Effect of process deformation history on mechanical performance of AM60B to AA6082 self-pierce riveted joints. (15th March 2019)

• Record Type: Journal Article
• Authors: Moraes, J.F.C.; Jordon, J.B.; Su, X.; Barkey, M.E.; Jiang, C.; Ilieva, E.
• Is Part Of: Engineering fracture mechanics. Volume 209(2019)
• Page Start: 92
• Page End: 104
• Publication Date: 2019-03-15
• Subjects: Self-pierce riveting -- Finite element analysis -- Dissimilar alloys -- Fatigue, fracture; Fracture mechanics -- Periodicals; Rupture, Mécanique de la -- Périodiques; Fracture mechanics; Periodicals; 620.112605
• DOI: 10.1016/j.engfracmech.2018.12.020
• Languages: English
• ISSNs: 0013-7944

Abstract

Highlights: Non-linear FEA of the SPR process achieved good agreement compared to experimental results. Predicted crack initiation and fracture location agreed with quasi-static experiments. Adding residual stresses and plastics strains improved numerical results. LEFM fatigue life predictions of SPR agreed with experimental results. Abstract: The role of the process deformation history on the mechanical performance of self-pierce riveted (SPR) joints is examined in detail. The SPR process was simulated using a non-linear finite element model coupled with a damage criterion sensitive to the stress state. Results of the FE simulation compared closely to cross-sectional images of experimental SPR joints. The FE simulation results suggest that the strain hardening due to the piercing process is the main mechanism for joint strength under quasi-static condition. Aspects of fatigue damage were modeled including crack initiation, propagation, fretting and number of cycle to failure, and compared to experimental results. Results from the study indicate that residual stresses and plastic strain may alter the crack initiation location and crack propagation plane, leading to more accurate results when included in finite element analysis of SPR joints. Including these residual stresses and plastic strains also changed the predicted fretting locations leading to better agreement with experimental results. Lastly, linear elastic fracture mechanics were performed to estimate fatigue lives based on initial crack location and angle of crack plane. The estimation of the number of cycles to failure that included residual stresses and plastic strains resulted in a better correlation to experimental data when compared to the calculations that excluded residual stresses and plastic strains.