Abstract
Understanding the actual effect of stress distribution within longitudinally submerged-arc welded (SAWL) large-diameter pipe is essential for reliable and practical execution of sulfide stress cracking (SSC) tests. To qualify SAWL pipe as SSC resistant, SSC tests are typically performed on four-point bend specimens pre-stressed to a certain level within the elastic regime. For a detailed evaluation of stresses present in pipe and in four-point bend specimens, subjected to typical loads with respect to AYS and SMYS, the local stress at the unmachined part of the inner weld was simulated by a FEM approach indicating local plastic strain in the heat-affected zone (HAZ) of the longitudinal weld. In the same approach, the stress distribution was compared to that of laboratory-scale four-point bend specimens subjected to typical loads defined with respect to actual yield strength (AYS) as well as to stresses in a full pipe body pressurized up to 72 % of specified minimum yield strength (SMYS). It is shown that stresses found in four-point bend specimens when loaded up to 90 % of AYS exceed the stresses in a pressurized pipe by up to 60 % and that residual stresses have a beneficial effect. The findings from the FEM simulation were backed up by experimental internal loading of a full ring test specimen according to ISO 3845 to different stress levels up to 100 % of AYS. For a SAWL large-diameter pipe of grade X65 with a diameter of 864 mm and a wall thickness of 30.4 mm, all strain gauge positions referenced in the standard and additional locations throughout the ring specimen length and circumference were considered to gain detailed understanding of the overall stress distribution and to identify potential local overstressing in base material and along the longitudinal weld. The observed effects of local stress variation should be considered in a performance analysis when defining suitable stress levels for SSC laboratory tests, depending on the type of test, such as four-point bend or full ring test.
