Defect-based fatigue model for additive manufacturing

Abstract

A novel defect-based fatigue model for the prediction of S–N (stress versus number of cycles) data points and curves is proposed in this paper. The model is capable of predicting the material fatigue performance based on defect size and location from the surface. A defect factor was introduced and obtained based on notch theory, which considers the notch sensitivity of the material as well as the stress concentration obtained using the finite element method. A newly developed equation was applied to represent the relationship between the defect factor, defect size and defect location from the surface. AlSi10Mg samples were manufactured using laser powder bed fusion, and then machined. The samples were tested under rotational bending cyclic loading until failure. The failed samples were analysed using scanning electron microscopy and it was found that cracks initiated from defects located at the surface. The measured defect size and location were used to predict the number of cycles for an applied stress using the proposed defect-based fatigue model. This model was validated by comparing the predicted and experimentally obtained S–N data. The proposed model has the potential to be applied to component-level fatigue assessment and integrated into industrial quality assurance workflows. For instance, defects can be measured for each produced industrial component and directly assessed against fatigue performance using the developed defect-based fatigue model. This could enable the rapid approval and certification of future additively manufactured industrial components, which can unleash the commercial potential of additive manufacturing for light-weight multi-functional component designs.