Issues Involved in Seismic Retrofit of Reinforced Concrete Frames Using Infilled Walls

Buildings in the United States, designed and constructed in the late 1950's and early 60's in accordance with prevalent standards, are often found to be inadequate to withstand major earthquakes. Gravity loading was the primary concern in the design of these buildings, and as a result, they may not possess adequate lateral strength and ductility. Addition of infill walls is a widely-used scheme for seismic retrofit of non-ductile reinforced concrete moment-resisting frames (RCMRF) which have been identified as a type of structural system that poses significant potential hazard to life safety and may result in major financial loss.

The current experimental study focused of issues associated with seismic retrofit of non-ductile RCMRF using infilled walls. Issues investigated were: (1) retrofit techniques for strengthening column lap splices which may control the performance of retrofitted RCMRF; and (2) the shear transfer mechanism across frame-wall interfaces.

Investigation of retrofit techniques for column lap splices involved strengthening column-splice specimens using a variety of selected retrofit techniques and testing the performance of retrofitted splices in terms of splice tensile strength and ductility. The retrofit techniques used included: confinement of the splice region with steel angles and straps or reinforcing bar ties; and welding of spliced bars. They were intended to supplement confinement in the splice region to delay/prevent splitting of concrete cover in the splice region or to provide continuity between spliced bars. Retrofitted column-splice specimens were subjected to axial load reversals simulating the action of seismic forces in the boundary elements (columns) of a frame-wall system. Providing confinement or continuity improved splice tensile strength and ductility in most cases.

The study of the shear transfer mechanism across a frame-wall interface involved constructing test specimens that were representative of a portion of the frame-wall interface and subjecting the specimen interface to different load patterns in direct shear. Variables investigated in the test program included: type of shear loading (reversed cyclic vs monotonic), level of external compression across the specimen interface, number of dowels (the dowels were not anchored to develop yield) as shear connectors across the specimen interface, strength of concrete in the frame segment of the test specimen, and procedure used for construction of the specimen interface. Test results provided information for establishing (1) the benefit of cyclic compression across frame-wall interfaces resulting from seismic loads, (2) the number of dowels that must be provided across the interface for satisfactory performance, (3) the influence of strength of concrete in the existing frame, and (4) the strength to be assigned to grouted (dry-packed) frame wall-joints. Test results were used to verify and extend the application of shear-friction provisions currently incorporated in Sect 11.7 of ACI 318-89.