Oil and gas transmission pipelines have a good safety record; however, like any engineering structure, pipelines do occasionally fail. The main causes of pipeline failures in North America and Europe are defects, such as damage due to external interference, corrosion defects, or material defects. Consequently, the pipeline industry developed its own ‘pipeline-specific’ methods for assessing the significance of these defects. These pipeline-specific methods have their origins in fracture mechanics, but the complexity of the underlying failure mechanisms was addressed through empiricism. Over the intervening years, more accurate ‘pipeline-specific’ methods have been developed that better model the underlying failure mechanisms. The toughness of line pipe steels is typically characterised in terms of the Charpy V-notch impact energy. This is a qualitative measure of toughness. The pipeline-specific methods use empirical correlations between Charpy V-notch impact energy and fracture toughness.
In parallel to the development of the pipeline-specific methods, fracture mechanics has been generalised in standards such as BSI 7910: 2005 and API 579-1/ASME FFS-1, using the ‘failure assessment diagram’ (FAD). The pipeline-specific methods and the methods in these general (‘generic’) standards have common roots.
The pipeline industry has used its pipeline-specific methods for 50 years, as there was little need to use the generic methods. This was because most of the defects detected in pipelines were corrosion or damage (e.g. dents), where pipeline-specific methods were well-researched and validated.
The increasing sophistication of in-line inspection methods (‘intelligent pigs’) in the pipeline industry mean that cracks in pipelines are now easily detected. Additionally, recent failures caused by crack-like defects (e.g. the San Bruno failure in the USA in 2010), mean that the industry needs guidance on how to assess these crack-like defects. There is little or no guidance on pipeline-specific methods for crack-like defects. Therefore, the industry has adopted the generic methods, without any evaluation of their relevance or accuracy.
The similarities and differences between the generic and pipeline-specific methods for crack-like defects in pipelines are illustrated in this paper through comparison with published full-scale test data for pipes containing notches and cracks. The significance of correlations between Charpy V-notch impact energy and fracture toughness is highlighted.
In conclusion, a commentary is given on how and when to take advantage of either approach. The critical gaps in the existing methods are also identified.