Flexural behavior of RC beams strengthened with fully or partially prestressed near-surface mounted FRP strips: An experimental investigation

The strengthening technique based on prestressed near-surface mounted (NSM) strips made of fiber-reinforced polymers (FRPs) is well known as an effective means to restore or enhance the flexural performance of reinforced concrete (RC) beams, notably in bridges. To better understand the mechanical behavior of RC beams strengthened with prestressed NSM FRP strips, 18 beams were tested under either static loading (15 specimens) or fatigue loading (3 specimens), with the focus on the failure mode, load-displacement response, stiffness, ductility, and fatigue life. The effects of the prestress level and bond length of the FRP strips were studied as well. The test results show that the bond length plays a major role in the flexural performance of the strengthened beams, since the greater the bond length, the higher the ultimate load, stiffness, and ductility. A strengthening method based on the partial prestressing of NSM FRP strips is proposed as well to reduce the bond stresses around the extremities of the strips and to increase the ultimate load of the beams by delaying the failure due to end debonding. According to the proposed method, the strengthening strips are prestressed along most of their length, but the segments close to the end sections are nonprestressed, because they are bonded to the concrete after prestressing has been released. Compared to the beams with fully prestressed NSM-FRP strips, partially prestressed strips improve the bearing capacity, ductility, and fatigue life of the strengthened beams. The role of the length of the nonprestressed segments is also discussed. Finally, a few strength models available in the literature, aimed at predicting the bearing capacity at the onset of beam failure (because of the separation of the concrete cover at the extremities of the strengthening strips), are applied to the beams tested in this project to check the reliability of the models vs. the test results.