Analytical and numerical study of plastic overstrength of shear links G. Della Corte a , M. D'Aniello b , R. Landolfo b, ⁎ a Department of Structures for Engineering and Architecture, University of Naples “Federico II”, Via Claudio 21, 80125 Naples, Italy b Department of Structures for Engineering and Architecture, University of Naples “Federico II”, Via Forno Vecchio 36, 80134 Naples, Italy abstract article info Article history: Received 7 February 2012 Accepted 13 November 2012 Available online 29 December 2012 Keywords: Eccentric bracing Finite element model Link Inelastic response Shear Shear links are widely used in eccentric bracing of steel buildings and, recently, for seismic protection of existing bridges and buildings. Experimental tests carried out for classic eccentric bracing of steel buildings have consistently shown that peak inelastic shear forces up to 1.4–1.5 times the plastic shear strength can de- velop at plastic link rotations of about 0.08–0.1 rad (plastic overstrength). However, more recent tests have shown that larger forces could be developed. Three basic parameters are devised as influencing shear overstrength: (i) axial forces acting on the link, (ii) the ratio of link flange over web area and (iii) the ratio between link length and cross section depth. In this paper only tensile axial forces induced by the presence of axial restraints and due to nonlinear geometric effects are dealt with. Numerical analysis of detailed finite element models has been carried out in order to ascertain the combined influence of these factors on the plastic overstrength of short links. A simple analytical model is proposed on the basis of finite element model analysis results. The analytical predictions are compared with the results of available experimental test results, showing good agreement. © 2012 Elsevier Ltd. All rights reserved. 1. Introduction Steel eccentric bracing (EB) was generated from an idea which is dated back to the 1970s–1980s [1–6]. Recently, a renewed interest has been addressed to the experimental and theoretical verification of former design suggestions [7–11]. Furthermore, either novel materials [12,13] or novel applications have been proposed, such as the use of shear links as hysteretic dissipaters for seismic protection of bridges [14,15] or buildings [16–18]. In EB systems, the use of short links is often preferred, because of the larger stiffness and ductility. Short links yield in shear and dissipate the earthquake input energy through cyclic plastic deformation, while developing some hardening. A good estimate of the level of hardening developing prior than buckling or fracture phenomena producing strength degradation is essential at the design stage for a reliable appli- cation of capacity design principles. Former tests carried out in the 1980s consistently showed that failure of shear links started as local shear buckling of panel zones at link ends, ultimately leading to fracture because of excessive local plastic deformation. The same tests consis- tently showed that the peak shear strength of short links is, in average, 1.4–1.5 times the yielding shear strength of the link web. Recent test re- sults [9], carried out on modern shear links made of higher strength ma- terials, showed a different type of failure, with web fractures occurring prior than any buckling phenomena taking place. Possible explanations of this new type of behaviour are discussed in reference [9], where the different welding processes and stiffener details, as respect to those implemented in the 1980s, are considered as the primary source of frac- ture initiation. The new type of failure mode exhibited in the tests reported in [9] was also responsible for reduced deformation capacity of shear links, which did not meet the standard requirement of 0.08 rad as minimum available plastic rotation under conventional cyclic loading history. However, within the research carried out by Richards and Uang [10], a new testing protocol was purposely developed for shear links, which resulted in a larger number of small-amplitude defor- mation cycles and a smaller number of large-amplitude deformation cy- cles. The modern shear links satisfied the 0.08 rad minimum plastic rotation requirement when the new loading protocol was considered, even though failure occurred almost always by web fracture before buck- ling. Notwithstanding this new type of ultimate failure mode, the peak inelastic strength was measured to be about 1.4 times the yielding shear strength of the link web, thus confirming the results from the for- mer tests carried out in the 1980s. On the contrary, signi ficantly larger values of plastic over-strength (ratio of peak inelastic to plastic strength) have been shown by links tested by McDaniel et al. [14] and Dusicka et al. [13]. In particular, McDaniel et al. [14] carried out cyclic tests on two full-scale built up shear links and found that the over-strength factors were 1.83 and 1.94. Dusicka et al. [13] performed experimental tests on shear links made of both conventional and special high-strength and low-strength steels, concluding that the over-strength factor may range from 1.50 to 4.00, with a value of about 2.2 obtained in the case of or- dinary carbon steel. One possible explanation, formerly proposed by McDaniel et al. [14], was the large ratio between the flange and web Journal of Constructional Steel Research 82 (2013) 19–32 ⁎ Corresponding author. Tel.: +39 0812538052; fax: +39 0812538989. E-mail addresses: gdellaco@unina.it (G. Della Corte), mdaniel@unina.it (M. D'Aniello), landolfo@unina.it (R. Landolfo). 0143-974X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jcsr.2012.11.013 Contents lists available at SciVerse ScienceDirect Journal of Constructional Steel Research