This article was downloaded by: [National Chiao Tung University 國立交通大學] On: 27 April 2014, At: 21:27
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of the Chinese Institute of Engineers
Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tcie20
Roles of non
‐structural walls in Chi‐Chi earthquake
Gin‐Show Liou a , Yi‐Feng Chiu a & Fu‐Ping Cheng aa Department of Civil Engineering , Chiao‐Tung University , Hsingchu, Taiwan 300, R.O.C. Published online: 03 Mar 2011.
To cite this article: Gin‐Show Liou , Yi‐Feng Chiu & Fu‐Ping Cheng (2002) Roles of non‐structural walls in Chi‐Chi
earthquake, Journal of the Chinese Institute of Engineers, 25:4, 473-477, DOI: 10.1080/02533839.2002.9670722
To link to this article: http://dx.doi.org/10.1080/02533839.2002.9670722
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
ROLES OF NON-STRUCTURAL WALLS IN
CHI-CHI EARTHQUAKE
Gin-Show Liou*, Yi-Feng Chiu, and Fu-Ping Cheng
Department of Civil Engineering Chiao-Tung University Hsingchu, Taiwan 300, R.O.C.
Key Words: n o n - s t r u c t u r a l w a l l , e a r t h q u a k e a n a l y s i s , C h i - C h i Earthquake, ductile failure
ABSTRACT
Presented are some close observations of failure modes of build-ing structures after the Chi-Chi Earthquake, which occurred on Sep-tember 21, 1999, and some numerical analysis of building structures that proves the conclusions of these close observations. According to the observations and analysis results, one can conclude that non-struc-tural walls play a major role in the damage and collapse of building structures in strong earthquakes. Therefore, some suggestions are made for the structural design industry and academic research in order to improve the earthquake resistance of existing and future building structures.
*Correspondence addressee
I. INTRODUCTION
On the early morning of Sept. 21, 1999, a cata-strophic earthquake attacked the central part of Taiwan, which caused more than two thousand casualties, the collapse of tens of thousands of build-ings and damage to hundreds of thousands of buildings. Most of the human lives were lost in the collapsed buildings. Therefore, the causes of build-ing collapse should be investigated in great detail in order to upgrade design practices for better seismic resistant buildings in the near future and to know how to improve the seismic resistant ability of existing buildings. After a tour of the earthquake stricken area, the causes of building damage and collapse, in general, can be categorized into two groups. One is lousy construction practice. For example, adding ex-cess water in concrete in order to obtain better workability, which will reduce the strength of
concrete, overlapping all the main reinforcement of columns at the same location which is against build-ing code provisions, and not followbuild-ing the construc-tion specificaconstruc-tions for seismic resistant structures. The other is malpractice by structural designers, for example, using an oversimplified analysis method to design very complicated structures, little care about feasibility of construction (e.g. too much reinforce-ment at beam-column joints), and neglecting some design factors which are important. To some degree, all of these causes may contribute to the collapse or damage of buildings.
However, after closer examination of the earth-quake stricken area, it seems that non-structural RC or brick wall, whose influence on structural be-havior is always overlooked in design and construc-tion practice, was one of the major culprits in caus-ing the collapse of or damage to buildcaus-ing structures during the earthquake. Therefore, this paper will
474 Journal of the Chinese Institute of Engineers, Vol. 25, No. 4 (2002)
investigate some patterns of structural failure of build-ings due to non-structural walls (brick wall or thin R.C. wall) that are a common feature of buildings in Taiwan. After some in-depth investigations of col-lapsed buildings, one finds that this kind of failure pattern was quite common in the 921 Chi-Chi earthquake.
II. NON-STRUCTURAL WALL PERFORMANCE
In reinforced concrete buildings, heavy non-structural brick walls (24cm in thickness) or R.C. walls are often used as partitions between two adja-cent residential units in order to avoid interference between the two residential units. Also, in Taiwan, one building is often used for both residential and commercial purposes. For low rise buildings, simi-lar to condos or town houses, the ground story is used for commercial purposes and upper stories are used for residential purposes. For high-rise collective apartment buildings, the ground or lower stories are usually used for a lobby or a department store. Therefore, the number of walls on lower stories is much less than on upper stories. This creates a soft story effect. Unfortunately, in the past fifteen years, using ductile moment resistant frames to resist hori-zontal seismic loads has been favored by structural engineers in Taiwan for R.C. structures without con-sidering the influence of heavy brick or R.C. walls. T h i s s i t u a t i o n c r e a t e s s e r i o u s p r o b l e m s i n earthquakes. The following investigation shows how the soft story effect dramatically increases the mo-ment and axial loads at certain columns.
To investigate a specific building structure, the commercialized computer program ETABS Version 6.2 was used. Linear static and dynamic analyses were performed (ETABS, 1997). However, in the analyses, the actual weight of the building was also considered in order to understand the wall effect on structural behavior more precisely. The Chi-Chi earthquake record (accelerograms) at the location close to the specific building with similar geological condition is chosen for the dynamic analyses. The structural damping ratio was assumed to be 0.05.
The building was 14 stories in height with a 2-story underground basement. The basement was used as a parking garage, the ground and first floors were used for commercial purposes and the second to four-teenth floors were used for residential purposes. The plan of the building with column number indicated is shown in Fig. 1. Fig. 1 also shows the locations of non-structural walls with ×’s. In the 921 Chi-Chi earthquake, the left two rows of columns (column Nos. 51~70) at the ground floor were crushed, and the non-structural walls above the first floor were
found almost intact.
Three seismic analyses were performed in or-der to compare results in different conditions. The first was a quasi-static analysis with horizontal seis-mic loads according to the seisseis-mic provisions of the building code published in 1992 (code, 1992), which the design of the building was required to follow. In the code, the horizontal seismic force is calculated by the formula V=ZKCIW, where Z=0.8 for the seis-micity zone of the region, K=1.0 for the structural type, C=0.143 for X-direction and C=0.129 for Y-di-rection which depend on structural first periods in X and Y directions respectively, I=1.0, and W is the weight of the structure. The second analysis was a dynamic time history analysis without considering the effect of non-structural walls. This is what the de-sign practice usually assumes. The third analysis was also a dynamic time history analysis considering the stiffness effect of the non-structural walls above the first floor in order to obtain the influence of the soft story effect. The locations of non-structural walls are indicated in Fig. 1 with ×’s. The reason for simu-lating the building structure without non-structural wall at the ground level is that the number of walls at the ground floor is much smaller than the number of walls above the first floor and the walls at the ground floor mostly collapsed after the earthquake while the walls above the first floor were relatively intact. Al-though the existence of walls at ground level will stiffen the structure and induce larger seismic forces, this increase due to a decrease of the structural Fig. 1 The plane of building with column numbers and wall
lo-cation (Marked with ×)
period is usually insignificant (e.g. the decrease of structural period is less than 5% in this case), and most importantly, the walls will confine the columns and share a large part of the axial force due to seis-mic force. The input accelerogram used for our analy-sis was recorded near the building with similar char-acteristics and geological situation. The response spectrum of the accelerogram for EW and NS direc-tions is shown in Figs. 2 (a) and (b). In order to re-flect the real situation, both X and Y direction seis-mic forces were applied simultaneously. In the dy-namic analysis, EW is the Y-direction and NS is the X-direction according to the orientation of the struc-ture shown in Fig. 1.
Figures 3 (a), (b), (c) show the comparisons of the results of the three analyses for columns 47~70 at the ground floor. From these figures, one can ob-serve that the results due to the code specified lateral seismic force is similar to that due to the Chi-Chi
earthquake, if one assumes the non-structural walls do not influence the dynamic behavior of the struc-ture. This is because the distribution of horizontal Fig. 2 Acceleration response spectra of free-field accelerograms
recorded at chung-hshin university
Fig. 3 Comparison of three analyses
476 Journal of the Chinese Institute of Engineers, Vol. 25, No. 4 (2002)
seismic force to all the columns is similar for the dy-namic analysis without the influence of non-structural walls and the quasi-static analysis without the influ-ence of non-structural walls. This gives some evi-dence that if the walls did not influence the struc-tural behavior, the structure would have withstood the Chi-Chi earthquake. However, if one takes the in-fluence of non-structural walls into account, the re-sults change dramatically as shown by the dashed-lines on these figures. From Figs. 3(b) and (c), one can see that the column moment is greatly increased (almost doubled) in both directions, if wall effects a r e t a k e n i n t o a c c o u n t . T h e r e a s o n f o r t h i s phenomenon, is that the structure becomes more stiff if walls are considered. For this specific structure, if wall stiffness is considered, the periods (T) of the first and second modes of the structure are 0.77 sec. and 0.71 sec., respectively. The periods (T) are increased to 2.38 sec. and 2.12 sec., respectively, if wall ef-fects are not considered. Referring to the response spectra of Figs. 2 (a) and (b), one can obtain much larger responses with T=0.77 sec. and 0.71 sec. respectively. Also, observing Fig. 3 (a) with refer-ence to Fig. 1, one can see that the axial force at the external columns (or corner columns) increases dra-matically and the axial forces in other interior col-umns do not change much, if the wall effect is considered. This occurs because the walls will act like shear panels and attract a larger share of hori-zontal seismic load in the story. This condition in turn will create huge axial forces in the adjacent col-umns in order to balance the overturning moment due to the shear in wall, and this axial force will be trans-mitted down to the ground floor column without dis-tributing to other frames or columns as shown in Fig. 4. This is the reason why the axial forces in Column Nos. 51 and 61 are so large, as shown in Fig. 3(a). For interior columns having walls at both sides, these increases of axial force will cancel out each other. This axial force may create a problem for exterior
columns, since the ductility of a column with a large axial force (gravity load plus earthquake load) will be difficult to be taken into account in design practice. From the analyses above, one can conclude that the soft story effect will not only increase the total seismic horizontal load, which will induce huge mo-ments in the columns, but also could increase the axial force in some columns. This situation will create very serious problems for columns. Also, as shown in Fig. 4, the design philosophy for the earthquake resistant frames should be strong-column-weak-beam design. The first reason for this philosophy is that it is much easier to design a beam with high ductility, since the axial force in beams is very small and no P−∆ effect occurs in beams. The second reason is that the fail-ures of beams would not create a catastrophic situation. However, from Fig. 4, we can conclude that the plastic hinge will occur at the column first, since the beams have been strengthened by the walls. This means the strong-column-weak-beam design breaks down and the actual structural behavior is strong-beam-weak-column.
Also, another situation causing column failure by non-structural walls is the well-known short-col-umn effect. The short-colshort-col-umn situation, as shown in Fig. 5, is created by large window openings. In the Chi-Chi earthquake, a lot of column failure was ob-served due to the short column effect, especially for the columns in school buildings. Referring to Fig. 5, the short column effect causes column failure in two ways. The first is the well-known reason in which the shorter columns have a large stiffness compared to the longer columns in the same story as Fig. 5 indicated. This situation will increase shear and mo-ment in the shorter columns. The second reason is that point A in Fig. 5 is located at the middle part of the column which needs less tie reinforcement in the original design. But, if the short column effect is included, point A should be at the end part of the Fig. 4 Soft story effect
Fig. 5 Short column effect
column which needs more tie reinforcement to con-fine the concrete core and to increase the ability to resist the shear force as specified in the building code. Therefore, shear cracks of × pattern are observed quite commonly in the region close to point A in short col-umns in the Chi-Chi earthquake.
III. CONCLUDING REMARKS AND SUGGESTIONS
In the Chi-Chi earthquake, one interesting sce-nario is observed. Most collapsed buildings with the loss of one or two stories had walls comparatively intact at the unfailed stories, while a lot of buildings with severe damage to the non-structural walls (including the walls in upper stories) were intact. This means that the building structures that lost all the walls act more in accord with design expectations. Therefore, structural failure during a strong earth-quake may not be expected, if most of the walls are severely damaged first.
After a thorough tour of the earthquake stricken area and some analysis of failed buildings, it seems that non-structural walls change the structural behav-ior dramatically during strong earthquakes. However,
the wall effect is always neglected in structural de-sign practice, and the research on the interaction of walls and frames is sometimes ignored just as if the effect did not exist. Therefore, the design industry at present should avoid the combination of heavy walls with ductile moment resistant frames. In academic, more research should be focused on nonlinear inter-action behavior of brick walls and frames. This re-search should be directed at both experimental and theoretical aspects, and research on seismic retrofits of existing wall-frame structures should also be done in order to determine how to improve the earthquake resistant ability of existing structures.
REFERENCES
1. Computer & Structures Inc., 1997, ETABS Ver-sion 6.2.
2. Ministry of the Interior, 1992, Code and Specifi-cation for Building Design and Construction in Taiwan.
Manuscript Received: Sep. 20, 2000 Revision Received: Jun. 25, 2001 and Accepted: Aug. 14, 2001
