8/2/2019 Beam Column Joint 01 http://slidepdf.com/reader/full/beam-column-joint-01 1/10 Paper: Cheung et al Ordinary Meeting A paper to be presented and discussed at the Institution of Structural Engineers on Tuesday 25 May 1993 at 5 pm Behaviour of beam-column joints in seismically- loaded RC frames P. c. Cheung, MSc, PhD, CEng, MIStructE, MHKIE, MASCE Professor T. Pauiay, OBE, DTechSc, BE, PhD, FRSNZ, FIPENZ, H~~MACI University of Canterbury, New Zealand Professor R. Park, ME, PhD, FEng, FIStructE, FICE, FRSNZ, FIPENZ,FACI, FASCE University of Canterbury, New Zealand Ove Arup & Partners, Hong Kong Pak Chi0 (Patrick) Cheung is a senior engineer with Ove Arup & Partners, Hong Kong. He studied engineering at the National C heng Kung University in Taiwan, Cornell University in the Zealand. He has worked for Moh & Associates, USA, and University of Canterbury in New Scott Wilson Kirkpatrick & Partners, and Australia’s Victoria University of Technology. who studied in Hungary and New Zealand, is the Thomas Paulay, a retired teacher and researcher, author and coauthor of numerous technical papers and three books. The behaviour of reinforced concrete buildings exposed o large earthquakes is his major interest. He is the recipient of national and international awards International Association for Earthquake and honours. He is the current President of the Introduction It is only since the 970s that the attention f structural engineers has een drawn to the critical role of beam-column joints in reinforced concrete frames subjected to earthquake effect^'.^. Traditiona lly, engineers had placed more emphasis n the esign of beams and columns, as can be seen, for instance, in a paper published in The Structural Engineer in 19344. Concern for the structural adequacyf beam-column joints,however, has been justified as a result of repeated field observations of joint failures in recent earthquakes’.’, an example being shown in Fig 1. Fig 1. Example of beam-column joint failures in the 1985 Mexico earthquakes6 (courtesy of the New Zealand National Society for Earthquake Engineering) ’ Robert Park tudiedatCanterburyUniversity College, New Zealand, and the University of Bristol. His reasearch work, related primarily to the design of concrete structures or buildings and bridges, has been published in over 200 technical papers, book chapters and two books, and these have been recognised by 16 national and international awards. He is currently Deputy Vice-Chancellor of the University of Canterbury. W- Synopsis The behaviour of beam-column joints is discussed in the context of current design procedures for reinforced concrete ductile frames subjected to severe earthquake motions. As plastic hinges are expected to develop in beams, the beam- column joints must be capable of transferring large shear forces across the joint cores. The mechanisms of shear resistance of joint cores comprise a diagonal concrete strut mechanism and a truss mechanism. A considerable amount of joint core shear reinforcement is necessary to sustain the truss mechanism if bond failure of longitudinal bars is avoided. The diameter of longitudinal beam reinforcement in joint cores needs to be restricted to ensure adequate anchorage in joint cores. The significant differences in detailing requirements of beam-column joints that exist between various concrete design Codes led to an international collaborative research project involving the testing of full-scale beam-column-slab joint subassemblages under quasi-static cyclic loading. The three subassemblages designed to New Zealand practice performed very well. The 1993 129
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Ordinary MeetingA paper to be presented and discussed at the Institution of Structural Engineers on Tuesday 25 M ay 1993 at 5 p m
Behaviour of beam-column joints in seismically-loaded RC framesP. c. Cheung, MSc, PhD, CEng, MIStruc tE , MHKIE, MASCE
Professor T.Pauiay, OBE, DTechSc, BE, PhD, FRSNZ, FIPENZ, H ~ ~ M A C IUnivers i ty of Canterbury, New Zealand
Professor R. Park, ME, PhD , FEn g, FIStructE, FICE, FRSNZ, FIPENZ, FACI, FASCEUnivers i ty of Canterbury, New Zealand
Ove Arup & Partners , Hong Kong
Pak Chi0 (Patrick) Cheung is a senior engineerwith Ove Arup & Partners, Hong Kong. Hestudied engineering at the National C heng KungUniversity in Taiwan, Cornell University in the
Zealand. He has worked for Moh & Associates,USA, and University of Canterbury in New
Scott Wilson Kirkpatrick & Partners, andAustralia’s Victoria University of Technology.
who studied in Hungary and New Zealand, is theThomas Paulay, a retired teacher and researcher,
author and coauthor of numerous technicalpapers and three books. The behaviour ofreinforced concrete buildings exposed o largeearthquakes is his major interest. He is therecipient of national and international awards
International Association for Earthquake
and honours. He is the current President of the
IntroductionIt is only since the970s that the attention f structural engineers haseen
draw n to the critica l role of beam-column joints in reinforced concrete
frames subjected to earthqu ake effect^'.^. Traditiona lly, engineers hadplaced more emphasis n the esign of beams and columns,as can be seen,
for instance, in a pape r published in The Structural Engineer in 19344.
Concern for the structural adequacyf beam-column joints,however, has
been justified as a result of repeated field observations of joint failures
in recent earthquakes’.’, an exa mple being shown in Fig 1.
Fig 1. Example of beam-column joint failures in the 1985 Mexicoearthquakes6 (courtesy of the New Zealand National Society for
Earthquake Engineering)
’ RobertPark tudiedatCanterburyUniversityCollege, New Zealand, and the University ofBristol. His reasearch work, related primarily tothe design of concrete structures or buildingsand bridges, has been published in over 200
technical papers, book chapters and two books,and these have been recognised by 16 nationaland international awards. He is currently DeputyVice-Chancellor of the Universityof Canterbury.
W-SynopsisThe behaviour of beam-column join ts is discussed in thecontext of current design procedures f o r reinforced concreteductile fram es subjected to severe earthquake motions. Asplastic hinges are expected to develop in beam s, the beam -colum n join ts m ust be capable of transferring large shearforces across the joi nt cores. The mech anisms of shearresistance of jointcores comprise a diagonal concrete strutmechanism and a truss mechanism. A considerable amou nt o fjoin t core shear reinforcement is necessary to sustain the trussmechanism i f bond failure of longitudinal bars is avoided. T he
diameter of longitudinal beam reinforce ment in joi nt coresneeds to be restricted to ensure adequate anchorage in joi ntcores. The significant differences in detailing requirements ofbeam-column joints that exist b etween various concrete designCode s led to an internationa l collabora tive research projectinvolving the testing of full-scale beam-column-slab joi ntsubassemblages under quasi-static cyclic loading. The threesubassemblages designed to New Zealand practice performedvery well.
Fig 8. Detailing requirements o r longitudinal beam bars in exterior
beam-column joints io
The role of transverse reinforcement in providing confinement in
reinforced concrete components is to develop significant passive lateral
pressure t o compressed concrete when strain ductility demand arises. This
pressure is mobilised by dilatation o f the concre te and is transverse to theapplied external compressive load. Confinem ent so achieved may then result
in two very desirable features of inelastic response of concrete’. Firstly,
it can convert th e relatively brittle material into a ductile one. Secondly,
it may enhance compressive strength so that , for example, the loss of the
con tribu tion to esistance of spalled concrete o utside a confined core ay
be more th an com pensated for by strength increase of the concrete within
the core. While there is need for confinement of the plastic hinge regions
of columns subjected to axial compression and bending above or below
a joint of a ductile frame, there s not the same need inside joint cores since
significant inelastic compression stra ins do not arise in the concre tewithin
a joint core.
Under seismic actions, the sign of both the column and thebeam bending
momen ts usually changes inside the joint core. T his feature is illustrated
in Fig 9 in which moment patterns f or a model structure idealised by line
members and a real frame with beam and column depths being taken intoconsideration can be compared. It is evident that the m oments within the
joint core of the real structure will always be less critical. Hence the need
to confine a joint core to the same extent as an adjacent potential plastic
hinge region of a column does not appear to be justified. Of course, it
is necessary to preserve the integrity of the core con crete which is subjected
to tensile strains in several directions. D iagonal sp litting cracks developing
in the non-prismatic diagonal co ncrete strut (Fig 7(a)) have to be restricted.
Usually, a nominal am ount of reinforcement, more appropriately referred
to as basketing or containment reinforcement, is used for this purpose.
It is evident that the main need for reinforcement within joint cores is to
provide shear resistance, not confinement.
Column Icentre-Line- Beam moments Column fac e
moments
(a) Model structure (b ) Real structure
Fig 9.A comparison of moment patterns fo r m odel and real
frame substructures
*it cal
4 Predicted resistance
Fig I O . A comparison of force-displacement hysteresis loops
It is sometimes argued that the presence of beams at the four faces of
joint cores of two-way frames results in an enhan cement of the performan ce
of joint coresi131’,This is indeed the case when the seismic forces act in
the direction of one axis of the building only, and hence plastic hinges do
not for m in all beams. However, the effects of earthquake groundmotions
in various directions cannot be ignored. With the form ation of plastic hinges
at all four sides of an interior rectangular column at various stages duringan eart hqu ake , he cracking in the beams at the colum n faces will reduce
the confinemen t of the joint core. Hen ce confinement by transverse beams
cannot be relied on . Laboratory tests which have demonstrated significant
confinement from transverse beams have generally loaded the specimens
in one direction only, leaving transverse stub beams unloaded or subjected
only to simulated gravity loads.
Anchorage of beam bars within joint coresThe above m echanisms of the shear resistance of beam-column joint cores
imply that the bond stresses due to the longitudinal bars of beams and
columns passing throug h joint cores play a very importan t role in the shear
behaviour of join ts. D uctile frames during a severe earthquake will develop
plastic hinges at the ends of the beams for the moment patterns shown
in Fig 9. It is necessary to ensure th at be am bars can develop tensile stresses
(with overstrength) on one side of an interior joint core and compressivestresses on the other side simultaneously, if the plastic hinges are to be
sustained. As a result, very high bond stresses can occur which could lead
to excessive slip or bond failure of the beam bars.
In the evaluation of ea rth qua ke resistance, energy dissipation capacity
of astructure is traditiona lly associated with theshape of the force-
displacement hysteresis loops’. The solidtline loop in Fig 10 therefore
represents a more desirable hysteretic response as opposed to the dashed-
line loop with ‘pinch ing’ characteristics. How ever, recent studiesz2 uggest
that som e variations in hysteresis oop shape may not have a major influence
on the inelastic dynamic response of a structure when subjected to severe
earthq uake excitations. Tha t is, hysteresis loops showing some pinching
or stiffness degrad ation, caused by, for instance, inelastic deform ations
du e to shear and bond mechanism s, will not necessarily lead o significantly
larger inelasticdisplacements, provided that the structure has some damping
of viscous ype and is capable of some further damping by hysteretic energy
dissipation . Th e inelastic response of structures with a sho rt fun dam ental
period of vibration depends to a greater extent on hysteretic energy
dissipation. Th us the extent to which shear an d bond mechanisms should
be perm itted to p articipate in the hysteretic behaviour is still a con troversial
matter . Mo reover, it is easier to repair dam age due to inelastic flexural
deformations a t a well-detailed plastic hinge of a member a nd topreserve,
albeit at a reduced level, structural stiffness than to restore shear and bond
strength within a joint core.
Bond degradation of beam bars in joint cores can be avoided as far as
possible by limiting the radio db/h, , where db s the beam bar diameter and
h, is the column (joint) depth, by the following relationship:
where CY is a coefficient. The m ost stringent current requiremen t is stipulated
in New Zealan d”, where it is required tha t
and the cylinder compressive strength of con crete,f , is not to be less than
20 MP a. Thu s it s required that d , /h , 1/36 when grade 430 U; = 430
MPa) steel is used for beam bars. It is often difficult to satisfy this bond
132 T h e Structural E n g i n e e r / V o l u m e 71 N o . 8 / 2 0 April 1993
(a ) One-way interior joint specimen to ductility of p = 10
~~
(b ) Two-w ay interior join t specimen to ductility of p = 8
. I
(c ) Two-w ay exterior join t specimen to ductility of p = 13
Fig 14 . One-w ay and two-way beam-column join t subassemblages with
froorslab tested under quasi-static cyclic oading simulating severe
earthquake forces
New Zealand test results
The New Zealand tests using three full-scale isolated beam-column-slab
joint subassemblages of one-way and two-way building frames have been
reported in de ta i l e l~ewhere~~*”.ig 13 shows the dimensions of two test
units. The third unit (ID -I) represented a n interior one-way joint and was
similar to unit 2D-I (see Fig 13(a)) except that the nor th and south beamswere omitted. To avoid furth er complexity in the construction of the loading
rig, no axial forces were applied to the columns. Thus the influence of
vertical axial compression stresses on joints, co nsidered to be beneficial,
were not explored in these tests.
The primary aim of the tests was to examine the behaviour of the test
units designed according to the New Zealand Codel’. In addition, he
effects of bidirectional displacements and of t he presence of transverse
beams an d floorslabs were to be investigated. The great m ajority of beam-
column join t tests eported since 19672.12 onsisted ofplane frame
subassemblages without floorslabs.
In the tests, all units designed according to the New Zealand Codel’
perfor med very satisfactorily in terms of strength , nergy dissipation, and
ductility capacity, when subjected to the cyclic lateral load o r displacement
history shown in Fig 12. Beam plastic hinges formed at the column faces,
but the joint remained fully functional, as can be seen from Fig 14 (a),
(b) and (C) howing the specimens at the final stages of testing. In the case
of unit 2D-I, deteriorationof the bottom beam bar anch orages within the
joint core eventually occurred, leading to bar slippage through the joint
core. Tests of theother two unitswere terminated after the imposition of
very large ductility demand s when the bottom beam b ars in the plastic hingeregions had buckled.
Fig 15 shows for the three test units the m easured horizontal force v.
displacement hysteretic responses n terms of the eq uivalent interstorey drifts
(displacements). The reference ideal strengths, expressed in terms o f the
column hearorces, Vi and V$ are based on measuredmaterial
properties. Vi includes also the contribution to flexural tension of the slab
reinforcement withi? the recommen ded” effective tension flange width
(see Fig 1 ), w hile I. :allows for the full participa tion of reinforcement in
tension over the entire slab width. Both the effects of strain hardening of
the steel and participation of slab bars in tension are evident. Also indicated
are displacement ductility levels p and corresponding interstorey drifts
expressed as ercentageof the storey height. Despitea gradual and inevitabledegradation of stiffness, the hysteresis curves exhibit stable energy
dissipation. D isplacement ductility factors of at least p = 8 and interstorey
drifts of at least 3.5 070 of storey height - ell in excess of usable limits
in ductile frames- ere attaine d, while streng th degrada tion was negligible.
The circled numbers represent the progression of cyclic displacements
following the loading history outlined n Fig 12. The hysteretic responses
shown, resulting fr om judiciou s detailing of critical regions, are considered
to be close to the optimal performance attainable in reinforced concrete
frames.
The ranges of response, of particular importance to a structural engineer,
are highlighted in Fig 15(a) for unit ID-I . In the elastic range, the dr ift
at first yield displacement (i.e. at p = 1) is ab out 0.45 070 (11220) of the
storey height. It has been dem onstrated2’ that the measured lateral
stiffnessof each test ubassemblagewas considerably ess than that estim ated
by conventional analysis techniques. I n addition to the deviation of actual
material properties fro m those specified, distortions of the beam-column
joint core made a majo r contribution t o the stiffness reduction. Indeed,’
the joint panels cannot be assumed to be infinitely rigid.
Limits fo r inelastic response (Fig 15(a)) are suggested in terms of th e
maximum likely ductility demand with p = 6 and a drift of 2.7 Yo,as well
as reserve displacement capacity a t which P 4 ffects are likely to become
critical. It is evident that, with appropriate detailing of beam plastic hinges
and provision of ade quate strengths of both the joint and column, ample
reserve ductility is available.
The full details of the test results may be seen elsewhere25. he following
brief review attempts to highlight the major observations.
(1) Deformations along the diagonals of the beam-column joint core of
unit ID-I are recorded in Fig 16. Along the diagonal 51-53, the observed
relatively large an d gradually increasing tensile (positive) strains an d theconsistently small compressive (negative) strains confirm that the joint core
gradually dilated. This expansion of the joint ore canbe readily explained
with the aid of the trussmechanism in Fig 7@). Jo int core expansions were
primarily the consequence of steel tensile strains developed within the join t
core, whereas the small compressive strains resulted from the essentially
elastic response of the concrete under diagonal compression forces.
(2) Distributions of beam bar strains for unit 2D-I from ductility p = 1
t o p = 4 are shown in Fig 17. Th e beam b ars were in compression on one
side of the joint core and in tension on the other side. Tensile strains in
the beam bars a t the central part of the joint core were consistently low.
Thus the bars were well anchored, allowing high bond stresses to be
sustained in the jointcore and theplastic hinges to be spread towards the
free ends of the beams.
(3) Typical strain U istributionsof column bars are shown in Fig 18 for barsC l an d C 2of unit 2D-I. For thecorner bar C l , residual tensile strains at
levels 1 and 4 were recorded, although loading conditions were expected
to impose compressive strains. The deviation from expected strains was
believed to be caused by the actionof intersecting beam bars a nd high local
bond forces introduced by the beam bars. For the intermediate bar C2,
consistent tensile strains of significant magnitudes, although below yield
level, were found from evels 1A to 4A. Thispattern did not conform with
the sense of flexural actions as implied by the m oment patterns in Fig 9,
The Structural Engineer/V olume 71/No.8/20 pril 199334
Park, R., Milburn, J. R.: ‘Comparison of recent New Zealand and
United States seismic design provisions for reinforced concrete beam-
column joints and est results from four units designed according to
the New Zealand Code’, Bulletin of the Ne w Zealand National Society
fo r Ear thquake Engineer ing , 16, No.1, March 1983, p3
Paulay, T.: ‘A critique of the special provisions for seismic design
of the building C ode requirements for reinforced co ncrete ACI
318-83)’, Journal of the American Concrete Institute,3, No.2, March-
April 1986, p274
Paulay, T.: ‘Seismic behaviour of beam-colum n joints in einforced
concrete space frames’, State-of-the-Art Report in Special Theme
Session SF on Inelastic Behaviour and Modelling of Concrete
Structural Compo nents under M ultidirectional Seismic Forces, Proc.
9 th Wor ld C onference on Earthquake Engineering, Tokyo-Kyoto,1988, VIII, p557
CEB: ‘Model Code for seismic design of concrete structuresBulletin
d’Information, No. 165, ComitC Eur o-Inte rnatio nal du B eton, April
1985
Eurocode 8: ‘Structures in seismic regions- esign- art 1 general
and building’, Report EUR 12266 EN , Luxem bourg, Com mission of
the Eu ropean Communities, May 1988 ed. (d raft for comment)
Rosenblueth, E. (ed.) Design of earthquakeresistantstructures,
Lon don, Pentech Press Limited, 1980
Par k, R.: ‘Capacity design of ductile reinforced concrete building
structures for earthq uake resistance’, The Structural Engineer, 70,
No.16, 18 August 1992
Paulay, T ., Priestley, M. J . N.: Seismic designof reinforced concrete
and masonry bui ldings, New York, John Wiley & Sons, 1992
Park, R.: ‘Ductility evaluation from laboratory and analytical testing’,
State-of-the-Art Report in Special Theme Session SG on Ductility
Evaluation a nd Design of Concrete Structures and Elements, Proc.
9th World Conference on Earthquake Engineering, Tokyo-Kyoto,
1988, VIII, p605
Paulay, T. : ‘Equilibrium criteria for reinforced concrete beam-co lumn
joints’, Structural Journal, American Concrete Institute, 86, No.6,
November-December 1989, p635
ACI SP-123: ‘Design of beam-column join ts for seismic resistance’,
Special Publication of American Concrete Institute, etroit, Michigan,
1991
Cheung, P. C., Paulay, T. , and Park, R.:Seismic design of reinforced
concre te beam-column jo in t s wi th f loor s lab , Department of Civil
Engineering Research Report 91-4, University of Canterbury,
Christchurch, New Zealand, 1991Cheung, P. C. , Paulay, T. , and Park , R .: ‘Some possible revisions
to the seismic provisions of the New Zealand concrete design Code
for mo ment-resisting frames’, Proc. Pacific Conference n Earthquake
Engineering, Auckland, 1991, 2, p79
Informal study groups
The purpose of the study group scheme is to createopportunities fo r mem bers of the Institution to exchange ideasand work on deepening and developing their knowledge ofstructural engineering, thu s stimulating a greater interest inand promoting the art and science of structural engineering.
or who require furthe r information about a study groupshould write to the appropriate Convener.
Members wishing to tak e part in th e work of a study group
History of Structural EngineeringConvener: Frank Newby, MA(Cantab) , FEng, FIStructE, HonFRIBA,
27 Mayfield Avenue, Londo n W4 1 PNThe Structural Engineer, Ma rch 1973, p1 10
Model Analysis as a Design ToolConvener: F. K. Garas, PhD, CEng, FIStructE, MICE,
Taylor Woodrow Construction Ltd., Taywood House, 345Ruislip Road , South hall, Middlesex UB1 2QXThe Structural Engineer, Febru ary 1977, p63
Qualitative Analysis of Structural
BehaviourConvener: D. Johnson, BSc(Eng), PhD, CEng, FIStructE, MICE,
Department of Civil & Structural Engineering, NottinghamTrent University, Burton Street, Nottingham NG1 4BUThe Structural Engineer, November 1978, p309
The Design of Steel Portal FramesConvener: L. J . Morris , BSc(Eng), PhD, ACGI, DIC, CEng, FIStructE,
Simon Engineering Laboratories, University of Manc hester,Manchester M1 3 9 PLThe Structural Engineer, Part A, June 1983, p170
Vibration Problems in StructuresConvener: J . W. Smith , BSc(Hons), PhD, ACGI, CEng, MIStructE,
Department of Civil Eng ineerin g, University of Bristo l, BristolBS8 1TRThe Structural Engineer, Part A , June 1983, p1 70
Computing in structural engineeringConvener: A. T. Humphrey, CEng, MIStructE, MIMechE,
Analysis & Test Division, G EC Re search, Marconi ResearchCentre, West Haningfield Road, Gt. Baddow, EssexThe Structural Engineer, Ma rch 1987, p83
Advanced Composite Materials and
StructuresConvener: P. R. Head, BSc(Eng), AC GI, CEng, MIStructE, MICE,