|
|
Failure Analysis of ERW Pipe Burst

|
|
|
|
The catastrophic burst of a piece of LS-125 ERW pipe was attributed to weld defects and low toughness along the fusion line. The fracture initiated at an ERW weld defect on the pipe inner surface. The microstructure was normal for quenched and tempered low alloy steel. Transverse mechanical properties across the ERW weld fusion line showed acceptable yield and tensile strengths but poor ductility.
|
My Saved ArticleThe catastrophic burst of a piece of LS-125 ERW pipe was attributed to weld defects and low toughness along the fusion line. The fracture initiated at an ERW weld defect on the pipe inner surface. The microstructure was normal for quenched and tempered low alloy steel. Transverse mechanical properties across the ERW weld fusion line showed acceptable yield and tensile strengths but poor ductility. In addition to the mating sections from the center of the fracture, three other sections of pipe were submitted for examination (Figure 3). The smallest section contained unfractured pipe which had survived the hydrostatic tests and yet showed indications on ultrasonic testing (Figure 4). Another section contained in the last six inches of the fracture and approximately two additional feet of unfracture pipe. Ultrasonic testing showed rejectable defects in the first foot of material beyond the end of the fracture. The third section had survived the hydrostatic test and showed no rejectable defects on the ultrasonic testing (Figure 5). The profile through the initial flaw was flat and non-crystallographic, showing no evidence of welding or subsequent fracture. A thin surface scale was observed on this area. Below a depth of approximately 0.1 inch from the surface, fracture surface features began (Figure 6). In profile the fracture was non-branching with a mixture of transgranular propagation and plasticity on a microscopic scale. Figure 7 shows a photomicrograph from the area rejected by ultrasonic inspection. A sharp, short mill discontinuity was present near the fusion line. The fusion line itself was visible only as a light-etching band less than 0.4 thousandths of an inch wide. Otherwise the structure through the weld seam was entirely uniform and typical of properly quenched and tempered low alloy steel (Figure 8). This structure was typical of all sections examined except section F, through the tip of the fracture. Microhardness tests were performed on several of the metallographic sections. The base material well away from the weld showed an average microhardness of KHN200g 331, which is approximately equivalent to Rockwell C 30. There were a few areas of apparent decarburization on the inner surfaces which showed hardnesses as low as KHN200g 220. Otherwise, hardness throughout the cross-sections showed little variation except in the light-etching fusion zones. There hardness averaged KHN200g 257, which is approximately RC18. This low fusion line hardness was uniform throughout the cross-sections tested, varying little from I.D. to mid-wall. On Section F, through the tip of the fracture, the crack ran precisely down the center of the fusion line in the center of the wall thickness. A thin (0.06 inch) bridge of unfractured steel still held at inner and outer surfaces. However, a shear lip had started to form on the inner surface, indicating that complete fracture of the remaining ligaments was imminent (Figure 9). The fracture profile was irregular and showed a number of voids and inclusions from the original weld (Figure 10). Unlike the other sections examined, section F did not have low fusion line hardness. In the two small ligaments remaining, fusion line hardness ranged from KHN200g 308-334, approximately Rc 27-30. The fracture in specimen 2 was in the base metal just outside the weld. The fractures in specimens 1 and 3 were both in the weld. | | | |