Improved Response Spectrum Analysis Procedure For Design Of Reinforced Concrete Tall Buildings

Abstract

Response spectrum analysis (RSA) procedure commonly used in design practice has been found to underestimate design forces in tall buildings. Failure and damage of reinforced concrete (RC) shear walls in mid-rise and high-rise buildings have been observed in recent earthquakes. In this dissertation, a modified RSA (MRSA) procedure to compute shear forces and strains in RC walls and columns to identify locations requiring ductile detailing was proposed. Six tall RC shear-wall buildings of 15 to 39 stories and four RC moment-frame buildings of 3 to 15 stories subjected to earthquake ground motions expected in Bangkok and Chiang Mai were first designed using seismic demands determined by conventional RSA procedure. Results from nonlinear response history analysis (NLRHA), which is the most accurate method, were used as reference values to evaluate the accuracy of the RSA and MRSA procedures. For floor displacements and story drifts, RSA provides good estimates for shear-wall buildings but underestimates for moment-frame buildings, and linear RSA is recommended for moment-frame buildings. The proposed MRSA method can significantly improve the underestimation of RSA in computing shear forces in vertical members for all cases, i.e., MRSA provides good accuracy when compared with NLRHA. In MRSA procedure, bending moments are computed and designed for in the same way as conventional RSA but ductile detailing in RC walls and columns is to be provided at the locations where combined axial-bending strains exceed a certain limit. The proposed method in MRSA to compute strains can predict well NLRHA results.