Comparative behaviour of ‘I beam- RHS column’ joints with and without web weld

This paper presents the results of an experimental and numerical comparative study for beam-column joints constructed of welding open section beams to rectangular hollow section (RHS) columns between welding the entire beam profile and welding only the beam flanges. . Since flanges of open section beams contribute most of the section to the joint stiffness and moment resistance, welding the flanges only has the potential of achieving nearly the same joint stiffness and moment resistance as welding the entire beam profile while reducing the cost of welding considerably. However, it is necessary to quantify the effects. Furthermore, since the beam web is conventionally considered to resist the beam shear force, it is also necessary to investigate whether premature shear failure will occur in the case of welding the flanges only.
This paper presents the results of 27 tests and accompanying detailed numerical modelling to provide further insight into the experimental observations. The 27 tests considered the effects of different welding arrangements (flanges+web, flanges only, web only), RHS dimensions, different beam sizes, different beam to RHS width ratios.
The experimental observations suggest that due to flexibility of the RHS face, the contributions of the beam web weld to the total joint stiffness and moment resistance are much lower than those of the beam web to the beam section. In fact, the beam web weld contributions to joint stiffness and moment resistance were no more than 1% for the tested samples and no more than 3% for a wide range of beam/RHS dimensions. Furthermore, if only the beam flanges are welded, the flange welds have much higher shear resistances than would be applied in realistic applications.
Based on these findings, the characteristics of welded open section beam to RHS column joints can be quantified by using the contributions of the beam flanges only.