CS M ON %D 1.0 1p M I'D NNN


00 v %0 ON ll- I'O NM Wl - v1 4411V*l
.-r .-r :4 t- \p V11 00 mmm V11 ON V1 W1 Ul M vl 00 00 00 00 00 00 00
et 't et -4 vl -+ 10 C; N
C-i t -i NNM
-, et vi tei vi vi vi
.. rr ..... . -ti .....

Iti v'1N1.0 NC
an tn t'ý -"' tl- N 4n eh ýM
vnNOýOýM"-rý0ý10,
00 ýt NNNý ."
I-
"" 00 00 00 00 "M ON ON M ON M Ch
00 OO ONNNN v1 l- M ON ON O, ON cN
%O 00 . -4 [ý Q\ ON ON N 1ý0 MMMM
Ný/ V1 v1hOýMMMONMCh "T NN 00 00 00 00 00
w° .--i . -ý .-i .--i "--i nj
v1v1v1
,..
ýC4NcalNcVMrtýtetetetet
Vl V1 N 00 Qý 01 Oý

Mt d' ON -4 ke)
ct ýO VO t1% N
^' -'+ NNh . -r -4 .r . -4 m ýo l- et ýt 'et et ýt
c OO
to --+ . -4 vl %0 O w"I ¢t vl O 'TY 00 00 00 00 vl " 00 n lý Cý I tý
0 hn
"r -4 in V1 0n
tý 00 00 I'O 00 MMMV
Cý 00 00 00 ýD et et It 'tt 'RT V1 M '40 ON 0N aý ON M
'ý7 4 I! 1 [-
U t ýi tM t ý1 MM ýi l Qi Qý . ""...,
OÖ oÖ oÖ OÖ oÖ O "-i "--ý . -ý .-ý ... ý .r
..., 1.4 "-4 -I V-4 ý,, ý , ,,,.. r N tV cV N fV eV N
I I

10 .-r -4 .r .-4 -4 IM `et Q\ 00 00 \D \ýO t- O% t-


OOO"r-4en N 'D \,C NN C\ N et vy 'rt r! ' tt
lL N et 't .--4Q\CD oo[, O\ MMOQ\v1ett/1 W1ýD[ý"rN..
v1 v1 '4O 00 O "r 00 CP%tt [ý lý kn tý OOOOOO ý. -ý. -r
cä ýDOOOOOO ; ýO "" ""ý . -ý ý--ý .. r
0 ýt..: C; cV NNNcV V viýýCýýCý
4 ''""NNNNNNNNNNNNN
i
fl
..00
ýN NN NN C14 NN "1 N-
o~o VO ^l+ (3N -r t- "r . -4 %D ON 00 00 ON ON 00
Ch NNNO Oý O% ON 4n en m 00 00 ON m"
ON O lý 00 `et \O "-+ W) v1 v1 N cn Itr IRT W) v1 It
It Q, a, 17, O1`ItI'- V) 00 00 00 00 00 00 Vt QN ON ON OS ON
cs ýM%6l-:
cd ,.. l-: l-z 000
riMMMM CI . -I . -I ..
E- S S
..
I
,I..
I
,. yI,.. ýNNNNNNNNNNNNN

C14 C\ qtt %D m qt W) t'- \0 - t ,tT


wIT O\ t- O\ Rt\01ýt -4 vt qt eq W) on It qT 4-1
soO\OO\OC\ O\ NýtOOOOton-4
NIýt', \D\Dtý
onW)W)W)kn4nOON-t'\ýCw1h,
6OdC\ ý0ýC\ Mon
. "ifV en en M

r- 0 C -
mNNNNO %, [- [- MMM-M Ch v-, V)
(> vi V1 G> %, O n [- 4 SS t-
(D V-) vý v) vi I-M(>ntOnt\io ONNO l- t- nNM t- r- K- t- nONNNMNN
My v1 v1výO`el-O\NýON0 mMen %V--4 r-r-r. (: cýcý
-rMchO\(:
OpNMMc+i
111111i111t1t111t1t~~~~~~~~~~
111i111111

192
10
Cd 01% t-
00 V41, W*l kn ") m %n eq Mt 1-t VII In tý
o" -4 -4 -4 --4 -4 00 00 r- t- -%D \0
ýo 10 It NT C14 (D C> "Q" C14 -- en m W) (n m t- t- C\ 0%
1-4 'TT 'tr
\C; %6 06 06 06 \6 \16 1ý6 %6 C-i C4 cli C-i t-: tz cl cn wl cn Cf! t- t- (7ý C\ en cn
C> C,4 eq C4 (14 " C4 " C,4 C\ C\ cp\ 0\ m en en 'D 0 \6 ýo 6 c; c; \ýo
r4 C14 ..... .0\. C; .o
C14 " -4 -4 -4 --1 --4 -. 4 - C\ a-, 0\ c-, C\ ýo ZZ ,0 "D 0ý C'R

CA

a Ei
kn ýt C\ en t- t- t- 00 kn 4n -4 - -4 - \, O D
%. r- ttl 00 W)
-TI
o
E C44mmmt
-. In kn kt'j mcnIRT M"\D\o
IýT 'Rt IT 'I* \D\Dm
ý: r men- t- ZtCYNON C) t-
IT --4 --4
-4 t- r- t- t- \ýD Itt Mt Tr M
\0 r
C\ eq C14

rA
J)
, co
fý "T
; 't
en 'Rt
en -t
M "D
IT
r-
W')%o\0 f-4
ON CN"
C-4
ON eq 00 00 00
00 \C) --4 %D t- 00 t-ý -. 4 -4 M cq

.6 en \411' IT It " C,4 C\ en m C,


.o ýo \0 -ýo C'4 " rn4 \0 tn \,\D
-t \0 O \0 o eq
tn \,
ýo 00 C,t:r4 0\ 00
0 C,4 eq I'l t
oU z C', 'ei Wi 'i Ci Ci Ci Ci tn tq -. 4 4

O
M qT C, 4
--4 00 kn "t W) -4 ON

öA
10 0 t-:
qT W"
00 W)411W)qT rn M en M en en en
Cý Cý (: 6
e4 m cl qtt kn eq on -,t W)tn t- a6C: ) 00 't
(2NC\ aNoN00 o ONC>a%%b
r-
)ý 0ý
Cý Cý Cý
Cý 0ý
Ile 0: \.
Cý 1 (ýl 1.6\6 116 \,6 \-6 ý6 %6C-;C'i C-ivi \Cý