Experimental and finite element study on RC beams retrofitted by full-length and spliced CFRP laminates

Fiber reinforced polymers (FRPs) in the form of laminates, fabrics, bars or rods have gained world-wide recognition and became a normal choice for reinforcing, repairing, and strengthening concrete structures. However, retrofit of members with long-spans or with limited accessibility (e.g. bridges over waterways and freeways) using continuous FRP laminates requires extensive labour and equipment. Splicing the laminate could provide an economical solution by reducing the cost of installation. This study consists of experimental testing of three full-scale RC beams subjected to four-point bending static loads. One beam served as a control specimen, while two beams were strengthened with Carbon FRP (CFRP) laminates (one beam by full-length laminate and the other by two half-length laminates made continuous with a splice). The under-reinforced control beam failed by yielding of steel in tension followed by crushing of concrete in compression. The strengthened beams failed by debonding, at the laminate ends for full-length laminate and at splice ends for the spliced laminate. The maximum load for the strengthened beams was 49% higher than the control beam for the full-length laminate; and 32% higher for the spliced laminate. In addition, three-dimensional finite element models were developed for the beams to provide an in-depth analysis including the simulation of debonding failure. Comparisons with the experiments, in terms of maximum load, load-deflection history and prediction of failure have confirmed the validity of the models and provided an excellent platform for future parametric studies intended to better design spliced laminates and reduce the likelihood of debonding.