Local buckling behaviour and capacities of Cold-Formed Steel Double-I-Box stub and short column sections

Abstract

This paper presents an experimental and numerical investigation on the local buckling behaviour of a new built-up Cold-Formed Steel (CFS) homogeneous and hybrid Double-I-Box Column sections (DIBCs and HDIBCs) under axial compression. In total, 24 experimental tests are conducted, out of those 12 tests are on stub columns and the remaining 12 are on short columns. The parameters considered are hybrid ratio (Φh), section’s breadth to depth ratio (B/D) and column height (H). The columns considered in this study are having an overall slenderness (λ) less than 0.20, and the flange plate slenderness (λpf) less than 1.85. Test specimens failed in either of the following three failure modes: local buckling, yielding or by the interaction of both. In all the tested specimens, typical Bottom Zone Local Buckling (BZLB) similar to the shape of Elephant’s Foot is identified. The deformation shape is recognized by expansion (outward buckling) along the major axis and contraction (inward buckling) along the minor axis at the bottom 1/4th height of the columns in opposite faces, followed by the collapse. This kind of failure mode is a clear representation of Poisson’s ratio effect at the ends causing a restraint which then creates a local imperfection under load which leads to a localized failure at bottom zones in the stub and short columns of DIBCs and HDIBCs. BZLB phenomenon is enhanced because of the tubular cross-section geometry, boundary conditions plate element slenderness, overall slenderness of these built-up sections. The ultimate capacities of both the stub and short columns are governed by their strength limits and slenderness. Nonlinear Finite Element Analysis (FEA) is performed on experimental FE models using ABAQUS software and verified against the test results and validated with the actual structural behaviour. Further FEA is performed on 96 ideal parametric FE models considering different material properties and slenderness values. Effect of the intermediate longitudinal stiffener and the influence of edge constraints on axial strengths are also reported. The axial strengths obtained from the experimental tests and FE parametric studies are compared with the predicted results using the Effective Width Method (EWM) as suggested in the Eurocode specifications to access the feasibility of the current design rules, which is found to produce reasonable predictions albeit slightly conservative. It is found that for stub columns with λ<0.1, the experimental values are found to be 1.25 times higher than the predicted values. It is also observed that for columns having λ≥0.1<0.2 with λpf>1.2 there is a reduction in local buckling resistance of around 2.5% for each increase of λpf by 10%. Hence, it is concluded that the concentric axial compression capacities of DIBCs and HDIBCs subjected to local buckling can be computed as the arithmetic summation of effective strengths of individual channel sections calculated by using the EWM design equations, as recommended in this study. The attained results prove that these sections have a considerable local inelastic reserve strength as specified in North American standards.