DNV-OS-F101 Appendix A provides procedures for the assessment of circumferential flaws located in subsea pipe girth welds using fracture mechanics methods, commonly referred to as engineering critical assessment (ECA). The purpose of the ECA approach is to provide critical flaw dimensions for given material properties and loading conditions in a conservative way. The results of the assessment are used to derive weld flaw acceptance (or weld repair) criteria to be used during pipeline installation. An ECA will typically consider flaws under installation and operation loading conditions, including fracture and fatigue crack growth (FCG) calculations. Internal and external surface-breaking flaws are assessed, along with embedded flaws.
DNV-OS-F101 provides guidance on the appropriate FCG law to be used for the assessment of each flaw type under operational loading. For internal surface flaws exposed to sour production fluids (i.e. containing H2S) FCG rates (FCGRs) are known to be significantly higher than in air and, in the absence of relevant published data, project-specific testing is commonly performed to quantify fatigue performance. The recommendation for the assessment of embedded flaws is to use an air curve, as long as it can be substantiated that the fatigue performance is not reduced due to the environment.
It has been demonstrated that the FCG behavior of C-Mn pipeline steels exposed to sour environments is dominated by bulk hydrogen charging effects, i.e. hydrogen charging by absorption from the exposed surfaces rather than the corrosion process at the crack tip. Therefore, it is expected that an embedded (or external) flaw in a sour pipeline will be located in steel containing absorbed hydrogen.
This paper describes the results of an investigation aimed at understanding and quantifying the FCG behavior of embedded flaws in sour pipelines. For the purposes of this work, an embedded flaw refers to a crack propagating in hydrogen charged material but whose crack tip is not directly exposed to the sour environment.
Hydrogen diffusion modelling and simulation studies were performed to predict the through wall hydrogen concentration in standard fracture mechanics specimens based on sour environmental conditions. Two novel test methods were developed to accurately measure FCGRs in hydrogen charged steel, one for single edge notched bend (SENB) specimens and one for compact tension (CT) specimens. FCGR tests were carried out using both methods. The FCGRs measured in hydrogen charged API 5L grade X65 pipeline steel were significantly higher than in air. In some cases, the observed FCGRs in hydrogen charged steel were higher than for specimens fully immersed in the sour environment. This is believed to be due to reduced environmental crack closure/blunting effects; the steel is charged with hydrogen, but there is no active corrosion process occurring inside the crack.
The results of the present study indicate that the use of an air FCG curve for embedded (or external) flaws located in hydrogen charged steel may be non-conservative. Further work is required to establish the relationship between FCGR and hydrogen concentration in steel and to evaluate the implications for pipeline ECA calculations.