Predicting fatigue S-N (stress-number of cycles to fail) behavior of reinforced plastics using fracture mechanics theory

The present study describes a model to predict fatigue S-N behavior, and thus fatigue life, of glass fiber reinforced thermoplastics by using a fracture mechanics approach. The model assumes the presence of an inherent initial flaw in the molded plastic parts and thus ignores crack initiation contributions. In this paper we describe how fatigue crack propagation rate data were obtained for the same three glass fiber reinforced plastics whose S-N behavior was previously described in detail. Using the measured constants from the crack growth data, and corresponding S-N data for uncracked specimens, the validity of the single initial flaw hypothesis was evaluated. From the analyzed results it is concluded that accurate S-N predictions are possible using this simple fracture mechanics model for some materials. The best results are obtained for glass filled polyamide, PA (nylon 66) and polycarbonate, PC; however, with polybutylene terephthalate, PBT, predictions were poor. It is also shown that S-N data for different glass fiber orientations can be predicted by combining the single flaw model with predicted fatigue crack propagation rate measurements. The latter are calculated from a generalized crack growth rate expression utilizing the strain energy release rate fracture mechanics parameter, which was previously described.