Microstructure-based fatigue modelling with residual stresses: Prediction of the fatigue life for various inclusion sizes

• Published in: International journal of fatigue 2019-12, Vol.129, p.105158, Article 105158
• Main Authors: Gu, Chao, Lian, Junhe, Bao, Yanping, Xie, Qingge, Münstermann, Sebastian
• Format: Article
• Language: English
• Subjects: Bearing steels; Calcium aluminate; Computer simulation; Crack initiation; Crack propagation; Defects; Fatigue failure; Fatigue life; Fracture mechanics; Heat treatment; Inclusion size; Inclusions; Materials fatigue; Metal fatigue; Microstructure; Microstructure-sensitive modelling; Predictions; Residual stress; S N diagrams; Size effects
• ISSN: 0142-1123; 1879-3452
• DOI: 10.1016/j.ijfatigue.2019.06.018

[Display omitted] •The S-N curves for differnet inclusion sizes are accurately predicted by the proposed model model.•The residual stresses between the inclusions and matrix is critical for the fatigue life prediction.•The effect of the inclusions size on fatigue strength is quantitatively analysed.•A general 3D analytical equation is proposed between fatigue life, stress amplitude and inclusion size. In this study, the inclusion induced fatigue failure of a high-carbon bearing steel was investigated. The experimental results showed that the size of calcium aluminate inclusions observed near the fatigue crack initiation site ranges from 12.5 μm to 33.2 μm and the size has a pronounced impact on the fatigue life. A microstructure-based model that considers the residual stresses between the steel matrix and inclusions induced by the heat treatment of the steel was developed and applied to investigate the effect the inclusion size on fatigue properties. The detailed model parameter calibration strategy and its validation were illustrated. It is concluded that the model considering the residual stress showed very good predictive capability of the S-N curves for different inclusion sizes, while the model without residual stresses failed to reflect the inclusion size effect on the S-N curve. In addition, based on the simulation data with accurate inclusion size control, an analytical relation between fatigue life, fatigue stress, and inclusion size was proposed for the investigated steel.