The Use of Notched Compact-Type Specimens for Crack Initiation Design Rules in High-Temperature Water Environments

This report describes the techniques used and results obtained for fatigue crack initiation of notched compact-type (CT) tests of SA333 Gr.6 carbon steel in air or in autoclaves mostly at 288°C (550°F), 1.05 kg/mm² (1500 psi) oxygenated water conditions. The fatigue crack initiation life N_i was quantitatively determined from a cycle-by-cycle crack length measurement as the number of cycles of produce 7.6 × 10⁻² mm (0.003 in.) of growth from the notch root. Results for several parameters of the experimental program (including temperature, environment, oxygen content, frequency of cycling, notch radius, R-value, and ΔK) are reported. A majority of the tests were run in 288°C (550°F) water at 8 ppm O₂ with 5.1 × 10⁻² mm (0.002 in.) notch radii, and showed a severe degradation in initiation life, especially at a frequency of 0.00021 Hz (0.0125 cpm), compared with 288°C (550°F) air tests. On the other hand, at 0.021 Hz (1.25 cpm) and 8 ppm O₂, decreasing the temperature to 232°C (450°F) restored the life to that of air. Reducing the oxygen level to 0.2 ppm in 288°C (550°F) water was also beneficial. These latter two findings are preliminary because of limited testing.

The test results were interpreted in terms of conventional procedures in which a notch root pseudostress amplitude S_a was calculated for each test and compared with the ASME fatigue data curve for carbon steel. The quantity S_a was obtained from the applied stress intensity range (ΔK) and Neuber notch analysis procedures to account for elasticplastic notch root conditions. The worst-case notch concept was introduced to account for the observed maximum in the fatigue notch factor k_f with decreasing notch radii and for the environmental notch sensitivity. For the present study, the worst-case notch radius ρ was 0.165 mm (0.0065 in.).

A procedure is described for calculating S_a and N_i from crack growth data. This procedure assumed a crack to be equivalent to the worst-case notch and provided a good prediction of the ASME fatigue data curve from the ASME crack growth curves in air, both for carbon steel. By using this approach and a conservative threshold stress intensity value for fatigue crack growth in 288°C (550°F), 8 ppm O₂ water, a value of S_a was determined below which crack initiation will not occur in this environment, regardless of defect sharpness or frequency of loading. This value of S_a was consistent with the notched CT crack initiation results obtained.