ABSTRACT
This paper presents a summary of an extensive study of cold-formed thin-walled steel columns at elevated temperatures, forming part of an EPSRC funded research project to study the behaviour of thin-walled steel structures in fire. Both experimental and numerical studies were conducted. Experiments include:
Fire tests on small panels, measuring 300×300mm and various thickness, consisting of a cold-formed thin-walled steel channel, one or two layers of 12.5mm gypsum plasterboard on both sides and either with or without interior insulation, exposed to the standard BS 476 fire on one side;
Short channels (400mm) in compression at various elevated temperatures up to 700°C;
Fire tests on six full-scale panels of 2.2×2.0m, each consisting of three cold-formed thin-walled channel sections, one layer of 12.5mm gypsum plasterboard on both sides and with interior insulation, exposed to the standard BS 476 fire on one side.
Numerical studies of heat transfer and structural analysis, using ABAQUS, were performed to model the tests to validate ABAQUS simulations. Afterwards, a number of numerical parametric studies were conducted to address different topics and to assess a number of current design methods. Numerical simulations were conducted to investigate:
Temperatures attained in different types of construction using cold-formed thin-walled steel channels;
The structural behaviour of short columns with and without service holes under uniform temperature;
The effect of non-uniform temperature distributions in the cross-sections of columns on their behaviour;
The effect of initial imperfections on the behaviour of uniformly heated long columns.
Finally, current design methods were assessed for their accuracy and complexity by comparing design calculation results against finite element and experiments results.
324It has been concluded that ABAQUS is a useful tool for studying detailed behaviour of cold-formed thin-walled steel structures in fire. The current design recommendations in Eurocode 3 Part 1.2, based on the effective width concept, are extremely complex to implement for thin-walled steel sections under non-uniform temperature and stress distributions. The current UK design method, based on the limiting temperature method for hot-rolled steel structures, is not accurate for thin-walled steel structures with non-uniform temperature distributions
