Effects of end distance on thin sheet steel bolted connections

•Thin sheet steel (TSS) single shear and double shear bolted connections subjected to tensile loading were tested.•TSS 0.42 mm G550, TSS 1.20 mm G500 and 1.90 mm G450 were used to fabricate the connection specimens.•Effects of end distance on the strengths and failure modes of the connection specimens were investigated.•Predictions by Australian/New Zealand Standard, North American Specification and European Code are generally conservative.•Design provisions of Australian/New Zealand Standard are reliable and probabilistically safe. The effects of end distance on thin sheet steel (TSS) bolted connections were investigated experimentally. The TSS grades G450, G500 and G550, with the respective nominal thicknesses of 1.90 mm, 1.20 mm and 0.42 mm, were used to fabricate the connection specimens. The connection specimens were assembled by one bolt in both single shear and double shear. The connection specimens in the single shear and double shear were designed in 5 series with different ratios of bolt diameter to connection plate thickness. In each series, the end distance of the connection specimen was varied. The connection specimens were subjected to tensile loading using the displacement control test method. The effects of end distance on the behavior of the specimens were obtained, including ultimate loads and failure modes. It was found that the ultimate loads were increased when the end distances were increased up to three times and five times the bolt diameter for single shear and double shear, respectively. Furthermore, the tearout failure mode progressed to bearing failure mode with the increment of the end distance. The experimental results were compared with the predicted nominal strengths and failure modes by using the Australian/New Zealand Standard (AS/NZS), North American Specification (NAS) and European Code (EC3-1.3) for cold-formed steel structures. Overall, it is found that both the NAS and EC3-1.3 specifications generally provide conservative predictions, while the AS/NZS generally provides unconservative predictions. However, the reliability analysis showed that the design provisions of AS/NZS are reliable and probabilistically safe for the TSS bolted connections in this study. The AS/NZS, NAS and EC3-1.3 generally predict accurate failure modes of the specimens. However, the EC3-1.3 provides more accurate predictions when specimens fail in plate bearing.