Experimental testing and numerical modelling of steel moment-frame connections under column loss

The beam-to-column connections of moment-resisting steel frames should exhibit capacities that allow them to transfer the forces that develop under normally expected loading conditions. However, when a column is lost owing to accidental loading, these conditions change, and the forces are redistributed to the adjacent beams and columns. In such cases, the connections must be capable of resisting the combined axial and flexural loads and allow for the redistribution of the loads, so that progressive collapse development is prevented. In this study, we investigated the performances of four types of beam-to-column connections, namely, the welded cover plate flange connection (CWP), the haunch end plate bolted connection (EPH), the reduced beam section welded connection (RBS), and the unstiffened extended end plate bolted connection (EP), against progressive collapse. Two span frames were constructed and tested under a central column removal scenario until failure. The results from the experimental tests were used to validate finite element models. The CWP, EPH, and RBS specimens showed good ductility, with the catenary action making a significant contribution to the ultimate load resistance. Further, the ultimate rotations of the beams were greater than the deformation limit given in the latest Unified Facilities Criteria guidelines for design of buildings to resist progressive collapse. Specimen EP showed the lowest ductility and ultimate load resistance, with the bolts in the rows under tension fracturing before the catenary action could develop. Further, the failure mode for specimen EP indicated that bolt strengthening is necessary for improving its progressive collapse resistance.