Strengthening of brick infilled RC frames with different aspect ratios by means af CFRP overplays

Abstract

The effects of aspect ratio on the engineering properties (stiffness, strength, energy dissipation, ductility…etc.) of infilled RC frames with or without CFRP were evaluated experimentally and analytically. Ten single-bay, two-story specimens at 1/3 scale were tested in a vertical position under quasi-static reversed cyclic seismic loading. Vertical and lateral loads were applied concurrently. Two out of ten specimens were control specimens (with no CFRP) and the rest of them were strengthened with CFRP unidirectional overlays in various configurations. The specimens had some deficiencies commonly seen in pre- 1998 Turkish construction. The specimens had detailing deficiencies; such as insufficient confinement, 900 hooks, no transverse reinforcement in the joints, design deficiencies; such as weak column-strong beam, and construction deficiencies; such as poor material quality, and inadequate lap splice length. To eliminate the existing deficiencies of the specimens and improve their seismic performance, unidirectional CFRP overlays were applied in the form of vertical strips and/or cross diagonals on both sides of the infill panel and anchored to the surrounding RC frames. Local rehabilitation techniques such as confinement of the frame members were also used for some of the specimens. The application of CFRP overlays improved the load carrying capacity and the energy dissipation capacity significantly. Generally, as the peak loads increased, their corresponding drifts also increased. Some stiffness improvements were also recorded depending on the configuration of the CFRP overlays and the aspect ratio of the frames. The test results were compared with the numerical results obtained from the nonlinear static pushover analysis. The analytical model incorporated the equivalent diagonal compression strut concept for the infill panel. A new stress-strain model for the infill panel was proposed and calibrated by using test data. The numerical model predicted global behavior satisfactorily.