Title: Cyclic Response of Composite Coupling Beams
Author(s): Bingnian Gong, Bahram M. Shahrooz and Arnold J. Gillum
Publication: Symposium Paper
Appears on pages(s): 89-112
Keywords: Beams; composite construction; connections; cyclic loads; embedment;
seismic analysis; walls
Adequate performance of coupled walls depends on sufficient stiffness, strength, and energy dissipation of coupling beams. To meet these goals, reinforced concrete coupling beams are often deep On the other hand, shallower steel beams can be used instead, and steel/composite coupling beams may be designed as shear-yielding members which have a more desirable energy dissipation characteristics. Such an option is not feasible for reinforced concrete beams. Well-established guidelines for links in eccentrically braced frames can be extrapolated to steel coupling beams. However, these provisions ignore the effects of concrete encasement which often surround the steel coupling beam. The reported research was conducted in an effort to till this void. Four one-third scale subassemblages of a wall segment and a coupling beam were extracted from a prototype structure, and were tested. The main test variables were the presence or lack of concrete encasement, and the number of web stiffeners. The encasement around the steel coupling beam increased the beam stiffness by 25%, and the shear strength by 18%. The additional stiffness enhances the level of coupling action which could lead into significantly larger wall axial loads, and would increase the demands on the foundation system. The increased stiffness needs to be incorporated in design. Although all the specimens could develop and exceed the expected capacities, the specimens failed due to less than desirable performance of the connection. Participation of the connection region towards energy dissipation became more substantial for the encased specimens, which is not desirable in view of post-earthquake repair. Connection design has to account for the increased capacity due to encasement, and details need to be improved to delay connection failure until plastic hinges in the coupling beam are fully mobilized. The encased specimens without any stiffeners performed as well as the specimens with stiffeners equal to or less than those required for steel link beams. No significant differences between the strength and stiffness characteristics of the encased specimens could be found. The experimental data suggest that nominal encasement is an effective means for preventing web buckling, and stiffeners are not needed. Current design codes need to be reevaluated for the cases where the steel coupling beam is encased.