ANALYTICAL STUDY ON AXIAL COMPRESSION ON RC SECONDARY FLAT WALLS UNDER SEISMIC LOADS

Abstract

 1. Introduction

 In recent earthquakes in Japan, significant damage to RC secondary flat walls (with rectangular cross sections), which have been regarded as non-structural walls, was repeatedly observed. The previous study⁵⁾ clarified that shear failure of flat walls can cause drift concentration in RC buildings and decrease the ultimate deformability. However, advanced numerical analyses were needed to evaluate such comprehensive behavior including the shear failure of flat walls affected by passive compression under seismic loads. Therefore, the present study investigates practical methods to evaluate the passive compression on the flat walls.

 2. Analytical models

 This study designed several idealized models which consisted of beams and flat walls representing typical RC residential buildings in Japan which had 1, 3, 6, 12, and 18 stories. The flat walls were designed referring to common construction in Reference 3), while structural members were designed according to Japanese practice⁶⁾. Figure 2 illustrates the idealized models for the following analyses.

 3. Numerical methods

 The beams were modeled by line elements, which had a roller support at the outer edge and an elastic or inelastic bending spring at the inner end (Fig. 2). The flat walls were modeled using IPE model³⁾ (Fig. 3), which could well simulate the strength deterioration. Such numerical methods applied to the following analyses were justified through simulation of the authors’ preceded test (Fig. 8).

 4 and 5. Results from numerical analyses and estimation on axial compression of the flat walls

 Pushover analyses were conducted for the analytical models. High axial compression of approximately 0.4 times as the compressive strength of the flat walls was found to act on those in the first story (Figs. 10 and 14). The mechanism of such high compression passively applied to the flat walls was illustrated based on their non-linear axial elongation (Fig. 11). The upper bound of axial compression was estimated by Eq. (11) based on a simple theoretical model (Fig. 12b), which well agreed with the numerical results for high-rise models with 12 and 18 stories. In addition, the other theoretical estimation of the axial compression using iterative cross-sectional analyses (Fig. 15) was presented, which modified the estimations by Eq. (11) for lower models.

 6. Conclusions

 The present paper investigated axial compression passively applied to RC flat walls in moment-resisting frame structures under seismic loads. Numerical analyses clarified that the passive compression on the flat walls attained to approximately 0.4 times as the compressive strength and had an upper bound. A theoretical equation representing an ultimate state of the cross section could estimate the upper bound particularly for high-rise models. The other theoretical estimation procedure considering iterative bending analyses was also presented and more precisely estimated the axial compression on the flat walls for lower models.