Practice-Accessible Methodology for Nonlinear Refined Analysis of Gusset Plate Connections for Steel Truss Bridges

Abstract

A new approach is proposed for estimating structural capacity of gusset plate connections of steel truss bridges. The approach involves nonlinear refined analysis of a truss model, consisting of a single truss made of shell elements for two gusset plates at a subject connection and frame elements for truss members. Connectors (bolts or rivets) are excluded from the proposed model in order to simplify the model, with the knowledge that their associated failure modes can be addressed by simplified design calculations. Only yielding and buckling failure modes of gusset plates are considered. The new approach is calibrated by comparison with laboratory test data from NCHRP Project 12-84. The primary intent of this paper’s approach is to reduce the complexity of the refined analysis developed under this National Cooperation Highway Research Program project and make it more user-friendly to load rating engineers seeking accurate estimation of gusset plate capacity. Further, an imperfection sensitivity curve characterizing reduction in capacity as a function of lateral imperfection (gusset plate out-of-flatness) in gusset plate geometry is introduced as an added tool for the load rating engineer, providing more confidence in capacity estimation. Based on a limited case study, the proposed refined analysis provides an estimate of gusset plate capacity that is approximately equal to capacity calculated by the truncated Whitmore method of the AASHTO Manual for Bridge Evaluation (2018). In other words, the proposed approach validates the truncated Whitmore method as compared to the more conservative partial shear method of the AASHTO Manual.