The seismic behaviour of steel braces with large sections

Since the 1994 Northridge earthquake, concentrically braced steel frames have gained in popularity as lateral load resisting system in seismic areas. Energy dissipation in the system is achieved through inelastic buckling in compression and yielding in tension of the bracing members. Brace fracture under cyclic inelastic loading is however a possibility and may govern the system ductility. In spite of this increasing popularity, very few large-scale tests have been performed so far on these systems and rigorous analytical modeling of braced frames structures with detailing requirements for the bracing members and the connections is also limited. Test validated analytical models of large size bracing members, as typically encountered in multi-storey structures, are needed in order to predict the performance of these structures under severe ground motions.; The main objectives of this project were: (1) to study the overall seismic performance, including global stability, of multi-storey Split-X braced steel frames designed according to NBCC 2005 and CSA-S16S1-05 seismic provisions; and (2) to develop a numerical database that can be used to propose loading protocols for the seismic testing of bracing members used in concentrically braced steel frames.; Parametric studies were carried out in order to evaluate the influence of modelling assumptions when simulating the hysteretic response of steel bracing members with the OpenSees computer software. Comparison between test and analytical results as well as nonlinear dynamic analysis of a simple structure confirmed the capacity of the OpenSees models to predict well the seismic behaviour of low-rise braced steel frame structures.; Thereafter, extended analytical studies were performed on steel structures with Split-X braces. Five building heights were chosen for the study: 2-, 4-, 8-, 12- and 16-storeys buildings located in Victoria, British Columbia. Twenty ground motions were selected to be applied on the structures in order to study their seismic performance and stability. The buildings were designed following the NBCC 2005 and CSA-SI6SI-05 capacity design provisions. Three-dimensional models were created using the OpenSees computer software. The structural performance of the buildings was examined through non-linear dynamic analysis and incremental dynamic analysis.; Good seismic performance was found for all the buildings. The median estimates of the: inter-storey drifts met the 2.5% limit prescribed in the NBCC 2005 for buildings of normal importance. For all buildings, however, these deformations exceeded the values predicted from response spectrum analysis and the 12- and 16-storey buildings exhibited significant concentration of storey deformations along their height. In spite of these relatively high and non-uniform storey drift values, the 8-, 12- and 16-storey structures, all demonstrated a very robust response against global collapse when subjected to incremental dynamic analysis. The confidence level against this failure mode was equal to 99.99%, well above the recommended minimum value of 90% in U.S. design guidelines.; The study also showed that the braces exhibit satisfactory performance against fracture. The most critical braces were those with a low slenderness ratio (KL/r ≈ 55) and relative lower demand was observed on the stockier braces (KL/r ≈ 45) and on the more slender ones (KL/r > 65). Finally, the analyses showed that the 8-, 12- and 16-storey buildings experienced significant permanent storey deformations after design level earthquakes, which could lead to major repair costs. A numerical database was created for use in the development of loading protocols for the seismic testing of bracing members used in concentrically braced steel frames.; Future studies are necessary in order to examine further the effects of residual deformations and the potential for the brace fracture. In addition the findings of this study should be verified for other braced...