Seismic behavior of hollow and concrete-filled square tubular bracing members.

Abstract

Cold-formed square tube sections are one of the most efficient sections for compression members and bracing members. According to past studies rectangular tubular bracing members with large width-thickness ratios (yet within the limits specified by current codes) showed early fractures due to local buckling when subjected to large cyclic deformation. Thus, square tubular bracing members can also have such a problem despite their structural efficiency. There are two possible ways to avoid such early fractures. One is reducing width-thickness ratio. The other is concrete infilling. However, no systematic experimental studies on full-scale hollow square tubular bracing members have been done to find the effect of width-thickness ratio and slenderness ratio on their hysteresis behavior. Moreover, no tests have been reported on concrete-filled square tubular bracing members. Thus, in the present study thirteen full-scale hollow and concrete-filled square tubular bracing members with different slenderness ratios and width-thickness ratios were tested under two different quasi-static cyclic loading histories. These loading histories were considered to be representative of the range of cyclic post-buckling deformation which may be expected to occur during a severe earthquake. Test results showed for hollow specimens that when the width-thickness ratio was reduced from 30 to 14, the initial and severe local buckling were much delayed and consequently, the fracture life increased. However, this was not enough to allow the specimens to survive the full loading history. Concrete-filled specimens with a width-thickness ratio of 14 behaved much better than the corresponding hollow specimens. However, concrete-filled specimens with a width-thickness ratio of 30 behaved only a little better than the corresponding hollow specimens. The difference in improvement gained by concrete infilling in the two cases was attributed to difference in local buckling mode depending on the width-thickness ratio. An empirical hysteresis model is proposed for hollow and concrete-filled tubular bracing members which has a clear and rational selection method for control points and can predict hysteresis loops very well. An empirical fracture criteria for hollow and concrete-filled tubular bracing members is also proposed to predict the fracture life of such members.