This randomized trial investigates compressive behaviour in steel truss bridge connections, indicating underestimation of resistances by existing codes.
Gusset plates are essential to the stability and load transfer of steel truss bridges, and their compressive behaviour can govern the connection capacity. Current design provisions in AASHTO and CSA S6 evaluate gusset plate compression using simplified column-based formulations developed primarily for interior connections. In contrast, the behaviour of corner gusset plate connections has received limited attention, despite their distinct restraint conditions and buckling characteristics. This study investigates the compressive behaviour of corner gusset plate connections through advanced nonlinear finite element (FE) analyses. The models incorporate geometric and material nonlinearities, realistic fastener behaviour, and initial geometric imperfections. The FE modelling approach is validated against large-scale experimental results reported by the FHWA Turner-Fairbank Highway Research Center. Following validation, a series of representative corner joint configurations is analysed to examine their buckling response under compression. The FE-predicted compressive resistances are compared with the nominal capacities determined using AASHTO and CSA S6 provisions over a broad range of slenderness parameters. These comparisons showed that both standards generally underestimate the nominal compressive resistance of gusset plates, with mean FE-to-code resistance ratios of approximately 1.48 for AASHTO and 1.5 for CSA S6, and that the degree of underestimation increases at moderate to higher slenderness levels. Effective length factors (K) are back-calculated from the FE results using both code formulations, and representative effective length factors are proposed specifically for corner gusset plate connections. The proposed values result in significantly improved agreement with FE-predicted critical compressive resistances compared to calculations based on the conventional K = 0.5 assumption.
No takes yet. Share an insight, caveat, or question.
Dalia et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: