A Math-Based Methodology for Fatigue Longevity Prediction of 3D Woven Fiberglass Reinforced Vinyl-ester Composites 2006-01-0001

In the DOE-Delphi Composite Chassis Cross-Member program, 3TEX 3Weave™ (3D woven fiberglass mat)/vinyl-ester (Dion 9800™) composites have been investigated as a candidate material. One of the most important mechanical properties for qualifying these composites for such applications is the mechanical fatigue longevity. In this work, a predictive math-based technology has been developed as a virtual engineering tool for the design of 3TEX 3Weave™/vinyl-ester composite parts by using a state-of-the-art simulator, GENOA™ (Generalized Optimization and Analysis) PFA (Progressive Failure Analysis), developed jointly by Alpha Star Corp and NASA. This math-based GENOA™ methodology effectively tracks the details of damage initiation, growth, and subsequent propagation to fracture, for composite structures subjected to cyclic fatigue, thereby predicting the fatigue life. The material database inputs are: the experimental data for the stress-strain curve and the S-N curve for the vinyl-ester resin, the experimentally measured volume fraction of voids in the matrix, and the Young's modulus and the S-N curve for the fiber. The last response was “reverse engineered” using GENOA™ to match values measured experimentally for a composite with a measured volume fraction of voids. The utility of the GENOA™ technology was demonstrated by predicting premature and extended fatigue lives in tensile mode of various 3TEX 3Weave (7-ply E-glass fiber)/Dion 9800™ vinyl-ester composites. The fatigue longevity of the 3D woven ISO coupons simulated using GENOA™ agrees well with those measured in actual tensile-tensile fatigue tests using the R (minimum-to-maximum stress ratio) value of 0.1. Furthermore, GENOA™ PFA simulations quantitatively predict the effect of the void content on premature fatigue failures. Indeed, a 10% volume fraction of void defects reduces the fatigue life of the 3-D woven composite by a factor of 40 at the tensile load of 30% composite ultimate strength. On the basis of these results, this math-based predictive methodology is currently being used by the DOE (Department of Energy)-NCC (National Composite Council) Composite Chassis Cross-Member program.