High-cycle and low-cycle fatigue tests were conducted on SA533-B1 steels with four levels of sulfur content at room temperature. For high-cycle fatigue (HCF) tests, the fatigue limit was observed around the yield stress (650 MPa) and showed little or no dependence on sulfur content. The lower the applied maximum stress (around the fatigue limit), the more scattered the fatigue life data. From the low-cycle fatigue (LCF) results, an excellent correlation between the SWT (Smith, Watson, and Topper) stress parameter versus reversals to failure was obtained. Fatigue life could be predicted from either the SWT parameter or Coffin-Manson equation for a specimen under constant strain amplitude loading conditions. A comparison of scanning-electron-microscopy (SEM) fractographic features of HCF- and LCF-tested specimens has been made. Obscured crack-initiation sites, wider striation spacing, and quasi-cleavage were observed to be characteristic of LCF, in contrast to the fractographic features of HCF-tested specimens.
To predict the fatigue residual life of a plate containing surface cracks, the associated stress intensity factor must be known. In this work, the stress intensity factor for surface-cracked plates was calculated by a three-dimensional finite-element method (FEM). As a verification practice, fatigue tests were conducted on surface-cracked specimens of SA533-B1 steels to study the crack growth behavior under a constant amplitude loading condition. The results show that it is feasible to calibrate the calculated stress intensity factor of a surface crack by the experimental backtracking method. From both FEM and experimental results, the empirically obtained stress intensity factors for surface cracks are reasonably with the numerical solutions with a maximum deviation of 21 % in the range of stress intensity studied. The crack-closure effect is proposed to account for discrepancies between the predicted and measured stress intensities.