Defect tolerance assessments are carried out to support the demonstration of structural integrity for high integrity components such as nuclear reactor pressure vessels. These assessments often consider surface-breaking defects and assess Stress Intensity Factors (SIFs) at both the surface and deepest points. This can be problematic when there is a high stress at the surface, for example due to the stress concentration at the root of a screw thread. In the past this has led to the development of complex and costly 3D finite element analyses to calculate more accurate SIFs, and still resulting in small apparent limiting defect sizes based on initiation at the surface point.
Analysis has been carried out along with supporting materials testing, to demonstrate that the increased SIF at the surface point is offset by a reduction in crack-tip constraint, such that the material exhibits a higher apparent fracture toughness. This enables a more simplistic assessment which reduces the effective SIF at the surface such that only the SIF at the deepest point needs to be considered. This then leads to larger calculated limiting defect sizes. This in turn leads to a more robust demonstration of structural integrity, as the limiting defect sizes are consistent with the capability of non-destructive examination techniques.
The high SIF at the surface location, and the concomitant reduction in crack-tip constraint, meant that it was not possible to demonstrate the material response with conventional tests, such as those using shallow-notched bend specimens. Instead it was necessary to develop modified specimens in which semielliptical defects were introduced into a geometry which replicated the notch acuity at the root of a screw thread. These feature tests were used to demonstrate the principle, prior to testing with more conventional specimens to fit more accurately the parameters required to represent the material response in a defect tolerance assessment.
Margins in defect tolerance assessments are usually measured against the initiation of tearing, even though the final failure for the material may occur at a higher load following stable crack extension. This work measured and assessed the benefit of reduced crack-tip constraint on both the point of initiation and on the development of the tearing resistance curve. This demonstrated that the effect of constraint was valid with tearing for this material and that there was additional margin available beyond the onset of tearing. The feature test geometry also provided evidence of the tearing behaviour at the surface and deepest points of a surrogate component under representative loading.
This paper provides an overview of the range of tests performed and the post-test interpretation performed in order to provide the R6 α and k constraint parameters.