IC debonding failure in RC beams strengthened with FRP: Strain-based versus stress increment-based models

• A new interfacial fracture energy model is proposed using beam-type tests.
• Fracture energy models are applied to an existing stress-increment based model.
• Calibration of the stress increment-based model was not conducted.
• The models are evaluated using the breakdown of residual sum of squares.
• The stress increment-based model yields a close agreement with experimental data.

Since the advent of the use of fibre-reinforced polymers (FRP) to strengthen reinforced concrete (RC) and its employment in civil engineering in the early nineteen nineties, much of the research has concerned the enhancement of bending capacity. Failure modes are well understood and can be predicted with some accuracy. The models adopted by the most highly reputed international standards for the specific case of inter-crack (IC) debonding are based on limiting the strain on the laminate. These methods, an offshoot of fracture mechanics analysis, are calibrated with the results of point load beam tests. This paper proposes a new interfacial fracture energy model by characterising the FRP-concrete interface with beam-type tests. The existing fracture energy models, as well as the proposed model, were applied to an existing stress increment-based formulation, in which no calibration was conducted with experimental results and yet proved to predict inter-crack debonding failure moderately better than the models in use. The performance of each IC debonding model is assessed through an exhaustive statistical analysis with experimental data of point load RC beams tests gathered from literature. The behaviour of the models in uniformly loaded RC beams is discussed and the implications of each approach are drawn.