Fatigue life prediction considering variability for additively manufactured pure titanium clasps.

Abstract

This study aims to develop a numerical prediction method for the average and standard deviation values of the largely varied fatigue life of additively manufactured commercially pure titanium (CPTi grade 2) clasps. Accordingly, the proposed method is validated by applying it to clasps of different shapes. The Smith-Watson-Topper (SWT) equation and finite element analysis (FEA) were used to predict the average fatigue life. The variability was expressed by a 95% reliability range envelope based on the experimentally determined standard deviation. When predicting the average fatigue life, the previously determined fatigue parameters implemented in the SWT equation were found to be useful after conducting fatigue tests using a displacement-controlled fatigue testing machine. The standard deviation with respect to stroke and fatigue life was determined for each clasp type to predict variability. The proposed prediction method effectively covered the experimental data. Subsequently, the prediction method was applied to clasps of different shapes and validated through fatigue tests using 22 specimens. Finally, the fracture surface was observed using scanning electron microscopy (SEM). Many manufacturing process-induced defects were observed; however, only the surface defects where the maximum tensile stress occurred were crucial. It was confirmed that the fatigue life of additively manufactured pure titanium parts is predictable before the manufacturing process considering its variability by performing only static elasto-plastic FEA. This outcome contributes to the quality assurance of patient-specific clasps without any experimental investigation, reducing total costs and response time.