Journal Earthqueke

Ingegneria Sismica - Anno XXXIII - N. 3 Settembre 2016 - Trimestrale - Poste Italiane S.p.A. - Sped. in Abb.
Postale DL 353/2003 ISSN 0393-1420

(conv. in Legge 27/02/2004 n. 46 Art. 1, Comma 1) - CN/BO - Pàtron Editore s.r.l. - Via Badini, 12 - Quarto Inferiore - 40057 Granarolo dell’Emilia (Bo)
3 2016
Trimestrale tecnico-scientifico
Pàtron editore
International Journal of Earthquake Engineering
www.iingegneriasismica .o
or g
trimestrale tecnico-scientifico
Anno XXXIII N. 3 – settembre 2016 – Pagine 1-94
Special issue - Numero Speciale

Italian Steel Days 2015 (CTA2015)
Giornate Italiane delle Costruzioni in Acciaio
Founder / Fondatore
Editors:
Duilio Benedetti – Politecnico di Milano Rosario Montuori, Dep. Civil Engineering, University of Salerno, Italy
Editor-in-Chief / Coordinatore editoriale Francesco Fabbrocino, Dep. of Engineering, Pegaso University, Naples, Italy
Gianmario Benzoni – University of California San Diego – benzoni@ucsd.edu
Associate Editors / Coordinatori associati

Sommario/Contents
Fernando Fraternali – Universita’ di Salerno, Italy
Panayotis Caridys – National Technical University Athens, Greece
Dina D’Ayala – University College London, UK
Metal shear panels for the seismic protec tion of buildings:
Rosario Montuori – Universita’ di Salerno, Italy recent findings and perspectives
Maria Adelaide Parisi - Politecnico di Milano, Italy
Lastre forate a taglio di acciaio per la protezione sismica
degli edifici
Scientific Board /Comitato scientifico
Raimondo Betti – Columbia University, USA G. BRANDO, G. DE MATTEIS Pag. 5
Alfredo Campos Costa – LNEC Lisbon, Portugal
Giacomo Di Pasquale – Dipartimento della Protezione Civile, Italy
Giancarlo Gioda – Politecnico di Milano, Italy Effetcs of Class B Site on the Seismic Reliability Of Base-
James M. Kelly – University of California Berkeley, USA
Sergio Lagomarsino – Università di Genova, Italy
Isolated Steel Systems
Giuseppe Lomiento – Bianchi-Lomiento, Italy Effetti di suolo di Categoria B sull’affidabilità sismica
Claudia Madiai – Università degli Studi di Firenze, Italy
Alessandro Martelli – ENEA Bologna, Italy di sistemi in acciaio isolati alla base
Mauro Mezzina – Politecnico di Bari, Italy
Maria Ofelia Moroni – Universidad de Chile, Santiago, Chile P. CASTALDO, G. AMENDOLA, B. PALAZZO ” 29
Elide Nastri - Universita' di Salerno, Italy
Stefano Pampanin – University of Canterbury, New Zealand
Floriana Pergalani – Politecnico di Milano, Italy Appraisal of Seismic Design Criteria for Concentric Bracing Steel
Yuri Ribakov – Ariel University, Israel
Paolo Rugarli – Castalia S.r.l, Milano, Italy
Structures According to Italian and European Codes
Gaetano Russo – Università di Udine, Italy Valutazione dei criteri di progetto per le strutture di acciaio a
Milutin Srbulov – Mott McDonald, U.K.
Miha Tomasevic – Slovenian Civil Engineering Institute, Slovenia controventi concentrici secondo le norme italiane ed europee
Mihailo Trifunac – University of Southern California, USA B. FAGGIANO, A. FORMISANO, C. CASTALDO, L. FIORINO, V. MACILLO,
Keh-Chyuan Tsai – National Taiwan University, Taipei, Taiwan
Thomas Wenk – Swiss Society for Earthquake Eng., Switzerland F.M. MAZZOLANI ” 42
Aspasia Zerva – Drexel University, USA
Director /Direttore responsabile Preliminary Analysis on the Influence of the Link Configuration
Fausto Giovannardi – Giovannardi e Rontini, Borgo San Lorenzo (FI), Italy
on Seismic Performances of MRF-EBF Dual Systems
Designed by TPMC
Analisi preliminare dell’influenza della configurazione nelle
performance sismiche dei sistemi accoppiati con controventi
eccentrici progettati mediante la TPMC
R. M ONTUORI, E. NASTRI, V. P ILUSO ” 52
The Non-Linear Response of Steel-Concrete Composite Joints

La risposta non lineare dei nodi
composti acciaio calcestruzzo
M.R. P ECCE ” 66
Importance of Link Models in the Assessment of the

Redazione, amministrazione, abbonamenti e pubblicità Seismic Response of Multi-Storey EBFs Designed by EC8
Pàtron Editore Influenza della modellazione del link nella valutazione
Via Badini 12, Quarto Inferiore
40057 Granarolo dell’Emilia, Bologna della risposta sismica di EBF
Tel. (051) 767003 – Fax (051) 768252 M. B OSCO, A. GHERSI, E.M.M ARINO, P.P. R OSSI ” 81
e-mail: info@patroneditore.com
Sito: www.patroneditore.com
Stampa: Rabbi S.r.l. Bologna per conto della Pàtron editore
Bologna, febbraio 2014
Subscriptions see URL:
www.ingegneriasismica.org
Sul sito www.ingegneriasismica.org nella sessione archivi sono pre-

senti tutti gli indici delle annate pubblicate.
Ingegneria Sismica
Registrazione Tribunale di Bologna n. 5139 del 20.2.84
Realizzazione grafica della copertina: Arturo Galletti
ASSOCIATO ALL’USPI
UNIONE STAMPA
PERIODICA ITALIANA
TO READERS AND AUThORS
We are pleased to announce that Ingegneria Sismica is covered by Compendex, one of Elsevier products. Compendex will index papers and possibly extract
data from the full text. Coverage increases dissemination of authors’ work: the benefits are high visibility to a global audience. Since 1970, Compendex has
been the engineering database of choice for researchers, students, faculty and engineering professionals around the world.
AI LETTORI ED AGLI AUTORI
Siamo lieti di comunicare che Ingegneria Sismica fa parte del database di Compendex, uno dei prodotti di Elsevier. Questo comporta l’indicizzazione degli arti-
coli pubblicati e la eventuale citazione
Anno di dati
XXIX – estratti
N. 4 –dalottobre-dicembre
testo. Verrà così fortemente
2012 accresciuta la visibilità internazionale degli autori e degli argomenti trattati. 3
Compendex costituisce, fin dal 1970 un importante database per ricercatori, operatori accademici, studenti e professionisti di tutto il mondo.
Anno XXXIII – Speciale CTA 2015 – Num. 3
Dear Readers,
this special issue comprises selected papers presented at the XXV edition
of “Italian Steel Days” held in Salerno on October 1-2, 2015. This
conference represents the place where Italian researchers present their
results and development activities in the field of Steel Structures.
Common papers' emphasis is about the structural performance under seismic
conditions. The Guest Editors, Rosario Montuori and Francesco Fabbrocino,
wishes to thank Dr. Gianmario Benzoni for this opportunity and for his
assistance.
Cari lettori,
Questo numero speciale è costituito da una selezione di lavori presentati alla
XXV edizione delle “Giornate Italiane della Costruzione in Acciaio” tenutasi
a Salerno il 1-2 ottobre 2015. Questa conferenza rappresenta il luogo dove i
ricercatori italiani presentano gli sviluppi e i risultati delle loro attività di
ricerca nel campo delle strutture metalliche. Il presente numero è stato
realizzato mediante una selezione degli articoli presentati al convegno

riguardanti problemi di sismica. Gli Editori, Rosario Montuori e Francesco
Fabbrocino, desiderano ringraziare il Prof. Gianmario Benzoni per
l'opportunità ricevuta e per la sua assistenza per la realizzazione di questo
numero speciale.
4
Num. …
METAL SHEAR PANELS FOR SEISMIC PROTECTION OF

BUILDINGS: RECENT FINDINGS AND PERSPECTIVES
Gianfranco De Matteis1, Giuseppe Brando2*

1
Department of Architecture and Industrial Design “Luigi Vanvitelli”,
Second University of Naples, Italy
2
Department of Engineering and Geology, University “G. D’Annunzio”, Chieti-Pescara, Italy
SUMMARY: This paper describes recent experimental researches carried out on metal shear
panels to be employed as dampers for seismic protection of new and existing buildings.
Three typologies of shear panels are presented, which have been conceived with different
strategies to mitigate the detrimental effects provoked by buckling phenomena.
For each solution, experimental results are provided together with design issues. Also, some
remarks on the technological aspects of the devices are highlighted in order to limit the
adoption of bad details that could jeopardize the structural performance of the system.
In the whole, the obtained outcomes provide interesting information opening new frontiers in
the field of research on dissipative metal shear panels.
KEYWORDS: Metal Shear Plates, Pure Aluminium Shear Panels (PASPs), Buckling Inhibited
Shear Panels (BISPs), Perforated Shear Plates (PSPs), Experimental tests
1. INTRODUCTION
In order to satisfy the performance levels commonly required by codes and provisions, a large
amount of research was conducted, over the second half of the last century until today, for
developing innovative earthquake-resistant systems aiming at improving the seismic
performance of structures while keeping either construction or retrofitting costs reasonable.
Therefore, the “response control” methodology was developed as alternative to the “traditional”
one. The former is based on controlling and limiting the dynamic effects on the structural
elements by means of added special devices [Brando et al., 2015]. Contrarily, the “traditional”
design methodology exploits the ductility resources of sections and connections [Giugliano et
al., 2010, Montuori et al., 2014a], leading to the seismic input dissipation by means of plastic
hinges developing.
For the response control methodology, different approaches can be distinguished, namely
Active Control, Semi-Active Control and Passive Control [Soong and Dargush, 2007]. With
particular regard to Passive Control, this approach consists in designing buildings in order to
dissipate energy by supplemental damping mechanisms and/or in limiting the transmission of
seismic energy to the main structure by decoupling the structure movements the ground
shacking. Supplemental damping systems are based on the use of special devices, also known
as dampers. They are arranged in the structure and are designed in order to start to dissipate
energy by hysteretic [De Matteis et al., 2009], friction [Montuori et al., 2014b] or viscous
___________________
*Corresponding author: Giuseppe Brando, Department of Engineering and Geology, University “G. D’Annunzio”,
Chieti-Pescara, Italy
Email: gbrando@unich.it
G. De MATTEIS and G. BRANDO
mechanisms as soon as the structure is shaken, hence reducing its overall dynamic response.
The advantages in using dampers may be synthetically listed as follows [Christopoulos and
Filiatrault, 2007]: (i) They may be designed and arranged in the construction in order to
conveniently change both the damping and the dynamic features of the structure, also reducing
possible non regularities; (ii) they may be conceived to dissipate energy under low movements
of the structure so that their protective function is activated when the other elements are still in
the elastic fields; (iii) after a strong ground motion event, they can be inspected and
conveniently replaced.
Among the existing dampers, in recent decades, dissipative metal shear panels have collected
an increasingly wide consensus. A shear metal panel is generally made of a metal plate working
in shear connected to a surrounding steel frame by means of bolted and/or welded connections.
The frame has the task of delivering the forces from the primary members of the protected
structure to the plate. This type of function can be carried out throughout a direct connection
of the panels to the structure (Full-Bay or Partial-Bay configurations; Figs. 1a and 1b) - this
leading to the concept of Metal Plate Shear Walls- or by way of steel braces (Bracing Type
configuration; Fig. 1c).
Apart from the high in-plane stiffness, that allows to solve easily some critical issues related to
the lateral deformability of buildings, in particular when they are made of steel moment
resisting frames, one of the main prerogatives of shear panels is the easiness in controlling the
shear resistance. In fact, unlike the more traditional braced frames, the use of extremely thin
elements is made possible by a stable post-buckling behaviour due to the onset of "tension
field" type resisting mechanisms [Shishkin et al., 2009].
a) b) c)
Figure 1. Metal Shear Panels arrangements in a frame: (a) Full-Bay configuration; (b) Partial-Bay
configuration and (c) Bracing Type configuration. Source: [De Matteis et al., 2007]
Furthermore, even in presence of reduced thicknesses, the possibility of combining the base
plate with other elements – for example, transversal stiffeners - allows the control of the level
of demand leading to buckling phenomena as well as to mitigate possible pinching effects
producing detriment on the hysteretic cycles.
Significant research efforts have been addressed to the definition of innovative metal shear
panels conceived in order to provide convenient solutions able to comply with the several
demands posed for their employment in seismic prone zone.
In this paper, three typologies of shear panels, investigated by the authors in the last ten years,
are presented. These are the so called Pure Aluminium Shear Panels (PASPs), Buckling
Inhibited Shear Panels (BISPs), Perforated Shear Plates (PSPs), which differentiate each other
because of the alternative strategies undertaken to mitigate the detrimental effects that may be
provoked by buckling phenomena, as well as to activate the dissipative capacity of the base
plate already for low shear strain demands. In particular, full-scale experimental investigations
6
INGEGNERIA SISMICA – INTERNATIONAL JOURNAL OF EARTHQUAKE ENGINEERING
are shown with the aim of giving a general overview on the structural performance parameters
that characterize the response of the studied panel typologies and to put in evidence pros and
cons of the proposed solutions. Moreover, for each typology, information about technological
details to be adopted are provided, in order to minimize possible issues that could arise for a
bad set-up of the proposed systems.
2. PURE ALUMINIUM SHEAR PANELS (PASPS)

2.1 General
PASPs fall in the family of shear panels made of low yield stress (LYS) metals. Their use,
joined to a rigorous application [Longo et al., 2009] of the capacity design criteria, could be
definitely effective for the seismic upgrading of new and existing buildings.
During the nineties, LYS panels found a large success in Asia, where a steel with low carbon
content characterized by a yield stress of 120 MPa and a ductility of 50% (Nakashima, 1995)
was used. The idea of employing the aluminium alloy EN-AW 1050A H24, which is an almost
pure aluminium, came from the fact that the steel described above is not produced and
commercialised in both the European and North-American market. On the other hand, the
considered alloy seems to be the perfect surrogate solution, as it can be characterized, if
properly treated (see Section 2.2), by a very low yield stress and a high ductility.
In this Section, experimental and numerical studies on two types of PASPs are presented. They
have been conceived for being installed in the structures according to a “full-bay” and a
“bracing type” configuration, respectively.
2.2 The mechanical features obtained downstream the applied heat treatment process
In order to minimize the conventional yield stress and, simultaneously, to maximize the
ductility, the used EN-AW 1050A H24 was subjected to an annealing process carried out by
means of four consecutive phases of heating at 150 °C, 230 °C, 280 °C and 330°C, each one
characterized by a duration of four hours. The applied process allowed a transformation of the
base material according to the following three successive phases: i) “Crystal Recovery”, aimed
at changing the internal structural equilibrium by acting on the ties between the grains; ii) “Re-
Crystallization”, which serves to establish and to develop new internal grains, to annul residual
stresses and to diminish the elastic strength; iii) “Grains Enlargement”, that complete the “Re-
Crystallization” phase, giving back an alloy with lower elastic strength and higher ductility [De
Matteis et al, 2012]. At the end of the annealing process described above, the elastic strength
of the base material significantly diminished, as it can be observed in Tab. 1, where the original
material is compared with the treated one in terms of mechanical features. Moreover, the
ductility increased of about eight times, this making the studied material particularly apt for
fabricating shear panels to be used as dampers.
Table 1. Mechanical features of the not treated and the treated EN-AW 1050A H24 aluminium alloys
f0.2 fu u E
Material E/f0.2 =fu/f0.2
[MPa] [MPa] [%] [MPa]
Not treated
60 120 5 70000 1166 2.00
EN-AW 1050A H24 alloy
Treated
20 65 40 70000 3500 3.25
EN-AW 1050A H24 alloy
7
The original value of the ultimate strength of the material (120 MPa) was lowered to 65 MPa,
providing a kinematic hardening factor  equal to 3.25 for the heat treated alloy. Nevertheless,
some uni-axial tests carried out in the cyclic field on a plate made of the treated alloy, for which
buckling phenomena were restrained by an external steel jacketing (Fig. 2a), put in evidence
that the ultimate strength can increase up to a value of 80 MPa ( =4.00) due to the isotropic
hardening (Fig. 2b), which is usually higher for low yield stress metals and that plays a very
important role when the material is used for producing structural elements that have to
participate to the inelastic cyclic response of the whole structure [Bosco et al., 2015].
2.3 Experimental tests on “Full-Bay” Pure Aluminium Shear Panels
Full bay shear panels have been investigated by performing cyclic tests under shear loads on
four different specimen configurations.
1st cycle
2nd cycle
3 rd cycle
a) b)
Figure 2. Cyclic Tests on the heat treated alloy: (a) the tested specimen; (b) isotropic hardening effects
on the compression/tensile stress.
For three of the studied coupons (which henceforth are labelled as shear panels “type B”, “type
F” and “type G”), welded longitudinal and transversal rectangular stiffeners, with depth of 60
mm and thickness of 5 mm. were used. Instead, for the fourth specimen, named “type H”, steel
channel shaped ribs, bolted to the basic aluminium plate, were considered. Finally, for
comparison purpose, specimens “type B” and panel “type F” made of the AW 5154A
aluminium alloy, which is more commonly used, were considered.
The selected configurations of the tested panels are described in Fig. 3, where the slenderness
parameter b/t (depth/thickness) of each sub-panel portion determined by the presence of
stiffeners is reported. It is to be noted that panel “type F” has the same slenderness panel “type
G” (excluding the corner portions), but different ribs arrangements. In fact, unlike the other
panel typologies, panel “type F” has stiffeners welded on the base plate in a staggered manner
on the two opposite faces of the shear plate.
In Fig. 4 the testing apparatus used for the quasi-static cyclic tests is shown. The studied panels
were inserted in a steel articulate frame and connected to the perimeter elements by means of
friction high-strength bolts (Fig. 4a). The external load was applied at the top beam of the frame
by means of a servo-hydraulic actuator. The actuator was connected to a very rigid lateral
support steel frame used as reaction frame (Fig. 4b).
The results of the experimental tests are provided in Fig. 5 in terms of hysteretic cycles and in
Fig. 6 in terms of hardening ratio (a), secant global stiffness (b) and equivalent viscous damping
8
ratio (c) vs. shear deformation (), consistently with the definition given in Fig. 6d. Based on
the obtained results tested specimens pointed out a very high ductility and a good structural
performance in terms of strength, stiffness and dissipative capacity. This proved that the
considered systems can be usefully adopted as passive seismic protection device.
Nevertheless, two negative behaviours were observed during the tests, with evident
repercussions on the panels response. In fact, some local buckling phenomena arose on the
bottom panel portions in the corners, because of the vertical forces produced by the moment
given by the shear force applied on the top beam of the perimeter frame, which was
characterized by a lever arm equal to the height of the panel. This type of secondary effects
produced some slipping phenomena (more evident for low shear strain) values that limited the
performance of the systems giving back a strong reduction of equivalent viscous damping at a
shear strain of about 2%.
Also due to the above described phenomena, it was noted that each panel exhibited its better
dissipative behaviour only for very high inter-story drift demands (6%:7%) (equivalent viscous
damping factor of around 30%), which are usually out of the range within which the studied
shear panels should work.
The experimental tests described above have been used in order to construct reliable FEM
models [De Matteis et al, 2008] that were used in order to carry out parametric numerical
analyses, which allowed to define design curves and to provide indication about the optima
configuration of stiffeners depending on the panels slenderness [Formisano et al., 2006].
panel type B (b/t=100) panel type F (b/t=50)
panel type G (b/t=50-25) panel type H (b/t=50)

Figure 3. The tested “full-bay” pure aluminium shear panels
9
a) b)
Figure 4. The testing apparatus for the “full bay” shear panels
50 P U R E A L U M IN IU M S H E A R P A N E L
C O N F IG U R A T IO N T Y P E F
40 40
30 30
sh ear stress (MPa
shear stress (MPa)
20
20
10
10
0
0
-10
-10
-20
-20
-30
-30 -40
-40 -50
- 0 .1 5 - 0 .1 2 - 0 .0 9 - 0 .0 6 -0 .0 3 0 0 .0 3 0 .0 6 0 .0 9 -0 .1 -0 .0 8 - 0 .0 6 -0 .0 4 -0 . 0 2 0 0 .0 2 0 .0 4 0 .0 6
s h e a r s tr a in ( m m / m m ) s h e a r s t ra in (m m /m m )
50 P U R E A L U M I N IU M S H E A R P A N E L 60 P U R E A L U M IN IU M S H E A R P A N E L
C O N F IG U R A T IO N T Y P E G C O N F IG U R A T IO N T Y P E H
40 50
30 40
30
s hea r s tress (M Pa
20
shear stress (MPa)
20
10
10
0
0
-1 0
-1 0
-2 0
-2 0
-3 0 -3 0
-4 0 -4 0
-5 0 -5 0
-0 .1 2 - 0 .0 9 -0 . 0 6 -0 .0 3 0 0 .0 3 0 .0 6 0 .0 9 - 0 .1 5 -0 . 1 2 - 0 .0 9 - 0 .0 6 - 0 .0 3 0 0 .0 3 0 .0 6 0 .0 9 0 .1 2 0 .1 5
s h e a r s t ra in ( m m /m m ) s h e a r s t r a in (m m /m m )
Figure 5. Hysteretic cycles provided by the tested full bay shear panels
10
2.4 Experimental tests and numerical analysis on “bracing-type” Pure Aluminium Shear
Panels
The above experimental tests highlighted that shear panels conceived in a full bay type
configuration may present some critical issues when they are subjected to low-middle shear
demands, whereas they are able to provide a significant dissipative capacity for very high
values of shear strains. In order to increase the local deformation of shear panels for reduced
inter-story drifts demands of the primary framed structure, exploiting the high ductility of
the base material, the bracing type configuration was proposed.
4 16000
14000
3 12000
G sec (Mpa)
10000
max/ 02
2 8000 AW 1050A-Panel type B

AW 1050A-Panel type F
AW 1050A-Panel type B 6000 AW 5154A-Panel type B
1 4000 AW 1050A-Panel type G
AW 5154A-Panel type B
AW 1050A-Panel type H
AW 1050A-Panel type G 2000
0 0
0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28%
shear strain  shear strain 
a) b)
35% 1
ES  F  s
8
30%
1 ED
 eq 
25% 4 ES
F
Gsec 
20% s
eq
15%
10% AW 1050A-Panel type F
5%
AW 1050A-Panel type G
0%
0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28%
shear strain 
c) d)
Figure 6. Comparison among tested panels (interpolation curves) in terms of hardening ratio (a),
secant global stiffness (b) and equivalent viscous damping ratio(c); definition of parameters (d).
It was characterised by reduced dimensions of shear panels with respect to the surrounding
structural frame in which panels should be installed, so to increase the ratio between the shear
deformation of the panel and the developed inter-story drift of the frame. Moreover, shear
panels connected to the structure by means of steel braces allow to avoid those detrimental
secondary effects described for the full by typology, when the shear force lead to a bending
moment development. Four cyclic tests were carried out under quasi-static cyclic loads. The
tested shear panels, which henceforth will be indicate as “type 1”, “type 2”, “type 3” and “type
4” were characterized by global dimensions of 500 by 500 mm and a plate thickness of 5 mm
(Fig. 7). The base material was the same considered for the “full-bay” specimens.
Rectangular ribs having a depth of 60 mm and a thickness of 5mm were applied on the base
plate. They have been positioned on both faces of the base plate and arranged in order to comply
11
with the rules given by Eurocode 9 [EN1999-1-1, 2007] in order to postpone shear buckling of
the plate in the inelastic field [Höglund, 1997].
The selected stiffeners arrangement lead to slenderness ratio b/t (where b is the distance
between two consecutive stiffeners and t is the base plate thickness) equal to 100 (no
intermediate stiffeners), 50, 33.3 and 25 for shear panel “type 1”, “type 2”, “type 3” and “type
4” respectively. The four specimens were inserted into a pin jointed steel frame and were
connected to its channel elements along their edge by means of tightened steel bolts. A MTS810
universal machine was used in order to load the investigated shear panels along one of the two
diagonals, according to the load protocol reported in Fig. 8. In Fig. 9, the hysteretic cycles
given back by tests, expressed in terms of diagonal force-diagonal displacement, are provided.
panel “type 1” (b/t=100) panel “type 2” (b/t=50)
panel “type 3” (b/t=33) panel “ type 4” (b/t=25)

Figure 7. The tested “bracing-type” pure aluminium shear panel configurations
80
60
Displacement [mm]
40
20
0
-20
-40
-60
-80
0 1000 2000 3000 4000
Time [s]
a) b)
Figure 8. Applied displacement history (a) and loading scheme (b)
12
The cyclic responses appeared to be appealing from the dissipative point of view due to the
large ductility offered by the devices. Nevertheless, it is evident that two different types of
hysteretic performances were obtained.
The first, typical of panels usually classified as semi-compact and characterizing both shear
panels “type 1” and “type 2”, gave back hysteretic cycles affected by pinching effects in the
inelastic field. The latter, which are related to both shear panel “type 3” and “type 4”, allowed
to have, despite of some local buckling phenomena observed during tests, large cycles and,
therefore, to get a larger dissipative capacity.
Therefore panels “type 3” and “type 4” behave a compact shear panels, also in accordance with
the fact that the ultimate shear strength did not result deteriorated with respect to the theoretical
one offered by a pure shear resisting mechanism. In fact, considering that shear areas equal to
A3=4900mm2 and A4=5500mm2 are detectable for panels “type 3” and “type 4” respectively,
the ultimate shear strengths (Vu,3 and Vu, 4) due to a pure shear mechanism (Eqs. (1) and (2))
are equal to the maximum strength read on the hysteretic cycles.
fu
Vu , 3  A3   198031 N (1)
3
fu
Vu , 4  A4   222279 N (2)
3
Further consideration may be fruitfully obtained by the comparison of the investigated panel
types in terms of energy dissipation, hardening ratio, secant global stiffness and equivalent
viscous damping ratio, according to the definition reported in Fig. 6d. In Figs 10 and 11 these
synthetic parameters, are expressed as a function of the applied shear strain value for semi-
compact and compact shear panels respectively. In the same figures the interpolation curves of
the above magnitudes, expressed by means of polynomial laws, are also depicted. In detail, as
far as the semi-compact panels are concerned, shear panel “type 2” provided a better dissipative
behaviour than panel “type 1”, due to both the larger cycles and resistant capacity, as it is
detectable by the comparison carried out in terms of cumulated dissipated energy. This is
testified also by the higher equivalent viscous damping factor (45% versus 37%).
However, panel configuration “type 1” exhibited higher values of both hardening ratio and
secant shear stiffness at collapse. In fact, the higher strength presented by panel “type 2”, due
to to the larger shear area including part of the stiffeners, provoked a higher engagement of the
plate-to-perimeter frame connection and, thus, an anticipated failure of the system.
Comparison between shear panels “type 3” and “type 4” allows to state that the two panels are
almost characterized by the same type of performances in terms of cumulated dissipated energy,
equivalent viscous damping factor and secant stiffness (Fig. 11). They presented an important
response from the dissipative point of view, giving back equivalent viscous damping factor
equal to about 50%.
Among the two panel typologies, the first results therefore the most convenient to be used, as
it requires a minor quantity of material to be employed.
The only differences may be observed for the comparison carried out in terms of hardening
ratio which put in evidence that, due to a larger shear area, panel “type 4” presented a shear
strength higher than panel “type 3”. As explained also for the comparison of panels “type1”
and “type 2”, this rebounds on the performances of the connecting system placed on the edges
of the panels which is more stressed for panel “type 4”, entailing a quicker decay of its strength
for very high shear strength demands.
13
250 250
200 200
150 150
100 100
Force [kN]
50
Force[kN]
50
0 0
-50 -50
-100 -100
-150 -150
-200 -200
-250 -250
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Displacement [mm] Displacement [mm]
a) b)
250
250
200
200
150 150
100 100
Force[kN]
50
Force[kN]
50
0 0
-50 -50
-100 -100
-150 -150
-200 -200
-250 -250
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
Displacement [mm] Displacement [mm]
c) d)
Figure 9. Hysteretic cycles: “type 1” (a), “type 2” (b), “type 3” (c) and “type 4”
300 8.00
"type 1" "type 1"
7.00
250 "type 2"
Hardening ratio 0.2
"type 2"
6.00
Cumulated dissipated
Poli. ("type 1")

200
Energy[kNm]
Poli. ("type 1") 5.00 Poli. ("type 2")
150 Poli. ("type 2") 4.00

3.00
100
2.00
50 1.00
0 0.00
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
Shear Strain [%] Shear Strain [%]
a) b)
50 0.6 "type 1"
"type 1"
Equivalent viscous damping factor
45 "type 2"
0.5 Poli. ("type 1")
Secant Stiffness[kN/mm]
40 "type 2"
Poli. ("type 2")
35 Poli. ("type 1") 0.4
30
Poli. ("type 2")
25 0.3

20
0.2
15
10 0.1
5
0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
c) d)
Figure 10. Results for semi-compact panels “type 1” and “type 2”: (a) cumulated dissipated energy,
(b) hardening ratio, (c) secant shear stiffness, (d) equivalent viscous damping factor
14
It must be noted that the four shear panels exhibited unexpected slipping phenomena for very
small shear forces, due to the presence of internal residual stresses produced by the welding
process of stiffeners. In fact, some experimental tests reported in [De Matteis et al., 2012]
proved that these residual stresses could be comparable to the nominal yield stress, and could
provoke hardening effects leading to an increase of yielding strength of about 10 MPa, which
corresponds to about 50% of the original value for not-welded specimen. However, it is to
pointed out that the detrimental effects due to residual stress does not influence significantly
the hysteretic response for medium-high shear demands. As for the “full-bay” typology, also
for the studied “bracing type” shear panels the proposed experimental tests served to develop
FEM numerical models. These models have been profitably used to check the stress states, the
deformed shapes and the main resistant mechanisms of the analysed shear panel typology for
different shear strain demands. Moreover, the obtained numerical models have been used in
order to implement parametrical numerical analyses and to provide design curves and
formulations [Brando and De Matteis, 2014].
300 8.00 "type 3"
"type 3" "type 4"
7.00
250 Poli. ("type 3")
Hardening ratio 0.2
"type 4"
6.00
Cumulated dissipated
Poli. ("type 4")

200
Energy[kNm]
Poli. ("type 3") 5.00

150 Poli. ("type 4") 4.00
3.00
100
2.00
50 1.00
0 0.00
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
a) b)
"type 3"
50 0.6
"type 3"
Equivalent viscous damping factor
"type 4"
45
Poli. ("type 3")
0.5
Secant Stiffness[kN/mm]
40 "type 4"
Poli. ("type 4")
35 Poli. ("type 3") 0.4
30
Poli. ("type 4")
25 0.3

20
0.2
15
10 0.1
5
0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
c) d)
Figure 11. Results for semi-compact panels “type 1” and “type 2”: (a) cumulated energy, (b)
hardening ratio, (c) secant shear stiffness, (d) equivalent viscous damping factor
3. BUCKLING INHIBITED PURE ALUMINIUM SHEAR PANELS (BIPASPS)

3.1 General
When low yield strength shear panels are used, one of the main critical issues is that very thin
plate are often necessary, even in presence of a base material characterized by a low yield stress
point. This could entail economical and technological counter-indications, as well as
technological problems related to an excessive use of weld for the transversal ribs. As a
15
convenient alternative to the use of stiffened shear panels, recently, the authors proposed a new
type of shear panel, based on the concept of buckling phenomena inhibition [Brando et al.
2013].
In particular, two different technological solutions for restraining the out-of-plane
deformations of the system have been proposed. They are based on the use of not connected
steel elements, able to restrain the first and more important critical modes of the panel. These
elements react only in the direction perpendicular to the base plate, leaving the base plate work
according to a pure shear resistant mechanism.
In the following, the main results of an experimental campaign carried out on two selected
configurations are shown. The first is a partially buckling inhibited solution, for which the out-
of-plane displacement that could develop along the two diagonal of the panel are restrained.
The latter is a totally buckling inhibited shear panels, for which the out-of-plane displacements
are restrained for the whole plate. Both the investigated solution are conceived to work in a
bracing-type configuration. Moreover, in order to prove the effectiveness of the proposed
solution, the performance of the tested panel is compared with the one obtained downstream
the previous presented experimental activity carried out on bracing type pure aluminium shear
panels.
3.2 The experimental behaviour
In Fig. 12 the two proposed buckling inhibited shear panels are shown. They were obtained by
inserting a 5 mm thick pure aluminium shear plate in the same square articulated steel frame
used for the experimental tests carried out on the PASP typology. For the first solution (p-BISP,
acronym of “partially Buckling Inhibited Shear Panels”), two cross shape 10mm thick/140 mm
wide steel elements were used for inhibiting the first four critical modes of the base plate. These
were arranged on both side oF the base plate along the two diagonal.
In the second technological solution (“t-BISP”, acronym of “totally Buckling Inhibited Shear
Panels”), two octagonal steel plates were mounted for restraining the out-of-plane
displacements of the whole plate in shear.
In both cases, lexan sheeting were employed in order to reduce the friction between the parts.
The hysteretic behaviour of the two panels are shown in Fig. 13. The obtained large hysteretic
cycles prove the high dissipative capacity of the devices, guaranteed, also for high shear
demands, by a substantial absence of significant pinching effects. However, it is to be
underlined that the t-BIPSP configuration behaved in a more performing way, as it was not
influenced by the secondary buckling phenomena that, contrarily, developed for the p-BIP
solution. Furthermore, larger cycles given from the t-BIP configuration were due to the
confinement effects produced by the more extended contact of the restraining plates. The
panels behaviour can be considered approximately elastic up to a diagonal displacement of
±0.50 mm (shear strain of 0.1%). After this threshold, an inelastic behaviour was registered.
The first buckling phenomena slightly arose for a shear strain demand of ±0.66% (diagonal
displacement of ±3.00 mm) on the not inhibited portions of the p-BIP configuration.
Nevertheless, these did not lead to any significant effects on the revealed hysteretic cycles. On
the contrary these instabilities were more eye-catching for a shear strain of 2.20%, (Fig. 14.a),
starting to influence the hysteretic response with slight pinching effects (Fig. 14.b) after the
first cycles. On the other hand, at this shear demand, no particular degrading phenomena were
measured for the t-BIP configuration and the corresponding hysteretic cycles did not present
any particular detrimental effects.
16
a) b)
Figure 12. The studied (a) Partially Buckling Inhibited Panel (p-BIP) and (b) Totally Buckling
Inhibited Panel (t-BIP)
80 80
60 60
Shear Stress (MPa)
Shear Stress (MPa)
40 40
20 20
0 0
‐2 0 ‐2 0
‐4 0 ‐4 0
‐6 0 ‐6 0
‐8 0 ‐8 0
‐1 8 ‐1 5 ‐1 2 ‐9 ‐6 ‐3 0 3 6 9 12 15 18 ‐1 8 ‐1 5 ‐1 2 ‐9 ‐6 ‐3 0 3 6 9 12 15 18
Shear Strain (% ) Shear Strain (% )

a) b)
Figure 13. The obtained hysteretic cycle: a) “p-BIP” and b) “t-BIP”
When a shear strain demand of ±4.40% (diagonal displacements of ±20 mm) was attained,
relevant buckle waves were noticed in the extent of the not restrained plate portions of panel
type “p-BIP” (Figure 15.a). Nevertheless, the shape of the hysteretic cycles was not still
influenced by these phenomena (Figure 15.b), this proving that the influence of the higher
critical modes is less important for higher shear demands.
80
60
Shear Stress (MPa)
40
20
‐20
‐40
‐60
‐80
‐2.5 ‐2.0 ‐1.5 ‐1.0 ‐0.5 0.0 0.5 1.0 1.5 2.0 2.5
a) Shear Strain (%) b)

Figure 14. “p-BIP” experimental response for a 2.20% shear strain demand: a) experimental
evidences and b) hysteretic cycles
17
80
60
Shear Stress (MPa)

40
20
‐20
‐40
‐60
‐80
‐5 ‐4 ‐3 ‐2 ‐1 0 1 2 3 4 5

Figure 15. “p-BIP” experimental response for a 4.40% shear strain demand: a) experimental
80
60
Shear Stress (MPa)
40
20
‐20
‐40
‐60
‐80
‐5 ‐4 ‐3 ‐2 ‐1 0 1 2 3 4 5

Figure 16. “t-BIP” experimental response for a 4.40% shear strain demand: a) experimental
At the same shear strain level, the first buckling were observed on the totally inhibited shear
plate. They arose on the small not restrained portion of the plate closed to the lateral connection
(Fig. 16a), provoking a slight pinching of the hysteretic cycles (Fig. 16b).
Some cracks developed at the four corners of the base plates when diagonal displacements of
±30 mm (shear strain of 6.74%) were imposed (Fig. 17). These were due to a sort of “clamp”
action of the inhibition plates on the two faces of the devices which, being pushed in the out-
of-plane direction at their centre tended to go toward each other at their ends.
These failure become important for a displacement of ±40.00 mm (shear strain of 9.04%), when
the two specimens presented the maximum shear strength, as it can be observed in Fig. 16. The
revealed collapse mechanisms are reported in Fig. 18.
a) b)
Figure 17. Local failure at the vertex of the systems for a) “p-BIP” and b) “t-PIB”
18
a) b)
Figure 18. Collapse modes of a) “p-BIP” and b)” t-BIP”
3.3 Performance of the studied shear panels
Based on the definition given in Fig. 19f, global response parameters obtained by the
experimental results are shown. Moreover, these are compared with the analogous results
obtained by the experimental of the bracing type pure aluminium shear panels “type 1” (the
unstiffened less performing among the testes BTPASPs) and “type 3” (the optimal solution
among the testes BTPASPs), which have been dealt with in Section 2.
In particular, in order to compare the dissipative capacities, an “energy efficiency factor”  has
been introduced (Fig. 19a). This parameter represents the ratio between the area of the actual
hysteretic cycle get for the generic shear strain and the area of an ideal cycle obtained, for the
same demand, by reversing an ideal bilinear elasto-plastic curve. The adopted bilinear
relationship, which is considered as the potential optimal response for the proposed devices,
has the first elastic branch characterized by the slope of the unloading branch of the cycle of
the real hysteretic response and is characterized by a post-elastic stiffness k2 (Fig. 19b) read as
the slope of the tangent of the envelope curve of each experimental cycle.
It is possible to observe that the tested panels with buckling restrained systems (both types “p-
BIP” and “t-BIP”), have an energy dissipation performance better than the one associated to
previously tested samples. In particular “t-BIP” shear panel gives  factor higher than 0.9 for
shear strains ranging between 1.5% and 6.5%, therefore behaving as a fully dissipative damper.
On the other hand, the “p-BIP” type with partial buckling inhibition device reaches a value
=90% only for a 2.2% shear strain demand; for larger strain demands, the dissipative capacity
is about 85% of the one corresponding to the ideal bilinear behaviour.
A similar behaviour can be noticed for multi-stiffened shear panel BTPASP “type 3”, which,
indeed, provides  factor slightly lower than 85%. On the contrary, the unstiffened shear plate
“BTPASP type 1”, as underwent relevant detrimental effects due to buckling for shear strain
larger than 1,0%, evidences a significant decrease of the  factor.
In addition, the analysis of the equivalent viscous damping factor eq given in Fig. 19c, which
retrieves a measure the fractional part of the strain energy (Es) dissipated during each
deformation cycle, evidences that i) both buckling inhibited and multi-stiffened shear panels
allow increasing of about 1.3 times the dissipative capacity of the simple plate in shear and ii)
the specific energy dissipated by “BTPASP type 3”shear panel is larger than the one offered
by the buckling inhibited plates for very large shear strain demands, namely greater than 4%
and 6.5% for p-BIP and t-BIP, respectively. With reference to the last comment, it is to be
pointed out that although shear panel “t-BIP” presents always larger hysteretic cycles with
respect to other panels, the lower eq factor registered for very large shear deformation is due
to the fact that the strain energy is higher as well. This is justifiable because of the higher
19
strength caused by the confinement adjustment effects which has been previously discussed.
For a better comprehension of this phenomenon, in Fig. 19d the variation of hardening ratio h,
defined as the ratio between the maximum attained shear strength for each shear demand and
the conventional yielding one, is shown. From the same figure it is also evident that the “t-BIP”
solution and the “BTPASP type 3” shear panel provide a similar strength for medium and high
shear strain levels, whereas the stiffened plate is clearly weaker for smaller strain levels.
Finally, in Fig. 19e the response of the four tested aluminium panels are compared in terms of
effective secant stiffness.
a) b)
c) d)
e) f)
Figure 19. Global response parameters obtained by the experimental tests and comparison with
PASPs “type 1” and “type 3”: (a) Energy efficiency factor; (b) Post elastic stiffness; (c) Equivalent
viscous damping; (d) Hardening ratio;; (e) Secant stiffness; (f) definition of the main behavioural
parameters.
20
Also according to this parameter, the buckling inhibited shear panels were more performing
than the conventional ones (shear plate with welded stiffeners). In fact, a larger initial stiffness
is retrievable for low shear demands, while in the large strain field the obtained responses are
quite comparable to each other. This result is evidently due to the absence of residual stresses
and other imperfections which have to be ascribed to welding processes of multi-stiffened shear
panels.
4. PERFORATED SHEAR PLATES (PSPS)

4.1 Basis
An alternative fruitful way for obtaining dissipative shear panels may consist in weakening the
base plate by removing some of its parts. This type of operation allows to obtain a variation of
the internal stress pattern and potentially to mitigate the negative effects generated by possible
buckling phenomena. On the other hand it leads to a reduction of the shear strength, also in
presence of thick plates, so to accomplish more easily capacity design criteria. A significant
literature concerning the use of weakened shear panels has been proposed in the last decade.
For example, panels weakened by means of vertical slits were extensively investigated by
Hitaka and Matsui [2003] and Pohlenz [2010]; for this type of panel, the shear force acting on
the whole system is commuted in a bending mechanism for the plate portions confined by a
couple of slits, which undergo large flexural deformations producing a significant dissipative
capacity. Vian, Bruneau and Purba [2009] carried out experimental tests on shear panels
weakened by holes arranged in a staggered configuration, providing design formulations of the
shear strength accounting for the holes diameters and spacing, also considering the influence
of the surrounding frame stiffness.
Valizadeh et al. [2012] quantified the influence of the holes diameter on the loss of dissipated
energy due to the pinching effects on the hysteretic cycles. Furthermore, they highlighted some
brittle failures around the perforations when very thin plates are adopted, due to the strong
concentration of stresses.
Alavi and Nateghi [2013] proved, by experimental tests on 1:2 scaled single-story SPSWs, that
perforated diagonally stiffened shear panels allow to obtain the same stiffness of un-ribbed
solid panels, with an increase of ductility of more than 14%. In addition, an extension of the
design formulation of the shear strength given previously by other Authors was provided,
accounting for the diagonal stiffeners contribution.
In this framing of research, in the following, the outcomes of two full scale tests carried out on
steel perforated shear plates are shown. The main goal is to highlight the most influencing
experimental evidences, with particular regard to those phenomena that could reduce the
dissipative capacity of the studied devices.
3.2 The experimental tests

The two tested perforated shear panels, namely PSP1 (Perforated Shear Panel 1) and PSP2
(Perforated Shear Panel 2), were made of 2.5 mm steel thick plates. Their geometric features
are described in Figs. 20a and 20b, where the sizes are expressed in mm. They are characterized
by nine perforations arranged in a rectangular pattern, which usually allows better
performances of a staggered configuration.
Each PSP was connected to the same perimeter articulated frame described for the other panel
typologies dealt with in previous Sections. The plate-to-perimeter frame connections were
21
realized again by 8.8 grade M14 steel friction bolts spaced by a pitch of 50 mm. In addition, in
order to increase the contact area between the plate and the built up members, double sided
internal 10 mm thick plates (two for each edge of the articulated frame) were applied, as it can
be seen in Fig. 20.c where the panel is shown during the assemblage process. The experimental
set-up was completed by two hinged steel jigs connecting two opposite vertices of the panel to
the MTS machine used for carrying out cyclic tests (Fig. 20d).
Uniaxial tensile tests on dog bone elements were performed in order to investigate the base
material mechanical properties of the plates, showing that the yield stress measured in the
lamination direction (about 300 MPa) differs from the one considered perpendicularly (about
270 MPa), whereas no significant differences were revealed in terms of tangential stiffness and
ductility. An anisotropic behaviour at yielding was therefore expected.
a) b) c) d)
Figure 20. The experimental set-up: The plates geometry of a) PSP1 and b) PSP2 panels; one of the
specimens (a) during the assemblage process and (b) during the tests
The two tested perforated shear panels were subjected to pseudo-static cyclic tests, considering
a diagonal displacements history consistent with the ECCS-CECM Provisions [1985]. The
testing apparatus was composed by a mechanical transducer for the measurement of the
diagonal displacements (Fig. 20d), by the loading cell installed in the testing machine, as well
as by four mechanical LVDT transducers to measure the potential relative motion between the
plate edges and the frame elements.
3.3 The obtained experimental evidences

Following an initial phase in which the panels worked in pure shear, the former “global”
buckling waves arose in the plate cores centred around the two diagonals, when diagonal
displacements of ±5mm (shear strain of ±1.1%; Fig. 21a) and ±2mm (shear strain of ±0.8%;
Fig. 21c) were attained, for PSP1 and PSP 2 respectively. Indeed, the amplitude of the diagonal
buckling waves described above did not change for the remaining part of the tests. In other
words, the applied perforation patterns resulted to be truly effective in modifying the internal
stresses and, therefore, in mitigating buckling effects. Nevertheless, for larger diagonal
displacements, the activation of higher critical modes was noticed. These modes consisted in
lateral-torsional buckling of the panel portions included within the perforations and were due
to not constant stresses that produced bending moment in the plane of the plate. These
phenomena were clearly visible for a diagonal displacement of 10mm (shear strain of 2.2%)
for both PSP1 and PSP2 specimens, as it is shown in Figs. 21b and 21d The main effects
consisted in a twisting of the buckled plates portions, with damage significantly cumulated
when the number of cycles increased.
22
a) c) c) d)
Figure 21. (a) the shear panel PSP1 at a displacement of ±5.00 mm (shear strain of ±1.1%) aand (b)
10.00 mm (shear strain of ±2.2%); the shear panel PSP2 at a displacement of ±2.00 mm (shear strain
of ±0.4%) and (b) 10.00 mm (shear strain of ±2.2%)
For diagonal displacement of 20 mm (shear strain of 4.4%), the local twisting described above
became permanently visible, as shown in Figs 22a and 22b. Finally, for higher shear strains,
low cycle fatigue provoked some tears closed the perimeter of the holes, provoking some losses
of shear strength for diagonal displacements higher than 30 mm (shear strain of 6.7%). In Figs.
22c and 22d the collapses evidenced at a diagonal displacement of 40 mm (shear strain of 9.6%)
are shown. The obtained cyclic responses of the two tested shear panels are plotted in Fig. 23
in terms of diagonal displacements vs. diagonal forces. As it can be observed, conspicuous
pinching effects were revealed in both the analysed cases.
a) b) b) d)
Figure 22. The experimental evidences registered for the specimens (a) PSP1 and (b) PSP2 at a
diagonal displacement of 20 mm (shear strain of ±4.4%). The collapse modes of the (a) PSP1 and (b)
PSP2 specimens
100 125
75 100
75
50
Diagonal Force (kN)
Diagonal Force (kN)
50
25
25
0 0
-25 -25
-50
-50
-75
-75
-100
-100 -125
-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60
Diagonal Displacement (mm) Diagonal Displacement (mm)
a) b)
Figure 23. The obtained hysteretic cycles for specimen (a) PSP1 and (b) PSP2
23
These were due to several detrimental phenomena, caused by cumulated plastic deformations
induced by the local buckling that led to a cyclic decay of the maximum strength for each shear
demand. This type of phenomena must be therefore mitigated, by applying a more stringent
control on the design of the panel portion included between two consecutive perforations.
The measured decay is highlighted in figure 24, where the maximum diagonal forces, measured
for each shear strain demand, are plotted. Moreover, in the same figure, the percentage
differences between the maximum and the minimum strength (normalised to the maximum
one) at each cycle is given.
The analysis of the obtained results put in evidence some significant outcomes; for instance, it
has been noticed that in order to recover the strength detriment achieved after three cycles at a
certain level of strain demand, it is necessary to impose an increase of diagonal displacement
of more than 5 mm.
Also, it must be observed that the panel characterized by larger perforations (PSP1) presented,
apart from an expected reduced strength, a lower ductility (30mm for specimen PSP1 –about
6% shear strain- versus 40 mm of panel PSP2 –about 9% shear strain). Furthermore, it has been
found that the strengths measured for each shear demands, when the number of cycles
increases, are basically aligned on a straight line. This allowed to determine a close form
analytical formulation, given in Eq. (3), able to reproduce, with good approximation, the
measured experimental values of diagonal forces, as is it also shown in Fig. 24.
Fdiag=-0.00243+0.143-1.55Fy,diag-an (3)
In the above equation Fdiag is the diagonal force corresponding to the diagonal displacement ,
Fy,diag is the diagonal force corresponding to yielding, n is the number of cycles carried out, a
is a coefficient that relies on the panel geometry. For specimen PSP1 it has been found that a
is equal to 6.52, 5.42, 11.37, 12.81 and 13.79 for diagonal displacements of 5 mm, 10 mm, 20
mm, 30 mm and 40 mm. Instead, for specimen PSP2, a was assumed as 8.75, 10.45, 12.69,
13.87, 18.87 for diagonal displacements of 5 mm, 10 mm, 20 mm, 30 mm and 40 mm,
respectively.
140 140
=20% =17% =32% =35% =41% =17% =22% =24% =26% =31%
120 120
test results
eq. (1)
100 100
Diagonal Force [kN]
Diagonal Force [kN]
80 80
60 60
40 40
20 20
0 0
5 10 20 30 40 5 10 20 30 40
Diagonal Displacement [mm] Diagonal Displacement [mm]
a) b)
Figure 24. Experimental vs. Analytical (eq.1) strengths for specimen (a) PSP1 and (b) PSP2.
The above equation is valid under the hypothesis, not investigated during the tests, that after
three cycles, the strength decay can be considered as negligible. Finally, in order to measure
the loss of dissipative capacity due to pinching effects, the ratio Fpinc/ Fmax (averaged on the
three cycles performed for each shear strain demands) of the diagonal force corresponding to
a zero displacement (Fpinc) by the maximum diagonal force (Fmax) measured on each cycle is
given in Tab 2.
24
Table 2. The average value of the Fmax/ Fpinc ratios for several diagonal displacement demands
Fpinc/Fmax (-)
Tested Specimen for each hysteretic cycle at displacement (mm)=
5 10 20 30 40
PSP 1 0.28 0.31 0.25 0.20 0.16
PSP 2 0.27 0.32 0.21 0.17 0.15
This parameter can be seen as a measure of the potential dissipative capacity that is (ideally)
maximum when a unitary value is attained. As it can be observed, the loss of dissipative
capacity is significant.
5. CONCLUSIONS
In this paper, experimental tests on three types of metal shear panels subjected to shear forces,
to be used as dampers, have been presented. The investigated metal shear panels have been
conceived in order to maximize their dissipative function. To this purpose, different strategies
adopted both for anticipating their inelastic response when low shear strains are demanded and
for mitigating possible detrimental effects due to buckling phenomena have been proposed.
In particular, for the first panel typology, named with the acronym PASP and proposed
according to both “full bay” and“bracing type” configurations, this goal has been achieved by
adopting the aluminium alloy EN-AW 1050A H24, an almost pure aluminium, which, when
subjected to an annealing process, presents a very low yield stress and a very high ductility.
The obtained experimental results proved that this type of solution could be particularly
performing if an adequate number of transversal stiffeners is adopted. Nevertheless, the
observation on the experimental evidences highlighted that some detrimental effects on the
hysteretic response could arise for low shear strain demands due to the negative influence of
residual stresses and imperfections provoked by the welding processes. Furthermore, for the
“full bay” solution, some parasitic effects have been noticed which can be substantially reduced
in the case of the “bracing type” configuration.
The second type of shear panel, namely BISP (Buckling Inhibited Shear Panel), has been
conceived with additional steel elements able to restrain the out-of-plane displacements of the
base plate, or at least of its parts that are more susceptible to the first and more important critical
modes. It presented a very suitable dissipative response, resulting even more performing than
traditional stiffened shear panels. A technological problem that could arise with the BISPs
typology regards the thickness to be adopted for the restraining elements, that should properly
designed depending on the base plate thickness and, therefore, to the expected shear demand.
Finally a steel perforated shear plate (PSP) has been proposed. The concept on which this
solution is based is that the applied perforations lead to a reduction of the shear strength, also
in presence of thick plates, and allows to obtain a variation of the internal stress pattern that
could mitigate the negative effects generated by possible buckling phenomena. Indeed the
experimental tests put in evidence that, although the main buckling phenomena that usually
influence the response of shear plates, could be conveniently controlled, nevertheless
detrimental effects could be produced by the local buckling of the plate portions included
between two consecutive perforations. For this reason, design formulas relating the holes
diameter with the plate thickness are necessary.
25
In the whole, the presented experimental investigation allows to outline the following main
conclusions:
 the three tested panel typologies represent very interesting and performing solutions to be
used as dampers;
 design formulations should be provided in the very next future for their use, also to control
the possible negative effects that could be provoked by possible poor technological details;
 it is necessary to develop a unitary framework that allows the designer to have the
possibility of selecting the most convenient panel typology and details according to the
design requirements.
6. ACKNOWLEDGEMENTS
This study was founded in the framing of the ReLUIS Italian research project.
7. REFERENCES
Alavi, E., Nateghi, F. [2013] “Experimental study on diagonally stiffened steel plate shear walls with
central perforation”. Journal of Constructional Steel Research, n.89, pp. 9–20.
Bosco, M., Marino, E.M., Rossi, P.P. [2015]. “Modelling of steel link beams of short, intermediate or
long length”. Engineering Structures, 84:406–418, ISSN: 0141-0296,
doi:10.1016/j.engstruct.2014.12.003
Brando, G., D'Agostino, F., De Matteis, G.[2013]. “Experimental tests of a new hysteretic damper made
of buckling inhibited shear panels”. Materials and Structures/Materiaux et Constructions 46 (12)
PP. 2121 – 2133.doi: 10.1617/s11527-013-0040-6.
Brando, G., De Matteis, G. [2014]. “Design of low strength-high hardening metal multi-stiffened shear
plates”. Engineering Structures 60 PP. 2 - 10 doi: 10.1016/j.engstruct.2013.12.005
Brando, G., D'Agostino, F., De Matteis, G. [2015] “Seismic performance of MR frames protected by
viscous or hysteretic dampers”, Structural Design of Tall and Special Buildings, 24 (9), pp. 653-671.
Christopoulos, C., Filiatrault, A. [2007] Principles of Passive Supplemental Damping and Seismic
Isolation. IUSS Press, Pavia. ISBN 88-7358-037-8.
De Matteis, G., Mazzolani, F.M., Panico, S. [2007]. “Pure aluminium shear panels as dissipative devices
in moment-resisting steel frames” Earthquake Engineering and Structural Dynamics, 36 (7), pp.
841-859.
De Matteis, G., Formisano, A., Panico, S., Mazzolani, F.M. [2008]. “Numerical and experimental
analysis of pure aluminium shear panels with welded stiffeners”. Computers and Structures, 86 (6),
pp. 545-555.
De Matteis, G., Brando, G., Panico, S., Mazzolani, F.M. [2009] “Bracing type pure aluminium stiffened
shear panels: An experimental study” Advanced Steel Construction, 5 (2), pp. 106-119.
De Matteis, G., Brando, G., Mazzolani, F.M. [2012]. “Pure aluminium: An innovative material for
structural applications in seismic engineering”. Construction and Building Materials, 26 (1), pp.
677-686.
EN1999-1-1:2007 [2007]. EUROCODE 9, “Design of Aluminium structures”.
Formisano, A., Mazzolani, F.M., Brando, G., De Matteis, G. [2006] “Numerical evaluation of the
hysteretic performance of pure aluminium shear panels”. Proceedings of the 5th International
Conference on Behaviour of Steel Structures in Seismic Areas - Stessa 2006, pp. 211-217.
26
Hitaka, T., Matsui, C. [2003]. “Experimental study on steel shear wall with slits” Journal of Structural
Engineering, 129 (5), pp. 586-595.
Höglund T. [1997]. “Shear buckling resistance of steel and aluminium plate girders”, Thin-Walled
Structures, Vol.29 (1-4), pp 13-30.
Pohlenz, A. J., [2010] “Development of Steel Slit Wall Dampers with Embedded Condition Assessment
Capabilities”, Ph.D. thesis, University of Kyoto, Japan.
Giugliano, M.T., Longo, A., Montuori, R., Piluso, V. [2014] “Plastic design of CB-frames with reduced
section solution for bracing members “ Journal of Constructional Steel Research 66 (5), pp. 611-
621
Longo, A., Montuori, R., Piluso, V. [2009] “Seismic reliability of chevron braced frames with
innovative concept of bracing members” Advanced Steel Construction 5(4), pp. 367-389.
Montuori, R., Nastri, E., Piluso, V. [2014a] “Theory of plastic mechanism control for eccentrically
braced frames with inverted y-scheme” Journal of Constructional Steel Research 92, pp. 122-135
Montuori, R., Nastri, E., Piluso, V. [2014b] “Theory of plastic mechanism control for the seismic design
of braced frames equipped with friction dampers” Mechanics Research Communications 58, pp.
112-123
Nakashima, M. [1995]. “Strain-hardening behavior of shear panels made of low-yield steel. I: Test”.
Journal of Structural Engineering (United States), 121 (12), pp. 1742-1749.
Shishkin, J.J., Driver, R.G., Grondin, G.Y. [2009] “Analysis of steel plate shear walls using the
modified strip model” Journal of Structural Engineering, 135 (11), pp. 1357-1366
Soong, T. T, Dargush, G. F. [2007] Passive Energy Dissipation Systems in Structural Engineering, John
Wiley & Sons, Inc, New York. ISBN-13: 9780471968214.
Valizadeh, H., Sheidaii, M., and Showkati, H. [2012] “Experimental investigation on cyclic behaviour
of perforated steel plate shear walls.” Journal of Constructional Steel Research. 70 308-316.
Vian, D., Bruneau, M., Purba, R. [2009]. “Special perforated steel plate shear walls with reduced beam
section anchor beams. II: Analysis and design recommendations” (2009) Journal of Structural
Engineering, 135 (3), pp. 221-228.
27
Num. …
LASTRE FORATE A TAGLIO DI ACCIAIO PER LA PROTEZIONE

SISMICA DI EDIFICI
Gianfranco De Matteis1, Giuseppe Brando2*

1
Department of Architecture and Industrial Design “Luigi Vanvitelli”,
Second University of Naples, Italy
2
Department of Engineering and Geology, University “G. D’Annunzio”, Chieti-Pescara, Italy
SOMMARIO: Il presente lavoro riporta i risultati principali di una attività sperimentale e

numerica effettuata al fine di indagare la capacità isteretica fornita da sistemi dissipativi
basati sull’uso di lastre forate a taglio. Dopo una breve introduzione, in cui i concetti di base
che hanno portato a considerare le piastre forate come soluzione conveniente per la protezione
sismica delle strutture, sono descritte le prove sperimentali effettuate su due prototipi. Esse
hanno messo in evidenza la possibile presenza di criticità del sistema a causa del prematuro
sviluppo di fenomeni di instabilità locale che potrebbero influenzare negativamente la risposta
dissipativa del sistema. Infine, sono riportati i risultati principali di un'analisi parametrica,
basata su modelli numerici agli elementi finiti. Tali risultati sono utilizzati per introdurre
formule di progetto utili per la selezione ottimale della geometria di perforazione della lastra
da scegliere in funzione della domanda strutturale prevista.
___________________
*Corresponding author: Giuseppe Brando, Department of Engineering and Geology, University “G. D’Annunzio”,
Chieti-Pescara, Italy
Email: gbrando@unich.it
Num. …
EFFETCS OF CLASS B SITE ON THE SEISMIC RELIABILITY OF

BASE-ISOLATED STEEL SYSTEMS
Paolo Castaldo*, Guglielmo Amendola, Bruno Palazzo
Department of Civil Engineering, University of Salerno, Italy
SUMMARY: Seismic reliability of steel structures isolated using the frictional pendulum
system bearings and subjected to artificial earthquake ground motions is studied herein. The
superstructure is idealised as a linear shear-type flexible building as well as the FPS devices
are described by adopting a widespread model which considers the variation of the friction
coefficient with the velocity. The uncertainty affecting both the seismic inputs, modelled as
non stationary random processes within the power spectral density method, and the friction
coefficient at large velocity is considered through appropriate probability density functions.
Incremental dynamic analyses are developed in order to evaluate the fragility curves related
to both superstructure and isolation level. Finally, considering the seismic hazard curve
related to a site near Sant’Angelo dei Lombardi (Italy), the seismic reliability of the overall
steel system is evaluated.
KEYWORDS: Seismic isolation, steel superstructure, power spectral density method, Monte
Carlo simulations, Latin hypercube sampling method, seismic reliability
1. INTRODUCTION
In the last decades, seismic dissipation and isolation through friction pendulum system (FPS)
0[Christopoulos and Filiatrault, 2006; Zayas et al., 1990] have emerged as very effective
techniques for the protection of building frames which, even if designed according to the
most advanced codes, suffer severe damages under strong earthquake events [De Iuliis et al.,
2010; Bonessio et al., 2012]. Over the years, many works have developed new design
strategies and methodologies [Castaldo, 2014; Castaldo and De Iuliis, 2014; Castaldo et al.
2016d; Castaldo et al. 2016e; De Iuliis and Castaldo, 2012; D'Ayala and Benzoni, 2012;
Longo et al., 2012; Palazzo et al. 2014b; Palazzo et al., 2015] as well as probabilistic analyses
in structural dynamics, structural reliability methods, and reliability-based analysis have been
presented [Giugliano et al., 2011a; Giugliano et al., 2011b; Longo et al., 2009]. Barroso and
Winterstein [Barroso and Winterstein, 2002] evaluated the seismic performance of steel
buildings isolated with FPS bearings by taking into account the variability of both the seismic
intensity and the record characteristics. Seismic reliability analyses of 2D and 3D systems
isolated by FPS bearings have been carried out in [Castaldo et al. 2015; Palazzo et al., 2014a;
Castaldo et al. 2016a; Castaldo et al. 2016b; Castaldo et al. 2016c] by accounting for the
randomness of both the isolator properties (i.e., coefficient of friction) and of the earthquake
main characteristics by defining a reliability criterion to assist the design of the isolator
dimensions in plan. In [Castaldo and Tubaldi, 2015], the influence of FPS bearing properties
and of the structural parameters on the seismic performance of base-isolated structures
___________________
*Corresponding author: Paolo Castaldo, Department of Civil Engineering, University of Salerno, Italy.
Email: pcastaldo@unisa.it
P. Castaldo et al.
through the nondimentionalitation of the motion of equations is analyzed by providing useful

results for seismic reliability analyses. In [Castaldo and Ripani, 2016], the optimal values of
the friction coefficient are defined as a function of the system properties and of the soil
condition in order to minimize the superstructure response in terms of displacement relative
the base mass.
In this work, seismic reliability of steel structures isolated using the frictional pendulum
system (FPS) bearings and subjected to artificial earthquake ground motions is studied herein
by considering both the friction coefficient and soil characteristics as random variables. The
superstructure is idealised as a linear shear-type flexible building as well as the FPS devices
are described by adopting a widespread model which considers the variation of the friction
coefficient with the velocity [Mokha et al., 1990]. The uncertainty affecting both the seismic
inputs, modelled as non stationary random processes within the power spectral density
method [Pinto et al., 2004; Shinozuka and Deodatis, 1991; Talaslidis et al., 2004], and the
friction coefficient at large velocity is considered through appropriate probability density
functions. Incremental dynamic analyses are developed in order to evaluate the probabilities
exceeding different limit states related to both superstructure and isolation level. According
to the PEER-like modular approach [Cornell and Krawinkler, 2000], in the final part of the
work, considering the seismic hazard curve related to a site near Sant’Angelo dei Lombardi
(Italy) [NTC08, 2008], regarding a structural steel system isolated by FP bearings with a
design life of 50 years, reliability curves are derived.
2. EQUATION OF MOTION OF BASE-ISOLATED STRUCTURES

The equation of motion governing the response of a single-degree-of-freedom (SDOF)
system on single concave FPS isolation devices to the seismic input ug ( t ) is:
ms ⋅ us ( t ) + cs ⋅ us ( t ) + k s ⋅ us ( t ) =− ms ⋅ ug ( t ) + ub ( t ) 
(1)
mb ⋅ ub ( t ) + fb ( t ) + cb ⋅ ub ( t ) − cs ⋅ us ( t ) − ks ⋅ us ( t ) = −mb ⋅ ug ( t )
where 𝑢𝑢𝑠𝑠 denotes the displacement of the superstructure relative to isolation bearing, 𝑢𝑢𝑏𝑏 the
isolator displacement relative to the ground, 𝑚𝑚𝑠𝑠 and 𝑚𝑚𝑏𝑏 respectively the mass of the
superstructure and of the base, 𝑘𝑘𝑠𝑠 and 𝑐𝑐𝑠𝑠 respectively the superstructure stiffness and inherent
viscous damping constant, 𝑐𝑐𝑏𝑏 the bearing viscous damping constant, 𝑢𝑢̈ 𝑔𝑔 (𝑡𝑡) the ground motion
input, the dot differentiation over time, and 𝑓𝑓𝑏𝑏 (𝑡𝑡) denotes the FPS bearing resisting force, that
is expressed as:
fb ( t ) = kb ⋅ ub ( t ) + µ ( ub )( m + mb ) gZ ( t ) (2)
where = kb ( ms + mb ) g / R , g is the gravity constant, R is the radius of curvature of the FPS,
µ ( ub ( t ) ) the coefficient of sliding friction, which depends on the bearing slip velocity ub ( t ) ,
and Z ( t ) = sgn ( ub ) , where sgn(∙) is the sign function. Experimental results [Constantinou et al.,
1990; Constantinou et al., 2007; Mokha et al., 1990] suggest that the coefficient of sliding
friction of Teflon-steel interfaces obeys to the following equation:
µ ( ub ) = f max − Df ⋅ exp ( −α ub ) (3)
in which f max represents the maximum value of friction coefficient attained at large velocities
of sliding, f=
min f max − Df represents the value at zero velocity. By dividing Eqn. (1a) by ms ,
30
and Eqn. (1b) by mb , and introducing the mass ratio

= γ ms / ( ms + mb ) [Kelly, 1997], Eqn. (1)
can be rewritten as:
us ( t ) + 2ξ sωs us ( t ) + ωs2us ( t ) =
−  ag ( t ) + ub ( t ) 
1 γ γ
(4)
ub ( t ) +  2ξbωb ⋅ ub ( t ) + ωb2 ⋅ ub ( t ) + µ ( ub ) g sgn ( ub )  − 2ξ sωs
 ⋅ u ( t ) − ωs2 ⋅ u ( t ) = −ug ( t )
1− γ 1− γ 1− γ
where the parameters ωs and ξ s denote the superstructure circular frequency and damping
factor, whereas the parameters =ωb kb / ( ms + m=
b) g / R and ξb denote the fundamental
circular frequency and damping factor for a rigid mass ( ms + mb ) on a linear frictionless
isolator of stiffness kb and viscous damping constant cb .
3. RANDOM VARIABLES
Seismic reliability assessment of a steel building structure, according to the structural
performance (SP) evaluation method [Bertero and Bertero, 2002], is based on the coupling
between structural performance levels [Vision, 2000] and associated exceeding probabilities
during its design life [Saito et al., 1998]. According to [Cornell and Krawinkler, 2000; Porter,
2003], the uncertainties related to the seismic input intensity are separated from those related
to the characteristics of the record (record-to-record variability) by introducing a scale factor,
i.e., an intensity measure (IM). The approach is based on calculating the probabilities of
exceeding different limit state thresholds given different values of the intensity measure with
the aim to define the fragility curves of the system. Afterward, the abovementioned fragility
curves integrated with the seismic hazard curve, expressed in terms of the same IM, related to
a reference site, lead to the mean annual rates of exceeding the limit states. Using a Poisson
distribution, it is possible to transform the mean annual rates of exceeding the limit states into
probabilities of exceedance in the time frame of interest (e.g., 50 years). The aim of this work
consists of evaluating the seismic reliability of steel structural systems equipped with friction
pendulum isolators (FPS) considering both the friction coefficient and earthquake
characteristics as random variables.
In order to evaluate the record-to-record variability of the structural system response, several
artificial earthquake excitations have been considered. In particular, each earthquake
excitation can be modeled as a Gaussian stationary process with mean value equal to zero and
two-sided power spectral density (PSD) function S ff (ω ) . It follows that the stochastic process
f (t ) can be simulated by the following series as N → ∞ :
𝑓𝑓(𝑡𝑡) = √2 ∑𝑁𝑁−1
𝑛𝑛=0 𝐴𝐴𝑛𝑛 𝑐𝑐𝑐𝑐𝑐𝑐(𝜔𝜔𝑛𝑛 𝑡𝑡 + Φ𝑛𝑛 ) (5)
where
= An (2 S ff (ωn )∆ω ) 1/2
, ω n = n∆ω for n = 0 N − 1 , ∆ω = ω u / N , having assumed
T0 2π / ∆ω =31.25s (NTC08) and ω u = 50rad / s , Φ 0 , Φ 1 , Φ 2 ,......, Φ N −1 are independent random
=
phase angles distributed uniformly over the interval [0, 2π]. A sample function f (i ) (t ) of the
simulated stochastic process f (t ) can be obtained by replacing the sequence of random
phase angles Φ 0 , Φ1 , Φ 2 ,......, Φ N −1 with their respective i-th realizations Φ i0 , Φ1i , Φ i2 ,....., Φ iN −1 ,
sampled through Monte Carlo simulations. In this study, 50 sequence of random phase angles
are sampled through Monte Carlo simulations in order to generate 50 input accelorometric
signals. The power spectral density function (PSD) of the embedded stationary process,
31
P. Castaldo et al.
described by the widely-used Kanai and Tajimi [Kanai, 1957; Tajimi 1960] and modified
according to Clough and Penzien [Clough and Penzien, 1993], applies:
ω g4 + 4ξ g2ω g2ω 2 ω4 (6)
= S f (ω ) ⋅ S0
(ω g2 − ω 2 ) + 4ξ g2ω g2ω 2 (ω 2f − ω 2 ) + 4ξ f2ω 2f ω 2
In the parametric study, with the aim to assume the uncertainty related to earthquake
characteristics in terms of soil dynamics parameters corresponding to site class B (according
to EC8 [CEN, 2004]), ω g and ξ g are modeled as random variables uniformly distributed,
respectively, in the intervals [3π,5π] (rad/sec) and [40%,60%] [Pinto et al., 2004; Talaslidis et
al., 2004], and sampled through Monte Carlo simulations. In order to obtain non-stationary
stochastic processes, a time-modulating function proposed by [Shinozuka and Sato, 1967] is
adopted. As regards the friction coefficient, the experimental data 0[Constantinou et al., 1990;
Constantinou et al., 2007; Mokha et al., 1990] on sheet type Teflon bearings, have pointed
out that friction is a complex phenomenon, not complying with the Coulomb friction law and
that several mechanisms contribute to its variability. In this study, a uniform density
probability function (PDF), ranging from 3% to 12%, has been assumed to model the sliding
friction at large velocity as random variable fmax. For the generation of the sampled values of
the friction coefficient fmax, within the stratified sampling techniques the Latin Hypercube
Sampling (LHS) method [Mckey et al., 1979] has been used. In particular, in the following
parametric study, 15 sampled values of the random variable fmax are employed and a ratio
fmax/fmin equal to 3, based on regression of experimental results, whereas the exponent α of
Eqn. (3) equal to 30 [Constantinou et al., 1990; Constantinou et al., 2007; Mokha et al., 1990]
are assumed.
4. INCREMENTAL DYNAMIC ANALYSES (IDA) RESULTS OF BASE-ISOLATED

STEEL STRUCTURES
Seismic reliability assessment of equivalent base-isolated systems is based on developing
incremental dynamic analyses (IDA) [Vamvatsikos and Cornell, 2002]. In this study, the
spectral displacement, SD(Tb, ξb) is assumed as intensity measure [Luco and Cornell, 2007].
Figure 1. IDA curves of the isolation level with γ=0.7, for R=4m
32
Figure 2. IDA curves of the superstructure with γ=0.7, for R=4m

The corresponding IM, hereinafter denoted as SD(Tb), in the IDA is considered ranging from
0m to 0.5m. The parameters ξb and ξs are assumed respectively equal to 0% and 2%, γ is
considered equal to 0.7, the radius R of the FPS is assumed to be equal 4m, the fixed-base
system period Ts is considered equal to 0.3s, 0.9s and 1.5s, respectively. The response
parameters us and ub are adopted as the engineering demand parameters (EDPs), assumed to
follow a lognormal distribution according to [Castaldo et al., 2015]. A lognormal distribution
can be fitted to the both response parameters (i.e., the extreme values of the EDPs), by
estimating the sample lognormal mean, 𝜇𝜇𝑙𝑙𝑙𝑙 (𝐸𝐸𝐸𝐸𝐸𝐸) , and the sample lognormal standard
deviation, 𝜎𝜎𝑙𝑙𝑙𝑙 (𝐸𝐸𝐸𝐸𝐸𝐸), through the maximum likelihood estimation method. Figs. 1-2 illustrate
the IDA results regarding both the isolation level response ub and the superstructure response
us. Each figure contains some surface plots, corresponding to different values of percentile
(50th, 84th and 16th). Fig. 1 shows the IDA results regarding the isolation bearings response ub.
The lognormal mean and dispersion decrease for lower values of Ts (high value of the
isolation degree [Palazzo, 1991]). Fig. 2 shows the IDA results regarding the superstructure
response us. The lognormal mean and dispersion slightly decrease for lower values of Ts. The
trend of the IDA curves is consistent with the results in terms of the median and response
percentiles obtained by [Barroso and Winterstein, 2002].
5. SEISMIC FRAGILITY OF BASE-ISOLATED STEEL STRUCTURES

With reference to performance levels of the steel superstructure, four discrete performance
levels or limit states (LS1,LS2,LS3,LS4), corresponding respectively to “fully operational”,
“operational”, “life safety” and “collapse prevention” are provided from Vision 2000 [Vision,
2000]. In accordance to [FEMA274, 1997; FEMA356, 2000; Castaldo et al., 2015], in Table
1, the LS1 and LS2 thresholds of fixed-base steel buildings as well as the corresponding
failure probabilities in a design life of 50 years are reported. The performance limit states for
the seismic fragility of base-isolated steel buildings, in accordance to NTC08 [NTC08, 2008]
provisions, have been defined by limiting the response of the lateral-load-resisting
superstructure system, IDI limits, to a fraction of the limits provided for designing
comparable fixed-base buildings. In particular, with regard to NTC08 [NTC08, 2008]
33
P. Castaldo et al.
provisions the limit response of the lateral-load-resisting superstructure system of a base-

isolated building is equal to about two-third of the IDI limit of a comparable fixed-base
building as also reported in Table 1. In the hypothesis of regular buildings, the Eqn. (7),
according to NTC08, is considered as relationship between the fixed-base building period and
its height H, and is employed to estimate the height H as the integer multiple of the inter-
storey height assumed equal to h=3m and, so, the corresponding total number of floors Nf.
3
Ts = 0.085 H 4 (7)
Table 1. Limit state thresholds for the superstructure [FEMA274, 1997; FEMA356, 2000; Castaldo et
al., 2015]
LS1 LS2
fully operational operational
Inter-story drift (ISD) index of fixed-base steel buildings 0.7% 1.5%
Inter-story drift (ISD) index of base-isolated steel buildings [NTC08, 2008] 0.47% 1.0%
pf (50 years) 5.0·10-1 1.6·10-1
For each 𝑇𝑇𝑠𝑠 = 2𝜋𝜋/𝜔𝜔𝑠𝑠 , assuming the building floor mass equal to ms,i = 1000 kNs2/m, for
i=1…Nf, it is possible to determinate the floor stiffness and vector Φ1 containing the floor
displacements of the first mode of the fixed-base structure normalized to the top floor
displacement. The base mass mb is assigned in order to respect the mass ratio γ [Kelly, 1997]:
Γ12 M s1 (8)
γ= Nf
∑ ms,i + mb
i =1
where Γ1 and M s1 represent respectively the participation factor and modal mass of the
fundamental mode of the fixed-base structure. It follows that the maximum absolute inter-
story drift of the 1st floor can be evaluated as us ,1,max = Γ1φ11us ,max , and this response parameter,
divided by the inter-storey height assumed equal to h=3m, corresponds to the overall
maximum interstorey drift index experienced over the different stories that controls the
performance of the superstructure.
Figure 3. Seismic fragility curves of the superstructure 1st floor with γ=0.7, for R=4m
34
At each value of the intensity measure IM, the probabilities pf exceeding different limit states
related to the superstructure have been numerically and fitted by a lognormal distribution. Fig.
3 shows the fragility curves regarding the superstructure 1st floor for three values of Ts (0.3s,
0.9s and 1.5s) The seismic fragility of the superstructure increases for higher values of Ts
(lower values of the isolation degree).
With reference to the performance levels of the isolation system, several different values for
the plan dimension of the isolator are considered. In Table 2, the limit state thresholds
assumed for the seismic fragility of the FPS isolation level are reported. Fig. 4 shows the
fragility curves regarding the isolation level for three values of Ts (0.3s, 0.9s and 1.5s) and the
different values of the limit state thresholds varying in the range LS1-LS9. The fragility
curves are plotted in logarithmic scale in the range between 100-10-6. The seismic fragility of
the isolation level increases for higher values of Ts (lower values of the isolation degree).
Table 2. Limit state thresholds for the isolation level
LS1 LS2 LS3 LS4 LS5 LS6 LS7 LS8 LS9
Maximum relative displacement [m] 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Figure 4. Seismic fragility curves of the isolation level with γ=0.7, for R=4m
6. SEISMIC RELIABILITY OF BASE-ISOLATED STEEL STRUCTURES

Considering a site near Sant’Angelo dei Lombardi (Italy) as the reference site, in Figure 5 the
seismic hazard curve, expressed in terms of the same IM=SD(Tb), related to the isolated
period (R=4m), is plotted according to NTC08. The abovementioned curve represents the
average values of the annual rate λ of exceeding the IM=SD(Tb) level. Integrating the fragility
curves related to both the superstructure and isolation level with the seismic hazard curve and
using a Poisson distribution, it is possible to evaluate the seismic reliability of the overall
steel system in the time frame of interest (50 years). The seismic reliability of the steel
superstructure increases for low values of Ts (high values of the isolation degree), as shown
in Figure 6.
35
P. Castaldo et al.
Figure 5. Seismic hazard curves related to a site near Sant’Angelo dei Lombardi (Italy)
Figure 6. Seismic reliability curves of the superstructure 1st floor for γ=0.7 and R=4m
Figure 7. Seismic fragility curves of the isolation level for γ=0.7 and R=4m
36
The seismic reliability of the isolation level decreases slightly depends on the values of Ts
(Figure 7). An exceeding probability of pf=1.5∙10-3 (related to the collapse limit state,
reliability index β=3 in 50 years) is achieved through a radius in plan r ranging from about
0.35 m to about 0.45 m depending on system properties. The results are consistent with those
obtained by Castaldo et al. [2016a] considering r.c. superstructures in ordinary conditions
[Etse et al., 2013; Etse et al., 2014; Etse et al., 2015; Etse et al., 2016; Mroginski et al., 2015;
Ripani et al., 2014; Vrech et al., 2015; Ripani et al. 2016] and L’Aquila as reference site.
7. CONCLUSIONS
This paper deals with the seismic reliability of steel systems equipped with friction pendulum
isolators (FPS) considering both the friction coefficient and earthquake characteristics as
random variables. The uncertainty in the seismic inputs is taken into account by considering a
set of artificial records related to class B site. Incremental dynamic analyses are developed to
evaluate the probabilities exceeding different limit states related to both superstructure and
isolation level for different structural properties.
In the final part, considering the seismic hazard curve related to a site near to Sant’Angelo dei
Lombardi (Italy), according to NTC08, and regarding steel structures isolated by FP bearings
with a design life of 50 years, reliability curves are derived.
8. REFERENCES
Barroso L.R., Winterstein S. [2002] “Probabilistic seismic demand analysis of controlled steel
moment resisting frame structures,” Earthquake Engineering and Structural Dynamics
31(12):2049–2066.
Bertero, R.D., Bertero, V.V. [2002] “Performance-based seismic engineering: the need for a reliable
conceptual comprehensive approach,” Earthquake Engineering and Structural Dynamics 31:627–
652.
Bonessio N., Lomiento G. and Benzoni G. [2012] Damage identification procedure for seismically
isolated bridges, Struct. Control Health Monit., 19:565–578.
Building Seismic Safety Council [1997] NEHRP commentary on the guidelines for the seismic
rehabilitation of buildings. Provisions (FEMA 274). Washington, DC.
Building Seismic Safety Council [2000] NEHRP commentary on the guidelines for the seismic
rehabilitation of buildings. Provisions (FEMA 356). Washington, DC.
Castaldo, P., [2014] Integrated Seismic Design of Structure and Control Systems. Springer
International Publishing: New York,. DOI 10.1007/978-3-319-02615-2.
Castaldo P., Amendola G., & Palazzo B. [2016a] Seismic fragility and reliability of structures isolated
by friction pendulum devices: Seismic reliability-based design (SRBD), Earthquake Engineering
and Structural Dynamics, DOI: 10.1002/eqe.2798.
Castaldo, P., Amendola, G. & Palazzo, B. [2016b]. Seismic reliability-based design of structures
isolated by FPS. ECCOMAS Congress2016, Crete Island, Greece, 5–10 June 2016.
Castaldo P., De Iuliis M. [2014] “Optimal integrated seismic design of structural and viscoelastic
bracing-damper systems,” Earthquake Engineering and Structural Dynamics 43(12):1809–1827,
DOI: 10.1002/eqe.2425.
37
P. Castaldo et al.
Castaldo P., Palazzo B., Della Vecchia P. [2015] “Seismic reliability of base-isolated structures with
friction pendulum bearings,” Engineering Structures 95:80-93.
Castaldo P., Palazzo B., Della Vecchia P. [2016c] “Life-cycle cost and seismic reliability analysis of
3d systems equipped with FPS for different isolation degrees,” Engineering Structures 125:349–
363, http://dx.doi.org/10.1016/j.engstruct.2016.06.056.
Castaldo, P., Palazzo, B., Perri, F., Marino, I., Faraco, M.M. [2016d]. Seismic retrofit of existing
buildings through the dissipative columns. ECCOMAS Congress2016, Crete Island, Greece, 5–10
June 2016.
Castaldo, P., Palazzo, B., Perri [2016e] “FEM simulations of a new hysteretic damper: the dissipative
column”. Ingegneria Sismica - International Journal of Earthquake Engineering, Anno XXXIII –
Speciale CTA 2015 – Num. 1-2:34-45.
Castaldo P., Ripani M. [2016] "Optimal design of friction pendulum system properties for isolated
structures considering different soil conditions", Soil Dynamics and Earthquake Engineering,
90:74–87, DOI: 10.1016/j.soildyn.2016.08.025.
Castaldo P., Tubaldi E. [2015] “Influence of FPS bearing properties on the seismic performance of
base-isolated structures,” Earthquake Engineering and Structural Dynamics 44(15):2817–2836.
CEN - European Committee for Standardization [2004] Eurocode 8—Design of Structures for
Earthquake Resistance. Part 1: General Rules, Seismic Actions and Rules for Buildings, Brussels.
Christopoulos, C., Filiatrault, A. [2006] Principles of Passive Supplemental Damping and Seismic
Isolatio, IUSS Press: Pavia, Italy.
Clough, R.W., Penzien, J. [1993] Dynamics of Structures, 2nd edn. McGraw-Hill, New York.
Constantinou, M.C., Mokha, A., Reinhorn, A.M. [1990] “Teflon Bearings in Base Isolation. II:
Modeling,” J. Struct. Eng. 116(2):455-474.
Constantinou, M.C., Whittaker, A.S., Kalpakidis, Y., Fenz, D.M., Warn, G.P. [2007] “Performance of
Seismic Isolation Hardware Under Service and Seismic Loading,” Technical Report MCEER-07-
0012.
Cornell, C.A., Krawinkler, H. [2000] “Progress and challenges in seismic performance assessment,”
PEER Center News 4(1):1-3.
D'Ayala D., and Benzoni G., [2012] Historic and Traditional Structures during the 2010 Chile
Earthquake: Observations, Codes, and Conservation Strategies, Earthquake Spectra, 28;S1:S425-
S451
De Iuliis M, Castaldo P. [2012] “An energy-based approach to the seismic control of one-way
asymmetrical structural systems using semi-active devices,” Ingegneria Sismica - International
Journal of Earthquake Engineering XXIX(4):31-42.
De Iuliis M., Castaldo P., Palazzo B. [2010] “Analisi della domanda sismica inelastica del terremoto
de L’Aquila su sistemi dimensionati secondo le NTC2008,” Ingegneria Sismica, Patron Editore
XXVII(3):52–65.
Etse, G.J., Ripani, M., Vrech, S.M. [2013] “Fracture energy-based thermodynamically consistent
gradient model for concrete under high temperature,” Proceedings of the 8th International
Conference on Fracture Mechanics of Concrete and Concrete Structures, FraMCoS 2013, 1506-
1515.
Etse, G., Ripani, M., Caggiano, A. & Schicchi, D.S. [2015] “Strength and durability of concrete
subjected to high temperature: continuous and discrete constitutive approaches,” American
38
Concrete Institute, ACI Special Publication American Concrete Institute, ACI Special Publication
2015-January (SP 305), 9.1-9.18.
Etse, G., Ripani, M. & Mroginski, J.L. [2014] “Computational failure analysis of concrete under high
temperature,” Computational Modelling of Concrete Structures - Proceedings of EURO-C 2014,
2:715-722.
Etse, G., Vrech, S.M. & Ripani, M. [2016] “Constitutive theory for Recycled Aggregate Concretes
subjected to high temperature,” Construction and Building Materials, 111: 43-53.
Giugliano, M.T., Longo, A., Montuori, R., Piluso, V. [2011a] “Influence of homoschedasticity
hypothesis of structural response parameters on seismic reliability of CB-frames,” Georisk 5(2):
120-131.
Giugliano, M.T., Longo, A., Montuori, R., Piluso, V. [2011b] “Seismic reliability of traditional and
innovative concentrically braced frames,” Earthquake Engineering and Structural Dynamics
40(13): 1455–1474.
Kanai, K. [1957] “Semiempirical formula for the seismic characteristics of the ground,” Bulletin of
earthquake research institute 35, 309-325.
Kelly, J.M. [1997] Earthquake-Resistant Design with Rubber, 2nd ed. Berlin and New York:
Springer-Verlag.
Longo, A., Montuori, R., Piluso, V. [2009] “Seismic reliability of V-braced frames: Influence of
design methodologies,” Earthquake Engineering and Structural Dynamics 38(14): 1587–1608.
Longo, A., Montuori, R., Piluso, V. [2012] “Failure mode control and seismic response of dissipative
truss moment frames,” Journal of Structural Engineering 138(11), 1388-1397.
Luco, N., Cornell, C.A. [2007] “Structure-specific scalar intensity measures for near-source and
ordinary earthquake ground motions,” Earthquake Spectra 23(2): 357-92.
Mckey, M.D., Conover, W.J., Beckman, R.J. [1979] “A comparison of three methods for selecting
values of input variables in the analysis from a computer code,” Technometrics 21:239-45.
Mokha, A., Constantinou, M.C., Reinhorn, A.M. [1990] “Teflon Bearings in Base Isolation. I:
Testing,” J. Struct. Eng. 116(2): 438-454.
Mroginski, J.L., Etse, G., Ripani, M. [2015] “A non-isothermal consolidation model for gradient-
based poroplasticity,” PANACM 2015 - 1st Pan-American Congress on Computational Mechanics,
in conjunction with the 11th Argentine Congress on Computational Mechanics, MECOM 2015, pp.
75-88.
NTC08 - Norme tecniche per le costruzioni. Gazzetta Ufficiale del 04.02.08, DM 14.01.08, Ministero
delle Infrastrutture.
Palazzo B., Castaldo P., Della Vecchia P. [2014a] “Seismic reliability analysis of base-isolated
structures with friction pendulum system,” 2014 IEEE Workshop on Environmental, Energy and
Structural Monitoring Systems Proceedings, Napoli, September 17-18.
Palazzo , B. [1991] “Seismic Behavior of base-isolated Buildings,” Proc. International Meeting on
earthquake Protection of Buildings, Ancona.
Palazzo, B., Castaldo, P., Marino, I. [2014b] “The steel column damper: a new hysteretic device
providing additional stiffness and damping,” EUROSTEEL 2014, September 10-12, Naples, Italy.
Palazzo, B., Castaldo, P., Marino, I. [2015] “The Dissipative Column: A New Hysteretic Damper,”
Buildings 5(1), 163-178; doi:10.3390/buildings5010163.
Pinto P., Giannini R., Franchin P. [2004] Seismic Reliability Analysis of Structures, Iuss Press.
39
P. Castaldo et al.
Porter, K.A. [2003] “An overview of PEER’s performance-based earthquake engineering

methodology;” Proceedings, Proceedings of the 9th International Conference on Application of
Statistics and Probability in Civil Engineering (ICASP9), San Francisco, California, 973-980.
Ripani, M., Etse, G., Vrech, S. & Mroginski, J.L. [2014] “Thermodynamic gradient-based poroplastic
theory for concrete under high temperature,” International Journal of Plasticity, 61: 157-177.
Ripani, M., Etse, G., Vrech, S. [2016] “Recycled aggregate concrete: localized failure assessment in
thermodynamically consistent non-local plasticity framework”, Computers and Structures, doi:
10.1016/j.compstruc.2016.08.007.
Saito T., Kanda J., Kani N. [1998] “Seismic reliability estimate of building structures designed
according to the current Japanese design code,” Proceedings of the Structural Engineers World
Congress.
Shinozuka, M., Deodatis, G. [1991] “Simulation of stochastic processes by spectral representation,”
Applied Mechanics Reviews 44(4): 191-203.
Shinozuka, M., Sato, Y. [1967] “Simulation of nonstationary random process,” J. Engrg.Mech.Div.
93(1), 11-40.
Tajimi, H. [1960] “A statistical method of determining the maximum response of a building structure
during an earthquake,” Proc. 2nd World Conf. on earthquake Engineering II, 781-798.
Talaslidis, D.G., Manolis, G.D., Paraskevopoulos E.A., Panagiotopoulos C.G. [2004] Risk analysis of
industrial structures with hazardous materials under seismic input, 13th World Conference on
Earthquake Engineering, Vancouver, B.C., Canada, August 1-6.
Vamvatsikos, D., Cornell, C.A. [2002] “Incremental dynamic analysis,” Earthquake Engineering and
Structural Dynamics 31(3): 491–514.
Vision 2000 - SEAOC - Committee. Performance-based seismic engineering. Report prepared by
Structural Engineers Association of California, Sacramento, 1995.
Vrech, S.M., Ripani, M. & Etse, G. [2015] “Localized versus diffused failure modes in concrete
subjected to high temperature,” PANACM 2015 - 1st Pan-American Congress on Computational
Mechanics, in conjunction with the 11th Argentine Congress on Computational Mechanics,
MECOM 2015, pp. 225-236.
Zayas, V.A., Low, S.S., Mahin, S.A. [1990] “A simple pendulum technique for achieving seismic
isolation,” Earthquake Spectra 6:317–33.
40
Num. …
EFFETTI DI SUOLO DI CATEGORIA B SULL’AFFIDABILITA’

SISMICA DI SISTEMI IN ACCIAIO ISOLATI ALLA BASE
Paolo Castaldo*, Guglielmo Amendola, Bruno Palazzo
SOMMARIO: Il presente lavoro valuta l’affidabilità sismica di sistemi in acciaio isolati alla
base con dispositivi attritivi a pendolo scorrevole e soggetti a registrazioni accelerometriche
artificiali corrispondenti a suolo di categoria B. La sovrastruttura è modellata con un
comportamento elastico mentre gli isolatori sono descritti con un modello che tiene conto
della variazione del coefficiente di attrito con la velocità. L’incertezza relativa sia all’azione
sismica che al coefficiente di attrito è considerata assumendo opportune funzioni di densità
di probabilità. Analisi dinamiche non lineari incrementali sono sviluppate al fine di definire
le curve di fragilità relative sia alla sovrastruttura che livello di isolamento. Infine,
considerando come sito di riferimento Sant’Angelo dei Lombardi (Italia), l’affidabilità
sismica dell’intero sistema è valutata.
___________________
*Corresponding author: Paolo Castaldo, Department of Civil Engineering, University of Salerno, Italy.
Email: pcastaldo@unisa.it
Num. …
APPRAISAL OF SEISMIC DESIGN CRITERIA FOR

CONCENTRIC BRACING STEEL STRUCTURES ACCORDING
TO ITALIAN AND EUROPEAN CODES
Beatrice Faggiano*, Antonio Formisano, Carmine Castaldo, Luigi Fiorino,

Vincenzo Macillo, Federico M. Mazzolani
Department of Structures for Engineering and Architecture,

University of Naples “Federico II”, Italy
SUMMARY: The critical review of design methods provided by the Italian codes for
constructions NTC2008, in agreement with the European seismic code (EC8), for steel
Concentrically Braced Frames (CBF) with X and chevron (or inverted V) diagonals, has the
aim at providing more efficient design criteria able to ensure adequate safety levels under
seism. As reference case studies, common structural configurations of CBF are designed
according to NTC2008, by both Linear Static and Dynamic analyses. The purpose is identifying
the weaknesses in the design criteria, with particular reference to both the applicability of the
proposed procedures and the actual possibility to size braces and connected structural
members, like beams and columns. The critical issues in the design process are evidenced. A
discussion on the obtained results has allowed to point out the pros and cons of the current
design approach and to propose some enhanced design criteria.
KEYWORDS: Seismic resistant steel structures, Seismic design criteria, Concentrically

braced frames, Behaviour factor, Over strength factor, Non linear static and
dynamic analyses
1. THE RESEARCH CONTEXT

In recent years the huge amount of research and advances in the field of seismic engineering,
together with the frequent occurrence worldwide of seismic events of high intensity and serious
consequences, not last the case of Italy, recently stroke by severe earthquakes, has motivated
the upgrading of technical codes for constructions. With particular focus on steel structures, a
crucial moment for the development of the national seismic design codes has been the draft of
OPCM 3274 and 3431 [2005] since 2003, which introduced an extensive chapter, in line with
EC8: Design of structures for earthquake resistance [EN 1998-1, 2005], with the addition of
some noticeable changes with respect to the European standard, integrating the evidences of
extensive studies on the seismic behaviour. However these amendments were not included in
the current technical standards, for the sake of symmetry with EC8 [Mazzolani and Della Corte,
2008]. What is more, the current Italian NTC2008 [M.D., 2008] has several cuts with respect
to EC8, which were partially recovered in the explicative Italian Ministerial Circular [M. C.,
2009]. A general overview of the design aspects and applications related to steel CBF structures,
among the other steel seismic resistant structures, is presented in Mazzolani et al. [2006], where
the modern approach is discussed, evidencing the conformity between the Italian and European
___________________
*Corresponding author: Department of Structures for Engineering and Architecture, University of Naples
“Federico II”, Italy
Email: faggiano@unina.it
codes. A more rigorous plastic design procedure for CBFs was proposed by Longo et al. [Longo
et al., 2008a; 2008b], more recently Marino has proposed a unified design approach for CBFs
[Marino, 2014]. A wide research activity, aiming at the progressive upgrading and optimization
of seismic design rule for constructions, including steel braced structures, is ongoing within
the Italian project RELUIS-DPC. In this context, a deep analysis of X and V braced structures
has been presented in Faggiano et al. [2014; 2015a,b] and in D’Aniello et al. [2013, 2015].
2. NTC2008 DESIGN CRITERIA FOR CBF STRUCTURES

2.1 General aspects
In Concentrically Braced Frames (CBF), the resistance against seismic actions is provided by
the contribution of both tensile and compression braces. The ideal design ultimate condition of
a dissipative braced system is the simultaneous buckling of compression bracings and yielding
of tensile ones, the braces being the dissipative elements. According to the commonly accepted
resistance hierarchy criterion, the other structural elements, such as columns, beams and
connections, have to remain in elastic range and, therefore, they should be designed to have an
adequate over strength as respect to braces. In Table 1 the Ultimate Limit State (ULS) design
rules for seismic resistant concentric braced systems are summarised.
Firstly, members should be ductile, thus belonging to Class sections 1 or 2, according to the
cross section classification defined in Eurocode 3 [EN 1993-1-1, 2005; Formisano et al., 2006]
and taken by NTC2008. In order to prevent the untimely collapse of beams and columns,
according to the hierarchy design criterion, the code requires to determine the overstrength
factor Ω, as in Table 1. This indicates how much the axial force and then the seismic force can
exceed the design value until the brace member reaches the complete plasticization.
Table 1. NTC08 design criteria for steel CBF structures
X-Brace Chevron Brace
Class sections 1 or 2; for circular hollow sections: d/t ≤ 36
N Ed N pl , Rd  1 (1) N Ed N b, Rd  1 (2)
Brace 1,3    2 (3)  2 (4)

 N pl , Rdi   max   min
   min    ;  1,25 (5)
 N Edi  min  min
Npl
Beam N Ed  N EdG  1,1 Rd  min N EdE (6)
pbNpl
 pb  0,3 (7)
Column N Ed  N EdG  1,1 Rd  min N EdE (8)
Ductility High 4 2,5
q-factor
Class Low 4 2
where: d and t are diameter and thickness of the circular hollow profile, respectively; NEd, Npl,Rd, Nb,Rd are the
brace design axial force, plastic resistance, buckling resistance;  is the brace normalized slenderness; Ω is the
overstrength factor; γRd is the steel overstrength factor that is the ratio between the average and the characteristic
values of the yielding strength; γpb is a factor representative of the residual brace strength after buckling; NEdG,
NEdE are the axial forces corresponding to non-seismic and seismic loads.
43
B. FAGGIANO et al.
It is not the same for all diagonals, it depending on the distribution of internal forces within the
structure and on some sources of oversizing, like the selection of structural members among
the standard profiles or the need to provide lateral stiffness for deformability check, further to
the imposed limitation of slenderness. The latter condition is particularly strict at the upper
stories, giving rise to  factors increasing along the height of the structures. As a consequence,
aiming at assuring a distribution in elevation as uniform as possible to promote the yielding of
all braces, the difference between the maximum and the minimum values should be limited to
25% (Table 1). Moreover, considering that diagonals do not plasticize together at the same
level of seismic forces, the  factor to be used is assumed as the minimum one, min,
corresponding to the first not linear event, such as the plasticization of the first brace. Once
designed the braces and calculated the  factor, the capacity design criterion is applied for
determining the design forces for beams and columns (Eq. 8, Table 1).
2.2 CBF-X
The design of CBF-X is performed by considering only the contribution of braces in tension,
assuming that at collapse braces in compression are already buckled and do not provide any
bearing capability. With this assumption, tensile braces are designed on the basis of the plastic
resistance (Eq. 1, Table 1). Moreover, the normalized slenderness ( ) of diagonals should be
limited within a prefixed range (Eq. 3, Table 1), where the upper limit has the aim to avoid
excessive distortions due to buckling of braces in compression, which could cause damage to
connections or claddings, while the lower limit ensures the validity of the structural model with
only active tensile braces as well as restricts the design internal forces in the columns, which
are commensurated with the plastic resistance of braces. The behaviour factor q for dissipative
structures is assumed as equal to 4 for both low and high ductility classes. In the ideal condition
in which the whole brace in tension is plasticized, ductility and dissipation capability of
members would be much greater than how quantified by such q value. However it is not
possible to be confident on the ideal behaviour due to the uncertainties related to the behaviour
of braces under the seismic cyclic actions. In fact braces undergo alternate states of tension and
compression, therefore if the brace in compression buckles, the unstable deformed shape in
bending is characterized by localized plastic deformation, thus the subsequent cycle in tension
finds a degraded member, with limited ductile capabilities.
2.3 CBF-V
In the CBF-V, the compressed brace provides a contribution to the overall system stability, thus
it cannot be neglected. For this reason, both braces are taken into account in the design model.
Therefore the design resistance is the buckling strength (Eq. 2, Table 1) and only the upper limit
of the normalized slenderness of diagonals is imposed (Eq. 4, Table 1). In addition, beams have
to be designed by considering the concentrated force at the middle-span due to the unbalanced
force between the plastic resistance of the tensile brace and the residual resistance of the
compressed one after buckling, the latter being set equal to 30% of the brace plastic resistance
(Eq. 7, Table 1). The beam is therefore subjected to bending moment, shear and axial forces. The
behaviour factor q is equal to 2,5 and 2 for High and Low ductility classes, respectively.
3. THE CASE STUDY

The study structures have typical configuration and size of a regular building. The reference
geometrical scheme, together with the main design features are given in Figure 1. For the sake
44
of simplicity, the elastic spectrum is obtained according to the code OPCM 3431 [2005] since
seismic parameters are independent from the geographic position, unlike the current NTC2008.
Each case study is designed through either the Linear Static (LS) or Linear Dynamic (LD)
analysis. For CBF-X the profiles used for the diagonal members are HE sections; in total 6 case
studies are examined. For CBF-V for columns two cross section types are used, namely welded
box sections and HE profiles, for beams HE profiles are used; while for braces two cross-
section types, namely Circular Hollow Sections (CHS) and HE profiles, are used; in total 12
case studies are examined. As far as the Damage Limit State (DLS) is concerned, the limitation
of the inter-story drift equal to 1% is considered, corresponding to infill panels not rigidly
connected to the main structure.
Seismic resistant systems

Posizione sistemi sismoresistenti
3,5m
L L L L L
3,5m
L
35,50
3,5m
L
3,5m
Plan view
Pianta piano tipo 4m
L L
Design features: 3, 6, 10 stories where:

h = 3,5m (hGF = 4m) h is the interstory height,
L = 6m hGF is the interstory heigth at the ground floor,
High ductility class L is the span,
ag = 0,35g soil B ag is the peak ground acceleration,
Gk=4,81 kN/m2 Gk is the characteristic value of the dead load,
Gk,T=5,20 kN/m2 Gk,T is the characteristic value of the dead load at the top story,
Qk=2 kN/m2 Qk is the characteristic value of the live load.
Figure 1. Geometry and design features of the study structures
4. DESIGN ASSESSMENT
4.1 CBF-X
The NTC08 design procedure shows a first critical issue in the ambiguity in the use of LD
analysis. For CBF-X, the code prescribes that at the ULS only braces in tension resist the
seismic forces, while the compressed braces are considered buckled and unable to provide
strength. Nevertheless the vibration properties of the structure, i.e. periods and vibration modes,
are strictly related to the linear behaviour and they should be determined considering the
contribution of both braces in tension and in compression, therefore they cannot be calculated
disregarding braces in compression. For this reason, for the structures examined the LD
analysis is performed by considering the presence of both braces not only for evaluating the
elastic vibration properties, but also for assessing seismic forces in the members. Then, in order
to consider the model with only one active diagonal, the design of braces is carried out by
assuming the axial forces as the double of the one calculated by means of the structural model
including both diagonals (Fig. 2).
45
B. FAGGIANO et al.
NEd NEd Nd,Ed =2·NEd
Schemadynamic
Linear analisi Sollecitazioni impiegate
Seismic forces per il
used for
dinamica lineare
analysis scheme dimensionamento delle
diagonal design diagonali
Figure 2. Structural scheme assumed for linear dynamic analysis [Faggiano et al., 2014].
In Table 2, the main design information on the study structures, such as W the structural weight,
T the fundamental period of vibration, Fh the design base shear and min the design over
strength factor, are reported. The design results show that the CBF-X designed by LS analyses
are generally subjected to seismic actions higher than those designed by LD analyses. This
difference is mainly related to the underestimation of the fundamental vibration period through
the empirical formula provided by NTC2008 in case of LS analyses. This issue is more evident
for taller buildings. For instance, in case of 10s structures, T calculated by the code formula is
smaller than LD T, with a consequent increment of the total seismic force. This issue also
influences the weight of the seismic resistant members, in fact, the structural weight of LS
structures is higher than the LDs.
Table 2. Design results for CBF-X [Faggiano et al., 2014]

Design Method N. storeys (s) W [kN] T [s] Fh [kN] Ωmin
3 37 0.30* 348 2.52
LS 6 110 0.50* 687 1.88
10 186 0.73* 781 1.44
3 37 0.31 354 2.40
LD 6 95 0.59 548 1.94
10 149 1.25 486 1.75
*T=C1H3/4 with C1 = 0.05, H= total height of the structure
Another critical issue observed in the design phase is the difficulty in selecting the bracing
profiles. In particular, the lower bound of  (1,3) strongly limits the HE profiles that can be
used. In addition, the low seismic demand at upper storeys implies oversized bracings with
corresponding very high Ω values. This especially occurs at the top storey, where the
uniformity of the Ω factor distribution along the structure height is hard to be satisfied (Eq. 3,
Table 1). Thus, for the 10s structures examined the top storey has not been considered in the 
check.
4.2 CBF-V
Commonly, the design of columns is conditioned by the gradual cross-section reduction
criterion. In fact, strong variation of the cross-section sizes along the building height should be
avoided, they being generally a source of localised damage. Moreover the structure story
stiffness variation at consecutive floors should be limited, in order to fulfil regularity
requirements devoted to assure the achievement of the most uniform state of stress and
46
deformation and in particular plastic hinges distribution along the structure height at the
ultimate limit states. Another influencing design aspect is that HE profiles larger than HEB300
are barely able to withstand high axial loads, because, as far as the depth increases, the base is
almost constant; thereby, being the increment of second moment of area extremely limited, for
profiles larger than HEB300, the axial buckling check is hard to be satisfied. For these reasons,
for 10s structures and 6s HE braces structures, the columns are realized with welded square
box sections, opportunely reduced along the building height, in order to absorb the high axial
loads deriving from the capacity design criterion. Columns are HE profiles for 6s frames in
case of CHS braces, as the same for 3s frames, where, due to the limited number of floors, the
hierarchy design criterion is not penalising. In Table 3 the main design information on the study
structures are reported. In Figure 3 the response spectra with the evidence of T for 3s, 6s and
10s in case of HE braces and CHS braces are shown. In general some observations come be
done by comparing on one hand CHS and HE braces structures, on the other hand LS and LD
design. CHS braces structures are lighter than HE braces structures. This is particularly evident
for LS 10s buildings, where the adoption of HE braces induces the use of welded double T
beam profiles at the lower storeys, due to the high forces transferred by the braces, oversized
for the limited availability of standard HE hot-rolled profiles. This limitation does not exist for
CHS profiles, which are produced with a large range of cross-sections.
Table 3. Design results for CBF-V [Faggiano et al., 2015b]
W [kN] T [s] Fh [kN] Ωmin
Design Method N. storeys
CHS HE CHS HE CHS HE CHS HE
3 46 63 0.3* 0.3* 549 549 2.15 3.90
LS 6 159 234 0.5* 0.5* 1085 1085 2.11 4.04
10 408 722 0.73* 0.73* 1254 1254 2.46 4.43
3 52 71 0.33 0.25 580 588 2.32 4.21
LD 6 140 215 0.58 0.44 1062 1064 1.86 3.41
10 309 602 1 0.68 988 1428 2.34 3.70
*T=C1H3/4 with C1 = 0.05, H= total height of the structure
The vibration periods T determined by LD analysis are higher for CHS braces structures than
for HE braces structures, in conformity to the previous observation. In the LS design the first
vibration period does not depend on the bracing details, it being calculated by means of the
simplified formula. Moreover, the LD T, as respect to the LS T is lower for HE braces structures,
is higher for CHS braces structures. In case of 10s and 6s structures the LD design gives rise
to lower base shears and weights as respect to the LS design, contrary in case of 3s frames.
Figure 3. The LS and LD first periods of vibration T for the study CBF-V [Faggiano et al., 2015b].
47
B. FAGGIANO et al.
This trend does not reflect the variation of the first periods of vibration T, due to the influence
of the superior modes of vibration, which could provide not negligible additional actions that
are ignored by the LS analysis.
Concerning the over-strength factor the following observations can be done:
 High values of Ωmin, ranging from 1,86 to 2,46 and 3,41 to 4,43 for CHS and HE braces
structures, respectively, are achieved. These imply a significant increment of design axial
forces in the columns according to the capacity design.
 The Ω variation ratio (Eq. 5, Table 1) is always governed by the top storey braces, whose
Ω values are generally larger than those at lower storeys. This is due to the use of brace
cross-sections, which are subjected to low seismic actions but should contemporary respect
the standard slenderness limit. As a consequence, elastic members are oversized, it
producing a weight increase.
 In case of HE braces, Ωmin is generally larger than the design behaviour factor q (2,5), what
is not acceptable.
5. CONCLUSIONS
Based on the previous observations possible improvements of the NTC2008-EC8 design
criteria for CBF can be related to the following items:
1. A possible improvement of the design criteria for CBF-X could be to clearly state the design
procedure in agreement with the ULS model with only tensile diagonal active.
2. The use of HE profiles for structural members can become more convenient if a wider
spectra of cross sections are produced, in order that profiles could best fit all the design
requirements in terms of strength and stiffness, reducing the over strength that alters the
effect of the design provisions as respect to the expectations. This could also either avoid
the scatter in the geometrical variability of members composed by different parts, as it
occurs for columns belonging to high rise buildings, or enhance the efficiency of the
capacity design, or simplify the connection among members.
3. The simplified formula for the determination of the first period of vibration, necessary in
case of LS design, should be better fitted according to the number of floors, taking also into
account the structural system type, it being differentiated in case of bracing systems and
for type of bracings.
4. The top story needs specific design criteria, which balance capacity design and slenderness
requirements. Some authors proposed a different approach based on the reduction of the
bracing members section at the ends to obtain =1 [Giugliano et al., 2010; 2011].
5. With regards to capacity design, it should be explicitly stated that the over strength factor
 should be in any case lower than the design behaviour factor q.
In general, results briefly presented in the current paper, in line with the literature references,
delineate some important issues and suggestions for improvements, related to the design
procedure and structural models. However they require more wide elaborations through further
extensive campaign of both experimental and numerical investigations aiming at both
optimizing the calculation models and providing simplification to the design methods.
6. ACKNOWLEDGEMENTS
The authors gratefully acknowledge the Department of Civil Protection for the support within
the RELUIS-DPC project.
48
7. REFERENCES
D’Aniello M., La Manna Ambrosino G., Portioli F., Landolfo R., [2013]. Modelling aspects of the
seismic response of steel concentric braced frames. Steel and Composite Structures, An International
Journal, Vol. 15, No. 5, 539-566 November 2013.
D'Aniello M., Costanzo S., Landolfo R., [2015]. The influence of beam stiffness on seismic response
of chevron concentric bracings. Journal of Constructional Steel Research. 112: 305-324. Doi:
10.1016/j.jcsr.2015.05.021.
Decree of the Minister Council Presidency 03/05/2005 n. 3431 (OPCM) [2005] Further modifications
and integrations to the Decree of the Minister Council Presidency 20/03/2003 n. 3274 (in Italian).
EN 1993-1-1 [2005] Eurocode 3: Design of steel structures Part 1-1: General rules and rules for
buildings, European Committee for Standardization, Bruxelles.
EN 1998-1 [2005] Eurocode 8: Design of structures for earthquake resistance – Part 1: General rules,
seismic actions and rules for buildings, European Committee for Standardization, Bruxelles.
Faggiano, B., Fiorino, L., Formisano, A., Macillo, V., Castaldo, C., Mazzolani, F.M. [2014]
“Assessment of the design provisions for steel concentric X bracing frames with reference to Italian
and European codes”, The Open Construction and Building Technology Journal 8, (Suppl 1: M3),
208-215.
Faggiano B., Fiorino L., Formisano A., Castaldo C., Macillo V., Mazzolani F. M. [2015a]. Appraisal
of the design criteria for concentric braced steel structures according to Italian and European codes.
In XXV Congresso C.T.A. Le giornate Italiane della costruzione in acciaio, 1-3 Ottobre 2015,
Salerno, ISBN 9788894008944, pp.445-452.
Faggiano, B., Formisano, A., Fiorino, L., Castaldo, C., Macillo, V., Mazzolani, F.M. [2015b]
“Assessment of the design criteria for concentric V-braced steel structures according to Italian and
European codes”. The Open Civil Engineering Journal, Betham Open. (in press).
Formisano, A., Faggiano, B., Landolfo, R., Mazzolani, F.M. [2006] “Ductile behavioural classes of
steel members for seismic design”, In: Mazzolani, F.M., Wada. A. Behaviour of Steel Structures In
Seismic Areas, p. 225-232, London: Taylor & Francis Group/ Balkema, ISBN: 9780415408240,
Yokohama, 14-17 August.
Giugliano, M. T., Longo, A., Montuori, R., Piluso, V. [2010] “Plastic design of CB-frames with reduced
section solution for bracing members”, Journal of Constructional Steel Research 66 (5), 611-621.
Giugliano, M. T., Longo, A., Montuori, R., Piluso, V. [2011] “Seismic reliability of traditional and
innovative concentrically braced frames”, Earthquake Engineering and Structural Dynamics, 40
(13), 1455-1474.
Longo, A., Montuori, R., Piluso, V. [2008a] “Failure mode control of X-braced frames under seismic
actions”, Journal of Earthquake Engineering 12 (5), 728-759.
Longo, A., Montuori, R., Piluso, V. [2008b] “Plastic design of seismic resistant V-braced frames”,
Journal of Earthquake Engineering 12 (8), 1246-1266.
Marino, E. M. [2014] “A unified approach for the design of high ductility steel frames with concentric
braces in the framework of Eurocode 8”, Earthquake Engineering and Structural Dynamics 43, 97–
118, DOI: 10.1002/eqe.2334, 2014.
Mazzolani, F.M., Della Corte, G. [2008] “Steel structures in seismic zone – M.D. 2008” (in Italian), Il
Sole 24 Ore (ed.).
Mazzolani, F., Landolfo, R., Della Corte, G., Faggiano, B. [2006] Steel structures buildings in seismic
zones (in Italian), Pavia: Iuss Press, ISBN: 9788873580348.
49
B. FAGGIANO et al.
Ministerial Circular 02/02/2009 n. 617 (M.C.) [2009] Instructions for application of the “New technical
codes for constructions” (in Italian).
Ministerial Decree 14/01/2008 (M.D.) [2008] New technical codes for constructions (in Italian).
50
Num. …
VALUTAZIONE DEI CRITERI DI PROGETTO PER LE STRUTTURE

DI ACCIAIO A CONTROVENTI CONCENTRICI SECONDO LE
NORME ITALIANE ED EUROPEE
Beatrice Faggiano*, Antonio Formisano, Carmine Castaldo, Luigi Fiorino,

Vincenzo Macillo, Federico M. Mazzolani
Department of Structures for Engineering and Architecture,

University of Naples “Federico II”, Italy
SOMMARIO: La revisione critica dei metodi di progetto presenti nelle norme tecniche italiane
per le costru-zioni NTC2008, allineate con l’Eurocodice 8, per le strutture a controventi
concentrici con diagonali a X e a V inversa, ha lo scopo di definire criteri di progetto più
efficaci nel garantire gli adeguati livelli di sicurezza sismica. I casi di studio sono
rappresentativi di tipiche configu-razioni strutturali di CBF. Essi sono progettati secondo le
NTC2008, mediante analisi lineari sia statiche sia dinamiche. L’obiettivo è l’identificazione
delle carenze delle regole di progetto, con particolare riferimento sia all’applicabilità delle
procedure proposte sia alla reale possibili-tà di dimensionare le diagonali e le membrature
strutturali ad esse collegate, quali travi e co-lonne. La discussione sui risultati ottenuti ha
consentito di evidenziare gli aspetti sia positivi sia negativi dell’approccio progettuale attuale
e di proporre alcuni miglioramenti dei criteri di progetto.
___________________
*Corresponding author: Department of Structures for Engineering and Architecture, University of Naples
“Federico II”, Italy
Email: faggiano@unina.it
Num. …
PRELIMINARY ANALYSIS ON THE INFLUENCE OF THE LINK

CONFIGURATION ON SEISMIC PERFORMANCES OF MRF-EBF
DUAL SYSTEMS DESIGNED BY TPMC
Rosario Montuori, Elide Nastri*, Vincenzo Piluso
SUMMARY: The work is devoted to the evaluation of the influence of link configuration on
seismic performances of Moment Resisting Frames-Eccentrically Braced Frames dual systems
(MRF-EBF dual systems) designed by means of Theory of Plastic Mechanism Control (TPMC).
As it is known TPMC assures the development of a collapse mechanism of global type, therefore,
the seismic performances evaluated by means of IDA analyses, are affected only by the
structural scheme configuration. However, in this paper only a 5 bays structure with 4, 6 and
8 storeys are reported, so that, the work configures as a preliminary evaluation. Additional
analyses on different structural schemes with different number of bays will be the natural
development of the work herein presented.
KEYWORDS: EBFs, eccentrically braced frames, seismic, kinematic theorem, TPMC
1. INTRODUCTION
The work herein presented is devoted to the evaluation of the seismic performance of Moment
Resisting Frame-Eccentrically Braced Frame dual systems (MRF-EBF dual systems)
considering all the possible configuration of braces and links as reported in the Eurocode 8
[CEN, 2015] (Fig. 1). As it is known, EBFs constitute a suitable compromise between MRF
and CBF [Longo et al., 2014; D’Aniello et al., 2010; D’Aniello et al., 2012; D’Aniello et al.,
2015], due both to their lateral stiffness given by the braces and adequate dissipation capacity
given by links able to develop wide and stable hysteresis loops [Della Corte et al., 2013; Della
Corte et al., 2007]. The number of the proposed link configuration is 4: three of which have a
horizontal link configuration (K-scheme, D-scheme, V-scheme) while the last one has a
vertical link configuration (Inverted Y-scheme). In this paper, MRF-EBF dual systems with 4,
6 and 8 storeys have been designed using the Theory of Plastic Mechanism Control for a total
number of 12 study cases. The scope is to investigate the structural seismic response of the
designed structure being assured, by the TPMC design approach, a collapse mechanism of
global type. In this way, the seismic performances are only affected by the structural scheme,
i.e. by the link configuration. For this reason both push-over and Incremental Dynamic
Analyses (IDA) have been carried out on the designed structure with the scope to point out, on
one hand, the accuracy of the TPMC design approach, and on the other hand, downstream the
validation, the different seismic performances achieved by the differently configured structure.
However, in this paper only a 5 bays structure is investigated, so that the work configures as a
preliminary evaluation. For a higher accuracy, additional analyses on different structural
schemes with different storey and bay numbers will be reported in further works.
___________________
*Corresponding author: Elide Nastri, Department of Civil Engineering, University of Salerno, Italy.
Email: enastri@unisa.it
Figure 1. EBFs typologies proposed by Eurocode 8
2. DESIGN APPROACH
TPMC is based on a rigorous approach assuring a collapse mechanism of global type [Piluso
et al., 2015; Montuori and Muscati, 2015; Longo et al., 2014] which exploits the kinemathic
theorem of plastic collapse extended to the concept of collapse mechanism equilibrium curve:
(1)
where is the first order collapse multiplier of horizontal forces, is the slope of the
linearized mechanism curve due to second order effects and is the plastic top sway
displacement. TPMC states that the mechanism equilibrium curve corresponding to the global
mechanism has to be located below those corresponding to all the undesired mechanisms until
a design displacement compatible with the local ductility supply (Figure 2). Column
sections at each storey needed to assure a collapse mechanism of global type are unknown of
the design problem, while link, beam and diagonal members are preliminarily determined as
known quantities of the design problem. More details about this design procedure devoted to
EBFs are reported in previous works [Longo et al., 2014; Montuori et al., 2013; Montuori et
al., 2014a; Montuori et al., 2014b; Montuori et al., 2016]
3. STUDY CASES
The study cases herein investigated are referred to a building whose plan configuration is
depicted in Fig. 3. In particular, considering only the perimeter frames whose secondary floor
warping is orthogonal to the beam of the frame, four structures have been designed for each
selected number of storey (4, 6 and 8) according to TPMC for a total number of 12 frames.
The corresponding seismic resistant schemes are depicted in Figure 3 with reference to the 6-
storey building where also the leaning column adopted in structural modelling to account for
second order effects due to the internal gravity load resisting system is reported. Regarding the
length of the links, at each storey, it is equal to 1.20 m for K-scheme and D-scheme EBFs while
is equal to 0.60 m for V-scheme EBFs. As regards the inverted Y-scheme, link length has been
assumed equal to 0.70 m at each storey because the link length has been defined in order to
assure that the ultimate design displacement, , for the application of TPMC is the same both
for the all the structural configuration, according to the following relations:
⁄ 0.08 1.2/6 0.016 for K-scheme and D-scheme (2)
2 ⁄ 0.08 2 0.6/3.5 0.016 for inverted V-scheme (3)
⁄ 0.08 0.7/3.5 0.016 for inverted Y-scheme (4)
where is the target link plastic rotation, is the link length, is the braced bay length,
is the interstorey height and is the building height.
53
R. MONTUORI et al.
Figure 2. TPMC statement

The characteristic values of the vertical loads are equal to 4.0 kN/m2 and 2.0 kN/m2 for
permanent and live loads, respectively. As a consequence, with reference to the
seismic load combination provided by EC8, (where is the coefficient for
the quasi-permanent value of the variable actions, equal to 0.3 for residential buildings), the
vertical loads acting on the floor are equal to 4.6 kN/m2. The structural material adopted for all
the members is S355 steel grade 355 . The beams have been designed to
withstand vertical loads accounting also for serviceability requirements and they are delivered
in Figure 3. The design horizontal forces have been determined according to EC8, assuming a
peak ground acceleration equal to 0.35g, a seismic response factor equal to 2.5 and a behaviour
factor equal to 6. It is important to observe that EC8 does not make any difference in term of
behaviour factor between the EBFs configuration. On the basis of such force distribution, the
design shear action of link members has been obtained by assuming that the storey shear is
completely entrusted to the link. In addition, for each structure the link length combined with
the resistance belonging to the selection of the section from the standard shape assure a short-
link behaviour, according to the classification given in EC8. Short links show many advantages
if compared with long and intermediate links due to their performance in terms of both stiffness
and ductility. In fact, the cyclic behaviour of short links is characterised by wide and stable
cycles allowing the development of high energy dissipation capacity and highest rotational
capacity if compared with long and intermediate ones [Bosco et al., 2015; Mastrandrea et al.,
2013; Ohsaki et al., 2012; Gulec et al., 2011]. Table 1 the link and diagonal sections for the
designed structures are reported. In addition, in the Table 2 column sections of each structure
are delivered. For sake of shortness, the numerical details for the application of TPMC
procedure are given elsewhere [Longo et al., 2014; Montuori et al., 2014a; Montuori et al.,
2014b] so only the results obtained are herein presented.
4. PUSH-OVER RESULTS
With reference to the seismic resistant system depicted in Figure 3, push-over analyses have
been carried out by means of SAP2000 computer program for all the four structural schemes.
The aim of these analyses is to check the collapse mechanism actually developed in order to
validate the accuracy of the TPMC application. Member yielding has been accounted for by
modelling the dissipative zones by means of hinge elements, i.e. with a lumped plasticity model.
54
Column, beam, diagonal and link members have been modelled with an elastic beam-column
frame element with two rigid-plastic hinge elements located at the member ends. With
reference to beams, plastic hinge properties are defined in pure bending (M3 hinge) while in
case of columns and diagonals plastic hinge properties are defined to account for the interaction
between bending and axial force (P-M3 hinges). Both of them have a rigid-plastic constitutive
model for the moment rotation behaviour. In addition, an axial hinge has been located in the
midspan of each column with the scope to check the out of plane buckling. Regarding link
members, as short links yielding in shear are of concern, plastic hinges in shear have been
considered with a shear force versus shear displacement rigid-hardening constitutive model
(1.50 of overstrength).
Figure 3. Plan configuration of the analysed structures

Table 1. Design seismic forces, link and diagonal sections for the designed structures
K-scheme D-scheme V-scheme Inverted Y-scheme
F [kN] LINKS DIAGONALS LINKS DIAGONALS LINKS DIAGONALS LINKS DIAGONALS
4-STOREY
1 96.097 HEB 240 CHS 323.9x20 HEB 240 CHS 323.9x20 HEB 160 CHS 323.9x20 HEB 300 CHS 244.5x20
6-STOREY
1 31.745 HE 240 B CHS 406.4x32 HEB 340 CHS 406.4x32 HE 180 B CHS 406.4x12.5 HEB 340 CHS 406.4x32
8-STOREY
55
R. MONTUORI et al.
MRF-EBF DUAL K-SCHEME - 6 STOREYS MRF-EBF DUAL D-SCHEME - 6 STOREYS

Fc1=62.10 kN Fc2=103.50 kN Fc3=82.80 kN Fc3 Fc2 e=120 cm Fc1 Fc1 =62.10 kN Fc2=103.50 kN Fc3=82.80 kN Fc3 Fc2 e=120 cm Fc1
Fl=(G +0.3Q )Ll·L=
k k
Fl=(G +0.3Q )Ll·L=
k k
1490.40 kNm 1490.40 kNm

q =(G +0.3Q )L =6.90 kNm
d k k 0
q =(G +0.3Q )L =6.90 kNm
d k k 0
6 e 6
IPE 330 IPE 270 IPE 300 e IPE 270 IPE 330 IPE 330 IPE 270 HEB 180 IPE 270 IPE 330
Fc1 Fc2 Fc3 HEB 140 Fc3 F c2 Fc1 Fc1 Fc2 Fc3 Fc3 Fc2 Fc1
3.50
3.50
Fl Fl
5 e 5
3.50
3.50
Fl Fl
4 e 4
3.50
3.50
Fl Fl
e 3 e 3
IPE 330 IPE 270 IPE 270 IPE 330 IPE 330 IPE 270 HEB 240 IPE 270 IPE 330
3.50
3.50
Fl Fl
e 2 e 2
3.50
3.50
Fl Fl
e 1 e 1
HEB 240
3.50
3.50
C1 C2 C3 C3 C2 C1 C1 C2 C3 C3 C2 C1
9.00 6.00 6.00 6.00 9.00 9.00 6.00 6.00 6.00 9.00
36.00 36.00
MRF-EBF DUAL K-SCHEME - 8 STOREYS MRF-EBF DUAL D-SCHEME - 8 STOREYS

Fc1=62.10 kN Fc2=103.50 kN Fc3=82.80 kN Fc3 Fc2 e=120 cm Fc1 Fc1 =62.10 kN Fc2=103.50 kN Fc3=82.80 kN Fc3 Fc2 e=120 cm F c1
Fl=(G +0.3Q )Ll·L=
k k
Fl=(G +0.3Q )Ll·L=
k k
1490.40 kNm 1490.40 kNm

q =(G +0.3Q )L =6.90 kNm
d k k 0
q =(G +0.3Q )L =6.90 kNm
d k k 0
e 8 HEB 140 e 8
IPE 330 IPE 270 IPE 270 IPE 330 IPE 330 IPE 270 IPE 270 IPE 330
3.50
3.50
Fl Fl
e 7 HEB 160 e 7
3.50
3.50
Fl Fl
e 6 HEB 180 e 6
3.50
3.50
Fl Fl
e 5 HEB 200 e 5
3.50
3.50
Fl Fl
e 4 e 4
3.50
3.50
Fl Fl
e 3 e 3
3.50
3.50
Fl Fl
e 2 e 2
3.50
3.50
Fl Fl
e 1 e 1
HEB 240
3.50
3.50
C1 C2 C3 C3 C2 C1 C1 C2 C3 C3 C2 C1
9.00 6.00 6.00 6.00 9.00 9.00 6.00 6.00 6.00 9.00
36.00 36.00
MRF-EBF DUAL V-SCHEME - 6 STOREYS MRF-EBF DUAL INVERTED Y-SCHEME - 6 STOREYS

Fl=(G +0.3Q )Ll·L=
k k Fl=(G +0.3Q )Ll·L=
k k
1490.40 kNm 1490.40 kNm

q =(G +0.3Q )L =6.90 kNm
d k k 0 q =(G +0.3Q )L =6.90 kNm
d k k 0
e e 6 6
.70
IPE 330 IPE 270 HEB 140 IPE 270 IPE 330 IPE 330 IPE 270 IPE 300 IPE 270 IPE 330
Fc1 Fc2 Fc3 Fc3 F c2 Fc1 Fc1 Fc2 Fc3 Fc3 Fc2 Fc1
3.50
3.50
Fl Fl
e e 5 5
.70
3.50
3.50
Fl Fl
e e 4 4
.70
3.50
3.50
Fl Fl
e e 3 3
.70
3.50
3.50
Fl Fl
e e 2 2
.70
3.50
3.50
Fl Fl
e e 1 1
.70
3.50
3.50
C1 C2 C3 C3 C2 C1 C1 C2 C3 C3 C2 C1
9.00 6.00 6.00 6.00 9.00 9.00 6.00 6.00 6.00 9.00
36.00 36.00
MRF-EBF DUAL V-SCHEME - 8 STOREYS MRF-EBF DUAL INVERTED Y-SCHEME - 8 STOREYS

Fl=(G +0.3Q )Ll·L=
k k Fl=(G +0.3Q )Ll·L=
k k
1490.40 kNm 1490.40 kNm

q =(G +0.3Q )L =6.90 kNm
d k k 0 q =(G +0.3Q )L =6.90 kNm
d k k 0
e e 8 8
.70
3.50
3.50
Fl Fl
e e 7 7
.70
3.50
3.50
Fl Fl
e e 6 6
.70
3.50
3.50
Fl Fl
e e 5 5
.70
3.50
3.50
Fl Fl
e e 4 4
.70
3.50
3.50
Fl Fl
e e 3 3
.70
3.50
3.50
Fl Fl
e e 2 2
.70
3.50
3.50
Fl Fl
e e 1 1
.70
3.50
3.50
C1 C2 C3 C3 C2 C1 C1 C2 C3 C3 C2 C1
9.00 6.00 6.00 6.00 9.00 9.00 6.00 6.00 6.00 9.00
36.00 36.00
Figure 4. Structral schemes of the longitudinal seismic resistant system
56
Push-over analyses have been led under displacement control taking into account both
geometrical and mechanical non-linearities. In addition, out-of-plane stability checks of
compressed members have been performed at each step of the non-linear analysis for both the
examined structures. The results provided by the pushover analyses are reported in Fig. 4, Fig.
5 and Fig. 6 for the 4-storey, 6-storey and 8-storey buildings, respectively. In particular, the
results provided by the analyses show that the softening branch of the push-over curve
corresponding to the structure designed by means of TPMC tends towards the mechanism
equilibrium curve obtained by means of second order rigid-plastic analysis. It is also useful to
underline that, in the examined cases, push-over curves exhibit a softening behaviour, because
the occurrence of strain-hardening in shear links does not counterbalance the softening due to
second order effects.
Table 2. Column sections

K-scheme D-scheme V-scheme Inv. Y-scheme
C1 C2 C3 C1 C2 C3 C1 C2 C3 C1 C2 C3
R=HEB 400
1 HEB 280 HEB 280 HEB 360 HEB 260 HEB 260 HEB 260 HEB 260 HEB 360 HEB 260 HEB 280 HEB 400
L=HEB 340
4-STOREY
R=HEB 400
L=HEB 340
R=HEB 340
L=HEB 280
R=HEB 340
L=HEB 240
C1 C2 C3 C1 C2 C3 C1 C2 C3 C1 C2 C3
R=HEB 550
L=HEB 500
6-STOREY BUILDING
R=HEB 500
L=HEB 450
R=HEB 450
L=HEB 400
R=HEB 450
L=HEB 400
R=HEB 450
L=HEB 340
R=HEB 400
L=HEB 260
C1 C2 C3 C1 C2 C3 C1 C2 C3 C1 C2 C3
R=HEB 700
L=HEB 700
R=HEB 650
L=HEB 650
8-STOREY BUILDING
R=HEB 600
L= HEB 600
R=HEB 600
L=HEB 550
R=HEB 550
L=HEB 550
R=HEB 550
L=HEB 500
R=HEB 500
L=HEB 400
R=HEB 400
8 HEB260 HEB260 HEB260 HEB 300 HEB 300 HEB 280 HEB 280 HEB 400 HEB260 HEB260 HEB260
L=HEB 280
57
R. MONTUORI et al.
2500 2500
Base Shear (kN)
Base Shear (kN)
4 st. K‐scheme 4 st. D‐scheme
2000 2000
1500 1500
ultimate design displacement
1000 1000
500 TPMC 500

TPMC
Top sway displacement (m) Top sway displacement (m)
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
2500 2500
Base Shear (kN)
Base Shear (kN)
4 st. V‐scheme 4 st. Inverted Y‐scheme
2000 2000
1500 1500
1000 1000
500 TPMC 500 TPMC
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
Figure 5. Push-over curves for 4-storey structures
5. IDA ANALYSES RESULTS

While previous works have investigated the different performances achieved by the structure
designed by means of TPMC and EC8, pointing out the higher performances of TPMC
designed strucutures [Montuori et al., 2015a; Montuori et al., 2015b] the scope of the IDA
analyses herein carried out is the evaluation of the seismic performances of the structures
considering the four link configuration proposed by EC8 given the number of storeys given the
same design approach. To this purpose IDA analyses have been carried out on the same
structural model already adopted for push-overs by assuming a 5% damping according to
Rayleigh with the proportional factors computed with reference to the first and third mode of
vibration (Table 3). Record-to-record variability has been accounted for by considering 10
recorded accelerograms selected from PEER data base [Pacific Earthquake Engineering
Research Center]. These recorded accelerograms have been selected to approximately match
the linear elastic design response spectrum of Eurocode 8, for soil type A. In addition, each
ground motion has been scaled to obtain the same value of the spectral acceleration, Sa(T1),
corresponding to the fundamental period of vibration T1 of the structure under examination;
successively Sa(T1) values have been progressively increased. The IDA analyses have been
carried out by increasing the Sa(T1)/g value until the occurrence of structural collapse,
corresponding to column or diagonal buckling or up to the attainment of the limit value of the
chord rotation of structural members that is assumed equal to 0.04 rad in the case of beams,
columns and diagonals and equal to 0.08 rad in the case of links [CEN, 2005].
58
2500 2500
Base Shear (kN)
Base Shear (kN)
2000 2000
Ultimate design displacement
1500 1500
1000 1000
500 TPMC 500 TPMC
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
2500 2500
Base Shear (kN)
6 st. V‐scheme
Base Shear (kN)
6 st. Inverted Y‐scheme
2000 2000
1500 1500
1000 1000
500 TPMC 500

TPMC
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
In Table 5, Table 6 and Table 7 the Sa(T1)/g values corresponding to the structural collapse for
each ground motion are reported. In particular, by comparing the average value of the collapse
Sa(T1)/g it is possible to observe that, given the number of storeys, the collapse condition is
always achieved, first by the V-scheme structures followed by the D-scheme structure and K-
scheme. Inverted Y-scheme structures always confirm the best performances which becomes
more relevant at the increasing of the structural height compared with the second classified of
the list, i.e. K-scheme. It is also possible to observe that the seismic performances of D-scheme
and V-scheme decrease as far as the structural height increase, therefore it is preferable not to
use D-scheme and V-scheme EBFs for tall buildings. Conversely, inverted Y-scheme increase
its performances as the number of storey increase, that makes it more suitable to tall building
application. Other benefits given by inverted Y-scheme EBFs, not belonging to the beam
member, are the chance to be easily substituted after destructive seismic events and to conceive
the scheme within the framework of supplementary energy dissipation strategy [Castaldo, 2014;
De Iuliis and Castaldo, 2012; Castaldo and De Iuliis, 2012; Palazzo et al., 2015; Castaldo et
al., 2016], by substituting the vertical link member with dissipative device, such as a friction
damper or hysteretic damper, which is able to exhibit a highly dissipative behaviour if
compared with traditional link members[Montuori et al., 2014b; Nastri, 2016].
In addition, such link configuration can be suitable also for the seismic retrofitting of existing
buildings [D’Aniello et al., 2006; Mazzolani et al., 2009].
59
R. MONTUORI et al.
2500 2500
Base Shear (kN)
Base Shear (kN)
2000 2000
1500 1500
1000 1000
500 TPMC 500

TPMC
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
2500 2500
Base Shear (kN)
Base Shear (kN)
8 st. V‐scheme 8 st. Inverted Y‐scheme
2000 2000
1500
1500
1000 1000
500 TPMC 500 TPMC
0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 1.2
Given the above, it is possible to conclude that all the EBF configuration show quite the same
performances for a lower number of storeys while the differences becomes more relevant as
the number of storeys increases. For this reason, a different behavior factor for the different
EBF configuration should be assumed. However, in this paper only a 5 bays structure with 4,
6 and 8 storeys have been investigated so that additional analyses on different structural
schemes with different number of bays should be carried out in order to provide a more
exhaustive research.
Table 3. First and third vibration mode period of buildings designed

4 STOREYS 6 STOREYS 8 STOREYS
TPMC EC8 TPMC EC8 TPMC EC8
T1 (s) T3 (s) T1 (s) T3 (s) T1 (s) T3 (s) T1 (s) T3 (s) T1 (s) T3 (s) T1 (s) T3 (s)
K-scheme 1.00 0.45 1.02 0.46 1.38 0.56 1.46 0.60 1.80 0.67 1.92 0.73
D-scheme 1.01 0.44 1.04 0.46 1.42 0.56 1.40 0.52 1.87 0.67 2.00 0.73
V-scheme 1.00 0.42 1.00 0.43 1.40 0.51 1.44 0.53 1.74 0.59 1.88 0.65
Inverted
1.01 0.44 1.08 0.47 1.39 0.54 1.47 0.58 1.62 0.60 1.70 0.63
Y-scheme
60
Table 4. Analyzed ground motion records

Earhquake Component Date PGA/g Length Step recording
(record) [-] [-] [-] (s) (s)
Victoria, Mexico Chihuahua) CHI102 1980/06/09 0.150 26.91 0.01
Coalinga (Slack Canion) H-SCN045 1985/05/02 0.166 29.99 0.01
Kobe (Kakogawa) KAK000 1995/01/16 0.251 40.95 0.01
Spitak, Armenia (Gukasian) GUK000 1988/12/17 0.199 19.89 0.01
Northridge (Stone Canyon) SCR000 1994/01/17 0.252 39.99 0.01
Imperial Valley (Agrarias) H-AGR003 1979/10/15 0.370 28.35 0.01
Palm Springs (San Jacinto) PALMSPR/H08000 1986/07/08 0.250 26.00 0.005
Santa Barbara (Courthouse) SBA132 1978/08/13 0.102 12.57 0.01
Friuli, Italy (Buia) B-BUI000 1976/09/15 0.110 26.38 0.005
Irpinia, Italy (Calitri) A-CTR000 1980/11/23 0.132 35.79 0.0024
Table 5. Sa (T1) values corresponding to the attainment of the collapse condition for the 4 storey
structures
4 st. K-scheme 4 st. D-scheme 4 st. V-scheme 4-storey Inv. Y
Sa(T1) Sa(T1) Sa(T1) Sa(T1)
Coalinga 0.90 g 0.70 g 0.70 g 0.80 g
Friuli, Italy 0.90 g 0.90 g 0.60 g 1.00 g
Imperial Valley 0.60 g 0.60 g 0.40 g 0.60 g
Irpinia, Italy 1.60 g 1.40 g 0.90 g 1.60 g
Kobe 0.80 g 0.90 g 0.50 g 0.90 g
Northridge 0.70 g 0.60 g 0.50 g 0.80 g
Palm Springs 0.70 g 0.80 g 0.60 g 0.80 g
Santa Barbara 0.80 g 0.70 g 0.50 g 0.90 g
Spitak Armenia 1.30 g 1.20 g 1.00 g 1.20 g
Victoria Mexico 0.60 g 0.50 g 0.40 g 0.80 g
Mean value 0.89 g 0.83 g 0.61 g 0.94 g
Table 6. Sa (T1) values corresponding to the attainment of the collapse condition for the 6 storey
structures
6 st. K-scheme 6 st. D-scheme 6 st. V-scheme 6 st. Inv. Y-scheme
Coalinga 0.70 g 0.70 g 0.50 g 0.80 g
Kobe 0.65 g 0.60 g 0.40 g 0.70 g
Northridge 0.50 g 0.50 g 0.40 g 0.70 g
Mean value 0.64 g 0.64 g 0.44 g 0.71 g
61
R. MONTUORI et al.
Table 7. Sa (T1) values corresponding to the attainment of the collapse condition for the 8 st. structures
8 st. K-scheme 8 st. D-scheme 8 st. V-scheme 8 st. Inv. Y-scheme
Coalinga 0.70 g 0.70 g 0.40 g 0.85 g
Kobe 1.00 g 1.00 g 0.60 g 0.90 g
Northridge 0.50 g 0.30 g 0.30 g 0.55 g
Mean value 0.60 g 0.55 g 0.42 g 0.82 g
6. CONCLUSIONS
In this paper, the evaluation of the influence of link configuration on seismic performances of
Moment Resisting Frames-Eccentrically Braced Frames dual systems (MRF-EBF dual systems)
designed by means of Theory of Plastic Mechanism Control (TPMC) is reported. TPMC is
based on the kinematic theorem of plastic collapse and its extension to the concept of
mechanism equilibrium curve and assure to design structure showing at the collapse a global
mechanism. For this reason, being assured and verified the collapse mechanism of global type
through push-over analyses, IDA have been carried out to check the actual performances of the
designed frames.
Results show that V-scheme structures always show the worse performances which even
decrease as far as the number of storey increase. On the contrary inverted Y-scheme confirms
the best performances. Notwithstanding, inverted Y-scheme increases its performances as the
number of storey increases, that makes this structural scheme more suitable for tall buildings,
while all the horizontal link EBFs decrease or at least keep constant their performances being
an increase of the number of storeys.
However, in this paper only a 5 bays structure with 4, 6 and 8 storeys have been investigated
so that additional analyses on different structural schemes with different number of bays should
be carried out in order to provide a more exhaustive research.
7. REFERENCES
Bosco, M., Marino, E.M., Rossi, P.P., (2015), “Modelling of steel link beams of short, intermediate or
long length”, Engineering Structures, Volume 84, February 01, Pages 406-418.
Castaldo P, De Iuliis M. Optimal integrated seismic design of structural and viscoelastic bracing-
damper systems. Earthquake Engineering and Structural Dynamics 2014;43(12):1809–1827.
Castaldo P. Integrated Seismic Design of Structure and Control Systems. Springer Inter-national
Publishing: New York, 2014. DOI 10.1007/978-3-319-02615-2.
Castaldo P., Amendola G., Palazzo B. (2016) “Seismic fragility and reliability of structures isolated by
friction pendulum devices: Seismic reliability-based design (SRBD)”, Earthquake Engineering and
Structural Dynamics, DOI: 10.1002/eqe.2798.
62
CEN (2005): “EN 1998-1-1: Eurocode 8 - Design of Structures for Earthquake Re-sistance. Part 1:
General Rules, Seismic Actions and Rules for Buildings", Comite Euro-peen de Normalisation,
CEN/TC 250.
D’Aniello M., Della Corte G., Mazzolani F.M., (2006). “Seismic Upgrading of RC Build-ings by Steel
Eccentric Braces: Experimental Results vs. Numerical Modeling”. STESSA 2006. Conference –
Steel Structures in Seismic Area. Japan, 14-17 August
D’Aniello M., La Manna Ambrosino G., Portioli F., Landolfo R., (2013). Modelling as-pects of the
seismic response of steel concentric braced frames. Steel and Composite Structures, An International
Journal, Vol. 15, No. 5, 539-566 November 2013
D’Aniello M., La Manna Ambrosino G., Portioli F., Landolfo R., (2015). The influence of out-of-
straightness imperfection in Physical-Theory models of bracing members on seismic performance
assessment of concentric braced structures. The Structural Design of Tall and Special Buildings, Vol.
24(3), 176-197.
D’Aniello M., Portioli F., Landolfo R., (2010). “Modelling issues of steel braces under extreme cyclic
actions”. COST-C26 Final Conference — Naples 16-18 September; pp. 335-341.
De Iuliis M, Castaldo P. An energy-based approach to the seismic control of one-way asymmetrical
structural systems using semi-active devices. Ingegneria Sismica - Interna-tional Journal of
Earthquake Engineering 2012; XXIX(4):31–42.
Della Corte G., D’Aniello M., Landolfo R., (2013). Analytical and numerical study of plastic
overstrength of shear links. Journal of Constructional Steel Research, 82, 19–32
Della Corte G., D’Aniello M., Mazzolani F.M., (2007). “Inelastic response of shear links with axial
restraints: numerical vs. analytical results” 5th International Conference on Advances in Steel
Structures, Singapore, 5 – 7 December.
Gulec C.K., Gibbons B., Chen A., Whittaker A.S.: “Damage States and Fragility Func-tions for Link
Beams”, Journal of Constructional Steel Research, Vol. 67, 2011, pp. 1299-1309, 2011.
Longo A., Montuori R., Piluso V. "Theory of plastic mechanism control for MRF–CBF dual systems
and its validation." Bulletin of Earthquake Engineering , pp. 1-31, 2014.
Longo A., Nastri E., Piluso, V. [2014],“Theory of Plastic Mechanism Control: State of the Art” The
Open Construction and Building Technology Journal, 2014, 8: 262-278
Mastandrea L., Nastri E., Piluso, V., “Validation of a Design Procedure for Failure Mode Control of
EB-Frames: Push-over and IDA analyses”, The Open Construction and Building Technology
Journal, 7, 193-207, 2013.
Mazzolani F.M., Della Corte G., D’Aniello M., (2009). “Experimental analysis of steel dissipative
bracing systems for seismic upgrading”. Journal of Civil Engineering and Management 15(1): 7-19.
Montuori R., Muscati R., “Plastic design of seismic resistant reinforced concrete frame”, Earthquakes
and Structures, Vol. 8, No. 1, January 2015 Pages 205-224.
Montuori R., Nastri E., Piluso V. (2013): “Rigid-Plastic Analysis and Moment-Shear In-teraction for
Hierarchy Criteria of EB-Frames with Inverted Y-Scheme”, Journal of Constructional Steel
Research, Volume 95, January 2014, pp. 71–80.
Montuori R., Nastri E., Piluso V. (2014a), “Theory of Plastic Mechanism Control for Eccentrically
Braced Frames with Inverted Y-scheme”, Journal of Constructional Steel Re-search, Volume 92,
January 2014, pp. 122–135.
Montuori R., Nastri E., Piluso V. (2014b) Theory of plastic mechanism control for the seismic design
of braced frames equipped with friction dampers - Mechanics Research Communications – Vol. 58.
pp. 112-123,.
63
R. MONTUORI et al.
Montuori R., Nastri E., Piluso V., (2015a) “Seismic Design of MRF-EBF Dual Systems with Vertical
Links: Ec8 Vs Plastic Design”, accepted for publication on Journal of Earthquake Engineering,.
Montuori R., Nastri, E., Piluso, V., (2015b) “Seismic Response of EB-Frames with Inverted Y scheme:
TPMC Versus Eurocode Provisions”, accepted for publication on Earthquakes and Structures.
Montuori, R., Nastri, E., Piluso, V., “Theory of plastic mechanism control for MRF-EBF dual systems:
Closed Form Solution”, Submitted for publication to Engineering Struc-tures, 2016.
Nastri E., “Eccentrically Braced Frames Designed for Energy Dissipation Optimization”, ECCOMAS
Congress 2016 VII European Congress on Computational Methods in Ap-plied Sciences and
Engineering, Crete Island, Greece, 5–10 June 2016.
Ohsaki M., Nakajima T.: “Optimization of Link Member of Eccentrically Braced Frames for Maximum
Energy Dissipation”, Journal of Constructional Steel Research, Vol. 75, 2012, pp. 38-44, 2012.
Pacific Earthquake Engineering Research Center, PEER Strong Motion Database,
http://peer.berkley.edu.smcat.
Palazzo B, Castaldo P, Marino I. (2015) “The Dissipative Column: A New Hysteretic Damper”.
Buildings;5(1):163-178; doi:10.3390/buildings5010163.
Piluso V., Nastri E., Montuori R., “Advances in Theory of Plastic Mechanism Control: Closed Form
Solution for MR-Frames”, accepted for publication on Earthquake Engi-neering and Structural
Dynamics, 2015
64
Num. …
ANALISI PRELIMINARE DELL’INFLUENZA DELLA

CONFIGURAZIONE NELLE PERFORMANCE SISMICHE DEI
SISTEMI ACCOPPIATI CON CONTROVENTI ECCENTRICI
PROGETTATI MEDIANTE LA TPMC
Rosario Montuori, Elide Nastri*, Vincenzo Piluso
Department of Civil Engineering , University of Salerno, Italy
SOMMARIO: Il lavoro riguarda l’influenza della configurazione del link nella valutazione
delle prestazioni sismiche di telai accoppiati con controventi eccentrici (MRF-EBF dual
systems) progettati mediante la Teoria del Controllo del Meccanismo di Collasso (TPMC).
Come è noto, tale teoria assicura che i telai progettati in accordo con essa manifestino al
collasso un meccanismo globale. Per tale motivo, con riferimento ai telai riportati nel presente
lavoro, l’unico elemento che influenza le prestazioni sismiche è la disposizione dei link. In ogni
caso, in questa memoria sono state riportate le analisi push-over e IDA per una struttura a 5
campate facendo variare il numero di piani (4, 6 e 8). Essendo essa solo un’analisi preliminare
risulta superfluo affermare che una campagna di analisi più approfondita, riguardante
strutture con diverso numero di piani e di campate sarà oggetto di un futuro approfondimento.
___________________
*Corresponding author: Elide Nastri, Department of Civil Engineering, University of Salerno, Italy.
Email: enastri@unisa.it
Num. …
THE NON-LINEAR RESPONSE OF

STEEL-CONCRETE COMPOSITE JOINTS
Maria Rosaria Pecce*
Department of Engineering – University of Sannio

Piazza Roma 21, Benevento, Italy
SUMMARY: In this paper the experimental results of two beam-column composite

connections (one welded and the other bolted) tested under cyclic loading are presented and
discussed. The two full-scale specimens were designed as sub-assemblages of a frame to
localize the damage at the end of the composite beam in respect of the capacity design,
therefore the joint contributed to the deformability of the system essentially in the elastic field.
The detail in the area around the column were realized according to the provisions of Eurocode
8 to assure the strength hierarchy by the local mechanisms of the concrete slab. The results
highlight the main role of the composite beam in terms of ductility; the rotational capacity and
an “equivalent” plastic hinge length are evaluated.
KEYWORDS: steel-concrete frames, nonlinear response, composite joints, plastic hinge

length
1. INTRODUCTION
In Moment Resisting Frame (MRF) solutions, an optimal seismic performance of the structure
can be attained when a global mechanism is achieved at the ultimate limit state. This result is
effective in RC structures when the dissipative mechanism involves all ends of the beams and
column bases (global mechanism), conversely for steel and steel-concrete composite
constructions the dissipation of the beams can be substituted by the one of the joint panels, as
clearly provided by Eurocode 8 [Eurocode 8,2004] and confirmed in various studies [as
example Bursi and Gramola, 2000, Thermou et al. 2004; Aribert et al, 2006;Braconi et al, 2008;
Braconi et al. 2008].
However, for the steel-concrete composite frames, the Italian code [NTC2008] allows only the
plasticization of the beam ends; therefore the capacity design method is suggested to obtain a
weak beam-strong column behaviour in the MRF global mechanism. This result is exploited if
a suitable hierarchy is realized between the beams and joints/columns, which ensures a ductile
and stable response of the dissipative elements [D’Aniello et al., 2012]; thus a detailed design
of the composite joint is of paramount importance to realize a reliable over strength respect to
the beam.
Currently the code provisions for the design of the composite joints and the knowledge from
the research are still lack and usually referred to bolted connection where the dissipation is
located in the steel parts of the joint (nodal panel, bolted plate) [Lee et al. 1989; Simões et al.
2001, Green et al. 2004; Salvatore et al, 2005; Vasdravellis et al, 2009].Therefore tests on two
composite beam-column connections have been carried out by the author addressing the
___________________
*Corresponding author: Maria Rosaria Pecce, Department of Engineering, University of Sannio, Benevento, Italy.
Email: pecce@unisannio.it
dissipation in the beam [Pecce and Rossi, 2015],within the activity of a research project funded
by the Italian Civil Defence (RELUIS project).This paper presents an analysis of the
experimental results of the two full-scale composite joints (one bolted and one welded) that
have been tested under cyclic loading, focusing on the relation between the design according
to Eurocode 8 and the effective performance, but also examining in detail the inelastic
performance of the beam.
2. THE EXPERIMENTAL TESTS

2.1 The specimens
The specimens were designed to represent a full scale sub-assemblage of a frame (Fig. 1),
which includes not only an external joint, but also the two half columns of the adjacent floors
and a part of the beam for a length longer than the critical zone and significant of the position
of the zero bending moment, when vertical and seismic loads are applied (i.e., of the gradient
of the bending moment).The forces loading the sub-assemblage in the real structure, the shear
and axial forces at the points of the zero moments, have been applied during the tests; therefore,
the experimental results provide the effective deformability of the entire frame in terms of the
rotation (θ) that governs the interstorey drift, and is due to various contributions in the elastic
and plastic field:
  b  c   j (1)
that are the rotation of the beam, the column and the joint, respectively.
An exterior joint was selected for the test because exterior joints have exhibited some pinched
hysteresis loops, whereas the cyclic behaviours of interior joints are usually more stable. Two
full-scale composite joints (Fig. 2a) were constructed, one flanged (FJ) and the other welded
(WJ). The specimen FJ was characterized by an extended end plate connection (Fig. 2b),
whereas the specimen WJ was characterized by the steel beam directly welded to the column
(Fig. 2a and 2c).
Figure 1. Scheme of the sub-assemblage
67
Table 1. Properties of the materials

design strength experimental average values
Material
(MPa) (MPa)
Concrete (C20/25) 20/1.5 24.3
fy ft εu
Steel for RC 450/1.15 460 549 15.3
Construction steel (flanges) 275/1.05 309 419 17.3
Construction steel (web) 275/1.05 394 481 13.6
The specimens were designed according to the European codes [Eurocode 4, 2004, Eurocode
8, 2004], applying the hierarchy of resistance for addressing the damage at the end of the beam.
The beam section is depicted in Fig. 2d); the column was realized with a wide flange steel
profile (HE280B) of height 280mm, web and flange thickness equal 10.5 mm and 18 mm
respectively.
The mechanical properties of the materials used for the construction of the specimen are
reported in Table 1 introducing their design value and the average experimental strength.
2.2 The design of the specimen

The steel profile of the beam is in class 1 [Eurocode 3, 2005], however the composite beam
becomes of class 2 when the hogging moment is applied and the height of web in compression
overcomes the half height of the steel section; anyway the plastic moment can be assumed as
flexural strength of the composite section. In Table 2 the plastic moments of the section are
reported assuming the design values of the materials strength (i.e. the nominal characteristics
values of the materials strength divided by the safety factors 1.5, 1.15 and 1.05 for concrete,
RC steel and construction steel, respectively); the plastic moment is evaluated also considering
the average values of the materials strength of Table 1 (the strength of the flange has been
assumed for the entire profile) and reported in the same Table 2.
Furthermore the flexural strength of the joint has been calculated considering the mechanisms
around the column provided by Eurocode 8 for an exterior joint without the transversal beam;
in fig. 3 the case of the positive and negative moments are schematized respectively. In
particular for the sagging moment the mechanism 1 and 2 are considered (fig. 3.a) that surely
are active when the beam slab has a concrete cantilever out of the column; analogously the
mechanisms of fig. 3.b are active for the hogging moment (Amadio et al. 2015). The steel
panel of the joint defined in the column web has been strengthen with plates on the two sides
only in FJ to assure the beam-joint hierarchy; these plates haven’t been applied to WJ to verify
the importance of this detail on the capacity design.
Table 2. Flexural resistance of the beam

Plastic moment (kNm)
design strength average strength
of materials of materials
 
M Rd M Rd M R M R
189 152 260 179
68
In Table 3 the joint strength is reported for the two specimens considering both the design and
average strengths of the materials, in the case of positive and negative moment. Comparing the
FJ and WJ specimen it is worth to notice that the strengthening of the joint panel (column web)
increases the flexural strength of the joint of about 20-30%.
Table 3. Flexural resistance of the joints.

Strength of FJ(kNm) Strength of WJ (kNm)
design strength average strength design strength average strength
of materials of materials of materials of materials
M j , Rd M j , Rd M j , R M j , R M j , Rd M j , Rd M j , R M j , R
235 235 307 278 177 177 237 210
a)
b) c) d)
e)
Figure 2. Details of the composite specimens: a) welded joint (WJ) and instrumentation
arrangement (LVDTs) used for both specimens; b) connection plate of the flanged joint(FJ); c)
connection plate of the welded joint (WJ); d) the beam section; e) the holes in the slab
69
Comparing the flexural strength of the beam and joint (Table 2 and 3) it results that the beam-
joint hierarchy (joint stronger than beam with safety factor according Eurocode 8) is respected
in the design of FJ for the positive and negative moment, while is respected in the design of
WJ only for the negative moment. The result is the same when the average strength is assumed
for the materials. The resistance of the welds in the specimen WJ and the bolts with the end
plate in the specimen FJ is greater than the beam resistance according to rules provided by
Eurocode 8 for the capacity design.
2.3 The load history and the set-up

The experimental program is more detailed in [Pecce and Rossi, 2015] while in this paper, a
brief description of the tests is provided to discuss the main results and the behavioural aspects.
The specimens were tested under a cyclic load applied at the end of the beam. The column of
the specimen FJ was loaded with an axial compression equivalent to 15% of its axial strength
by a hydraulic handling jack; the axial load wasn’t applied to the specimen WJ in order to avoid
the influence of the holes realized in the concrete slab to allow the crossing of the bars for its
application (Fig.2.e). The test procedure was based on the ATC-24 protocols (Report No.
ATC-24 – 1992);the cycles were established in 11 steps: three cycles per step until step number
five and two cycles per step thereafter. The designed sequence of the step amplitude was based
on the theoretical yielding displacement (dy); 3 equal cycles have been applied for
displacement amplitudes of 0.25dy ,0.5dy, 0.75dy; dy, 2dy then 2 equal cycles have been
applied for each assigned displacement.
Figure 3. Failure mechanisms of the slab at the joint. a) positive moment. b) negative moment
70
The yielding displacement, i.e. when the yielding moment My is achieved in the composite
beam, was evaluated by a simple FE mono-dimensional model in SAP2000, where the
deformability of the joint was introduced as estimated by the component method [Amadio et
al., 2011],obtaining an higher value for the specimen FJ that is more deformable, but the
yielding displacement of the specimen WJ under sagging moment (that is approximatively
13mm), i.e. the lower one, was assumed for both cases and assigned symmetrically for the
positive and negative load. The experimental tests were stopped at a cyclic history of
approximately 5-6dy, because local failure or global distortion of the specimens were attained,
as discussed in the next paragraph. The vertical load was applied to the beam by a 500 kN-
capacity hydraulic actuator with an available stroke of +/-250 mm. Many parameters were
recorded during the tests; in particular a combination of linear transducers and strain gauges
was deployed to identify the experimental behaviour of the joints. The general arrangement of
the LVDTs is depicted in figure 2.a.
2.3 Cyclic behaviour and failure mode
In Figure 4 the cyclic response of the two specimens is drawn as the relation between the
applied load and the displacement at the beam end. It can be observed that the behaviour of WJ
is stiffer than the one of FJ in the elastic range especially under the positive moment;
furthermore the shape of the hysteresis curves shows a greater phenomenon of pinching for the
specimen FJ due to the slippage of the bolted flange. The collapse of the specimen FJ was due
to the fracture in the bottom flange (fig. 4.a) near the weld and in the web after local buckling.
Also for the specimen WJ, the local buckling of the flange in compression occurred (fig. 4.b);
the last cycle was also characterized by the lateral-torsional buckling of the specimen due to
the extensive damage in the concrete slab and the loss of its effective restraint, therefore the
test was stopped. The local buckling wavelength for both specimens was measured and resulted
similar to the height of the steel section (240 mm). Table 4 summarizes the main experimental
results of the global behaviour: the displacement and load at yielding (dy and Fy), the maximum
load and the corresponding displacement (Fmax and dmax), the ultimate displacement (du) that
is the maximum displacement measured with a reduction of the load lower the 15% of the
maximum one. The yielding load was identified by the yielding of the rebars for the hogging
moment and the bottom steel flange for the sagging moment.
Figure 4. Cyclic response of the joints: a) the specimen FJ (fracture and local buckling), b)
the specimen WJ (local buckling)
71
Table 4. Main experimental results

dy+ dmax+ du+ du+/dy+ Fy+ Fmax+ My+ Mmax+
ID [mm] [mm] [mm] [-] [kN] [kN] [kN] [kN]
FJ (M+) 23 35 90 3.91 119 137 198 227
WJ (M+) 12 66 73 6.08 115 165 191 274
dy- dmax- du- du-/dy- Fy- Fmax- My- Mmax-
Id [mm] [mm] [mm] [-] [kN] [kN] [kN] [kN]
FJ (M-) -30 -88 -88 2.93 -85 -118 -141 -196
WJ (M-) -28 -50 -98 3.50 -101 -115 -168 -191
The maximum load was identified with the pick load. The sign indicates the sign of the bending
moment in the composite beam: positive when the RC slab is in compression and negative
when the RC slab is in tension. The maximum load results quite equal for the two specimens
under the hogging moment while it is lower for FJ respect to WJ under sagging moment since
the holes realized in the concrete slab to apply the axial load to the column, reduced the slab
strength in compression i.e. the flexural strength of the section. The yielding and maximum
load were multiplied by the distance of the force from the edge of the column, which is 1.66
m, to calculate the corresponding experimental moments (My and Mmax).
2.3 Comparison with the theoretical strength of the beam and joint
The experimental maximum moments can be compared with the plastic moments evaluated
using the average strength of the materials. Comparison of the values of Tables 2 and 4 reveals
that the maximum experimental moment is greater than the theoretical plastic moment of the
beam, except than in the case of the sagging moment for the specimen FJ, presumably due to
the local crashing of the concrete weaker for the presence of the holes. This higher values of
the experimental strength is surely due to the hardening of steel that was neglected in the
theoretical calculation; in all cases the average strength of the joint (Table 3) was not reached
except than for the specimen WJ under the sagging moment. It means that the beam-joint
hierarchy was respected in the FJ specimen, and also in the WJ for the hogging moment, as
designed, while for the sagging moment the yielding of WJ occurred before the beam
plasticization. However, the plasticization of the joint and beam under sagging moment was
attained about contemporary, furthermore the degradation of the concrete slab and the
hardening of the steel joint, allowed the complete plasticization of the beam preventing the load
increasing and stopping the plastic deformation of the joint. Finally the theoretical yielding
moment of the beam was evaluated with the elastic distribution of the stresses along the
composite section; in the case of sagging moment the steel yielding (fy) at the bottom fiber
of the steel profile is reached, whereas in the case of the hogging moment the yield of the
section corresponds to the yielding of the rebars. Assuming the average values of the material
properties (Table 1), the yielding moments result 187 kNm and 139 kNm for the positive and
negative one respectively; in both cases the experimental results (Table 4) of the two joints are
higher of the theoretical ones.
2.4 Ductility
As previously introduced, the deformability of the two sub-assemblages in bending, the
rotation θ, is the sum of the three contributions due to the column, the joint panel and the beam.
72
The rotation due to the panel zone and the column is evaluated by the transducers (LVDTs 3
and 4) placed along the column (Fig. 2.a), the distortion of the panel is defined by LVDTs 5
and 6, which are arranged on the column panel and allow to isolate the contribution of the
column as a difference. The rotation of the entire specimen (θ) is evaluated by the LVDTs
along the beam (LVDTs 7, 8, 9 and 15). The three contributions and the global response
skeleton curve of the specimens are graphed in Fig. 5 both for sagging and hogging moments.
It is worth to notice the predominant role of the composite beam in the global response, which
highlights the greater deformability of this element in the elastic and plastic field, especially
for the specimen FJ and the effectiveness of the design approach that addressed the damage in
the beam. However, in the case of the specimen WJ, a small contribution of the shear panel in
the plastic field can be observed (fig. 4.b); in fact in this case the design was carried out without
the details to ensure the beam-joint hierarchy (without the strengthening plates of the nodal
panel), and the beam plasticization was allowed by the over-strength of the joint, i.e. the
hardening of the steel. To define the ductility of the specimens, the ideal elastic-plastic curves
equivalent of the moment-rotation curves were developed. The bilinear relations were based
on the criterion of the “equivalent area” [Eurocode 8, 2004]; the elastic branch of the bilinear
curve is secant to the experimental curve and intersect it at 60% of the maximum moment, the
ultimate curvature is the same of the experimental curve. The contribution of the beam to the
post-elastic deformability of the connection was obtained by the bilinear equivalent curve of
the specimen deducting the column and joint contributions:
b    c   j (2)
As a result the ductility of the beam-column connection can be attributed to the ductility of the
critical zone of the beam because the column and the joint panel remain in the elastic field or
have limited plasticization. The rotation capacity of the beam underlines the formation of an
efficient plastic hinge, as shown in Table 5. The ductility (μ) in terms of rotation is given by
the following expression:
  
  u y  p (3)
y y
The ductility of the beam results greater than 4, with the exception of specimen WJ under the
positive moment (approximately 3), which is influenced by the participation of the joint panel
in the plastic field and the premature halt of the test due to the loss of the correct restraint of
the beam end. According to the information provided by FEMA 356, and Eurocode8-part3
(2005), the rotational capacity of the plastic hinges corresponds to the different limit states,
which are defined as function of the yielding chord rotation for each limit state (SL):
 SL  k SL y (4)
therefore the ratio between the ultimate rotation and the yielding one is represented by the
factor kSL The values of kSL for steel beams and columns realized with profiles of class 2
according to §Eurocode8-part3 Table B.1, i.e., section with a satisfactory rotational capacity,
are:
k SL  2 ; k CP  3 (5)
Respectively, for the “life safety” and “collapse prevention”. The values of these factors have
to be compared with the experimental results of the last column of Table 5; it is worth noticing
that in all cases of the ultimate limit state, the capacity of the composite beam confirms the
high capacity in ductility of the composite beam under both positive and negative moments.
73
Figure 5. Skeleton curves: contribution of the column, panel zone and beam to the total
response. a) FJ; b) WJ
Table 5. Yielding moment, rotations and ductility of the beam

My y u p μ kSL
[ID] [kNm] [rad] [rad] [rad] [-] [-]
FJ (M+) 192 0.0092 0.0519 0.0427 4.66 5.64
FJ (M-) 181 0.0084 0.0491 0.0407 4.85 5.85
WJ (M+) 258 0.0097 0.0379 0.0281 2.88 3.91
WJ (M-) 178 0.0119 0.0527 0.0408 4.42 4.43
3. EVALUATION OF THE EQUIVALENT PLASTIC HINGE LENGTH

The evaluation of the plastic rotational capacity of the structural elements, useful for non-linear
analysis, is quite simple using the concept of the equivalent plastic hinge length.
In order to evaluate the experimental equivalent plastic hinge length, the theoretical plastic
curvature of the beam was calculated as the difference between the ultimate and yielding
curvatures both for the positive and negative moment:
χ p  χu  χ y (6)
The moment-curvature relationship of the composite beam was developed by the fibers method
assuming the following data:
 a constitutive relationship of the concrete in compression,-, elastic-plastic with a
maximum strength of 24.3 MPa, that is the experimental average value, and an ultimate
deformation εcu of 0.01, which accounts for the confining condition of concrete due to
the pluri-axial stress condition of the slab;
 an elastic-plastic (EPP) law with a yield strength of 460 MPa and an actual ultimate
strain of 0.15 for the steel reinforcement;
 an (EPP) law with a yield strength of 309 MPa for the construction steel, which
corresponds to the experimental value listed in Table 1 for the steel flange, and a
conventional ultimate strain of 0.15;
 furthermore, an elastic-plastic with hardening (EPH) model and a critical stress approach
was assumed for the construction steel in compression to consider the effect of local
buckling on the moment-curvature relationship. The formulation of (Mazzolani et al.,
74
1992) was considered for the critical stress approach.Really for the evaluation of the
numerical moment curvature relationship under hogging moment the criterion of the
achievement of the critical stress is a reasonable compromise for engineering
applications, that allows to avoid the implementation of a more sophisticated method
that considers the post-buckling response of the steel section.
In Fig. 6 the numerical and experimental moment-curvature relationships of the composite
beam are compared. The experimental one is evaluated by the LVDTs located at the bottom of
the composite beam near the joint connection, i.e. LVDTs 7, 8 and 9 (Fig. 2a). The comparison
shows a good agreement between the experimental and numerical results. The equivalent
plastic hinge length is defined as the ratio between the experimental plastic rotation, as listed
in Table 5, and the numerical plastic curvature:
 (7)
L pe  p
χp
The numerical results are listed in Table 6. The equivalent plastic hinge length of the composite
beam is approximately 0.6h, where h is the height of the steel profile for the hogging moment
and approximately h for the sagging moment case. These results correspond with the provisions
for the steel elements but are different from the few provisions suggested for composite beams
for the plastic hinge length: Lp=1,75htot [Chen and Jia, 2008] for the hogging moment and Lp=h
[Bruneau et al, 1998] for the sagging one, where htot represents the total height of the composite
section. The discrepancy for Lpe obtained under the sagging and hogging moment is due to the
difference in terms of plastic curvature (p). For the sagging moment, the collapse of the
composite section is governed by the ultimate strain of the concrete (approximately 0.01) as
highlighted in the comparison with the numerical curve in Fig. 6. Under negative bending, the
ultimate condition is identified with the achievement of the critical deformation of the structural
steel in compression, which is generally larger for a compact steel section than for concrete.
Table 6 lists the equivalent plastic hinge length for both tests. The specimen WJ showed a
lower ductility for the sagging moment due to the interruption of the test.
4. CONCLUSIONS
In this paper, the experimental results of two full-scale composite sub-assemblages of beam-
column connections (a bolted joint, FJ, and a welded joint, WJ,) provide information about
their nonlinear performance to verify the failure mechanisms of the joint assumed in the design
process and calibrate an equivalent plastic hinge length of the beam.
Table 6. Equivalent plastic hinge length

Numerical Experimental
y+ u+ p+ y+ u+ p+ Lpe+ Lpe+/h
[1/mm] [1/mm] [1/mm] [rad] [rad] [rad] [mm] [-]
FJ (M+) 0.0000194 0.000159 0.000140 0.0092 0.0519 0.0427 305 1.27
WJ (M+) 0.0000194 0.000159 0.000140 0.0097 0.0379 0.0281 201 0.84
Numerical Experimental
y- u- p- y- u- p- Lpe- Lpe-/h
[1/mm] [1/mm] [1/mm] [rad] [rad] [rad] [mm] [-]
FJ (M-) 0.000016 0.000281 0.000265 0.0084 0.0491 0.0407 153 0.64
WJ (M-) 0.000016 0.000281 0.000265 0.0119 0.0527 0.0408 153 0.64
75
Figure 6. Comparison of the experimental and numerical moment-curvature relationship of the

beam section for WJ
The results of the experimental program agree with the adopted design approach of a “weak
beam-strong column” without the plasticization of the joint, and the post-elastic deformation
due to the beam. In particular, the following features have been highlighted:
 the design of the connection considering the beam-joint hierarchy according Eurocode 8
resulted effective since the plasticization of the beam was attained;
 the mechanisms of the concrete slab due to the presence of the concrete cantilever edge strip
were developed both under sagging and hogging moment;
 the strengthening of the nodal panel can have a certain influence on the resistance of the
joint, in the case of WJ, without the strengthening of the nodal panel, the yielding of the
steel panel was arisen;
 the elastic deformability of the entire system is influenced by the type of joint; the flanged
joint (FJ) was less stiff due to the deformability of the bolted connection;
 the ductility of the composite beams in terms of rotational capacity, exceeded 3; in one case,
this limit is lower because the test was halted. The values of the rotation capacity overcame
the acceptance criteria of Eurocode8 for a section of class 2 and FEMA356 at ultimate limit
states, for both life safety and collapse;
 the equivalent plastic hinge length Lpe of the composite beam, which was evaluated from
the experimental tests, is approximately 0.6h under hogging moment and approximately h
under sagging moment, where h is the height of the steel profile.
The efficiency of the design procedure of the composite joints according Eurocode 8 can be
considered a general result both for bolted and welded type; conversely the results about the
rotational capacity and plastic hinge length are reference values that need of further
investigations to be generalized for various composite beams.
5. ACKNOWLEDGEMENTS
This work was supported by research funding DPC-RELUIS 2013-2015. Thank you to Eng.
Fernando Rossi for his contribution to the research development and Eng. Giuseppe Logorano
for the work of his degree.
76
6. REFERENCES
Min. LL.PP, DM 14 gennaio 2008. Code design for construction (NTC2008).Gazzetta Ufficiale della
Repubblica Italiana, n. 29 [in Italian].
Amadio C, Bella M, Bertoni V, Macorini L. [2011] “Numerical modeling and seismic assessment of
steel and steel–concrete composite frames”. In: The line 5 of the ReLUIS-DPC 2005–2008 Project,
Doppiavoce (Napoli, Italy). p. 409–48.
Amadio C., Fasan M., Pecce M., Rossi F. [2015]. “Composite steel-concrete joints design and
modelling guidelines”, XXV CTA, Salerno, 2015. (in Italian)
ATC-24. [1992] Guidelines for cyclic seismic testing of components of steel structures for buildings.
Report no. ATC-24, Redwood City (CA): Applied TechnologyCouncil.
Aribert J-M, Ciutina AL, Dubina D.[2006] “Seismic response of composite structures including actual
behaviour of beam-to-column joints”. In: Composite construction in steel and concrete V. ASCE,
2006.
Braconi A, Bursi OS, Fabbrocino G, Salvatore W, Tremblay R. [2008a] “Seismic performance of a 3D
full-scale high-ductility steel–concrete composite moment-resisting structure. Part I: Design and
testing procedure”, Earthquake Engineering Structural Dynamics; 37:1609–34.
Braconi A, Bursi OS, Fabbrocino G, Salvatore W, Taucer F, Tremblay R. [2008b] “Seismic
performance of a 3D full-scale high-ductility steel–concrete composite moment-resisting structure.
Part II: Test results and analytical validation”, Earthquake Engineering Structural Dynamics,
37:1635–55.
Bruneau M., Whittaker A., and Uang C.M. [1998] “Ductile design of steel structures”, McGraw Hill
Edition.
Chen S., Jia Y. [2008] “Required and available moment redistribution of continuous steel–concrete
composite beams”, Journal of Constructional Steel Research, n° 64, pp. 167–175.
Bursi O.S., Gramola G. [2000] “Behaviour of composite substructures with full and partial shear
connection under quasi-static cyclic and pseudo-dynamic displacements”, Material
Structure/Material Construction; 33:154–63.
Chen S., Jia Y. [2008] “Required and available moment redistribution of continuous steel–concrete
composite beams”, Journal of Constructional Steel Research, n. 64, pp. 167–175.
D’Aniello, M., Landolfo, R., Piluso, V., Rizzano, G.[2012]“Ultimate behavior of steel beams under
non-uniform bending”, Journal of Constructional Steel Research. Vol. 78, pp. 144-158.
Eurocodice 3 [2005].“Design of steel structures”. Part 1–8: Design of joints, 2005.
Eurocode 4, [2008]. “Design of composite steel and composite structures”. Part EN 1993-1-8:2005.
Eurocode 8 [2004] “Design provisions for earthquake resistance of structures. Part 1.1. and 1.3: Federal
Emergency Management Agency. [2000] FEMA 356: prestandard andcommentary for the seismic
rehabilitation of buildings. Washington (DC,US).
General rules. Specific rules for various materials and elements”, European Committee for
Standardisation, Brussels, Belgium.
Green TP, Leon RT, Rassati GA. [2004] “Bidirectional tests on partially restrained, composite beam-
to-column connections”. Journal Structural Engineering; 130(2).
Lee S-J, Lu L-W. [1989] “Cyclic tests of full-scale composite joint subassemblages”, Journal Structural
Engineering,115(8).
Mazzolani, F. M., Piluso, V. [1992] “Evaluation of the rotation capacity of the steel beams and beam-
columns”, 1st Cost C1 Workshop, Strasbourg, pp. 28-30.
77
Pecce M. and Rossi F. [2015]“The experimental behavior and simple modeling of joints in composite
MRFs”. Engineering Structures; 105: 249-263.
SAC Steel Project [1997]Protocol for the fabrication, inspection, tests and documentation of beam–
column connection tests and other experimental specimens. Report no. SAC/BD 97-02 version 1.1,
California University.
Salvatore W, Bursi OS, Lucchesi D. [2005] “Design, testing and analysis of high ductile partial-strength
steel–concrete composite beam-to-column joints”, Computer Structure; 83(28–30):2334–52.
Simões R, Simões da Silva L, Cruz P. [2001] “Experimental behaviour of end-plate beam-to-column
composite joints under monotonical loading”, Engineering Structures; 23:1383–409.
Thermou GE, Elnashai AS, Plumier A, Doneux C. [2004] “Seismic design and performance of
composite frames”, Journal of Constructional Steel Research; 60:31–57.
Vasdravellis G., Valente M., CastiglioniC.A.. [2009] “Behavior of exterior partial‐strength composite
beam‐to‐column connections: Experimental study and numerical simulations”, Journal of
Constructional Steel Research, 65 23–35.
78
Num. …
LA RISPOSTA NON LINEARE DEI NODI COMPOSTI

ACCIAIO CALCESTRUZZO
Maria Rosaria Pecce*
Department of Engineering – University of Sannio

Piazza Roma 21, Benevento, Italy
SOMMARIO: In questo articolo si presentano e discutono i risultati di prove sperimentali

cicliche relative a due nodi trave-pilastro composti acciaio-calcestruzzo (uno saldato e l’altro
bullonato). I due provini sono in scala reale e sono stati progettati, in accordo con il criterio
sismico di gerarchia delle resistenze, come un sottosistema di un telaio composto in cui il
danno si deve verificare all’estremità della trave composta; pertanto il collegamento trave-
colonna contribuisce alla deformabilità del sistema essenzialmente in campo elastico. I
dettagli delle armature nella zona intorno al collegamento sono stati realizzati secondo le
indicazioni dell’Eurocodice 8 per assicurare la resistenza dei meccanismi locali della soletta
in calcestruzzo. I risultati evidenziano il contributo principale della trave composta alla
duttilità del sistema, e sono stati utilizzati per valutare la capacità rotazionale e lunghezza di
cerniera plastica della trave.
___________________
*Corresponding author: Maria Rosaria Pecce, Department of Engineering, University of Sannio, Benevento, Italy.
Email: pecce@unisannio.it
Num. …
IMPORTANCE OF LINK MODELS IN THE ASSESSMENT OF THE

SEISMIC RESPONSE OF MULTI-STOREY EBFS DESIGNED BY EC8
Melina Bosco, Aurelio Ghersi, Edoardo M. Marino, Pier Paolo Rossi*
Department of Civil Engineering and Architecture, University of Catania, Italy
SUMMARY: The nonlinear seismic response of eccentrically braced frames is strongly

correlated to the cyclic response of links, i.e. the dissipative members of this structural type. In
this paper, a simple but refined link model recently proposed by the authors is used to assess
the effectiveness of some common and simpler models in which the effect of the isotropic
hardening is not taken into account explicitly. The comparison is carried out by comparing the
dynamic response of 12 multi-storey eccentrically braced frames designed as per Eurocode 8.
The structures are characterised by different number of storeys and link lengths. The
investigation considers both global and local response parameters. The results show which
model gives the best estimate of the ultimate response of the frames. It is also shown that the
major differences between the responses predicted by the models are recorded for systems with
long links.
KEYWORDS: modelling, isotropic hardening, kinematic hardening, eccentrically braced

frames, non-linear dynamic analyses
1. INTRODUCTION
Several models have been used in the past to replicate the cyclic behaviour of links. When the
nonlinear response of eccentrically braced structures is determined, the adopted link models
are generally simple and usually based on the Euler-Bernoulli beam with concentrated flexural
plastic hinges [Roeder and Popov, 1977; Ricles and Popov, 1987; Ramadan and Ghobarah,
1995; Richards and Uang, 2003]. Ending springs with an elastic plastic with kinematic
hardening or an elastic perfectly plastic behaviour are often considered to simulate the inelastic
shear and flexural response of the link [Rossi and Lombardo, 2005; Bosco and Rossi, 2009;
2013b; Badalassi et al, 2013; Montuori et al., 2014a-b; 2015a-b]. On the other hand, finite
element models are also developed to predict the response of links in more detail and generalize
the results of the laboratory tests to ranges of geometric and mechanical properties not fully
investigated yet. [Ghobarah and Ramadan, 1991; Itani et al., 2003; Richards and Uang, 2005;
Chao et al., 2005; Berman and Bruneau, 2008; Prinz and Richards, 2009; Daneshmand and
Hashemi, 2012; Della Corte et al., 2013]. However, because of the computation burden and
complexity of these models, they are not suitable for inelastic dynamic analyses of structures
with eccentric bracings. Recently, the authors have proposed a simple but refined link model,
which consists of an elastic beam with concentrated flexural and shear plastic hinges [Bosco et
al., 2015a]. The uniaxial material model proposed by Zona and Dall’Asta [2012] dictates the
response of the hinges.
___________________
*Corresponding author: Pier Paolo Rossi, Department of Civil Engineering and Architecture, University of
Catania, Italy.
Email: prossi@dica-unict.it
M. BOSCO et al.
M M
EL2 EL2
V V
EL0
EL1 EL1
e
Figure. 1. Elements of the link model
This uniaxial material was proposed as an improvement over the elastic-plastic models with
only kinematic hardening used in studies on buckling restrained braces [Kim and Choi, 2004;
Bosco and Marino, 2013]. The proposed link model takes into account both kinematic and
isotropic hardening [Formisano et al., 2006; De Matteis et al., 2012; Brando and De Matteis,
2014; Rossi, 2015] but neglects the effect of the axial force that may develop in the link
[D’Aniello et al., 2006; Della Corte et al., 2007; Mazzolani et al., 2009]. Thus, this model is
proper to replicate the cyclic response of links arranged in the split K configuration. In a
previous investigation [Bosco et al., 2015a], the proposed model and other simple existing
models have been adopted to replicate the cyclic response of short, intermediate and long links
for which laboratory test data are available [Hjelmstad and Popov, 1983; Malley and Popov,
1983; Engelhardt and Popov, 1989; Arce, 2002; Galvez, 2004; Okazaki et al., 2005; 2006;
2007; 2009; Drolias 2007]. The obtained results have proved the effectiveness of the proposed
model. Based on these results, the new model is considered here to assess the effectiveness of
some common and simpler models (in which the effect of the isotropic hardening is not taken
into account explicitly) in predicting the dynamic response of multi-storey structures subjected
to artificial accelerograms. The structures are characterised by different number of storeys and
link lengths and are designed as per Eurocode 8 [2003]. The investigation considers both global
and local response parameters, namely ultimate peak ground acceleration, plastic rotations of
links and residual drifts.
2. DESCRIPTION OF THE MODELS

Three different models are compared in this paper. All of them consist of elements connected
in series (Fig. 1). The central element (EL0) has the same length and moment of inertia of the
link and simulates the flexural elastic behaviour of the link (the shear stiffness of this element
is infinite). The two elements at each end of the link (EL1 and EL2) are zero length and connect
the beam segments outside the link to the central element EL0. The first of these two elements
(EL1) simulates the elastic and inelastic shear behaviour of half a link, while the second (EL2)
simulates the inelastic flexural behaviour of the ending part of the link (a very high value of
the elastic stiffness is assigned to this element). The nodes of EL1 may have only relative
vertical displacements while those of EL2 only relative rotations. The models differ because of
the inelastic response of the elements.
2.1 Model 1: isotropic and kinematic hardening

The first model (M1) has been proposed by Bosco et al. [2015a] and considers isotropic
hardening explicitly. The response of the two elements EL1 and EL2 is defined by means of
the uniaxial material model proposed by Zona and Dall’Asta [2012] for buckling restrained
braces. This uniaxial material model considers a simple rheological scheme where a spring “0”
82
is connected in series with a friction slider in parallel with a spring “1”. The stiffness k0 of the
spring “0” is equal to the initial elastic stiffness of the element while the stiffness k1 of the
spring “1” influences the kinematic hardening. The hysteretic response is described by the
nonlinear relationship between force and deformation in the friction element. The full
description of this model requires that values be given to the stiffness k0 and k1, to the initial
yield force Fy0 and to the maximum yield force for the fully saturated isotropic hardening
condition Fymax. In addition, the values of two positive non-dimensional constants have to be
specified. The first constant β controls the rate of the isotropic hardening; the second, α,
controls the trend of the transition from the elastic to the plastic response.
The elastic stiffness kL1,0 of element EL1 is calculated by means of the equation:
GA
k L1, 0  (1)
e 2
where A is the area of the link cross-section, G is the tangent modulus of elasticity, e is the link
length and χ is the shear coefficient. This latter parameter is calculated [Bosco et al., 2015a] as:
b d  t f   2 
5
b2
 15  10   5  2  (2)
240 A  i 4   d  t f  2

where i is the radius of gyration of the cross-section,   t w b and   2 t f d  t f  , b is the
width of the flange, d is the depth of the section, tf is the thickness of the flange, tw is the
thickness of the web. The initial yield force Vy of element EL1 is calculated taking into account
only the contribution of the web, i.e.
Vy  Vp  0.6 f yw t w d  2tf  (3)
where fyw is the tensile yield strength. Based on several numerical tests on links, the following
values of the parameters above have been suggested [Bosco et al., 2015a] to simulate the cyclic
response of links in which stiffeners are disposed as required by modern seismic codes
k L1,1  0.442 % k L1,0 V y ,max  1.308 V y (4a)
 L1  0.485  L1  0.113 (4b)
The ultimate shear force of short links (i.e. the shear force corresponding to a plastic rotation
angle equal to 0.08 rad) corresponding to the fully saturated isotropic hardening is:
0.08  e 2  k L1,0
Vu  1.308  Vp  (5)
k L1,0 k L1,1  1
Element EL2 is characterised by a very high elastic stiffness kL2,0 because the elastic flexural
deformability of the link is simulated by element EL0. The bending moment at yield My is
assumed equal to the plastic moment Mp of the entire cross-section:
M p  f yf b t f d  t f   f yw w d  2t f 
t 2
(6)
4
where fyf is the tensile yield strength of the flanges. The suggested values of the post-elastic
stiffness kL2,1, that of the fully saturated bending moment My,max, and those of the parameters
αL2 and βL2 are:
6 EI
k L2,1  0.795% M y ,max  1.212 M y (7a)
e
83
M. BOSCO et al.
 L2  0.025  L2  0.054 (7b)
V M
Vu Mu
Vymax kL1,1 Mymax kL0,1
Vp Mp
kL1,0 EI
(a) 0.08 e/2  0.02 / epl χ

V M
Vu Mu
Vymax kL1,1 Mymax kL0,1
Vp Mp
kL1,0 EI
(b) 0.08 e/2  0.02 / epl χ
Fig. 2. Response of element EL1 and plastic hinge of element EL0 according to
model: (a) M2, (b) M3
Owing to this, the ultimate bending moment of long links (i.e. the bending moment
corresponding to a plastic rotation angle equal to 0.02 rad) is:
M u  1.212  M p  0.02  k L2,1 (8)
2.2. Models 2 and 3: kinematic hardening

In the second model (M2) and in the third model (M3) the behaviour of elements EL1 is elastic-
plastic with kinematic hardening (Fig. 2a, 2b). The effect of the isotropic hardening is not
considered explicitly in these models. In particular, in model M2 an equivalent kinematic strain
hardening is used to include the effects of both isotropic and kinematic hardening while in
model M3 the effect of the isotropic hardening is considered assuming that the yield shear force
and the yield bending moments are higher than the plastic values provided by Eqs. (3) and (6).
Specifically, in model M2, the elastic stiffness and the plastic shear force of element EL1 are
equal to those assigned to model M1. The post elastic stiffness is such that the shear force
corresponding to a plastic rotation angle equal to 0.08 rad is equal to that provided by model
M1 at the same plastic rotation angle, i.e.
Vu  Vp
k L1,1   k L1,0 (9)
0.08  e 2  k L1,0  Vu  Vp 
84
L = 8.0 m
Y
bracing
24.0 m X
C2
C0 C1
24.0 m
Figure 3. Plan of the building
Elements EL2 are not included in the model because element EL0 is a beam with hinge element.
The plastic hinges of element EL0 are characterised by a length epl equal to 1/100 e and by an
elastic-plastic with kinematic hardening moment-curvature relationship. This moment-
curvature relationship is characterised by a plastic bending moment equal to Mp, an elastic
flexural stiffness equal to EI and a post elastic stiffness equal to:
Mu  Mp
k L0,1   EI (10)
0.02  EI epl  M u  M p 
The flexural stiffness in the equation above is such that the bending moment corresponding to
a plastic rotation angle equal to 0.02 rad is equal to that provided by model M1 at the same
plastic rotation angle. Instead, in model M3, the yield shear force and the yield bending
moments are equal to the values corresponding to the fully saturated isotropic hardening
condition, i.e. Vy = Vy,max and My = My,max. In addition, the post yield stiffness kL1,1 of element
EL1 is equal to the corresponding stiffness of model M1 while the post yield stiffness of the
plastic hinge of element EL0 is:
M u  M y,max
k L0,1   EI (11)
0.02  EI epl  M u  M p 
3. ANALYSED MULTI-STOREY STRUCTURES

The eccentrically braced frames investigated in this paper were designed in previous research
[Bosco et al., 2014, 2015b] and constitute simplified models of the structure of an apartment
building having squared plan (24m x 24m) and geometric and mass properties equal at all
floors (Fig. 3). The structure of this building is defined by the intersection of two groups of
four three-span plane frames oriented along two orthogonal directions and located
symmetrically to the geometric centre. Frames on the perimeter of the building are designed to
resist the entire horizontal force and are endowed with eccentric braces disposed in the central
span according to the split K-braced configuration. The interstorey height is equal to 3.3 m. To
nullify the interaction between deck and links, two beam members are introduced at each level
of the eccentrically braced frame instead of the traditional single section. Systems with
85
M. BOSCO et al.
different dynamic and mechanical properties are obtained by varying the number of storeys ns
from 4 to 12 (in step of 4) and the link length e. This length is equal to either 0.1 to 0.3 times
the length L of the braced span. Vertical dead and live loads are constant on every floor level
and are defined by characteristic values (Gk and Qk) equal to 4.4 and 2.0 kN/m², respectively.
The structures stand on soft soil (soil C according to Eurocode 8) and are designed assuming a
peak ground acceleration equal to 0.35 g and a behaviour factor equal to 5. The design internal
forces on members are determined by either the modal response spectrum analysis (MRSA) or
the lateral force method of analysis (LFMA). The non-dissipative members are designed
according to the capacity design principles as per Eurocode 8. All the members that do not
belong to the braced frames are designed to sustain gravity load only. In the following, the
frames are identified by a label obtained by adding the number of storeys (04, 08 or 12), the
link length (10 or 30 for systems with e/L equal to 0.10 or 0.30, respectively) and the design
method of analysis (M for MRSA or S for LFMA).
4. NUMERICAL ANALYSES AND RESPONSE PARAMETERS

In this section, the response of the multi-storey structures is determined by incremental
nonlinear dynamic analysis. The numerical analyses are performed by means of the program
OpenSEES [Mazzoni et al., 2003]. The numerical model includes both the braced frames and
the gravity columns, which are pinned at the base. To promote a more uniform distribution of
drift demand along the height of the building [Tremblay, 2003; Brandonisio et al., 2006] the
columns are continuous over the entire height of the building. Links are modelled by each of
the considered models; beam segments outside links, braces and columns are non-dissipative
members and for this reason they are modelled by means of elastic elements. As a consequence,
the differences between the responses of the corresponding frames are only due to the
considered model of the link. Beam-to-column connections are assumed perfectly pinned; the
numerical model neglects the effect of the panel zone deformation because yielding and
buckling of the panel zone are assumed to be prevented by a proper choice of its geometrical
slenderness [Brandonisio et al., 2011, 2012]. The connections at the end of braces are welded.
The Rayleigh formulation is used to introduce damping. Mass and stiffness proportional
damping coefficients are defined so that the first and the third modes of vibration of the
structure are characterised by an equivalent viscous damping factor equal to 0.05. In
accordance with Ricles and Popov [1987], no stiffness proportional damping is considered for
the zero length elements of the link. The seismic input consists of 10 artificially generated
accelerograms [Amara et al., 2014]. P-Δ effects are included in the numerical analyses. The
peak ground acceleration is scaled in steps of 0.04g. For each considered value of peak ground
acceleration, seismic demands of dissipative and non-dissipative elements are verified not
exceed the corresponding capacities in terms of deformations and strengths, respectively. Thus,
the peak ground acceleration is scaled up to the value agu that leads to the first achievement of
the ultimate rotation capacity φu of links or to yielding or buckling of non-dissipative members.
Specifically, the plastic rotation capacity φu of links is calculated according to Eurocode 8 as:
0.08 1.6
0.02 3.0 (12)
1.6 3.0
86
The safety level of non-dissipative members is calculated with reference to both yielding (SLy)
and instability (SLb). In particular, at each time t of the time history, the bending moments M(t)
and axial forces N(t) are calculated. Then, SLy is calculated as the maximum ratio of the
bending moment to the flexural strength MN,Rd reduced because of the axial force while SLb is
determined as the maximum ratio of the axial force to the buckling resistance Nb,Rd(M) reduced
because of the bending moment. Yielding and buckling resistances for combined values of
axial force and bending moment of non-dissipative members are calculated according to
Eurocode 3 [2005] assuming that the partial safety factors γM0 and γM1 are equal to 1.
For each accelerogram scaled to agu, the heightwise distribution of the normalised plastic
rotation of links and that of the residual drift angle is investigated. Specifically, the normalised
plastic rotation of links is calculated as the ratio of the maximum plastic rotation φ required at
the link of the i-th storey to the corresponding rotation capacity φu. The residual drift angles
Δres, i.e. the ratio of the residual drifts to the interstorey height, are calculated according to the
procedure described in Bosco et al., [2015c].
5. COMPARISON OF THE MODELS

In this section, the values of agu, φ/φu and Δres predicted by the considered models are averaged
over the number of the considered accelerograms and compared. The average values of agu
obtained for the considered models are compared in Fig. 4. In the figure, different colours are
used to represent results obtained by the considered models. The figure shows that the models
provide values of agu very close to each other when applied to predict the response of low-
storey systems with short links (0410M, 0410S) or high storey systems designed by LFMA
(0810S, 1210S, 0830S, 1230S). The greatest differences between the values of agu are obtained
when considering systems designed by MRSA, especially in the case of long links (0830M,
1230M). Model M2 (dark grey bars) gives the values of agu closest to those provided by model
M1 (black bars), although it significantly underestimates the ultimate peak ground acceleration
in the case of systems with long links designed by MRSA. Model M3 (light grey bars) provides
the lowest values of agu. In order to investigate deeper on the above-mentioned differences, it
is worth noting that in systems with short links the value agu is always related to the first
achievement of the ultimate rotation capacity of links; instead, in systems with long links, the
bending moment of the beam segments outside links may reach values higher than the flexural
strength (SLy larger than unity). In particular, for the considered structures with long links,
models M2 and M3 always predict the achievement of the yielding resistance of the beam
segment outside the link while, according to model M1, the achievement of the ultimate plastic
rotation of the link is recorded only for some accelerograms. Thus, the differences on the
predicted values of agu are partially due to the different failure mechanism and are related to
the different values of the bending moment that, according to the considered models, are
transmitted by the link for assigned values of plastic deformation. Fig. 5 shows the comparison
between the normalised plastic rotations of links predicted by the models for the considered
12-storey structures. The solid line is used to represent the results predicted by model M1,
whereas different types of hatches are used for the other models. For the structure with short
links (e/L = 0.10) designed by MRSA (Fig. 5a), the normalised plastic rotation of links is
significant at all storeys. However, the values predicted by the models are scattered especially
at the lower storeys. As an example, at the third storey, the percentage difference between the
normalised plastic rotation provided by models M2 and M3 and that given by model M1 is 30%
and -58%, respectively. In the corresponding system designed by LFMA (Fig. 5b), the
87
M. BOSCO et al.
normalised plastic rotation is very high only at the upper storey while the plastic behaviour of
the links of the other storeys is limited. Indeed, this system is characterised by scattered values
of the overstrength and low damage distribution capacity (Bosco and Rossi, 2013a). For this
reason, all the models provide a similar distribution of the considered response parameter. In
the systems with long links designed by either MRSA or LFMA (e/L = 0.30), the plastic
rotations predicted by models M2 and M3 are very low because they are obtained for ultimate
peak ground accelerations that are significantly lower than those corresponding to the model
M1. Even if not shown in any figure, these results are confirmed when considering 4- and 8-
storey frames. Fig. 6 shows the comparison between storey drift angles predicted by the three
considered models for the same structures analysed in Fig. 5 in terms of normalised plastic
rotations of links. When short links are considered, model M2 is that providing the results
closest to those of model M1. However, the percentage differences between the predicted
residual drift angles are significantly greater than those obtained with reference to the
normalised plastic rotation of links. Model M3 provides larger residual drifts, especially in the
upper storeys. When long links are considered, models M2 and M3 give residual drifts close to
zero. This result is not surprising owing to the low values of the plastic rotations of links
corresponding to the ultimate peak ground accelerations predicted by these models.
6. CONCLUSIONS
This paper investigates the effectiveness of some simplified models of steel link beams
commonly used in the seismic assessment of split K eccentrically braced frames. The
benchmark is represented by the seismic response obtained by means of a simple but refined
link model (M1 model) that has recently been proposed by the writers and that takes into
account both kinematic and isotropic hardening. The comparison is carried out on the seismic
response at collapse of 12 eccentrically braced frames designed as per Eurocode 8. The seismic
response at collapse of the analysed frames is determined by incremental nonlinear dynamic
analysis. The frames differ because of the number of storeys, the geometric link length and the
design method of analysis. The seismic response is expressed in terms of the ultimate peak
ground acceleration, link normalised plastic rotations and residual storey drift angles.
agu0.5
/g
0.4
0.3
0.2
0.1
M1
0
M2
0410S
0430S
0810S
0830S
1210S
1230S
0410M
0430M
0810M
0830M
1210M
1230M
EBF M3
Figure 4. Comparison between the predicted ultimate peak ground accelerations
88
e/L = 0.10 e/L = 0.30

12 12
M2
M3
M1
8 8
storey
storey
(a)
4 4
0 0
0.00 0.25 0.50 0.75 φ/φu 0.00 0.25 0.50 0.75 φ/φu
12 12
8 8
storey
storey
(b)
4 4
0 0
0.00 0.25 0.50 0.75 φ/φu 0.00 0.25 0.50 0.75 φ/φu
Figure 5. Comparison between the predicted normalised plastic rotations of links: systems designed
by (a) MRSA, (b) LFMA
e/L = 0.10 e/L = 0.30
12 12
M1 M3
8 8
storey
storey
(a)
4 4
M2
0 0
0.00 0.25 0.50 0.75 Δres (%) 0.00 0.25 0.50 0.75 Δres (%)
12 12
8 8
storey
storey
(b)
4 4
0 0
0.00 0.25 0.50 0.75 Δres (%) 0.00 0.25 0.50 0.75 Δres (%)
Figure 6. Comparison between the predicted residual storey drift angles: systems designed by (a)
MRSA, (b) LFMA
89
M. BOSCO et al.
The results of incremental nonlinear dynamic analyses show that the simplified model in which
the effect of isotropic hardening is represented by an equivalent (increased) kinematic
hardening (model M2) gives the values of ultimate peak ground acceleration closest to those
provided by the reference model for systems with short links. Instead, the simplified model in
which the effect of isotropic hardening is represented by increasing the value of the yielding
shear force and bending moments (model M3) underestimates significantly the same response
parameter.
The major differences between the responses predicted by the analysed models are recorded
for systems with long links. Indeed, in this case, models M2 and M3 predict the yielding of the
beam segment outside link while this behaviour is recorded only for some accelerograms when
the reference model is adopted. The greatest differences between the models are obtained for
residual drift prediction.
In order to generalize the results obtained in this study, the comparison should be extended to
eccentrically braced frames designed according to different design procedures. Finally, the
variability of the response related to the model should be compared to that related to the seismic
input.
7. REFERENCES
Amara, F., Bosco, M., Marino, E. M., Rossi, P. P. [2014] “An accurate strength amplification factor for
the design of SDOF Systems with P-Δ Effects”, Earthquake Engineering and Structural Dynamics
43, 589–611.
Arce, G. [2002] Impact of higher strength steels on local buckling and overstrength oflinks in
eccentrically braced frames, Master's thesis, Dept. of Civil and Environmental Engineering,
University of Texas at Austin, Austin, TX.
Badalassi, M., Braconi, A., Caprili, S., Salvatore, W. [2013] “Influence of steel mechanical properties
on EBF seismic behaviour”, Bulletin of Earthquake Engineering 11, 2249–2285
Berman, J. W., and Bruneau, M. [2008] “Tubular links for eccentrically braced frames. I: finite element
parametric study”, Journal of Structural Engineering 134(5), 692-701.
Bosco, M., and Rossi, P. P. [2009] “Seismic behaviour of eccentrically braced frames”, Engineering
Structures 31, 664-674.
Bosco, M., and Rossi, P. P. [2013a] “A design procedure for dual eccentrically braced systems:
Analytical formulation”, Journal of Constructional Steel Research 80, 440-452.
Bosco, M., and Rossi, P. P. [2013b] “A design procedure for dual eccentrically braced systems:
Numerical investigation”, Journal of Constructional Steel Research 80, 453-464.
Bosco, M., and Marino, E. M. [2013] “Design method and behavior factor for steel frames with buckling
restrained braces”, Earthquake Engineering and Structural Dynamics 42(8), 1243-1263.
Bosco, M., Marino, E. M, and Rossi, P. P. [2014] “Proposal of modifications to the design provisions
of Eurocode 8 for buildings with split k eccentric braces”, Engineering Structures 61, 209-223.
Bosco, M., Marino, E. M., Rossi, P. P. [2015a] “Modelling of steel link beams of short, intermediate or
long length”, Engineering Structures 84, 406-418.
Bosco, M., Marino, E. M, and Rossi, P. P. [2015b] “Critical review of the EC8 design provisions for
buildings with eccentric braces”, Earthquakes and Structures 8(6), 1407-1433.
Bosco, M., Marino, E. M., and Rossi, P.P. [2015c] “Design of steel frames equipped with BRBs in the
framework of Eurocode 8”, Journal of Constructional Steel Research 113, 43–57.
90
Brando, G., and De Matteis, G. [2014] “Design of low strength-high hardening metal multi-stiffened
shear plates” Engineering Structures 60, 2-10.
Brandonisio, G., De Luca, A., Grande, E., and Mele E. [2006] “Non-linear response of concentric
braced franmes”, Proc. of the International Conference STESSA 2006, Japan, 399-405.
Brandonisio, G., De Luca, A., and Mele E. [2011] “Shear instability of panel zone in beam-to-column
connections”, Journal of Constructional Steel Research 67, 891–903.
Brandonisio, G., De Luca, A., and Mele, E. [2012] “Shear strength of panel zone in beam-to-column
connections”, Journal of Constructional Steel Research 71, 129–142.
Chao, S. H., Khandelwal, and El-Tawil, S. [2006] “Ductile web fracture initiation in steel shear links”,
Journal of Structural Engineering 132(8), 1192-1200.
D’Aniello, M., Della Corte, G., and Mazzolani, F. M. [2006] “Seismic Upgrading of RC Buildings by
Steel Eccentric Braces: Experimental Results vs. Numerical Modeling”, Proc. of the International
Conference STESSA 2006, Japan, 809-814.
Daneshmand, A., and Hashemi, B. H. [2012] “Performance of intermediate and long links in
eccentrically braced frames”, Journal of Constructional Steel Research 70, 167-176.
De Matteis, G., Brando, G., and Mazzolani, F. M. [2012] “Pure aluminium: An innovative material for
structural applications in seismic engineering”, Construction and Building Materials 26(1), 677-686.
Della Corte, G., D’Aniello, M., and Mazzolani, F. M. [2007] “Inelastic response of shear links with
axial restraints: numerical vs analytical results”. Proc. of the 5th International conference on
Advances in Steel Structures, ICASS 2007, 651-656
Della Corte, G., D’Aniello, M., and Landolfo, R. [2013] “Analytical and numerical study of plastic
overstrength of shear links”, Journal of Constructional Steel Research 82, 19-32.
Drolias, A. [2007] Experiments on link-to-column connections in steel eccentrically braced frames,
Master's thesis, Department of Civil Engineering, University of Texas, Austin.
Engelhardt, M. D., and Popov, E. P. [1989] Behavior of long links in eccentrically braced frames,
Report No. UCB/EERC-89/01, Earthquake Engineering Research Center, University of California,
Berkeley, 1989.
Eurocode 3. [2005] Design of steel structures – Part 1–1: General rules and rules for buildings, ENV
1993–1–1. European Committee for Standardization, Bruxelles.
Eurocode 8. [2003] Design provisions for earthquake resistance of structures. European Committee for
Standardisation, ENV 1998-1-1/2/3.
Formisano, A., Mazzolani, F. M., Brando, G., and De Matteis, G. [2006] “Numerical evaluation of the
hysteretic performance of pure aluminium shear panels”, Proc. of the International Conference
STESSA 2006, Japan, 211-217.
Galvez, P. [2004] Investigation of factors affecting web fractures in shear links, Master’s thesis,
Department of Civil Engineering, University of Texas, Austin.
Ghobarah, A., and Ramadan, T. [1991] “Seismic analysis of links of various lengths in eccentrically
braces frames”, Canadian Journal of Civil Engineering 18, 140-148.
Hjelmstad, K. D., and Popov, E. P. [1983] Seismic Behavior of Active Beam Links in Eccentrically
Braced Frames, Report No. UBC/EERC-83/15, Earthquake Engineering Research Center,
University of California, Berkeley.
Itani, A. M., Lanaud, C., and Dusicka, P.[2003] “Analytical evaluation of built-up shear links under
large deformations”, Computers and Structures 81, 681-696.
91
M. BOSCO et al.
Kim, J., and Choi, H. [2004] “Behavior and design of structures with buckling-restrained braces”,
Engineering Structures 26(6), 693–706.
Malley, J. O., and Popov, E. P. [1983] Design Considerations for Shear Links in Eccentrically Braced
Frames, Report No. UBC/EERC-83/24, Earthquake Engineering Research Center, University of
California, Berkeley.
Mazzolani, F. M., Della Corte, G., D’Aniello, M. [2009] “Experimental analysis of steel dissipative
bracing systems for seismic upgrading”, Journal of Civil Engineering and Management 15(1), 7-19.
Mazzoni, S., McKenna, F., Scott, M. H., Fenves, G. L., and Jeremic, B. [2003] OpenSEES Command
Language Manual. Pacific Earthquake Engineering Research Center. University of California,
Berkeley.
Montuori, R., Nastri, E., and Piluso, V. [2014a] “Theory of plastic mechanism control for eccentrically
braced frames with inverted y-scheme”, Journal of Constructional steel research 92, 122-135.
Montuori, R., Nastri, E., and Piluso, V. [2014b] “Rigid-plastic analysis and moment–shear interaction
for hierarchy criteria of inverted Y EB-Frames”, Journal of Constructional steel research 95, 71-80.
Montuori, R., Nastri, E., Piluso, V., [2015a] “Seismic Design of MRF-EBF Dual Systems with Vertical
Links: Ec8 Vs Plastic Design”, Journal of Earthquake Engineering 19(3), 480-504
Montuori, R., Nastri, E., Piluso, V., [2015b] “Seismic Response of EB-Frames with Inverted Yscheme:
TPMC Versus Eurocode Provisions”, Earthquakes and Structures 8(5), 1191-1214
Okazaki, T., Arce, G., Ryu, H. C., and Engelhardt, M. D. [2005] “Experimental study of local buckling,
overstrength and fracture of links in eccentrically braced frames”, Journal of Structural Engineering
131(10), 1526-1535.
Okazaki, T., Engelhardt, M. D., Nakashima, M., and Suita, K. [2006] “Experimental performance of
link-to-column connections in eccentrically braced frames”, Journal of Structural Engineering 132
(8), 1201-1212.
Okazaki, T., and Engelhardt, M. D. [2007] “Cyclic loading behavior of EBF links constructed of ASTM
A992 steel”, Journal of Constructional Steel Research 63(6), 751-765.
Okazaki, T., Engelhardt, M.D., Drolias, A., Schell, E., Hong, J. K., and Uang, C. M. [2009]
“Experimental investigation of link-to-column connections in eccentrically braced frames”, Journal
of Constructional Steel Research 65(7), 1401-1412.
Prinz. G. S. and Richards, P. W. [2009] “Eccentrically braced frame links with reduced web sections”,
Journal of Constructional Steel Research 65(10–11), 1971-1978.
Ramadan, T., and Ghobarah, A. [1995] “Analytical model for shear-link behaviour”, Journal of
Structural Engineering 121(11), 1574-1580.
Richards, P., and Uang, C. M. [2003] Development of testing protocol for short links in eccentrically
braced frames, Report No. SSRP-2003/08. Department of Structural Engineering, University of
California, San Diego.
Richards, P., and Uang, C. M. [2005] “Effect of flange width-thickness ratio on eccentrically braced
frames link cyclic rotation capacity”, Journal of Structural Engineering 131(10), 1546-1552.
Ricles, J. M., and Popov, E. P. [1987] Dynamic analysis of seismically resistant eccentrically braced
frames, Earthquake Engineering Research Center, Report No. UCB/EERC-87/07. University of
California, Berkeley.
Roeder, C. W., and Popov, E. P. [1977] Inelastic Behavior of Eccentrically Braced Steel Framed
Systems Under Cyclic Loading, EERC Report No. 77-18. Earthquake Engineering Research Center,
University of California, Berkeley.
92
Rossi, P. P., and Lombardo, A. [2007] “Influence of the link overstrength factor on the seismic
behaviour of eccentrically braced frames”, Journal of Constructional Steel Research 63, 1529-1545.
Rossi, P. P. [2015] “Importance of isotropic hardening in the modeling of buckling restrained braces”,
Journal of Structural Engineering 141(4). 04014124, 1-11.
Tremblay, R. [2003] “Achieving a stable inelastic seismic response for multi-story concentrically
braced steel frames”, Engineering Journal AISC 40,111–129
Zona, A., and Dall’Asta, A. [2012] “Elastoplastic model for steel buckling-restrained braces”, Journal
of Constructional Research 68(1), 118-125.
93
Num. …
INFLUENZA DELLA MODELLAZIONE DEL LINK NELLA

VALUTAZIONE DELLA RISPOSTA SISMICA DI EBF
Melina Bosco, Aurelio Ghersi, Edoardo M. Marino, Pier Paolo Rossi*
Department of Civil Engineering and Architecture, University of Catania, Italy
SOMMARIO: La risposta ciclica dei link nei telai con controventi eccentrici è stata spesso
modellata attraverso una trave elastica con molle di estremità che simulavano il
comportamento inelastico flessionale ed il comportamento elastico ed inelastico a taglio. Il
comportamento di tali molle era elasto-plastico con incrudimento cinematico o elastico-
perfettamente plastico. L’utilizzo di tali modelli era giustificato dalla carenza di modelli in
grado di simulare con accuratezza l’incrudimento isotropo. Recentemente, gli autori di questo
articolo hanno proposto un modello in cui la risposta delle cerniere flessionali e a taglio è
definita dal legame costitutivo proposto da Zona e Dall’Asta per i controventi ad instabilità
impedita. Questo modello simula sia l’incrudimento cinematico sia quello isotropo e, come
mostrato da precedenti analisi, è in grado di riprodurre la risposta ciclica dei link ottenuta
nell’ambito di indagini sperimentali. Sono stati, inoltre, suggeriti i valori da utilizzare per i
parametri del modello per simulare il comportamento di link provvisti degli irrigidimenti
richiesti dalle attuali norme antisismiche. Nel presente lavoro, il modello proposto è assunto
quale riferimento per verificare l’affidabilità dei modelli più semplici utilizzati in passato. Il
confronto tra i modelli è condotto analizzando la risposta dinamica non lineare di un insieme
di telai con controventi eccentrici caratterizzati da diverso numero di piani e lunghezza
meccanica del link. La riposta è espressa in termini di enti di risposta locali e globali.
___________________
*Corresponding author: Pier Paolo Rossi, Department of Civil Engineering and Architecture, University of
Catania, Italy.
Email: prossi@dica-unict.it

Journal Earthqueke

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Journal Earthqueke

Caricato da

Copyright:

Formati disponibili

Ingegneria Sismica - Anno XXXIII - N. 3 Settembre 2016 - Trimestrale - Poste Italiane S.p.A. - Sped. in Abb.

Postale DL 353/2003 ISSN 0393-1420

International Journal of Earthquake Engineering

Anno XXXIII N. 3 – settembre 2016 – Pagine 1-94

Special issue - Numero Speciale

Associate Editors / Coordinatori associati

The Non-Linear Response of Steel-Concrete Composite Joints

Importance of Link Models in the Assessment of the

Sul sito www.ingegneriasismica.org nella sessione archivi sono pre-

realizzato mediante una selezione degli articoli presentati al convegno

l'opportunità ricevuta e per la sua assistenza per la realizzazione di questo

METAL SHEAR PANELS FOR SEISMIC PROTECTION OF

Gianfranco De Matteis1, Giuseppe Brando2*

2. PURE ALUMINIUM SHEAR PANELS (PASPS)

panel type B (b/t=100) panel type F (b/t=50)

panel type G (b/t=50-25) panel type H (b/t=50)

shear stress (MPa)

2 8000 AW 1050A-Panel type B

panel “type 1” (b/t=100) panel “type 2” (b/t=50)

panel “type 3” (b/t=33) panel “ type 4” (b/t=25)

Poli. ("type 1")

Poli. ("type 1") 5.00 Poli. ("type 2")

150 Poli. ("type 2") 4.00

Poli. ("type 4")

Poli. ("type 3") 5.00

3. BUCKLING INHIBITED PURE ALUMINIUM SHEAR PANELS (BIPASPS)

Shear Stress (MPa)

Shear Strain (% ) Shear Strain (% )

a) Shear Strain (%) b)

Shear Stress (MPa)

a) Shear Strain (%) b)

a) Shear Strain (%) b)

4. PERFORATED SHEAR PLATES (PSPS)

3.2 The experimental tests

3.3 The obtained experimental evidences

Diagonal Force (kN)

Diagonal Force [kN]

LASTRE FORATE A TAGLIO DI ACCIAIO PER LA PROTEZIONE

Gianfranco De Matteis1, Giuseppe Brando2*

SOMMARIO: Il presente lavoro riporta i risultati principali di una attività sperimentale e

EFFETCS OF CLASS B SITE ON THE SEISMIC RELIABILITY OF

Paolo Castaldo*, Guglielmo Amendola, Bruno Palazzo

Department of Civil Engineering, University of Salerno, Italy

through the nondimentionalitation of the motion of equations is analyzed by providing useful

2. EQUATION OF MOTION OF BASE-ISOLATED STRUCTURES

and Eqn. (1b) by mb , and introducing the mass ratio

4. INCREMENTAL DYNAMIC ANALYSES (IDA) RESULTS OF BASE-ISOLATED

Figure 2. IDA curves of the superstructure with γ=0.7, for R=4m

5. SEISMIC FRAGILITY OF BASE-ISOLATED STEEL STRUCTURES

provisions the limit response of the lateral-load-resisting superstructure system of a base-

pf (50 years) 5.0·10-1 1.6·10-1

6. SEISMIC RELIABILITY OF BASE-ISOLATED STEEL STRUCTURES

Porter, K.A. [2003] “An overview of PEER’s performance-based earthquake engineering

EFFETTI DI SUOLO DI CATEGORIA B SULL’AFFIDABILITA’

Paolo Castaldo*, Guglielmo Amendola, Bruno Palazzo

Department of Civil Engineering, University of Salerno, Italy

APPRAISAL OF SEISMIC DESIGN CRITERIA FOR

Beatrice Faggiano*, Antonio Formisano, Carmine Castaldo, Luigi Fiorino,

Department of Structures for Engineering and Architecture,

KEYWORDS: Seismic resistant steel structures, Seismic design criteria, Concentrically

1. THE RESEARCH CONTEXT

2. NTC2008 DESIGN CRITERIA FOR CBF STRUCTURES

Brace 1,3    2 (3)  2 (4)

3. THE CASE STUDY