Composite bonded joints under mode I fatigue loading

Abstract

Experimental investigation on fatigue behavior of carbon–epoxy composite bonded joints under mode I loading was performed in this work. The objective is to evaluate the performance of different data reduction schemes to obtain the energy release rate ( G I) in the fatigue crack growth (FCG) rate using double cantilever beam (DCB) specimens. This law relates the evolution of the crack along time as a function of the energy release rate ( G I) and is generally composed of three different regions: damage nucleation, stable propagation and abrupt final failure. The second phase corresponding to stable propagation leads to a linear trend on the Paris law representation (log–log scale) and must be well characterized to define the fatigue behavior of the structure. During fatigue tests the classical methods require rigorous monitoring of the crack length during its propagation, which is cumbersome and not easy to perform in some materials. In this work, an alternative data reduction scheme based on specimen compliance and crack equivalent concept is proposed to overcome this difficulty. The results provided by the proposed method, namely Compliance Based Beam Method (CBBM), are compared to the ones obtained from the polynomial and Beam on Elastic Foundation Method (BEFM), both of which require crack monitoring. The first is a compliance calibration method that fits a third-order polynomial curve to the experimental results (compliance ( C) versus crack length ( a)). The second one uses the beam theory to establish the C= f( a) relationship taking into account the properties of the adhesive. One additional advantage can be pointed to the proposed CBBM relative to the other classical methods. In fact, the equivalent crack is related to the specimen compliance, thus taking into account the influence of fracture process zone on specimen behavior. This issue is particularly important when adhesives with some ductility are being characterized in fatigue tests.