A Practical Approach to Interpretation and Consideration of Tool Tolerance for In-Situ Strength Testing

Abstract This paper outlines a methodology for employing in-situ test results, considering measurement tolerances, to reliably validate material grades as part of the Material Verification Program for MAOP reconfirmation. It addresses the recent amendments in 49 CFR 192 and the accompanying FAQs by the Pipeline and Hazardous Materials Safety Administration (PHMSA). These revisions mandate material testing, as stipulated in 49 CFR 192.607, in certain scenarios where U.S. pipeline operators do not have complete, traceable, and verifiable records of physical pipeline characteristics, including diameter, wall thickness, seam type, and grade. The proposed methodology, which aligns with the requirements of 49 CFR 192.607(d), ensures conservative accounting for measurement inaccuracies and uncertainties, providing a practical approach to comply with regulatory standards to verify pipe grade. To meet the material testing requirement, pipeline operators are using recently developed in-situ and nondestructive testing in lieu of destructive methods whenever possible. The results provide a measurement of the yield strength of the steel and may then be used as an input to determine pipe grade. Pipe grade is considered one of the most challenging material attributes to determine using non-destructive methods for strength measurement due to measurement variability and overlap of actual material strengths between different grades that is allowable in API Specification 5L. The purpose of the material verification as related to §192.624 MAOP reconfirmation is not to identify the actual yield strength of the pipe but instead to use estimates of yield strength based on established test methods to identify the grade. Pipeline engineering methods rely upon conservative models (i.e., equations) for design and defect assessment. Inputs to these models (e.g., yield strength, wall thickness, diameter) have known variability that is already accounted for by the inherent conservatism of the models. The conservatism of these models is discussed to demonstrate that the precise identification of yield strength is not necessary if the pipe can be demonstrated to be consistent with an expected grade.