Thin metal patch to FRP propeller: Experimental testing of bonding systems

Abstract: To enable the use of carbon fibre reinforced polymer (CFRP) propellers for merchant vessels, the problem of cavitation must be solved. The matrix of the FRP is generally brittle and more sensitive to cavitation erosion than the more conventional metal propeller blades. Nakashima Propeller Co. wants to solve this problem by attaching a metal patch adhesively in the high-risk areas of the blade. The bonding of FRP and metal is a classic problem when manufacturing hybrid components, and even more difficult when the bond is subject to fatigue loads, in a marine environment. The thin metal plate makes determining the crack resistance from the energy release rate difficult and a novel approach is necessary.This report focuses on outlining the problem and establishes an experimental method for conducting the initial investigations of the robustness of an adhesive bond for this application. Four different adhesives were tested in quasi-static mode I loading with constant cross-head velocity: Denatite 2204, co-curing, Newport 102 and Plexus AO420. All adhesives were epoxy based except for the Plexus. The co-cured system used resin Epolam 5015, the same was used for manufacturing the FRP in the other specimen. Instead of computing the energy release rate, the crack tip angles were compared between the different adhesive systems. These measurements are only internally comparable. Based on the results from the quasi-static loading, the Denatite and the Plexus were brought into a smaller fatigue testing series. The two others were deemed as too weak to be of interest.The experiments showed that the methacrylate based Plexus AO420 had a significantly higher strength, in the quasi-static loading it had a maximum recorded load almost ten times as high as the denatite. The fatigue test was inconclusive, although no drastic or unexpected behaviour was recorded. It was noted during the process (unsurprisingly) that the manufacturing process and the skill of the worker is very much affecting the bond strength, sometimes even more than the material properties and the surface treatments applied.The experimental method, including the measurements of the crack tip angle could be concluded as a simple but effective method to determine the basic adhesive strength and general characteristics of the different types. The second series of fatigue testing can be performed to see general fatigue behaviour for the adhesives, but as the actual load case on the propeller is unknown it can be difficult to directly relate the fatigue behaviour in the test to the real world. Both test series can be used to determine general behaviour, but due to the complexity of the geometry, deformation modes and loads on the real propeller, it is suggested to perform final testing on a live propeller, rather than trying to draw to large conclusions from isolated experiments.