Micromechanics and experimental methodology for defect and strutuctural integrity assessments of gas and oil transmission pipelines

Abstract

Predictive methodologies aimed at quantifying the impact of defects (e.g., cracks, blunt corrosion, inclusions, weld flaws) in oil and gas pipelines play a key role in fitness-for-service analyses including, for example, repair decisions and life-extension programs of onshore and offshore facilities. Conventional procedures used to assess the integrity of piping systems generally employ simplified failure criteria based upon a plastic collapse failure mechanism incorporating the tensile properties of the pipe material. These methods establish acceptance criteria for defects based on limited experimental data for low strength structural steels which do not necessarily reflect the actual failure mechanism (e.g., stable crack growth of the macroscopic defect prior to pipe collapse) nor do they address specific requirements for the high grade steels currently used.For high strength pipeline steels, the material failure (leakage or sudden rupture) is most often preceded by large amounts of slow, stable crack growth. Under sustained ductile tearing of the (macroscopic) crack-like defect, large increases in the load-carrying capacity of the structure, as characterized by J-R resistance curves (R-curves), are possible beyond the limits given by the crack driving force at the onset of crack growth (JIc). Simplified methods for defect assessment which utilize the often significant increases in toughness of these materiaIs during ductile crack growth were incorporated in the so-called R6, BS7910 and API 579 procedures. These methods rely on the direct application of R-curves measured using small, laboratory specimens to surface defects. However, laboratory testing of fracture specimens to measure resistance curves (J-R) consistently reveals a marked effect of absolute specimen size, geometry, reIative crack size (a /W) and loading mode (tension vs. bending) on R-curves. For the same material, deep-notch bend, SE(B), and compact tension, C(T), specimens yield low R-curves while shallow-notch SE(B), single-edge notch tension, SE(T), and middle-crack tension, M(T), specimens yield larger toughness values at similar amounts of crack growth. These geometry and loading mode effects on R-curves arise from the strong interaction between microstructural features of the material which govern the actual separation process and the loss of near-tip constraint in the crack front region due to large-scale yielding... (AU)