The aim of the paper is to investigate the shear failure mechanisms in T-shape, single span and simply supported beams exclusively reinforced with longitudinal glass fiber reinforced polymer (GFRP) bars. Usually the critical shear crack in RC beams without stirrups develops through the theoretical compression strut reducing the shear strength following the shear failure. The main parameter affecting the crack pattern and the shear strength of the beams is the shear slenderness. However, the test results presented in the paper indicated the new arching effect due to the bond losing between the GFRP flexural reinforcement and concrete. This failure mode revealed unexpected critical crack pattern and failure mode. The research of concrete beams flexurally reinforced with GFRP bars without stirrups indicated two failure modes: typical shear-compression and a new one leading by the bond losing between the ordinary reinforcement and concrete.

The aim of this paper is a comparative analysis of the experimental test results of twenty T-section beams reinforced with glass fiber reinforced polymer (GFRP) bars without stirrups with predicted values of the shear capacity according to the following design guidelines: draft Eurocode 2, Japanese JSCE, American ACI 440, Italian CNR- DT-203/2006, British BS according to fib Bulletin 40, Canadian CSA-S806-12 and ISIS-M03-07. Standard procedures for FRP reinforced beams based on traditional steel reinforced concrete guidelines. The longitudinal FRP reinforcement has been taken into account by its stiffness reduction related to the steel reinforcement. A basis of this modification is the assumption that the FRP-to-concrete bond behaviour is the same as it is for steel reinforcement. To assess the compatibility of predicted values (Vcal) with the experimental shear forces (Vtest) the safety coefficient η = Vtest / Vcal was used. The results corresponding to values η < 1 indicates overestimation of the shear capacity, but η > 1 means that shear load capacity is underestimated. The most conservative results of the calculated shear capacity are obtained from the ACI 440 standard. In contrast to them the best compatibility of the calculated shear values to the experimental ones indicated British BS standard, fib Bulletin 40 and Canadian CSA-S806-12 standard.