Laboratory coupon joints for fatigue debonding tests usually have narrow width and a through-width initial disbond. However, realistic structural joints are much wider and may contain process-induced defects and accidental damage; both are much smaller than the joint width. Small and discrete damage may behave differently from the idealised through-width disbond crack. This has brought a question on whether the laboratory coupon joint can accurately represent the fatigue behaviour of wider structural joints. This paper presents an experimental and numerical study of fatigue behaviour of a wide bonded lap joint with a process-induced defect of semi-circular shape. Fatigue debonding propagation was monitored by ultrasound inspection. Fatigue life was predicted using a normalised strain energy release rate parameter calculated by finite element method, and the adhesive material fatigue crack growth rate data measured under single and mixed mode conditions. Simulation of process-induced defect and validation by experiments have brought a better understanding of fatigue debonding behaviour in wide joints containing realistic damage. Suggestions are given for fatigue fracture tests of bonded joints.
|Journal||International Journal of Fatigue|
|Publication status||Published - 17 Mar 2019|