Fire Protection System Coupling Bolt Failure
A bolt in a fire protection system coupling failed within 24 hours of installation resulting in flooding of a warehouse.
The failed bolt and coupling were examined in the laboratory. The markings on the bolt indicated that the bolt was manufactured to the requirements of SAE J429 grade 8.2.
The bolt was electroplated with zinc.
Visual examination of the head of the bolt showed that the bolt had been subjected to installation forces in excess of the manufacturer's recommendations.
SEM examination of the fracture surface revealed areas of intergranular fracture on approximately 43% of the cross section. The remainder of the surface had features of microvoid coalescence, indicative of tensile overload. The presence of intergranular cracking and the short service history of the bolt were indications of hydrogen embrittlement.
SEM image of fracture surface showing Brittle Region (Intergranular) and Ductile Region
SEM image of fracture surface showing intergranular features. The "rock candy" like structure is indicative of hydrogen embrittlement.
SEM image of fracture surface showing ductile overload features. The structure is microvoid coalescence with some post fracture smearing of the surface.
The hydrogen induced delayed fracture of the subject bolt occurred as a result of residual hydrogen in the material and tensile stress. The installation of the bolt loaded the bolt material in tension. The hydrogen migrated to high stress regions at the root of the thread of the bolt. As the hydrogen concentrated at the high stress regions, grain boundary decohesion occurred, resulting in the formation of an intergranular crack. The tip of the crack became the new high stress region and the process repeated itself, allowing the crack to propagate and reducing the intact cross-sectional area. This continued until approximately 43 percent of the cross-sectional area had cracked intergranularly, as estimated from the SEM inspection of the fracture surface. At that point, the remaining intact cross-sectional area was insufficient to carry the applied load and the bolt failed.
The source of the hydrogen was traced to the zinc plating process. The plating vendor was aware the plating process could introduce hydrogen into the bolt but did not perform the required post plating baking procedure to remove the hydrogen.
The conclusion of the failure analysis was that the bolt failed as a result of hydrogen embrittlement. The bolt was defective in that it contained excessive amounts of hydrogen which resulting in a delayed fracture of the bolt shortly after installation. The high stresses introduced by tightening the bolt in excess of the manufacturer’s instructions exacerbated the rate of fracture, but would not have caused a failure without the presence of excessive hydrogen in the bolt material. Properly manufactured fasteners should not contain sufficient residual hydrogen to cause embrittlement and delayed fracture.