OUT OF PLANE STRENGTH OF INFILL PANELS

Transcription

1 October 1-17, 008, Beijing, China OUT OF PLANE STRENGTH OF INFILL PANELS M. Mohaadi Ghaziahalleh 1 1 Professor Assistant,Structural Research center, International Institute of Earthquake Engineering and Seisology Tehran, Islaic Republic of Iran Eail: M.Mohaadigh@IIEES.AC.IR ABSTRACT : This paper presents the results of an analytical investigation on out-of-plane strengths of infill panels. Effects of in-plane daages on the panels out of plane strengths are studied herein. For the purpose, soe infill panels with different nubers of cracks are odeled by finite eleents. It is shown that FEMA forula can accurately predict the out of plane strength of an infill panels, having good connectivity to the surrounding fraes. Nevertheless, for infills with a gap between frae and infill, which are practically created in noral earthquakes, infill out of plane strength will be ignorable. In this condition, required strength should be supplied by other eleents or devices, such as reinforceents. Based on experiental results of this study, during in plane loading, interface cracking will be observed in low drifts. For bigger ones, the corners of copression diagonal reain only in contact with the frae; however when the frae returns back to the noral position (zero drift), the gap can be seen all around the infill adjacent to the frae. In this case, infill has iniu out-of-plane strength, which has not been considered yet. Therefore, the out of plane strength of infill panels should be practically less than that proposed by FEMA. Tests on concrete speciens showed that infill ay lean outward just for in-plane loading, even in the absence of out of plane acceleration. KEYWORDS: Infill, In-plane, Out of plane, FEMA, Strength 1. INTRODUCTION Out of plane strength of walls affects structural behavior and earthquake itigation, therefore, the designers should supply enough out of plane resistance against seisic loads. Furtherore, the out-of-plane strength of walls especially those which are not confined by fraes should be checked [1]. Checking for asonry buildings with horizontal and vertical ties, having high stiffness and short period of vibration is ore essential. This will lead to high acceleration which ay lead to in plane cracking and out of plane instability of walls. X pattern of cracks, resulting fro in-plane forces, is siilar to the crack pattern for a square panel subjected to out-of-plane forces, iplying that the transverse strength can be substantially weakened by in-plane cracking. Therefore, out-of-plane strength, evaluated for a cracked infill, is often surised to be quite low []. In this regard, Mendola et.al, translated the proble into the analysis of a fixed-free ended prisatic colun, undergoing static

2 October 1-17, 008, Beijing, China horizontal forces equivalent to the axiu inertia actions [3]. In order to evaluate the perpendicular forces effects on asonry infilled fraes behavior, researches have perfored a lot of in-plane racking tests. In these tests, speciens were loaded in out of plane direction, and then loaded in plane. The results showed that the speciens, subjected to cyclic out of plane drift displaceent within the range a typical infill ight experience, are stable and in case of being loaded in plain, only inial degradation of in-plane stiffness and strength will happen. In other studies, soe infill panels were tested using sequential in-plane and out-of-plane pressure. They showed zero to 15% reduction in out-of-plane capacity after being initially loaded by 75% of ultiate in-plane strength. Alternatively, another specien was siultaneously loaded by in-plane and out-of-plane forces, showing no significant interaction at low to oderate loading levels [4]. A shaking table test, with bidirectional excitation, on two-story, square in plan, concrete frae structure, showed that infill panels with a clear height of.5 and thickness of 11.5 or 8.0 c, would sustain lateral accelerations of about 1.75g or 1.3g, respectively, showing no out of plane expulsion or significant daage [5]. The arching action, a source of infill out of plane resistance, depends on the infill aterial, twisting stiffness of frae ebers and infill to frae connection status. The gaps are developed at the top of the walls soon after being constructed, due to the ortar shrinkage. The walls cannot withstand the out-of-plane forces in the earthquakes and would collapse due to the inadequate binding at the top. However, the walls surrounded by rigid supports, can display very high resistance to out-of-plane forces through the developent of arching action in flexure. Therefore, in order to achieve an optial load-bearing behavior- or rather stability, the reaining gaps at the top of the walls are recoended to be filled with non-shrinkable ortar a few days after their construction [6]. The author believes that out of plane strength of a asonry infill, when it is first loaded in plane, ay be practically lower than that calculated by the proposed forula of previous researches, such as Angel [], or specified by Federal Eergency Manageent Agency docuents [7]. This is because in the worst case, the axiu lateral force ay happen during in plane excitation, when zero in-plane drift occurs. In such a condition, the connection between infill and frae is the iniu, specially under the top bea, because of horizontal ortar layers crubling, caused by in plane oveent. Therefore, it is suggested to ignore the arching action and supply the required out of plane resistance by reinforced bars or esh. - PROPOSED FORMULA TO EVALUATE OUT OF PLANE STRENGTH There are soe relations to estiate the out of plane strength of asonry infilled frae. The first ethod was as epirical relationship developed by Dawe and Seah (1989). In this ethod the unifor lateral capacity, q (kpa), of an individual infill panel deterined fro [8]:

3 October 1-17, 008, Beijing, China q = 4.50 f ' 1 α = ( EsI ch h 1 β = ( EsIbl l 0.75 t + G J th) + G J tl) s s α β ( + ).5.5 l h b c < 50 < 50 q=unifor lateral load capacity (kpa) f =Masonry gross copressive strength (kpa) t=infill thickness () h=infill height () l=infill length () E s =Steel frae odulus of elasticity (MPa) G s =Steel frae shear odulus (MPa) I c =steel colun oent of inertia ( 4 ) J c =Steel colun torsional constant ( 4 ) ( 1) ( ) ( 3) The second analytical ethod was developed by Angel et al. (1994), which is specified by Federal Eergency Manageent Agency docuents [7]. In this ethod the out of plane capacity of infill wall is deterined as []: f ' q = R1R λ (-4) ( h / t) where R1 can be deterines with table 1 or forual -5: ( / cr ) h h h R1 = ( ) ( ) ( ) t t + t (-5) Siilarly R can be calculated by: 14 R = EI 1.0 (-6) EI = Flexural stiffness of the sallest eber of the confining frae at the panel edge with no continuity (N. ) The value of λ is given in a table as a function of h/t ratio. For values of h/t between 10 and 40, which is the typical range, an exponential function can be fitted to the data as follows: h / t λ (-7) = e

4 October 1-17, 008, Beijing, China Table 1- Paraeters of Angle forula h/t λ R1 Moderate Daage Sever Daage The third analytical ethod was developed by Cohen and Laing (1956) and extended by Klinger et al. (1996) to two way arching. The unifor lateral load is obtained as [8]: 8 x yv [ ] l q = M yv ( l h) + h ln() + M yh ln l (-8) h l xyh l h / where M x yv M x 0.85 f ' 4 yv = ( t xyv ) (-9) t f ' = h 1000E1 ( h / ) + t 0.85 f ' 4 (-10) yh = ( t xyh) (-11) yh t f ' = l 1000E1 ( l / ) + t and E is the infill Modulus of elasticity (MPa) (-1) Flanagan et.al. have shown that the best correlation between the experiental results and the predicted results was for the epirical ethod of Dawe and Seah. Table presents a coparison of the load predicted by the three analytical ethods with the observed experiental load [4].

5 October 1-17, 008, Beijing, China Table - Coparison between experiental results and analytical ones for out of plane resistance of infill walls Specien Analytical ethod (kpa) Panel Panel Panel height Experiental Angel Klinger Thickness (c) length (c) (c) Load (kpa) Dawe & Seah et al. et al The author of the present study believes that out of plane strengths of a asonry infills should be practically considered lower than those calculated by the proposed forula. Because in certain cases the axiu transversal force ay happen when zero in-plane drift occurs. In such a condition, there is the iniu frae to infill contact, because soe ortar layers have been crubled of previous excitation. Therefore, it is suggested to ignore the arching action resistance and supply the required one by reinforced bars, esh, etc. 3- EXPERIMENTAL AND ANALYTICAL INVESTIGATION Nonlinear finite eleent ethod is applied here to calculate out of plane resistance of infill panels. For which, one strip of the wall is considered, which can be horizontal or vertical, based on the wall aspect ratio. The effects of in plane daage on the out of plane resistance of the wall are regarded as soe cracks in the strip, shown in Fig. 1a. In this figure a odel of 0 c thick, 300 c high wall is shown, having a crack at the iddle. As shown in Detail A, the connection of the two segents of a crack is supplied by copression vertical Link eleents; however the friction is considered high enough that the segents can be regarded coupled in horizontal direction. In this odel, it is assued that there is not any gap between the wall and frae eleents and the wall is in coplete contact with the top and botto frae eleents. It is also assued that sequential in-plane and out-of-plane pressure is applied on the wall; the wall is loaded in-plane, and then at the axiu drift, out of plane load is operated. This analysis leads to high out of plane strengths, very close to one estiated by Klinger forula. For exaple for a 0 c thick, 300 c high infill wall, the strength was calculated as 4500 kg/. Taking into account the above-entioned assuptions as well as odeling ore daages on the wall (by soe cracks on the strip), does not affect out of plane strength considerably. In the other words, the out of plane resistance of the wall with one, two (shown in Fig 1b) or 15 cracks are alost the sae. This analysis confirs the results of the experiental tests in which infills are loaded transversely, at the axiu drift of being loaded in in-plane direction. Siilar analysis shows that in the presence of top gap, the out of plane strength of the strip is ignorable. Considering the rando nature of earthquakes, the infill is probable to suffer fro axiu transversal

6 October 1-17, 008, Beijing, China acceleration at frae zero drift, after having been daaged during previous in-plane cycles. Therefore it is highly recoended to ignore out of plane resistance of infills and supply reinforceents or other devices to stabilize walls. a) A strip of an infill panel, with a b) A strip of an infill panel, with two crack at the iddle cracks, eleents and deforation Fig. 1. Strips of an infill panel, with cracks Fro the other point of view, in soe cases as observed in the concrete speciens of this research, infills ay becoe transversally unstable, leaning outward, during in-plane loading, even in the lack of out of plane acceleration. In this study soe asonry and concrete infill panels were tested. As shown in Fig a, each specien was coposed of a IPE-140 frae, with rigid connections, by the length and height of 65 and 00 c respectively, with a 10 c thick infill panel inside. The specien was subjected to seven cycles of loading, with aplitudes of 1,, 3, 4, 5, 7.5 and 10 c, respectively. Siilar to previous experiental test results on infill panels, in this study the interface cracking was observed in asonry and concrete speciens in drifts less than 0.01% and 0.1% in average, respectively. Many speciens- including asonry, concrete and ultilayer infill panels- were tested in this study, with the sae procedure. However in concrete ones (ordinary concrete and fiber concrete infill speciens, in which thickness of the infills and the fraes were the sae), the out of plane oveent of the infill were observed after cycle 3, shown in Fig. b. It should be noticed that in these tests, no out of plane loadings were applied and the specien were loaded only in in-plane direction. Therefore, the author believes that soe eleents should always be supplied for infill panels to connect it to the frae, in

7 October 1-17, 008, Beijing, China order to avoid such a oveent, based on nuerical and experiental of this study. a) A asonry specien b) Outward oveent of a concrete specien Fig. Loading of speciens and outward oveent of a concrete infill specien 4- CONCLUSION Based on the results of nuerical and experiental studies of the present paper, the following conclusions can be drawn: Daw & Seah relation results correlates to the experiental results better than the other forula. Since this relation and the other ones ignore the worse case, in which the axiu out of plane acceleration occurs when the iniu contact is between frae and infill, it is believed that all the proposed relations overestiate out of plane strength of infill panels. Furtherore, based on the experiental observation, infills -especially concrete ones- ay becoe soeties transversally unstable during just in-plane loading (in the lack of out of plane acceleration), by leaning outward. Therefore soe eleents should always be supplied to prevent such a oveent to stabilize infill panels. REFERENCES 1- G. De Felice, R. Giannini, Out of plane seisic resistance of asonry walls, Journal of Earthquake Engineering, 001, Vol. 5, No., pp R. Angle & D.P Abras, Out of plane strength of URM infill panels, NCEER , P 1-9 to L.L. Mendola, M. Papia, G. Zingone, stability of asonry walls subjected to seisic transverse forces, Journal of structural engineering, Vol. 11, No. 11, R.D Flanagan, M.A Tenbus & R.M. Bennett, Nuerical odeling of clay tile infills, NCEER , P 1-63 to M.N. Fardis, S.N. Bousias, G. Franchioni & T.B Panagiotakos, Seisic response and design of RC structures with plan-eccentric asonry infills, Earthquake Engineering and Structural Dynaics, Vol. 8, 1999, P

8 October 1-17, 008, Beijing, China 6- Athanasios Dafnis; Holger Kolsch; and Hans-Gunter Reierdes, Arching in Masonry Walls Subjected to Earthquake Motions, Journal of Structural Engineering, Vol. 18, No., 00, PP FEMA 356, Prestandard for the Seisic Rehabilitation of buildings, Federal Eergency Manageent Agency, Second Draft, March, R.D. Flanagan, R.M, Benette, Bidirectional behavior of structural clay tile infilled fraes. Journal of structural engineering, Vol. 15, No. 3, March, 1999.

9 October 1-17, 008, Beijing, China