Modeling Stress-Activated Creep at Axial Cracks in Pipelines

The failure of a pipeline that had just passed a proof-pressure test as it was being re-pressurized for its return to service (a so-called pressure reversal) reflects the stable growth due to stress-activated creep of a near-critical anomaly that had remained stable as the proof test ended. In the same way that stable growth of a near-critical anomaly can lead to a pressure reversal, stable tearing (cracking) can occur and remain stable at the pressure first imposed upon the pipeline’s return to service, and so pose concern for in-service failure. Ductile failures that are absent evidence of time-dependent degradation mechanisms, like corrosion, and show the traits of stable tearing have been termed time-delayed failures. As time passed, the reasons for time-delayed failures became clear, and criteria to prevent such failures through a pressure reduction were established. The advent of much tougher steels opened to the potential for crack initiation and stable tearing at service pressures under circumstances that differed from that for the early line pipe steels. The 2004 incident at Ghislenghien involving a modern high-toughness X70 pipeline raised the need to better understand how to manage time-delayed failures in such steels. This paper develops a model to quantify stable tearing and possible instability at axial part-through-wall defects as a function of the steel, the length and depth of the defect, and the operating pressure. The theoretical basis for this nonlinear fracture mechanics (NLFM) model is outlined first. Case-specific finite-element analysis were used to benchmark NLFM Handbook results, which extended the use of predictive technology developed previously for lower toughness steels. As before, this solution is recast for time-marching analysis that is coupled with isochronous stress-strain response and NLFM resistance curves. Finally, the model is used to make blind predictions of cracking and instability in step-load and hold testing, and found to be viable in that context. Companion papers at this conference present the details of related work.

Findings From an Investigation of Hydrotest Protocols

Experience has shown that in-service failures can occur in Electric Resistance Welded (ERW) line pipe shortly after a hydrotest as a part of line rehabilitation or in retesting to return to service. This can occur due to near-critical features that grew but remained in the line, or due to a pressure-reversal due to stable defect growth during the pressure test. The objective of this study was to improve hydrotesting protocols by developing and optimizing procedures for conducting hydrotests of pipe with ERW or Flash Welded (FW) seam defects, and validating the practical utility of the proposed procedures. This work has enhanced and adapted both failure and growth models to predict the behavior of defect shapes based on collapse and fracture theories, to assess idealized cold welds, hook cracks, and selective-seam weld corrosion (SSWC), and parametrically quantify differences and similarities in their response to increasing pressure. Based on insight from these analyses, this work has established a pressure-time sequence that incorporates the spike concept to expose near-critical defects to pressures that will cause them to fail and it has evaluated the sensitivity as a function of defect size with respect to time-dependent growth. In turn, insights from the outcomes of these analyses were used to identify potential hydrotest protocols.