Dr. Núñez-Castellanos, Eduardo

In this research an experimental study to assess the cyclic behavior of bolted moment connection in racks structures is performed. The effect of bolt pretension in the response is evaluated. Sixteen full-scale steel rack joint configurations were subjected to cyclic load according to the protocol established in AISC Seismic provisions. The cyclic performance was evaluated in terms of hysteretic response, failure mechanism, energy dissipation, stiffness, and rotation on the components. Two different configurations were studied. The results showed that the steel rack connections using a 70 % of bolt pretension can accommodate a 0.8 My at 0.04 rad of drift angle, while the joints without bolt pretension reached values below 0.7 My at 4 % of rotation. The failure mechanism was controlled by weld fracture at 4 % of the rotation. A high dispersion in the energy dissipation pattern was obtained and a drop in energy dissipation of up to 4 times in all specimens tested for a 4 % rotation was developed. This phenomenon is due to the welding rupture between the beam and the L-connector. A degradation of the secant stiffness reached up to 60 % for 2 % rotation. Finally, the most important effect of bolt pretension on the cyclic response of steel rack connections was achieved in the increase of flexural resistance and rotation developed.

The study of moment connections in steel structures subjected to cyclic loads has been extensively studied, providing a great number of requirements, including the strong column-weak beam relationship, to guarantee a satisfactory cyclic performance. However, investigations on the cyclic performance of moment connections considering the bidirectional and axial load effects simultaneously with tubular columns are limited. This study aims to assess and validate the strong column-weak beam relationship of 3D steel moment connections using reduced order models. The simplified model (reduced order model) approach was employed to extend the range of beam and column elements sizes and reduce the experimental and computational costs. These models were calibrated from full-scale experimental studies. A great number of configurations with different beam and column sizes without loss of reliability and structural representativeness of the studied phenomenon were studied. A total of 13640 simplified models were developed. Results show a cyclic behavior controlled by the strong column-weak beam relationship to modify the joint’s failure mechanism. The increasing of strong column-weak beam relationship and the biaxial effect caused degradation of the strength and stiffness as well as in dissipated energy. An optimal strong column-weak beam relationship was obtained for all joint configurations analyzed. Finally, a robust design procedure is proposed, ensuring the cyclic behavior of end-plate moment connection with built-up box column including biaxial effect and axial load. Therefore, the use of this type of moment connection can be used in special and intermediate moment frames designed according to Seismic provisions.

In this numerical research, the variation of cyclic behavior of beam-to-box column connection was studied. Dimensional and load conditions were parametrically evaluated with the goal of assessing the applicability and use of this biaxial moment connection according to the Seismic Provisions, such as bending strength and rotation capacity, secant and tangent stiffness, dissipated energy and strong column-weak beam relationship. A total of 83 different models of 3D connections were developed using in ANSYS software with the load at the top of the column. Results show a cyclic behavior not controlled by axial load. However, the variations of clear span to depth beam ratio caused degradation of the strength, secant and tangent stiffness as well as in dissipated energy. The 80% of plastic moment of beam and rotation at 4% interstory drift were reached for all models analyzed according to criteria established in AISC 341. Finally, the configurations designed with low levels of axial load are controlled by the design of the web panel zone shear, while configurations designed for high levels of axial load are controlled by the strong column-weak beam criterion.