Crankshaft Failure and Why It May Happen Again

This case study involves two continuously cast steel crankshaft failures. Three parties performed their own failure analyses: (1) the engine manufacturer responsible for component design, specification, and application; (2) the steel supplier and forging supplier responsible for making the steel, forging the shape, and preliminary heat treatment; and (3) a supplier that provided induction hardening, finish machining, and inspection. An independent fourth party engineering firm was subsequently involved, but because each party had their own agenda, there was no agreement on the metallurgical failure cause and therefore no continued analysis to pin-down and eliminate the root cause. A classic case showing how we may be doomed to repeat our failures because sound engineering was not allowed to proceed. Failure analysis has long been a special field of study for metallurgical and materials engineers. Much of what a metallurgical engineer is taught is intended to develop his observational and reasoning skills so that he or she may understand the interrelationship between observable features and properties or performance. To many, the process of a metal-lurgical failure analysis is almost an art form since so much of the analysis may be based on the ability of the analyst to identify and recognize important features. The myth that failure analysis is an art rather than a science is seemingly supported when two failure analysts come up with differing conclusions about the same event. Twenty-five years ago, the first paragraph of the text Analysis of Metallurgical Failures stated, " The primary reasons for conducting an analysis of a metallurgical failure are to determine and describe the factors responsible for the failure of the component or structure. This determination may be motivated by either sound engineering practice or by legal considerations. " [1] The first sentence of that paragraph very correctly states that the reason is to determine the factors responsible for the failure, but unfortunately the metal-lurgical engineer doesn't always have access to the information necessary to assess all the factors that may contribute to the failure. Additionally, the engineer may not have the knowledge and resources to analyze all the potential factors. This is particularly true for an independent heat treat supplier, but can also be true for captive heat treaters. So instead of performing a complete failure analysis to determine the factors responsible for failure, the engineer may simply determine if the material met the customer's specifications. This approach has become even more …