PSI - Issue 78

Alessandro Mei et al. / Procedia Structural Integrity 78 (2026) 120–127

125

Fig. 5. Moment-rotation curves of all monotonic tests

Fig. 6. (A) Normalized stiffness according to EN15512. (B) Normalized collapse moment

Analyzing Fig. 5 and Fig. 6 it is possible to evaluate the increase in terms of stiffness and capacity moment of all the specimens. It is worth noting that, in every specimen, the weld between the beam and the connector is invariably the first element to fail; however adding brackets, especially 90° brackets, creates a new joint configuration where, after the weld failure, the connection continues working thanks to the self-drilling screws or bolts. However, this is not the case for the rivets (r04), where failure occurs instantaneously by punching rather than gradually by pull-out as with the screws. This behavior is well described by the curves of r03, and r04 (Fig. 5), where, after the peak moment and the subsequent drop, the r03 specimen starts to take load again, as shown by the curve increase, and it only fails when all four screws are extracted or undergo tension failure, while r05 do not significantly take load again. Finally, although the r05 reinforcement was designed to reduce connector deformability and improve joint stiffness and performance, in the tested configurations the weld between beam and connector consistently fails before any tab pull out, rendering the reinforcement ineffective. 4. Cyclic tests The moment-rotation curves obtained by applying a cyclic displacement history are reported in Fig. 7. It is worth highlighting that the shape of the hysteresis loops depends on the number of cycles applied. Specifically, the peak bending moment recorded in the initial cycle of each set is higher compared to the following two cycles, due to the progressive degradation experienced by the connector and a similar trend is evident for the energy dissipated per cycle.