114

Nima Parsibenehkohal1, Rachel Brossman1, Ron Thompson2, Richard Kania3, Guillermo Solano2, USA.2, Canada.3

1PBF Energy, Cerritos, USA. 2Novitech, Toronto, Canada. 3KanEnergyPartners, Calgary, Canada

This paper summarizes the testing of the magnetic based in-line inspection (ILI) systems for the detection and sizing of circumferential and off-axis stress corrosion cracking (CSCC) and crack like and metal loss features. The objective of the testing program was to determine the optimal testing setups, types of defects and manufacturing methods and their applications. Pipe sections utilized for testing contained synthetic and man-made crack-like features and metal-loss features as well as natural defects from cutouts, which were then used to test and validate the tool specification to demonstrate compliance with API 1163.

During the evaluation and development of the ILI system for detecting CSCC, the use of high sensor density and high sampling rates from the combined AMFL and CMFL modules showed signal changes that were potentially attributable to the variations in material permeability.

During the testing program it was noted that the detection, discrimination, sizing and repeatability of the synthetic features were less than optimal, due to the nature of the manufacturing process, which altered pipe magnetic properties and permeability, negatively affecting the AMFL and CMFL tool capabilities. The permeability of the pipe when these defects were made may not be consistent with natural corrosion, or real-life SCC further hindering the analysis process, since the software algorithms were developed and trained on naturally occurring CSCC and off-axis cracks. In contrast, several live runs reported CSCC within severe corrosion proving the tool capability in real live runs.   This is further supported by PBF findings of real world CSCC and metal loss features correlating accurately to the values predicted in the ILI report.

This study includes pull test inspection results for NPS 6, 8 and 10 pipe segments with artificial features and corresponding ILI system responses as well as examples of real-life CSCC and metal loss findings for comparison.

Read/Download PaperGradeAvailable after the event