This paper presents a numerical investigation of the flexural behavior of timber beams externally strengthened with carbon-fiber-reinforced polymer (CFRP) sheets. At first, the accuracy of linear elastic and elastic-plastic models in predicting the behavior of bare timber beams was compared. Then, two modeling approaches (i.e., the perfect bond method and progressive damage technique using the cohesive zone model (CZM)) were considered to simulate the interfacial behavior between FRP and timber. The models were validated against published experimental data, and the most accurate numerical procedure was identified and subsequently used for a parametric study. The length of FRP sheets varied from 50% to 100% of the total length of the beam, while different FRP layers were considered. Moreover, the effects of two strengthening configurations (i.e., FRP attached in the tensile zone only and in both the tensile and compressive zones) on load-deflection response, flexural strength, and flexural rigidity were considered. The results showed that elastic-plastic models are more accurate than linear elastic models in predicting the flexural strength and failure patterns of bare timber beams. In addition, with increasing FRP length, the increase in flexural strength ranged from 10.3% to 52.9%, while no further increase in flexural strength could be achieved beyond an effective length of 80% of the total length of the beam. Attaching the FRP to both the tensile and compressive zone was more effective in enhancing the flexural properties of the timber beam than attaching the FRP to the tensile zone only.

This paper reports an investigation on the damage tolerance of adhesively bonded pultruded glass fiber–reinforced polymer (GFRP) joints under fatigue loading. Double-strap joints with and without bondline defects were subjected to constant amplitude fatigue loading. The study investigated the effects of the defect location, size, and number on the fatigue performance of the joints (i.e., failure mode, fatigue life). Results showed that the failure mode was a combination of delamination and adhesive fiber–reinforced polymer (FRP) interface debonding at the defect vicinities. The fatigue life of the joints was significantly reduced (up to 90%) when the defect was at the overlap end. A defect size larger than 10% of the total bonded area could yield a substantial reduction in fatigue life. Small defects (less than 10% of the total bonded area) could yield a reduction in fatigue life of up to 73% when their number exceeded four. The joint exhibited better damage tolerance under static than fatigue loading conditions.