Seismic Performance of Reduced Web-Section Beams

Abstract

This study investigates an innovative method of avoiding failure at the beam–column connection during earthquakes. Reduced web section (RWS) and reduced beam section (RBS) methods utilize the capacity design principle to move yielding points away from vulnerable beam–column connections to avoid the occurrence of such fractures. The RWS method shifts the area of inelasticity away from the connections by introducing large openings into the web. The configuration of the openings determines the type and capacity of the inelastic mechanism in the beam. This thesis reports the experimental results for 10 RWS specimens that were subjected to quasi-static cyclic-loading tests. Half of the samples were designed to develop Mode-V mechanisms; three had a single unique opening at midspan, whereas two others had two openings near the beam–column connections. The development of Mode B mechanisms was induced in five other specimens, four of which contained web posts created from more than five openings. The remaining sample was fabricated with a wide opening, and two brass plates were clamped to the web to induce a Mode-H mechanism without web-post buckling. The intended inelastic mechanisms were successfully realized using the specific web openings; inelastic deformation occurred because of yielding, buckling, and/or fracture of the webs around the openings and the plastic hinging of the T-sections located above and below the openings. In most Phase 1 cases, a stable hysteretic behavior was demonstrated up to a story drift of approximately 6%. The most stable load–drift responses were exhibited by the three specimens with single openings at their midspans; local buckling, web fracture, and the subsequent transition from full to S-shaped hysteretic loops resulted in a loss of strength during 3% or 4% drift cycles, although full strength was regained by the end of testing at story drifts up to 7%. This study also introduces the concept and potential inelastic modes of RWS beams and derives beam shear equations corresponding to the assumed plastic mechanisms. Finite element (FE) analysis was performed as a supplementary assessment and was confirmed to be a successful complement to physical experiments. This approach made it possible to determine that the buckling phenomenon occurred both before and after yielding. Additionally, a close correspondence was confirmed between the FE analysis and experimental results for drifts up to 4% prior to the occurrence of fractures in most RWS beams. For Specimens 2, 4, and 5, it was proven that even before fractures occurred, their 6% drift was consistent with the results of the FE analysis. However, the approach was unable to determine the cyclic behavior after fracture. Finally, a nonlinear time–history analysis was performed by applying the RWS beam to a building, and a fragility curve was derived. The seismic waves in the ordinary beam and RWS beam before yielding revealed no difference in safety. However, when the RWS beam was applied, the maximum strength of the roof drift decreased because of the load reduction from buckling. The ductility of the RWS beam increased the roof drift until the time of destruction by about 1.5 times compared with that of the ordinary beam building, allowing it to withstand earthquakes of higher intensity.