COMPDYN 2015 5 th ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics and Earthquake Engineering M. Papadrakakis, V. Papadopoulos, V. Plevris (eds.) Crete Island, Greece, 25–27 May 2015 CYCLIC RESPONSE OF A STEEL BEAM-TO-COLUMN CONNECTIONS – AN EXPERIMENTAL AND NUMERICAL STUDY G.C. Manos 1 , A. Nalmpantidou 2 , V. Kourtides 3 , A. Anastasiadis 3 1 Professor and Director of the Lab. of Strength of Materials and Structures, Aristotle University e-mail: gcmanos@civil.auth.gr 2 Postgraduate student, Lab. of Strength of Materials and Structures, Aristotle University e-mail: a.nalbantidou@gmx.de 3 Dr. Civil Engineer, Research Ass., Lab. of Strength of Materials and Structures, Aristotle University e-mail: anastasiadisa@hol.gr Keywords: Steel beams, steel beam-to-column connection, cyclic response, plastic hinge, low-fatigue tests, strain rate effects, earthquake performance of steel connection Abstract. The cyclic response of steel beam-to-column connections under a cyclic sinusoidal loading history was studied. The experimental investigation included two full-scale specimens having a cross section of IPE240 and a height of 1.0m. This part of the specimen was rigidly connected to a much stiffer steel beam thus representing a steel beam-to-column connection under investigation. These specimens were tested to failure being subjected to a cyclic sinu- soidal point load of continuously increasing amplitude that produced a similarly varying bending moment at the steel beam-to-column connection. Instrumentation was provided to record both the applied moment as well as the deformation of the specimen in the region of the connection in terms of plastic hinge rotation. Each specimen experienced yielding and plastic rotation in this region that resulted initially in a plastic stable response till the final damage, which occurred either in the form of local instability of the flanges or fracture. The response of this beam-to-column connection was numerically simulated employing commer- cial software that tried to reproduce all the stages of the observed non-linear response. By comparing the numerical predictions with the measured response the success of the employed numerical simulation could be examined. The employed numerical simulation could yield re- alistic predictions of the limit state local buckling of the flanges of the used IPE 240 cross section as well as good agreement with the observed measured response, in terms of flexural capacity, energy dissipation and deterioration characteristics, provided that the material model in this numerical simulation was equipped with the “proper” parameters. 519