Numerical and experimental studies of cold-formed steel floor systems made of hollow flange section joists in fire

The cold-formed steel utilization in buildings has increased globally due to its higher strength to weight ratio, ease of transportation and rapid erection and dismantlement. However, cold-formed steel buildings must be designed with adequate Fire Resistance Ratings (FRR). Hence cold-formed Light gauge Steel Frames (LSF) are assembled using channel sections and lined with fire resistive plasterboards to provide load-bearing wall and floor systems. There is an industry need to develop LSF floor systems with improved FRR. Adding multiple layers of plasterboard to increase the FRR of LSF floor systems is not an efficient method. Past research has focused on investigating the behaviour of LSF floor systems made of Lipped Channel Section (LCS) joists. No attempt has been made to use an improved joist section in LSF floor systems. The Hollow Flange Sections (HFS) with torsionally rigid hollow flanges and no free edges have higher local and lateral distortional buckling capacities than the conventional LCSs. This research focuses on investigating the structural and fire performance of LSF floor systems made of HFS joists with a goal to improve their FRRs. Four full scale standard fire tests were undertaken on non-insulated dual and single plasterboard lined LSF floor panels and cavity insulated dual plasterboard lined floor panel made of welded HFS joists known as LiteSteel beams (LSB). Fire tests of these panels undertaken for varying load ratios provided valuable results, which included failure times, joist temperatures and modes, and deflection versus time curves. The floor panels failed due to the section failures of joists. Both non-insulated and cavity insulated LSF floors made of LSB joists showed a significant improvement in the FRRs in comparison to Baleshan's (2012) results for LSF floors made of LCS joists. Another experimental study was undertaken to determine the elevated temperature mechanical properties of the steel used in LSB web and flange elements. The mechanical property reduction variation of LSB steel elements was found to be quite different to that of normal cold-formed steels and was even dissimilar amongst them. The yield strength reduction factors of Eurocode 3 Part 1.2 (ECS, 2005) were proposed for the web elements since they closely followed them whereas a new yield strength reduction factor model was proposed for the flange elements. An identical variation was proposed for the elastic modulus reduction factors of both web and flange elements. Suitable modifications were made to Dolamune Kankanamge and Mahendran's (2011) stress-strain model for improved predictions of LSB web and flange elements' stress-strain curves. A Finite Element (FE) model of an individual simply supported LSB joist was developed and validated using the cold-formed steel design standards and Anapayan et al.'s (2011b) section moment capacity test results. By using the accurate mechanical property reduction factors of LSB steel elements, the FE model was then extended to simulate the full scale fire tests. Finite element analyses (FEA) showed reasonably good agreements in terms of failure times, temperatures and modes, and the mid-span deflection versus time curves. Such good agreements verified the accuracy of the developed FE model to simulate the LSF floor panels made of HFS joists under fire conditions. Thermal FE models of LSF floor systems made of HFS joists were then developed and the time-temperature profiles were compared with the fire test results. They showed better agreements for Tests 1 and 4 whereas there were some discrepancies for Tests 2 and 3. Thermal FEA results obtained using appropriate thermal properties of plywood showed a reasonably good agreement with Baleshan's (2012) fire test results. Parametric studies using the validated model showed that joist section depth and profile had no significant impact on the thermal performance of LSF floor systems whereas steel joist thickness had a significant influence. An extensive FEA based parametric study was then undertaken to investigate the effects of joist thickness, depth, section profile, steel grade and mechanical property reduction factors, and web openings on the structural and fire performances (FRR) of LSF floor systems. Steel joist thickness significantly influenced the FRR of LSF floor systems due to different temperature developments in the steels for varying thicknesses. Joist section depth, section profile and web openings had no significant impact on the FRRs of LSF floor systems. Steel type affected the FRRs of LSF floor systems significantly due to different mechanical property reduction factors, especially different yield strength reduction factors. It was shown that Baleshan's (2012) critical average joist temperature method can be used to determine the FRR of non-insulated dual and single plasterboard lined floor panels made of HFS joists. However, it can be used for cavity insulated floor panels when the load ratio is less than 0.3. Fire test and FEA results showed that LSF floor panels made of LSB joists gave higher FRRs due to improved elevated temperature mechanical properties of LSB plate elements and lower temperature development due to thicker joists. Fire design rules were developed to predict the FRRs of LSF floor systems made of HFS joists based on Eurocode 3 Part 1.3 (ECS, 2006), AS/NZS 4600 (SA, 2005) and Direct Strength Method (DSM). For this purpose, Baleshan's (2012) three fire design rules of LCS joists were used and suitable modifications were made in order to use them for HFS joists. A good agreement was observed between the FRR predictions using two design methods and FEA, and thus they were recommended. In addition, the FRR predictions of HFS joists using the fire design method developed based on DSM were modestly conservative and therefore they were also recommended. Finally, the spread sheet based design tool was developed to undertake the complex calculations in predicting the FRR of LSF floors made of HFS joists with varying sizes and steel types, and subjected to varying load ratios. In summary, this research has significantly improved the knowledge and understanding of the fire performance of LSF floor systems made of hollow flange section joists and developed accurate fire design rules. Structural and fire design engineers can use the developed spread sheet based design tool to predict the fire performance of LSF floor systems made of HFS joists with varying sizes and steel types for a range of applications in commercial and residential buildings