Most of the previous research on thickness effect was carried out on relatively thicker walled joints typically used in the offshore industry and in ship structures. Results of previous research showing the effect of plate thickness on the fatigue strength of welded plate joints will be summarised. Fatigue design guidelines from different standards will be reviewed. Recent research performed on welded thin-walled joints made up of sections of thicknesses less than 4mm which show that the conventional concept of thickness effect may result in unsafe design of structures composed of thin-walled tubular joints under fatigue load will be discussed.
The thickness effect is the phenomenon in fatigue of welded connections where the fatigue strength decreases as the wall thickness of the member in a welded joint where fatigue cracks initiate and propagate to final fracture or failure increases.
Gurney (1989) pointed out that thickness effect could be demonstrated using both fracture mechanics theory and experimental work. This had led to the introduction of a thickness correction factor in the revised version of the UK Department of Energy Guidance Notes in 1984. Gurney (1989) also noted that a lot earlier than the introduction of the thickness effect on fatigue of welded connections, Phillips and Heywood (1951) had demonstrated the size dependence of fatigue strength of unwelded specimens. Gurney (1989) also pointed out that it had long been known that plate thickness was likely to be a relevant variable for fatigue strength under bending stresses, because the stress gradient through the thinner specimen would be steeper and therefore less damaging than that in thicker specimens. Gurney (1977) showed with the use of fracture mechanics theory, that fatigue strength of welded joints could be affected by plate thickness even when they were subjected to axial loading.
