Numerical Investigation of the Failures Observed in Eccentrically Braced Frames in the Aftermath of Christchurch Earthquake Series (2010–2011)

Reza Imani, A.M.ASCE1; and Michel Bruneau, F.ASCE2

1Structural Engineer, Thornton Tomasetti, 650 California St., San Francisco, CA 94108. E-mail: RImani@thorntontomasetti.com 2Professor, Department of Civil, Structural and Environmental Engineering, University at Buffalo, 130 Ketter Hall, Buffalo, NY 14260. E-mail: mbruneau@buffalo.edu Abstract Finite element analysis was used to investigate the effects of the misalignment of the brace flange-to-beam connection point with the link-end stiffener on the ductility of the EBF frames. The misalignment was speculated to be a possible reason for the unexpected EBF failures observed in the aftermath of the Christchurch earthquake series of 2010 and 2011. EBF models with different detailing at the offset area were analyzed under monotonic and cyclic displacements. Results showed severe stress and strain concentration in the offset area, preventing the EBF from developing its expected ductility, and suggested possible initiation of a failure from the part of link flange located in the offset area. Results from analyses on different detail configurations showed that removal of the offset by modifying the brace section to build an ideal case, or by a simple change in the location of the link stiffener, can mitigate the problem of possible premature failure, with the latter solution being slightly less effective but much easier to be used in practice. INTRODUCTION AND BACKGROUND A survey of damage caused by the Christchurch earthquake series of 2010 and 2011 on steel structures showed an unexpected failure type in a number of EBFs used in a parking garage in the city of Christchurch (Clifton, et al. 2011). This is considered to be the first observed failure of EBF systems during an earthquake worldwide. Despite the satisfactory seismic performance of the building due to the presence of six EBF frames in each of the structure’s principal directions, the fractures were observed in two braced bays. A typical fracture is shown in Fig. 1. Clifton et al. speculated that

Forensic Engineering 2015 468

© ASCE

Forensic Engineering 2015

the factstifin c

Fig

The(20to deftestfinistrubetwAccbuiandet aEBbratresThifact COFouthiscyc

cause of thit that the b

ffener. This hcases where

g. 1. Fracture

e reported E012). Materia

Charpy-V ficiencies in t results revite element uctures wereween the lincording to lding, which

d showed a sal. 2011). ImF frames wice to beam ss/strain distris paper sumtors contribu

ONNECTIOur connectios and investiclic displacem

is fracture mbrace flange hypothesis wthe flanges o

e in EBF Ob

BF fracturesal samples wnotch toughmaterial pro

vealed satisfanalyses o

e also condnk stiffener athe survey h did not hasignificant am

mani and Bruith different connection ributions andmmarizes thuting to the E

ON DETAILon detail alteigated usingments.

might due towas conne

was supporteof the brace

bserved in th(Clifto

s were studiewere extracthness tests operties coulfactory duction EBF moducted, showand the brace

of damage,ave the menmount of yieuneau (2015)

geometry darea and assd overall beh

he key findiEBF failure r

L ALTERNAernatives (sh

g finite elem

a local streected to the ed by the oblined-up wit

e Aftermathon, et al. 201

ed in a forened from theand tensil

ld explain thile behavior

odels with dwing stress e flange. , there were

ntioned offseelding in the) conducted

details regardsessed the efhavior of theings from threported from

ATIVES hown in Fig

ment analysis

ess concentrabeam at a

bservation thth the stiffen

h of Christch11)

nsic examinae fractured ste tests, to he observed r for the extdetailing simconcentratio

e other EBFet in the brae link withou

detailed finding the locffects of cone frame. hat study anm the Christ

gure 2) weres when subje

ation resultinan offset frohat fracture dners.

hurch Earthqu

ation by Kantructures anestablish i

fractures. Ttracted coupmilar to thons at the

F frames inace-to-beam ut any fractunite element ation of stifnnection det

nd concentrtchurch earth

e arbitrarily ected to mon

ng from theom the link didn’t occur

uake Series

nvinde et al. nd subjected if potential

The material pons. Some

he fractured eccentricity

n the same connection

ures (Clifton analyses on

ffener at the tail on local

ates on the hquake.

selected in notonic and

Forensic Engineering 2015 469

© ASCE

Forensic Engineering 2015

Casstifverconstifbra341 FinGenEBnodto bframbralinkMisbrainit0.1simto tsup

F

se 1 is similffener is placrtically line nfiguration; ffener has bece is welded

1-10.

nite Elementneral finite F frames wi

de 3D brick be finer aroume was moce were setk) and the stses yield crinch going totial stiffness5 to 0.2 to

mple connectthe frame bypported. Fig.

ig. 2. Detail

ar to the conced at the inup with theThe third caeen moved d to the beam

t Modeling element sofith differentelements (Cund the areadeled. To

t to remain tiffeners weriteria. Post-yo the strain v). The matecrudely moion at the ba

y pushing an3 shows dif

ing alternati

nfiguration untersection oe edge of tase, EBF-3, to align verm. Note that

ftware, ABAt connection

C3D8R) and a of interest.eliminate uelastic throure modeled yield behavivalue of 0.15erial was model damage

ase of the cold pulling top

fferent views

ves for brac

used in the rof the bracethe brace flais similar to

rtically with t all four co

AQUS, was details. Thea non-unifo Due to sym

unnecessary ughout the awith a simpor was mod5 (hardeningodeled to lin

in the elemlumn, a cyclp of the colus of the finite

e-to-beam c

reported frace and beam ange; Case o EBF-1, exthe point w

onfigurations

used to stue analysis m

orm mesh pammetry, onlycomplexitieanalysis. Th

ple bilinear sdeled with a g slope was dnearly lose

ments at highlic lateral disumn while ite element m

onnection

ctured EBFscenterlines 2 represent

xcept that thwhere the flas are allowed

udy the behamodel was battern that wy half of a tes, the columhe beam (insteel materia

linear straindefined to bstrength froh strains. Asplacement wts other end

model.

s in that the but doesn’t ts the ideal he link end-ange of the d per AISC

avior of the uilt with 8-as designed typical EBF mn and the

ncluding the al with von-n hardening e 3% of the m strain of

Assuming a was applied was simply

f

Forensic Engineering 2015 470

© ASCE

Forensic Engineering 2015

Fig

FAFig1 ucur0.2 direin reladire Figdefinteshooffs

g. 3. Finite E

AILURE INVg. 4 shows thunder monotrves show sig

rad (expecection A caudirection B atively smalection B.

g. 5 shows finformations aermediate stows excessivset area whic

Element Modconnect

VESTIGAThe graphs of tonic appliegnificant strcted for prouses tensile f

causes the ller plastic

nal deformatand elementiffener (typi

ve distortionsch can be co

deling of Thetion detail; (

TION (CASEbase shear v

ed displacemrength loss aoperly stiffenforces in the

opposite. Rlink rotation

tion plots frot distortions ical behavios for a differ

onsidered as

e EBF Systec) Case 2 co

E 1) versus plasti

ments in twoat plastic linkned EBF fr

e link bottomResults shown when mo

om the samein the web

or of EBF frrent group oa sign of im

em: (a) Componnection de

ic link rotatio opposite dk rotation varames). App

m flange andw that EBF

onotonic loa

e analyses. Cb of the linrames). On tof elements w

mproper beha

plete Modeletail.

ion for EBF directions. Balues in the plying displd applying diF-1 loses strading is app

Case A shownk close to the other hawhich are loavior.

; (b) Case 1

frame Case Both of the range 0.15-

lacement in isplacement rength at a

plied in the

ws excessive one of the

and, Case B cated in the

Forensic Engineering 2015 471

© ASCE

Forensic Engineering 2015

Fi

TheoppFigLoocomareapusarea

Fig. 4. P

ig. 5. Final d

e difference posite directig. 6 shows thoking into nmbined axiala to surpass sh over curvea remain be

Push-over andispl

deformation direc

in the mecions is due t

he loads applumerical soll and comprethe Von M

e. For Case elow the fa

nalysis resultlacements in

results fromctions: a) dir

hanisms leato the differlied to the belutions and Mession stressises failure cA, on the ot

ailure limit

ts for EBF-1 n two opposi

m push-over arection A; b)

ading to streent stress aneam flange inMohr circlesses in Case Bcriteria leadther hand, Vuntil failure

half frame uite directions

analysis of E) direction B

ength loss und strain disn the offset as of the two B pushed th

ding to a sooVon Mises str

e occurs in

under monots

EBF-1 in twoB

under loadingtributions inarea for case cases reveae elements i

oner strengthress values i

n the web o

tonic

o opposite

g from two n two cases. es A and B.aled that the in the offset h loss in the in the offset of the link.

Forensic Engineering 2015 472

© ASCE

Forensic Engineering 2015

Occfor

REFigof Casrecomu

F

currence of both cases i

Fig. 6. Lo

ESULTS FRg. 8 shows thall four moses 2 and 3ommendatioch lower pla

Fig. 7. Equiva

failure in diin Fig. 7.

ads applied t

ROM ALTEhe resulting bodels subjec3 have reac

on) without astic link rota

alent plastic

ifferent loca

to the segme

RNATIVE base shear vcted to cyclched the 0.failure. Howation limits.

strain distribtwo opp

ation is also

ent of the be

DETAILSversus plasticic displacem.08 rad plawever, case

bution for Eposite directi

shown with

eam flange lo

c link rotatioments with astic rotations 1 and 4 h

EBF-1 under ions

h plastic stra

ocated at the

on values froincreasing

n limit (AIhave reache

monotonic l

ain contours

e offset

om analyses amplitudes.

ISC 341-10 d failure at

loading in

Forensic Engineering 2015 473

© ASCE

Forensic Engineering 2015

Theinsiin tif tsatisuc

COEcclocaanaandfoll Res(mipoidue Mothe wel

e superior beide the EBFthe offset arethe geometryisfactory ducch as to elimi

Fig. 8. B

ONCLUSIOcentrically bation of link

alyzed with fd overall behlowing the C

sults from lisalignment)nt is respon

e to severe st

odifying the beam and b

ll as with th

ehavior of c link, while ea, leading toy of EBF-2ctile behavioinate the stif

Base shear ve

N braced framk end stiffenfinite elemenhavior of theChristchurch

limited mon) of link stiffnsible for thetress and stra

section of thbrace centerlhe link-end s

cases 2 and cases 1 and

o a prematur cannot be

or of the framffener-to-bra

ersus plastic

mes with difer at the brant method toe frame. Spe

earthquake

notonic and feners with re observed pain concentr

he brace to alines line upstiffener was

3 is due to pd 4 have sevre failure. Praccomplish

me can still bace flange of

link rotation

fferent geomace to beam o study the efecial attentioseries of 20

cyclic analrespect to thpremature farations in the

achieve a con vertically ws shown to b

proper distriere stress anroper behavi

hed (due to be ensured bffset.

n results from

metry configconnection ffects of con

on was paid t10 and 2011

lyses showehe beam-to-bailures outsie offset area.

ndition in wwith the edgebe effective

ibution of pnd strain conior of Case 3limits in br

by locating t

m cyclic ana

gurations regarea were m

nnection detato EBF failu

1.

ed that the brace flange ide of the lin.

which the intee of the bracin solving t

lastic strain ncentrations 3 shows that race depth), the stiffener

alyses

garding the modeled and ails on local ure reported

eccentricity connection

nk. This is

ersection of ce flange as the fracture

f

Forensic Engineering 2015 474

© ASCE

Forensic Engineering 2015

problem by properly distributing the plastic strain inside the link. Results showed that in case of geometric limitations to achieve the perfect alignment, simply moving the stiffener to remove the offset can also lead to an acceptable EBF behavior. REFERENCES AISC (2010). “ANSI/AISC 341-10: Seismic Provisions for Structural Steel

Buildings.” American Institute of Steel Construction, Chicago, IL. Clifton, C., Bruneau, M., MacRae, G., Leon, R., Fussell, A., (2011), “Steel Structures

Damage from the Christchurch Earthquake of February 22, 2011, NZST,” Bulletin of the New Zealand Society for Earthquake Engineering, Vol.44, No.4, pp.297-318.

Imani, R., Bruneau, M., (2015). “Effect of Link-Beam Stiffener and Brace Flange Alignment on Inelastic Cyclic Behavior of Eccentrically Braced Frames.” AISC Eng. Journal, Vol. 52, No. 2, pp 109-124.

Kanvinde, A. M., Grilli, D. A., and Marshall, K. (2012), “A Framework for Forensic Examination of Earthquake Induced Steel Fracture Based on the Field Failures in the 2011 Christchurch Earthquake,” Proceedings of Fifteenth World Conference on Earthquake Engineering, Lisbon, Portugal.

Simulia (2012), “Abaqus Documentation.” (Url: http://www.3ds.com/products-services/simulia/support/documentation/).

Forensic Engineering 2015 475

© ASCE

Forensic Engineering 2015