Seismic evaluation of all-steel buckling restrained braces using finite element analysis Sh. Hosseinzadeh, B. Mohebi ⁎ Faculty of Engineering, Imam Khomeini International University, Qazvin, Iran abstract article info Article history: Received 26 January 2015 Received in revised form 26 November 2015 Accepted 6 December 2015 Available online xxxx All-steel buckling restrained braces (BRBs) are a newly developed variation of ordinary BRBs with enhanced characteristics in terms of weight and curing of the mortar core. Finite element (FE) models of all-steel BRBs with varied geometries were subjected to cyclic analyses in this study. The satisfactory brace geometries that minimized instability of the core section while maximizing energy dissipation capacity were then identified. Bi- linear FE-derived back-bone curves of the selected BRBs were subsequently used in the representative truss ele- ments to retrofit three 4-, 8-, and 12-story frames. The advantages of these braces were highlighted by drawing performance comparisons against ordinary braces. Nonlinear static and dynamic responses of the frames with all-steel BRBs were also assessed in terms of parameters such as maximum inelastic deformation demand. © 2015 Elsevier Ltd. All rights reserved. Keywords: All-steel buckling restrained braces Back-bone curve Inelastic response parameters Retrofit 1. Introduction Buckling restrained braces are the new generation of concentric braced frames (CBFs) which solved the buckling problem and enhanced the ductility and stiffness of their frames. Conventional configuration of these braces consists of a central core plate encased in a mortar-filled tube, which restrains the core plate form buckling in compression. Com- pressional behavior of the core plate is dominated by yielding, rather than buckling, which is similar to tensional loading procedure [1], and results in a stable hysteretic curve accompanied by enhanced ductility. Qiang [3] investigated the practical application of these braces in Asian buildings [3]. Component testing was carried out by Black et al. [2] that revealed a symmetric and stable hysteretic curve for these braces. Investigation of the seismic performance of BRBs was widely conducted by Sabelli et al. [5] and design criteria of BRBs were provided in AISC 341-10 (Seismic Provisions for Steel Structures) [6]. Wakabayashi et al. [4] introduced the panel BRB which consisted of one or two steel core plates embedded in a reinforced concrete panel. Fahnestock et al. also conducted the pseudo-dynamic numerical analyses of large-scale BRBs [7]. Optimization studies on steel core lengths for damper BRBs were carried out by Mirtaheri et al. [8] that showed the significance of low cycle fatigue, at which short brace lengths were used. They also uttered that materials with considerable work hardening, such as stain- less steel, might be appropriate alternatives, instead of ordinary carbon steel. Prasad [9] claimed that BRBs require smaller beam sections than conventional CBFs with chevron bracing configuration. Takeuchi et al. [10] studied the local buckling of core plate and discussed the restrainer thickness and its effect on the local (global) buckling of BRBs. They also declared that, due to the fact that BRBs will experience large inelastic deformations during strong ground motions, it is not logical to study their behavior in the elastic range. Performance-based design (e.g. following FEMA 440 [11]) should be used instead, as a reliable way for obtaining a design capable of achieving the intended performance goals. Conventional configuration of BRBs suffers from the heavy weight and curing problem of the mortar core. To address these inefficiencies, a new type of BRBs, called all-steel BRBs [1], is introduced. The concept behind the new configuration is the same; but, the unbounding agent is not mandatory in this type; i.e. the core plate will be encased in a steel tube without any mortar and unbounding material surrounding it, which causes all-steel BRBs to be lighter, easier and faster to fabricate without needing mortar. Thus, this type becomes more economic and practical than the conventional BRBs. In addition, the proposed BRBs can be easily inspected after earthquakes by disassembling. The hyster- etic behavior of all-steel BRBs was experimentally investigated by Tremblay et al. [12]. An important factor which affects the buckling be- havior of all-steel BRBs is the ratio of Euler buckling load, P e , to the yield strength of the core, P y . Effect of P e /P y ratio was first noted by Wananabe et al. [13] and was suggested to be considered greater than unity in order to protect the brace from global (local) buckling. However, the P e /P y ratio of 1.5 was proposed for design purposes [14]. P e P y ≥1:0 ð1Þ Journal of Constructional Steel Research 119 (2016) 76–84 ⁎ Corresponding author. E-mail address: mohebi@ENG.ikiu.ac.ir (B. Mohebi). http://dx.doi.org/10.1016/j.jcsr.2015.12.014 0143-974X/© 2015 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Journal of Constructional Steel Research