Ansys has a Fatigue object in Workbench that allows stress-life and strain life fatigue analysis. This can be found by searching the help. I've never used it. Looks like the trick is getting your fatigue curve into the material database correctly.
Fatigue of thermoplastics is not straight-forward. Plastics behave very differently from metals due to their viscoelastic nature, and the type of loading influences the fatigue data you use. For instance, when plastics are stressed at high strain rates, they produce heat, which affects their properties. If tensile loads are applied to the specimen, the entire cross section heats, so the cycle rate must be low. This drags out the test duration, and creep effects will occur. If bending loads are applied, the heating is somewhat localized, and the temperature increase is lower, so the testing can be faster, and creep effects are reduced. The point is that the fatigue data is highly dependent on the testing method and parameters. The test method should be matched to the type of loading in the actual part. If the part has complex geometry and loading, it may see bending or tensile or both, which would require multiple fatigue curves for different locations in the part. This is just the tip of the iceberg. You cannot grab an S-n curve off the internet, plug it in and go and expect to get good results. You need to review fatigue of plastics. Two sources are Structural Analysis of Thermoplastic Components by Trantina and Nimmer, and Fatigue of Engineering Plastics by Hertzberg and Manson. There is also a treatment I think in Hertzberg's Deformation and Fracture Mechanics of Engineering Materials.
Rick Fischer
Principal Engineer
Argonne National Laboratory
