1416
14th SEE-2010
LIMITATIONS OF INDIAN SEISMIC DESIGN CODES FOR RC BUILDINGS Vijay Namdev Khose1, Yogendra Yogendra Singh 2 and Dominik Lang 3
ABSTRACT
Past experience, and traditional design and construction practices have an important role, even in the modern codes; and seismic design codes are no exception. The main objective of design codes is to providee guideline provid guideliness to design designers ers and set minimum minimum desig design n criteria. criteria. Past Past earthqua earthquakes kes have have demonstrat demonstrated ed that buildings designed by seismic design codes are not always safe against earthquake. Therefore, it is necessary that codes should be updated time to time. Code updating is done considering current state of the art, its understanding by design engineers and the construction practices in the country so that it can be easily implemented on site. Indian seismic design code IS 1893 (Part 1): 2002 and IS 13920: 1994 are obsolete in many respects, as compared to major national seismic design codes. This paper identifies the limitations of the present Indian seismic design code and proposes some topics for updating according to the current state of the art.
INTRODUCTION During the last few decades, earthquake engineering has undergone significant development. Before the development of earthquake engineering, very few structures were designed for earthquake loading. It was observed during past earthquakes that that structures designed for some lateral load due to wind loading or due to some other reason, performed better than structures designed for gravity only. This experience, initiated design of structures for lateral force equal to a fraction of seismic weight. Later, with understanding of structural dynamics, concept of period dependent design lateral forces was developed. With the development of inelastic time history analysis and understanding of seismic response, it was found that during an earthquake structures are capable of taking many times larger seismic forces than its design lateral strength, which lead to consideration of ductility in seismic design of structures. Later “equal energy” and “equal displacement” principles were developed and design base shear on structure is reduced by a force reduction factor to account for the inelastic energy dissipation in ductile structures. Initially the structures were designed only for force but later considering the importance of displacement in control of non-structural damage, concept of force based design design with with displacemen displacementt was introduc introduced. ed. With With the the development development and and understandi understanding ng of seismic seismic 1
Research Scholar, Department of Earthquake Engineering, IIT Roorkee, e-mail: vijaynkhose@g
[email protected] mail.com Associate Professor, Department of Earthquake Engineering, IIT Roorkee, e-mail: yogenfeq@iitr
[email protected] .ernet.in 3 Research Engineer, NORSAR, P.O. Box 53, 2027 Kjeller, Norway, e-mail:
[email protected] 2
1417
Vijay Namdev Khose, Yogendra Singh and Dominik Lang
behaviour of buildings and other structures, concept of capacity design was introduced, which aims at avoiding undesirable (brittle) modes of failure by designing a hierarchy of strength of different components. This is identified as start of performance-based seismic design of structures (Priestley et al., 2007). In 1990’s researchers found that there is need for developing design methodology for achieving multiple performance objectives. Significant efforts have been made and are still being made in development of Performance Based Design. Past earthquakes have demonstrated that codes should be updated time to time. Code updating is done considering current state of the art, its understanding by design engineers and the construction practices of the country, so that it can be easily implemented on site. Indian seismic design code IS 1893 (Part 1): 2002 and IS 13920: 1994 are obsolete in many respects, as compared to major national seismic design codes. This paper identifies the limitations of the present Indian seismic design code and proposes some topics for updating according to the current state of the art.
SITE CLASSIFICATION Amplification, de-amplification and shape of design response spectrum are governed by soil type and amplitude of ground motion. Most of the seismic design codes classify the site based on one or more parameters such as average shear wave velocity, V S (generally for top 30m soil deposit), SPT Value, N, unconfined shear strength, low amplitude natural period, etc. IS 1893 classifies soils into three types, namely, Type I, Type II and Type III, which represent rock or hard soil, medium soil and soft soil, respectively. However, ASCE 7-05 (2006) classifies soils into mainly five types - A, B, C, D and E. Type A represents the hard rock and type E represents soft soil. The ASCE 7 classification is based on average shear wave velocity, V S , SPT Value, N and unconfined shear strength. However, IS 1893 classifies soils only on the basis of SPT value, N. Table 1 shows the comparison of IS 1893 soil classification with ASCE 7 based on N values. Corresponding average shear wave velocity for ASCE 7 classes are also shown. It can be observed that IS 1893 classification is much coarser and Table 1. Comparison of site classification of IS 1893 with ASCE 7
IS 1893 classification Site Class
Type I (Rock or Hard soil)
Type II (Medium soil) Type III (Soft soil)
N - value
>30
ASCE 7 classification Site Class
Shear Wave Velocity (m/s)
N - value
Type A (Hard Rock)
> 1524
NA
Type B (Rock)
762-1524
NA
Type C (Very dense soil and soft rock)
366-762
>50
Type D (Stiff soil)
183-366
15-50
Type E (Soft clay soil)
< 183
<15
10-30
<10
1418
14th SEE-2010
Fig. 1: Comparison of normalized response spectra of IS 1893 and ASCE 7 a) ASCE 7 site class A, b) ASCE 7 site class B, c) ASCE 7 site class C, d) ASCE 7 site class D, e) ASCE 7 site class E
is based on only N values, whereas shear wave velocity is a more direct characteristic parameter for estimating soil amplification of seismic waves. Further, type I of IS 1893 represents a very broad range of site classes, ranging from ASCE 7 class A to D. Type III site class of IS 1893 represents very soft soils (even softer than site class E of ASCE 7).
DESIGN SPECTRUM Traditionally, acceleration spectrum has been used by codes for design of structures. The seismic design codes generally specify 5 % damped elastic acceleration spectrum as the design spectrum. But, displacement is also an important design criteria, particularly for safety of non-structural
Vijay Namdev Khose, Yogendra Singh and Dominik Lang
1419
components. Further, the recently developed displacement based seismic design methodology is also gradually becoming popular. Therefore, some codes provide the information to construct design displacement spectra also. The following sections examine the design spectra of the Indian code in comparison with ASCE 7.
Acceleration spectra All the seismic design codes provide standard shape of design response spectrum based on soil type, which is scaled to the local hazard, which is expressed in terms of zone factor (Effective Peak Ground Acceleration, EPGA) or spectral ordinates at short and long periods. IS 1893 specifies three spectral shapes for three different site classes and scales them by zone factor. On the other hand, ASCE 7 scales the design spectrum by two spectral ordinates at 0.2 sec and 1.0 sec periods. Soil amplification, de-amplification and spectral shape are governed by soil type and amplitude of ground motion. As per IS 1893 soil amplification is considered only in long period range and it is a function of only soil type. The soil amplification in short period range and effect of amplitude of motion on soil amplification is well known now, but it is completely ignored by Indian code. Figure 1 shows the comparison of normalized response spectra of IS 1893 and ASCE 7 for ASCE 7 site classes. To reflect the effect of PGA on soil amplification/de-amplification in case of ASCE 7, normalized spectra for 0.2g and 0.5g, have been plotted. The value of corner period, T L , is considered as 4 sec in present study, which is the minimum mapped value specified by ASCE 7. The design spectra for rock site (ASCE 7 site class B) are almost same for both the codes, but there are considerable differences for other site classes. There is a large variation in spectra for ASCE 7 site class E (soft soil). The maximum difference is 113% at short period and is 92% at 2 sec period. Soil de-amplification with increasing PGA (0.5 g) in medium and soft soils can be clearly observed in case of ASCE 7 (Figure 1(c)-(e)). Further, it is interesting to note that IS 1893 spectra are on non-conservative side, as compared to ASCE 7, for almost all the site classes.
DISPLACEMENT RESPONSE SPECTRA Displacement is now widely recognized (Moehle, 1992; Calvi and Kingsley, 1995; Kowalsky, 1995; Medhekar and Kennedy, 2000; Xue and Chen, 2003; Priestley et al., 2007) as the most important design parameter and damage indicator. In the conventional force based design also, interstorey drift is an important criteria, which may govern the design in many cases. Priestley et al. (2007) have shown that fundamental periods of even moderately tall buildings lie in the displacement controlled range of spectra. A comparison of displacement spectra (Figure 2) obtained from the design spectra of IS 1893 and ASCE 7 show even more remarkable differences, which are not as prominently visible in case of acceleration spectra. One of the parameters governing the displacement spectrum is corner period between velocity controlled range and displacement controlled range. The corner period depends on the source mechanism, magnitude of earthquake and distance (Tole and Faccioli, 1999; Bommer and Elnashai, 1999; Faccioli et al., 2004; Akkar and Bommer, 2007) and hence on the local seismo-tectonic setup. ASCE 7 has provided maps for estimation of corner periods at different locations in the US where the mapped values vary from 4 sec to 16 sec. IS 1893 specifies design spectra up to 4 sec period only, without any mention about corner period and displacement controlled range. Figure 2 shows the displacement spectra for ASCE 7 and IS 1893 for 0.2g PGA. For ASCE 7, the value of corner period has been considered as 4 sec, to plot the displacement spectra. The figure shows that IS 1893 yields much lower spectral displacement for almost all the
1420
14th SEE-2010
Fig. 2: Comparison of Displacement Spectra of IS 1893 and ASCE 7 for 0.2g a) ASCE 7 site class A, b) ASCE 7 site class B, c) ASCE 7 site class C, d) ASCE 7 site class D, e) ASCE 7 site class E
site classes and there is an urgent need to review and revise the IS 1893 design spectra to get reliable estimates of displacement.
MODELLING GUIDELINES Analysis of structures for seismic actions is a complex. Computer Aided analysis is invariably used in design offices nowadays, mostly due to availability of a large number of commercial software. However, reliability of analysis results depends on the modelling of structure. An error in modelling will end up with erroneous designs. Many practicing designers in India are not adequately aware of modelling rules, which may result in erroneous models and unsafe design of structures, even if the
Vijay Namdev Khose, Yogendra Singh and Dominik Lang
1421
design codes are faithfully followed. So the codes of practice need to provide adequate guidelines for proper modelling of structures. Several national codes (EC8, 2004; ACI 318M-08, 2008; ASCE 41-06, 2006), and document (ATC 40, 1996) have included detailed and updated modelling guidelines, whereas the Indian code is silent on this issues. Detailed guidelines need to be developed for modelling of members, joints and boundary conditions and incorporated in the revised Indian code. As the performance based design is gaining popularity, guidelines also need to be developed for nonlinear modelling of structures.
DUCTILE DETAILING Currently, all the seismic codes in the world are based on force-based design methodology. In force based design, structures are analysed using elastic models and inelasticity in structures is accounted for by the force reduction factors, which considers ductility, over strength and other unforeseen factors which cannot be directly accounted for in modeling. Major contribution to reduction factor comes from the ductility in structure. Codes classify buildings into various ductility classes and specify corresponding reduction factors. ASCE 7 classifies RC frame buildings into three ductility classes, namely Ordinary Moment Resisting Frame (OMRF), Intermediate Moment Resisting Frames (IMRF) and Special Moment Resisting Frame (SMRF) and corresponding reduction factors are 3, 5 and 8, respectively. However, IS 1893 classifies RC frame buildings into only two classes, namely Ordinary Moment Resisting Frame (OMRF) and Special Moment Resisting Frame (SMRF) and corresponding reduction factors are 3 and 5, respectively. Both codes provide constant reduction factor for all design periods. However, it is well known (Priestley and Pauley, 1992) that reduction factors should be different in short period and long period range of design spectrum, as the two ranges are governed by equal energy and equal displacement principles, respectively. Further, ductility in RC structures is achieved through proper detailing of reinforcement. Ductile behaviour of RC buildings is ensured through capacity design, providing special confining reinforcement in potential plastic hinge regions, providing adequate anchorage in joints and by adequate joint shear strength. For ductile detailing of RC frame buildings ASCE 7 refers to ACI 318 and IS 1893 refers to IS 13920 for specified ductility classes. As per both codes, in case of OMRF buildings, earthquake forces are considered in design but no special provision is made for ductility. Ductile detailing provisions for IMRF of ASCE 7 and SMRF of IS 1893 are identical and justify the use of same response reduction factor for the two cases. In both the cases, ductility of RC frame buildings is enhanced through capacity shear design of columns and beams, and special confinement reinforcement (closely spaced hoops) in potential plastic hinge zone of columns and beams. In case of SMRF of ASCE 7 two additional provisions as compared to IMRF are made: (i) strong column weak beam design, (ii) design of beam-column joints to avoid shear failure. The current Indian detailing code IS 13920 completely ignores these two aspects and need to be revised to yield satisfactory designs for high seismicity areas.
DRIFT CONTROL Performance of structural as well as non-structural components of a building is controlled by interstorey drift. Inter-storey drift also controls the P-Δ effects. Therefore, it is also one of the most important design parameters in all the seismic design codes, and governs the member sizes in many cases, particularly in tall buildings. As per ASCE 7, elastic displacement at floor level is calculated and amplified by deflection amplification factor provided for different types of buildings. IS 1983
1422
14th SEE-2010
provides the drift control limits directly on the elastic displacement at the design load, without any amplification for ductility demand. ASCE 7 limits the total storey drift within 1.5-2.5% depending on the occupancy category for multistory RC frame buildings. IS 1983 limits the interstorey drift to 0.4% at design load level, which renders it dependent on the ductility class of the building (Haldar and Singh, 2009). Considering the ductility factor approximately equal to the response reduction factor, the effective limits at ultimate drifts are 1.2% and 2% for OMRF and SMRF, respectively. This leads to some discrepancies (Haldar and Singh, 2009) in the design. Consideration of effective stiffness of RC members is obviously the most crucial step in estimation of building deformations and inter-storey drifts. Major national design codes such as ACI 318, EC8, NZS 3101 (2006) have fairly detailed provisions for effective stiffness of RC members. In addition to the code provisions, several proposals for effective stiffness of RC members under seismic loads are available and vary significantly (Kumar and Singh, 2010). Indian code is, however, silent on this crucial issue.
CONCLUSIONS Various limitations of the Indian seismic design and detailing codes, IS 1893 and IS 13920 as compared to the current state of the art have been identified. A comparative study of various provisions related to site classification, design response spectrum, modelling guidelines, drift control criteria and ductile detailing has been made with ASCE 7. The Indian site classification is based on single parameter, i.e. SPT value. However, a more direct characterization can be made using shear wave velocity. Type I soil of IS 1893 is a very broad class covering from Classes A to D of ASCE 7. The Indian code specifies design spectrum up to 4 sec period only, without any mention about the displacement controlled range of the spectrum, whereas the design period of medium rise and high rise buildings may be longer than 4 sec. In Indian code, soil amplification in short period range and de-amplification for higher PGA. The ignorance amplification in short period results in highly non-conservative design spectra. Although, it is well known that reliability of seismic design depends to a large extent on the accuracy of the modeling, the Indian code does not provide modeling guidelines, leaving it to the skills of individual designers. Ductile detailing provisions of the current Indian code are obsolete and important issues like strong column and weak beam and joint shear design are ignored. Code limits the interstorey drift to 0.4% at design load level, which renders it dependent on the ductility class of the building. Further, the code does not provide any guidelines about effective stiffness of RC members, rendering the check on interstorey drift, meaningless. Therefore, there is an urgent need to revise the relevant provisions of the Indian code in the light of the new research and the state of the art in the other major national codes.
REFERENCES 1. 2. 3. 4.
ACI 318M-08, 2008. Building code requirements for structural concrete and commentary, American Concrete Institute (ACI), USA. Akkar, S. and Bommer, J. J., 2007. Prediction of elastic displacement response spectra in Europe and the Middle East, Earthquake Engineering and Structural Dynamics , 36, 1275-1301. ASCE 41-06, 2006. Seismic rehabilitation of existing buildings, American Society of Civil Engineers (ASCE), Reston, Virginia. ASCE 7-05, 2006. Minimum design loads for buildings and other structures, American Society of Civil Engineers (ASCE), Reston, Virginia.
Vijay Namdev Khose, Yogendra Singh and Dominik Lang 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
1423
ATC 40, 1996. Seismic Evaluation and Retrofit of Existing Concrete Buildings, Applied Technology Council, Redwood City, California 94065. Bommer, J. J., Elnashai, A. S., 1999. Displacement spectra for seismic design, Journal of Earthquake Engineering, 3(1), 1-32. Calvi, G. M., and Kingsley, G. R., 1995. Displacement-based seismic design of multi-degree-of-freedom bridge structures. Earthquake Engineering & Structural Dynamics, 24(9), 1247-1266. Eurocode 8 (EC8), 2004. BS EN 1998-1: Design of structures for earthquake resistance- Part 1: General rules, seismic actions and rules for buildings, British Standard (BS), UK. Faccioli, E., Paolucci, R., and Rey, J., 2004. Displacement spectra for long periods, Earthquake Spectra, 20(2), 347-376. Haldar, P. and Singh, Y., 2009. Seismic performance and vulnerability of Indian code designed RC frame buildings, ISET Journal of Earthquake Technology, 46(1), Paper No. 502. IS 13920, 1993. Ductile detailing of reinforced concrete structures subjected to seismic forces-code of practice, Bureau of Indian Standards (BIS), , New Delhi 110002. IS 1893 (Part 1), 2002. Criteria for earthquake resistant design of structures Part 1 General provisions and buildings, Bureau of Indian Standards (BIS), New Delhi 110002. Kowalsky, M. J., Priestley, M. J. N. and MacRae, G. A., 1995. Displacement-based design of RC bridge columns in seismic regions, Earthquake Engineering & Structural Dynamics, 24 (12), 1623-1643. Kumar, R. and Singh, Y., 2010. Stiffness of reinforced concrete frame members for seismic analysis, ACI Journal of Structural Engineering , to appear in - 107(5) . Medhekar, M. S. and Kennedy, D. J. L., 2000. Displacement-based seismic design of buildings-theory, Engineering Structures, 22(3) , 201-209. Moehle, J. P., 1992. Displacement-based design of RC structures subjected to earthquakes, Earthquake Spectra, 8(3), 403-428. NZS 3101, 2006. Concrete structures standard, Part-1 The Design of Concrete Structures and Part-2 Commentary on Design of Concrete Structures, New Zealand Standard (NZS), Wellington 6020. Pauley, T. and Priestley, M.J.N., 1992. Seismic design of reinforced concrete and masonry buildings, Wiley, New York. Priestley, M.J.N., Calvi, G.M., Kowlsky, M.J., (2007). Displacement-Based Seismic Design of Structures, IUSS Press, Pavia, I taly. Tole, S. V. and Faccioli, E., 1999. Displacement design spectra, Journal of Earthquake Engineering , 3(1), 107-125. Xue, Q., and Chen, C. C., 2003. Performance-based seismic design of structures: A direct displacement based approach. Engineering Structures, 25(14), 1803-1813.
