Assessment of Significance of Embedded Defects Caused by Cold Cracking in Welded Joints

Abstract

Prevention of cleavage fracture is one of the most significant problems in securing the structural integrity of welded structures. To simulate the fracture behavior and plastic constraint condition of structural components, wide-plate tensile tests have been conducted for experimental fracture assessments. Artificial surface flaws, through-wall flaws, and fatigue pre-cracks are generally applied. Embedded flaws from cold cracking should be considered in an integrity assessment of a structure. Nevertheless, experimental fracture evaluations based on wide-plate tensile tests using welded joints with embedded flaws from cold cracking have never been conducted, because a method for intentionally preparing a welded joint with a cold-cracking embedded flaw has not been established yet. In this study, we develop a means of manufacturing joints with cold cracking embedded flaws, based on the y-groove cracking test, and perform a series of cracking tests. We determine that deeper flaws crack more stably with more heat and more diffusible hydrogen. In addition, no relationship is apparent between the welding conditions (heat input/diffusible hydrogen), and the relative position of the embedded flaw to the fusion line in the cross-section. Accordingly, we establish a technique for intentionally and stably preparing a welded joint with a cold-cracking embedded flaw of sufficient height. We perform a wide-plate tensile test for the welded joint with the embedded flaw using the established technique to evaluate the fracture load and fracture behavior. The embedded flaw is found to be nearly rectangular and near the fusion line in the heat-affected zone. A fracture-origin survey of the wide-plate tensile specimen reveals that the cleavage fracture occurred from a local brittle zone at the tip of the introduced embedded flaw. Consequently, it is confirmed that this test meets the requirements for application to an evaluation on the significance of defects. Finally, we perform fracture evaluation by comparing the fracture stress with allowable stress and verify the applicability of the evaluation method using a failure assessment diagram (FAD) of the cold-cracking embedded flaw. The fracture load is found to be higher than the allowable stress; therefore, fracture will not occur even if similar embedded flaws exist in the structures. Furthermore, we confirm that a BS7910-based FAD evaluation can be applied to cold-cracking embedded flaws with a specific safety margin. Although the load observed in the embedded flaw near the cross-section exceeds the plastic collapse load, the fracture load estimated by the FAD evaluation is less than the plastic collapse load. This difference can be explained by an appropriate correction of critical crack-tip opening displacement and through-thickness residual stress distribution.