GECM /#ì EUROPEAN CONVENTION FOR CONSTRUCTIONAL STEELWORK CONVENTION EUROPEENNE DE LA CONSTRUCTION METALLIOUE EKS I lllllll EUROPAISCI.IE KONVENTION FUR STAHLBAU
ECCS - Technical Committee 1 - Structural Safety and Loadings
Technical Working Group 1.3 - Seismic Design
Recommended Testing Procedure
Íor Assessing the Behaviour oÍ Structural Steel Elements under Gyclic Loads
FIRST EDITION
1 986 N" 45 PREFACE
The behaviour of steel structural elements under cyclic
loads is inportànt to know in the context of earthquake resistant desl-gn, because real behaviour may ilif f er by far f,rom the ideal reference of the perfect elasto plastie element. Testing may be necessary to prove the adequacy' of the element (substructure, connection, detail) to the demand of seismic recommendations. The testing procedure is a reference vray to carry out' and interpret Èests, intending to cover the lack of such reference in European countrj.es and most of other countries. This testing procedure has been prepared by the Technical tlorking Group 1.3 : Full Members :
Prof. Dr . Ing. F. Mazzolani- Italy (Chairnan)
Prof. J.M. Aribert France Prof.Dr.ïng. G. Ballio ltaly Mr R. Pepin Luxembourg Dr. Ing. A. Plumier Belgium Dipl. Ing. ETH R. Sàgesser Switzerland Prof. Dr. Ing. G. Sedlacek Fed. Republic of Germany Prof. C. Thomas Great BriÈain Dr. D. Tordof f rr
Mr H. B. Walker rr
Mr K. Thomsen Denmark
Corresponding Members
Prof.Ing. A. Giuffre (Ita1y), Prof. Ben Kato and
Prof.Dr. T. Naka iJapan), Dr" F'. Nahler (Austria), Mr R. Siirila (Fin1and.) and Prof. A. Lamas (Portugal), t
Svrnbol-s.
Preliminary remark.
ïn the following definitions, the word foad or force is Èo be taken in a
general meaning. It may be a classical tensile force load. ft may be a bending moment, if bending is the normal work of the structural element. ït also may be shear. Accordingly the word displacement is to be taken as an elongation for tensile force, rotation X for bending monent, rotation 'y for shear.
+ A:1 : area of the positive force range half cycla in tle load - dj-splacement diagram.
Al : same, negative.
F : force.
F.+1--1: positive force corresponding to displacement e_. in cycle i.
Fl : same, negative.
F+ : yield load in Èhe positive force range.
v F : same, negative. v e : absolute value of the displacement. + e.l_ : absolute value of the'maximum positive displacement in the i& cycle.
: same, negative. "i +-+ e : absolute va}-e of the displacement defined as rj /tU ar' r
e : same, negative. v value of Lhe maxlÍìum dlsplacement in the positive force 1.1t r absorute ..ng. in the ith cycle- lel '1 : samer neÇative.
i : index of the nurnlcer of achieved cycles in a test.
z-
+ tg oi : slope of the tangent to the (F - e) curve when F changes from ' negative to positive at the ith cycle.
tg si : same, reversed in sign-
tS o]- : slope of the tangent at the origin of the (r - e) curve, when
v F increases on the positive side.
a9 oo : same, negative.
Uo : general synbol for partial ductility, see next line.
u^] ' oJ- : partial ductility at the ith .y.le for the positive dJ-splacements.
Ëoi : saÍ!e, negative.
+ _,- ductility, : full th Ui at the i---cycle for the positive displacements.
U; : full duct.ility, at the i'ti"y.t., for the negative displacements.
.lL- : resistance ratj.o, at the ith.y"l., for Èhe positive force range.
,,t-- : resistance râtio, at the ith .y.1", for the negative force range.
e (Uo) : relative resisÈance function.
ult* (uoì+r : resistance drop ratio on the F positive side at the ith .y"te. .t -* E -fu I : same, negative. ' 01 + e-'(uo) : resistance drop function + El : rigidity ratio on the positive side of forces at the ith .ycle.
.1{l : samer neÇative.
€ (uo) : re.Lative rigidity function.
úl : full ductility ratio on the positive side of forces at the ith cyrl..
ü; : same, negative. J.
ú (uo) : full ductility function.
+ rì, t : absorbed energy ratio on the positive side of, forces at the .th CyIe. 1
n.I : same, negative.
n (p^) : rel-ative absorbed energy function,
o 4.
1. INTRODUCTION.
The following testing proôedure is intended as a reference, to produce
an adequate and, as much as possible, an unified htay to carry out tests in order to characterize the structural behaviour of strUctural componenLs, substructures or even complete structures.
The necessity of a testing procedure and of having test reSults is unques-
tionable, because the reaÌ cyclic béhaviour of stfuctural element may differ by far from the ideal referénce of the peff,ect elasto-plastic behaviour.
Testing may be necessary in order to provè the adequacy of a substrucLure
or a structural detail to fulfill the demande of local ductility as specified by seismic recommendations -
The testing procedure explained here consists in defining the way to
apply on a:itructural el-ement to be testr:ci t.hc part of the tcstjnq acLion correspondi.ng to seismic action-
The testing procedure should in particular help to verify the common design relation between a pseudo-static horizontal force and a specified ductility or dlsplaceinent given by Codes arid Recolrmandatlons, Such as, for instance the ECCS Recommandati-ons for Steel St.ructures in Seismic Zones.
The procedure has been chosen td set forward the characteristics of
the element in that peculiar context. The complete definitlon of the test aÌso requires datas on the combination of seismic and non seismic loads. Paragraph 6 is devoted to that aspect of the definition of the testing procedure.
2. ASPECTS OF THE TESTING PROCEDURE FOR ASSESSING THE BEHAVIOUR OF STRUCTURÀL
STEEL EI,EMENTS.
The testing procedure mây include preliminary classical monotonic
displacement increase testd or obviate them. fn the firsÈ case, it is calted complete testing procedure - Paragraph 3. fn the opposite case, it is called short testing procedure. Paragraph 4. q
The possibilities for restricted reversal tests are mentioned at
paragraph 6.
The procedure can be applied to plane or 3 dimensional tests. The
possibility of havi-ng varioQs law for increasing various forces 1n various directions is not considered in this reference procedure.
3. COMPLETE TESTING PROCEDURE.
Introductufy remarl< ,l Each of the three follor.ring tests vrill'be performed dn a diffetent spec imen.
3.1. First test.
The first test performed on the stïucturaÌ element is a classical
monotonic displacement increase test- The increases are on the tension range defined as posítive. From the recorded. F*e curve, the conventional i;lmit of elastlc ranqe F+ and the corresponding ctisplacenreht.e+ may be deduced as foÌlor.rs lfif,ure 1) :
- evaluate the tangent at the origin of t,he F-e curve; it gives a tangent modulus E+ = tq o*, t"y- -, locate tletangent that has à slope ot Xl/rc;
- the intersection of the two tangents defines the level of F+;
+ - e is the displacement corresponding to that intersection v
Figure 1 6.
The above definition is the gênêral definition of fl.
v Any other definition of rl may be used, if propet'fy justifÍéd by the v desÍgn of testing context. A commentary on possible ilefinitidns of F v is gÍven at paragraph T. 3 .2. -cecond test-
The second test also is a classical mohotonic displacement inerease test,
but it is pérforrned on the comprêssion (hegatj,.ve) range" The procedure j-s the as in the first test. F' same vvand e are deduced.
-J-J. 'l'nrrd t'est.
The third test is a cyclic test with increase of displacement, which has the following charácteristics :
- one cycre in the "!-/s,
yy e---/4 interval; - one cycle in the 2"+/4, 2e--/4 j-nterval; vy - one cycle in the 3 vy 3 e,,/4 interval; "+/q, - one cycre in the e]-, yv el inte/vat; three cycres in the Z - v'y2 e-l intervâl; "!-, - three cycJ-es in the (2 + 2 n).1 , Q + 2 n) e--- interval (n = 1,2,...). vy More cycles ôr more intervals may be used if necessary.
3.4. Parameters of interpretatiqn for one cycie.
The absolute vaÌues of the f,ollowing quantities are deduced from the F: ê diagram after each cyci.e - Fi gute 2 - in the range of e > er.
- the extrenes of displacehent .111u,la u. ;
- the vaÌues of the forces fl and f. corresponding to the e>ttremes of displacement I1 and e]; "] 7. the extremes of displacements i.fr the positiVe and negative range of the applied forces, O.ï and Aei i the tangent nnodulus corresponding tô thê change of, the sLgn of the âpplied load, tS clrl_and tg a]; J- the areas A.l1 and A. of the positive and negativè hald cyôles; Figüre 3.
F a"Ì I
t,: I
Vt t/ Fi
relI a
'WI"" 7:;--r1 Figuíe FiOure, 3
The following quantiti-es, consideréd as charâctêriziÍtg parametêrs are
then computed :
!erçtll-qre!lU!v :u oJ- + *"ï,1 ry .s uoi =.1/"- 1y
+ + Igll-9gg!1U!r ' lr Ì = Âe. 1. /"+v tr- Àe, /e- 1 ;t v
Full DuctilitY ratj-os : üis
+ Ae. + (.i+-+ (.i -e )) 1 v -+ (e, + (e. -e + )1 {,; = Ae. L l- v l_
Resistance ratios : + I ,it,I +
1 rlI tr v .+ + + $siqilv ratios : Bi= tq-l- s. /t-9dy qi tg o; / tg s.- + À + À Àbsorbed EnergY ratios Íl . F++ : = I yr.. (e, T e-) v
A. .t n,l" = "1ry- FyÀ.(ei -eT *+) v 8.
ific rameter for a of3 cles of al displacement.
group of The behaviour of the tested piece ls characterized after each 3cyclesofeqqaldisplacqmentbythefollowingparameters.Thepartial ductility , ol , where i is the lndex of the last cycle of the group' being taken as the variable, the parameteçs are :
- ú++. + (uot_.), minimum value of the three ú.l- evaluated in the grouP of 3 cycles. ++ + eval-uated in the grouP of - e (uo1.), IiliTgS value of the fhree ç I 3 cycles. I+ - ã'(u -o1 :), minimum value of the three E] evaru-ted in the grouP of 3 cycles. - 'oIfu n* l) , urr"t.ge value of the three n1 "t'alu-ted in the group of 3 cycles. * - ,* {u ol) = ,:/ rï _ , , deflned as the resistance drop ratio of the g'r@uP of 3 cycles in the rangg 9f çhe posit'ive forces' the mlnimum value of the sirni.larly, ú (uo;), e- íuo, ) ' €-luoll are three rf . , ç1 and Ç. evaLuated i n Çhe grouP of 3 cYcles'
n(u oLll is the average of the n.1 of the grouP.
-:* u /= p-/r.^:-" e( -i L't the resiEtance drop ratio of Lhe goup OI
of 3 cycles in the range of the negatlve forces.
'
3.6. Ppramqters oF interprstation foI the whole tsst'
The partial ductiliÈy u beinq taken as the variable, the test is
characterized by the foLlowinE functions, which are continuous functions defined on the basis of a limited number of values established in 3 5 -
- FuÌl ductility functÍon ü (uo) .
- Relative resistance function e (uo) .
- Relative rigidity function E (uo)
- Relative absorbed energy functlQn n (u^)
'o -
- Resistance drop function ,*{uo) -
The number of cycles n up to the end of test must also be given.
9.
4. SHORT TESTING PROCEDURE.
In that câs€r thê first and séôond tests of the complete testing procedure are not executed. The third test is executed aÌone, but F*, F-, e* and e- yyvv are not known at the beginning of the test, so that the procedure is as folLows :
The test should be performed with steps of displacement sufficiently small
to ensure that at IeasÈ 4 leveJ.s of displacement are reached before +- eande. vv The tangent rnoduli at the origin o'+-and tg o are evaluated tg-Y-v frorn the first c1zcle curve.
- As in the complete procedure, F+ is defined b1r the intersection of,
tÌ^to tangent linesor by any other Justified definition. In the first case, one ransenr ìs evaluarect ar rhe oÌigin (.slope .n , one is ."if :n: ".nï the tangent to the envelope curve of the cycles ltith a 0r l tg o; slope - Figure 4. + Again, e is the displacement corresponding to the intersect. v
- Similar definitions are used on the negative side for f.lvv
and e---.
- As soon as these F*, F-, e* and e are defined, the testing procedure vyyv becomes the same as in 3.3.
- The paramet,ers of interpretation are the same as in 3.4, 3.5 and 3.6.
Figure 4 10.
5. END OF TEST.
The test may be stoptrrd at any level of displacement decj-ded with regard to a specific code or reseêrch requirements.
6. COMBINATION OF LOÀDS.
The principÌes uses in tests for combined loads are those of the design codes. In generaÌ, they are as follows :
the seismic action should be considered as an accident situation; hence,
the values assj-gned to variable actions of long duratiÒn should be "most probable vafuestl while other short duration actions (such as wind forces) should not be considered. The combination of seismic action with long duration actions which have no reversal in sign may bring unsymmetrical demand on structural elements. ïn this case test may be performed with a partial reversal of displace- ment. This partial- reversal can be of varj-ous forms and must be properÌy j usti fied
7. COMMENTARY ON POSSIBLE DEFINITIONS OF F
v
There are many possibÌe definitions of a conventional limit of the
elastic range F . Let us mentj-on, amongst others : v
a) the value corresponding to first yield somewhere in the tested piece -
Flgure 5 a ; b) the maximum reached load. Figure 5.b ;
c) the value corresponding to a deformation a certain time the
deformation which would have been obtained in a purely elastic behaviour. Figure 5 c gives an example with 2 times the efastic cleformation.
d) figure 5 d recalls the definition recoÌÌÌmended in the present document.
e) in some cases, FO could be a reference load deduced from computation.
Each definition offers some advantages and disavantages.
11
Defintion a) ignores the post elastic resources.
Definition b) i-s interesting in the buckling context, but may correspond to exagerated deformation in flexural behaviour of beams or joints. On the contrary, definition c) applies well to beams or joints but not to buckling problems. The ECCS recommended definition - Figure 5,d. applies in aIÌ cases. Definition e) corresponds to the case where design resistances would have been defined previously to any test.
