Book Description

The chevron braced frame is a widely used seismic force resistant system in North America in areas subjected to moderate-to-severe earthquakes. However, the chevron braced frame system is limited in term of lateral loads redistribution over the building height. Khatib et al (1988) proposed to add zipper columns to link together all brace-to-beam intersecting points with the aim to drive all compression braces to buckle simultaneously and as a result to enlarge the energy dissipation capacity of the system. Although the Commentary of AISC Seismic Provisions for Structural Steel Building (AISC 2002) contains recommendations regarding this innovative zipper steel frame systems, no design provisions are included yet. The scope of this thesis is to refine the design method for the Zipper Braced Frame System which was initially proposed by Tremblay and Tirca (2003) and to study the system's behaviour under seismic loads by means of accurate inelastic time-history analysis. The main objective of this research project is three-fold: To develop accurate computer brace models by using Drain2DX and OpenSees and to validate the accuracy of computations with experimental test results for slender, intermediate and stocky braces; To refine the existing design method for CBFs with strong zipper columns; To validate the refined design method by studying the performance of CBF systems with strong zipper columns in Drain2DX and OpenSees environment for low-, middle- and high-rise buildings. Through this research, the overall understanding of the CBF system with strong zipper columns is improved by means of accurate numerical predictions. The outcome of this study will be further used as input data for experimental tests. The design procedure has been divided into two phases: design of braces, columns and beams according to NBC 2005 and CSA-S16-09 and design of zipper columns. A spreadsheet was developed for a 4-, 8- and 12-storey buildings and six different pattern loads related to the distribution of internal brace forces over the structure height were proposed. Based on this study, the best suited pattern load distribution is selected and considered for zipper column design. In order to evaluate the accuracy of modeling assumption in OpenSees, parametric studies were carried out. Comparisons between analytical and available test results have validated the accuracy of the computer models and analysis results. Three ground motion ensembles such as: regular, near-field and Cascadia were scaled to match the design spectrum for Victoria, B.C., have been considered in these analyses. In conclusion, good seismic performance was found for all studied buildings. The forces in the zippers were equal to or lower than predicted in the design method. All zipper columns performed in elastic range while buckling of braces propagated upward or downward within seconds. It was clearly demonstrated that by using CBF's with zipper columns the storey mechanism was mitigated and in almost all cases the interstorey drift was uniformly distributed over the structure height. In addition the median estimations of the interstorey drifts were below than 2.5% hs limit prescribed in the NBC-05 code for buildings of normal importance. The outcomes of this research project will be further used as input data for a future experimental test planned to be conducted on an 8-storey braced frame with zipper columns sample.