Effectiveness of novel methods to increase the FRP-masonry bond capacity

The paper is devoted to an experimental campaign on fifteen perforated masonry walls, which were strengthened with external FRP reinforcement, bonded using three different installation methods. Each wall was composed of two piers and two spandrels. The reinforcement consisted of carbon-FRP strips, which restricted opening of cracks already formed (sometimes referred to as repairing). In three tests, the strips were installed using the traditional method (i.e., epoxy resin applied to the masonry substrate, then layers of fabric impregnated using a roller, so that the composite and bond were formed at the same time). In six tests, the traditional method was enhanced by installing the strips in vacuum condition (i.e., the space around the impregnated strips was sealed and vacuum was created by removing air from that space using a vacuum pump). In six tests, the traditional method was enhanced with anchor spikes (i.e., the bond was provided by both epoxy resin and FRP spikes connected to the FRP strips). Each perforated wall was loaded by two vertical constant forces, each one directed along the axis of a pier, and an increasing in-plane lateral force, applied at the top spandrel. The load-carrying capacity of every wall was dictated by debonding of the external FRP reinforcement. The paper presents the experiments and the analysis of the results, including a discussion where the test results are compared and interpreted. The results are also related to existing knowledge on the topic, including code provisions. The bond strength (force that caused debonding of the FRP) with the vacuum method resulted to be double that with the traditional method, and the deviation lower. The bond strength with the anchor spike method resulted to be nearly double that with the traditional method, and the deviation the lowest of the three installation methods. The novel methods did not increase substantially the optimal bond length. The results also demonstrated that the code prescribed formulas for bond strength with the traditional method are overly conservative