Local buckling of hot-rolled stainless steel channel section stub columns after exposure to fire

This paper presents experimental and numerical investigations into hot-rolled stainless steel channel section stub columns after exposure to fire. The experimental programme used two hot-rolled austenitic stainless steel channel sections C 80 x 40 x 5 and C 100 x 50 x 5. For each channel section, seven standard coupons and seven geometrically identical stub column specimens were prepared and tested after exposure to different levels of elevated temperature (30 degrees C, 300 degrees C, 450 degrees C, 600 degrees C, 750 degrees C, 850 degrees C and 1000 degrees C). The numerical modelling programme included a validation study, where finite element models were developed and validated against the post-fire stub column test results, and a series of parametric studies, where the validated finite element models were adopted to generate further numerical data over a wide range of cross-section dimensions. Given that there are currently no established design standards for stainless steel structures after exposure to elevated temperatures, the relevant ambient temperature design rules, as set out in the European code, American design guide and continuous strength method, were assessed for their applicability to hot-rolled stainless steel channel section stub columns after exposure to elevated temperatures. The assessment results revealed that (i) the two sets of codified ambient temperature slenderness limits are accurate when used for cross-section classification of hot-rolled stainless steel channel section stub columns after exposure to elevated temperatures, (ii) the mean test and numerical to predicted resistance ratios for various temperatures (including ambient temperature and elevated temperatures up to 1000 degrees C) from the two design standards are between 1.15 and 1.20, indicating conservatism, and (iii) the continuous strength method provides a higher degree of design accuracy, with the corresponding mean test and numerical to predicted resistance ratios ranging from 1.04 to 1.08, owing to the consideration of material strain hardening and plate element interaction.