Keywords: buckling, cold-formed sections, residual stresses, finite element analysis.

The buckling behaviour of rack uprights subject to compression is influenced by many different factors (length of the column, end section boundary conditions, geometric imperfections, presence of perforations, etc.). For the uprights obtained by cold roll forming, the changes introduced in the material by this process (residual stresses and strain hardening) have also an important effect. In this paper the attention is focused on the influence of the residual stresses arising from the cold roll forming process in the buckling behaviour of non perforated uprights with open thin-walled steel sections. The effect of strain hardening in the curved parts has not been taken into account. The authors are involved in evaluating the effect of including it in compression buckling analysis.

This paper compares two different methodologies to obtain the ultimate load through finite element analysis with experimental results. The first one is the classic two step method. First, a linear elastic buckling analysis is performed in order to obtain an initial deformation shape to be used in the second step. Then, a compression analysis is completed with geometric and material non linearity. The prediction of the ultimate load is sensitive to the shape of the initial geometrical imperfection. The use of the first buckling mode from the linear analysis to generate the initial perturbation not always leads to acceptable ultimate loads, especially where the failure is clearly distortional. On the other hand, it is possible to achieve good predictions if the appropriate buckling mode is used.

The second methodology uses the residual stresses produced by the cold roll forming process as the initial state for the nonlinear compression buckling analysis. In order to obtain the residual stresses, the roll forming process has been simulated using specific finite element software. It is necessary to define the flower pattern, the characteristics of the forming line and the plastic behaviour of the material, because several zones of the cross-section undergo large strains. The predicted results obtained in compression buckling analysis are in good agreement with the experimental ones. The main advantage of this method is that no initial geometrical imperfection has to be used for nonlinear compression analysis. Consequently, it is not necessary to choose the right buckling mode and to decide the size of the initial geometrical imperfection.

purchase the full-text of this paper (price £20)

go to the next paper
return to the table of contents
return to the book description
purchase this book (price £65 +P&P)