The 5 th Asia Conference on Earthquake Engineering October 16-18, 2014 TESTING AND MODELING OF A36 AND STAINLESS STEEL BUCKLING-RESTRAINED BRACES UNDER NEAR-FAULT LOADING CONDITIONS Joel Lanning 1 and Chia-Ming Uang 2 ABSTRACT This paper summarizes results from a research project including a testing program of six full-scale buckling-restrained braces (BRBs). These braces were subjected to large-amplitude loading protocols, statistically representing deformation demands obtained from a suite of near-fault ground motions applied to a finite element model of a long-span bridge. Four BRBs possessed stainless steel (SS) yielding cores and two BRBs used conventional mild steel yielding cores. All BRBs performed well under the severe loading, demonstrating the ability of BRBs to sustain multiple consecutive near-fault protocols. Two BRBs, one of each steel type, were tested dynamically to study the strain rate effect. A summary is provided of the increase in brace force due to the significant strain hardening properties of SS and due to high strain rate, which was observed of both steel types. Further, the large nonsymmetrical loading results made clear the inconsistencies between current BRB prequalifying testing procedures, design assumptions, and actual BRB seismic force effects for capacity design. An alternative testing method is proposed. For numerical simulation of BRB response, a commonly used bilinear model is shown to be insufficient, especially for large nonsymmetrical loading like that experienced by structures near seismic faults. A modified Menegotto-Pinto material model is shown to provide excellent correlation to test results when the following features are incorporated: (1) a larger post-yield stiffness in compression than in tension, (2) an appropriate isotropic hardening relation for SS BRB which includes the effect of cumulative ductility, (3) the strain rate effect. Keywords: buckling-restrained brace, near-fault ground motion, mild steel, stainless steel, cyclic modeling TEST PROGRAM Introduction Buckling-restrained braces (BRBs) are an attractive energy dissipation device due to their excellent cyclic inelastic capacity, simple construction, and low maintenance requirements. Primarily they consist of a yielding steel core surrounded by, and de-coupled from, concrete mortar within a hollow structural section, as shown in Fig. 1(a) along with a schematic representation of the typical stable hysteretic response. As the yielding core of a BRB (with a yield length L y in Fig. 2) experiences multiple inelastic excursions, the material undergoes strain hardening and causing the brace force to exceed the initial yield force. Furthermore during compression excursions, Poisson expansion and restrained high-mode inelastic buckling of the yielding core result in contact friction between the core and the restraining assembly. Consequently, compression forces are somewhat larger than tension forces at equal and opposite deformations. Hence, for capacity design a crucial aspect of AISC Seismic Provisions (AISC 341) BRB qualification testing is the determination of the compression strength adjustment factor, β, shown in Fig. 1(b) along with the tension strength adjustment factor, ω. The value of β is limited to 1.3 in AISC 341 (AISC 2010), as measured within each of the symmetric cycles of the AISC loading protocol (Fig. 3c) from a single BRB test, in an attempt to regulate the unbalanced brace forces making BRBs more amenable to capacity based design of the adjoining structural members. 1 Assistant Professor, California State University, Fullerton, CA 92831, USA, joellanning@fullerton.edu 2 Professor, University of California, San Diego, La Jolla, CA 92093, USA, cmu@ucsd.edu