Angles exhibit a complex structural behaviour, responsible for the fact that, in the current North American Specification for Cold-Formed Steel Structures, short-to-intermediate equal-leg angle columns are (i) not yet pre-qualified for the Direct Strength Method (DSM) design and (ii) excluded from the application of the LFRD resistance factor ϕ=0.85, valid for all other cold-formed steel compression members. Recently, the specific behavioural features exhibited by the above angle columns were incorporated into the proposal of a novel DSM-based design approach, for both fixed-ended and pin-ended columns, and it was shown that this added rationality goes along with quite accurate and reliable failure load predictions. However, the investigation leading to this design proposal also unveiled that there are no available experimental failure loads of slender pin-ended columns with intermediate-to-high slenderness values, which implied that the design procedure was validated for such columns exclusively on the basis of numerical failure loads. The research work reported in this paper provides a contribution towards filling this gap, since it mainly consists of an experimental study, carried out at the Federal University of Rio de Janeiro, on the behaviour and collapse of short-to-intermediate slender pin-ended cold-formed steel equal-leg angle columns. After addressing the selection of the columns to be tested, the experimental set-up and test procedure are described in detail and the results obtained are presented and discussed. Such results involve (i) initial imperfection measurements, (ii) equilibrium paths relating the applied load to key column displacements, (iii) deformed configurations (including the collapse mode) and (iv) failure loads. Next, those same experimental results are used to validate a shell finite model previously developed by the authors, which is subsequently employed to obtain additional numerical failure load data concerning the pin-ended angle columns under scrutiny. Then, attention is turned to assessing the merits of the novel design approach. The comparison between the experimental and numerical values obtained in this work and their estimates provided by the design equations shows a very good correlation, perfectly in line with that observed in the recent studies available in the literature−this means that the validation and calibration of the above design approach may be deemed (successfully) completed. Finally, the paper closes with the presentation and assessment of small alterations to the existing design expressions, aimed at improving their accuracy and rationality, thus paving the way towards codification in the near future.
Read full abstract