Sizing limitations of ultrasonic array images for non-sharp defects and their impact on structural integrity assessments

Existing structural integrity assessment procedures typically assume flaws to be infinitely sharp when they cannot be considered as local thinned areas. This assumption is often over-conservative, resulting in a pessimistic assessment of structural components and a significant underestimation of their margins of safety against fracture. One of the main challenges while adopting non-destructive evaluation techniques is distinguishing between sharp cracks (e.g., fatigue) and non-sharp defects and identifying the more severe ones. Towards this broader challenge, the present work aims to examine the sizing limitation and accuracy of ultrasonic array image-based techniques for non-sharp defects (surface breaking u-notches) and investigate how these measurements would affect the structural integrity assessment of components. Parametric numerical simulations and experimental measurements are performed to generate full-matrix capture datasets, which are then processed using the total focusing method to form an image. The image-based sizing approach is shown to perform efficiently for notch depths higher than the inspection wavelength (λ_L), i.e. as small as 0.2 mm, and semi-notch widths as small as 0.1 mm. The influence of ultrasonic measurements on structural integrity assessments is highlighted using different case studies in the context of non-sharp defects of fatigue and fracture strength estimates. For the cases under analysis, resolving non-sharp defects led up to 5× and 3× values of effective fracture toughness and fatigue strength, respectively. We have also seen that a 30 % uncertainty in semi-notch width sizing would result in a 30 % and 20 % error in fatigue strength and fracture toughness estimations, respectively.