Analytical fatigue life prediction of shot peened AA 7050-T7451. (January 2019)

• Record Type: Journal Article
• Authors: Bianchetti, C.; Delbergue, D.; Bocher, Philippe; Lévesque, M.; Brochu, M.
• Is Part Of: International journal of fatigue. Volume 118(2019)
• Page Start: 271
• Page End: 281
• Publication Date: 2019-01
• Subjects: Fatigue life prediction -- Fatigue crack propagation -- 7050-T7451 aluminum alloy -- Residual stress -- Relaxation -- Navarro-Rios model -- Shot peening -- Cyclic yield -- Cold work -- Diffraction (XRD); Materials -- Fatigue -- Periodicals; Materials -- Fatigue; Periodicals; 620.1122
• DOI: 10.1016/j.ijfatigue.2018.07.007
• Languages: English
• ISSNs: 0142-1123

Abstract

Highlights: Fatigue life of shot peened samples were predicted with the Navarro-Rios model. The original (monotonic) and the new (cyclic) models were compared. The cyclic damage model predicted fatigue life more accurately. Perspectives for future work were proposed in the discussion section. Abstract: Fatigue lives of as-polished and shot peened specimens were predicted using the original Navarro-Rios model (monotonic damage model) and a more recent version of the same model (cyclic damage model). Crack closure, residual stress relaxation, cold work and cyclic yield stress were experimentally characterized and implemented within the cyclic damage model. Two shot peening conditions were studied (S230 and Z425 shots at 8A and 100% coverage). The in-depth profiles of the relaxed residual stress and cold work were experimentally characterized and implemented within both versions of the model. Two stress amplitudes leading to high cycle fatigue (HCF) and low cycle fatigue (LCF) were studied at a stress ratio of 0.1. Predictions of the cyclic damage model were globally closer to experimental fatigue lives than those predicted by the monotonic damage model. The monotonic damage model significantly overestimated fatigue life of shot peened samples in LCF. A detailed list of the used parameters and their characterization is provided. Finally, potential improvements and uses of the model for future studies are proposed.