Abstract
The majority of fatigue-crack growth data for specimens containing surface-hardened layers is in the form of S/N curves. In hardened specimens, however, many factors, such as residual stress levels in ‘case’ and ‘core’, relaxation of these stresses by specimen distortion during hardening, and microstructural and (in some cases) chemical variations, affect the processes of initiation and growth. In addition, it may happen that the effect of these factors on initiation and on propagation of a fatigue crack is not the same. The present paper describes experiments involving the monitoring of short fatigue cracks which grow from machined notches through hardened surface layers of different kinds, and examines the effects of residual stresses as the crack extends into the ‘core’ material. Observations of crack-growth rates in the important situation of an asymmetrically hardened component show the effects of specimen distortion in relaxing residual stresses, and in some situations permit an estimate to be made of the variation of residual stress throughout the hardened layer. In addition, there is described a means of correcting for the effect of notch shape on the growth rate of a short fatigue crack.
Get full access to this article
View all access options for this article.