It is well known that surface condition has a strong effect on fatigue life, and that most surfaces produced by conventional manufacturing operations such as machining and forging have poorer fatigue behaviour than polished surfaces commonly used for laboratory specimens. As yet, there are no reliable quantitative models to predict the behaviour of such surfaces; the problem is a multi‐parameter one, involving surface roughness, surface microstructure and residual stress. High‐cycle fatigue data was obtained for En19 steel, using four types of machined surface, produced by: polishing, grinding, milling and shaping. Residual stress was eliminated by heat treatment. Fatigue limit data were plotted as a function of roughness parameters using Kitagawa‐type diagrams, and compared to simple notch‐based and crack‐based models. It was found that, whilst both theories tended to be overly conservative, fracture mechanics approaches are useful for relatively low roughness, when the surfaces can be modelled as a series of short cracks. For higher roughness a notch‐based approach is appropriate.
No takes yet. Share an insight, caveat, or question.
Taylor et al. (1991) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: