Peeling failure of reinforced concrete beams with fibre-reinforced plastic or steel plates glued to their soffits

Abstract

In recent publications by the first author and his associates, a theoretical model (backed by extensive test data by others) for premature plate peeling failures of reinforced concrete beams strengthened in flexure by gluing steel plates to their tension sides was reported. The primary purpose of the present paper is to extend this theoretical model to cases where fibre-reinforced plastic (FRP) (as opposed to steel) plates are used for upgrading reinforced concrete beams in flexure. As in the case of steel, due to large variations in the spacings of stabilized cracks within the concrete cover zone (by a factor of, say, 2) a unique solution for the FRP plate peeling load does not exist, and one needs to resort to theoretical upper/lower bounds, with the lower bound being the appropriate one for design purposes. For steel plates, a simple preliminary design approach is to keep the plate width to thickness ratio above 60. However, no such simple value for the plate width/thickness ratio has been found for externally bonded FRP plates (which, unlike steel plates, are available with a wide range of ultimate strengths and Young's moduli); hence the pressing need for a reliable theoretical model. Finally, with the theoretical predictions for FRP and steel-plated beams backed by nearly 170 (mainly large-scale) test results from a number of independent sources, covering a wide range of beam design parameters, the proposed model is thought to be reliable and generally applicable.