Bond behaviour of externally bonded FRP to concrete in the case of 3-point-bending tests

By conducting 3-point-bending tests on flexural strengthened girders with FRP bonded over the entire length of the specimens, debonding of the external applied laminates is observed caused by intermediate crack debonding. Herewith the debonding mechanism is induced at a flexural-shear crack under the point load, which is a zone with a combination of high shear forces and high moments. By predicting the debonding load of this mecanism following fib Bulleting 14, two different models have to be concerned. First here is "peeling-off caused at shear cracks" and secondly there is "peeling-off caused at flexural cracks". Herwith, the first model gives a theoretical debonding load which is considerably higher compared to the experimental debonding load. One of the possible explanations of this phenomenon can be found in the fact that this model is a simplified model calibrated on 4-point-bending test results, while in this study 3-point bending is considered. Another motivation is the fact that the origin of intermediate crack debonding is not solely situated at (steep) shear cracks, but also at flexural cracks. Herewith, the second debonding model of fib Bulletin 14 can be used, which gave accurate predictions of the experimental debonding loads. Concerning these two different models, it would be more appropriate to have a model which combines both peeling-off caused at shear cracks and at flexural cracks. Such a model is proposed by Oller et al., which calculates a maximum shear force versus bending moment relationship to prevent peeling failure. By using the latter model, a good and save prediction of the debonding load is observed