Investigation of trapezoidal cladding with wide pans subject to splitting-type pull-through failures

Profiled steel roof and wall cladding systems used in Australia and neighbouring countries are made of thin, high-strength steel and are typically crest-fixed. When subjected to wind suction/uplift loading, the currently used trapezoidal steel claddings with wide pans often experience premature splitting-type pull-through failures at their screw connections due to significant stress concentrations under the screw heads. These failures, primarily initiated by transverse splitting at the screw fastener holes, challenge accurate determination through analytical studies owing to the less ductile nature of the high-strength steel. Consequently, the design of crest-fixed profiled steel claddings relies on costly laboratory tests, lacking adequate design rules. This study proposes and calibrates a new ductile damage material model specifically for very thin, high-strength cold-formed steel. This model enables accurate prediction of both splitting- and dimpling-type pull-through failures through detailed simulation of damage initiation, evolution, and failure. The validated finite element model incorporating this material formulation is used to perform an extensive parametric study, systematically evaluating the influence of geometric and material parameters under static wind suction loading. Using these results, accurate design capacity equations are proposed, and response surface optimisation technique is used to highlight performance improvements within realistic geometric bounds offering improved guidance for the safe and economical design of crest-fixed trapezoidal cladding systems. • Developed and validated an FE model of crest-fixed trapezoidal cladding with closely spaced ribs under wind uplift. • Proposed and calibrated a ductile damage material model to predict both dimpling- and splitting-type pull-through failures. • Conducted over 400 FE simulations to evaluate the effects of key geometric, material, and fastener parameters. • Proposed a new pull-through capacity equation, predicting numerical results with a maximum variance of only 5%.