Cross-section capacity of RHS and SHS at elevated temperatures: Comparison of design methodologies

The accuracy of different fire design methodologies for hollow cross-sections subject to compression and major-axis bending is studied and compared in this paper. For that purpose, a numerical model based on shell finite elements is developed and validated against existing experimental results available in the literature. Comprehensive numerical parametric studies are then carried out by means of the validated numerical model, including different cross-section dimensions, loading conditions, steel grades and temperatures, to obtain a large set of results of the ultimate resistance of hollow sections in fire. Based on these results, the existing design rules of the Eurocode 3 Part 1-2, the AISC 360 specification and the recent design proposals of Couto et al. [1,2] and the Direct Strength Method adapted to fire are assessed in terms of their accuracy for predicting the section capacity, including the resistance against local buckling at elevated temperatures. It is demonstrated that the cross-sectional classification procedure for the case of fire, available in the Eurocode 3 and AISC specifications, results in inconsistencies and should be further investigated and revised in the future. To overcome the Eurocode 3 limitations it is proposed that the design methodology of Couto et al. [1,2] is adopted irrespective of the crosssection classification and for the AISC specification it is suggested the use of a reduced stress based on the 0.2% proof strength until more efficient design methods are developed.