Failure Analysis Takes Time
by Jesse A. Grantham, Welding Forensic Expert
Published: January 10, 2007
 When I hold a broken metal part in my hands, or when I am standing before a broken metal structure, I talk with the part and I listen to what it has to tell me. I ask questions, I look, I turn it over, upside down and around, I listen, I taste, and I smell. I use my senses, education, experiences, and knowledge to find a one-on-one communion with the part. I ask where it hurts. Why, when, how, who and what happened to cause the break-down. I process data obtained from my observations and intuitions. Later, I can use enhanced techniques.
The term “failure” is relative. This definition is important and worth a lot of money. Similarly, the extent of “my expert analysis” is relative to how many resources are used to investigate and describe the “failure”.
Mediocre results often come from cursory reviews. What is the expectation of my analysis and report? Often a 200 word report merely presents the 1 or 2 possible causes, while listing other topics of interest for future examinations. I cannot accomplish a worth end by unworthy means.
The topic, the investigator, the author and the audience are a few of the key items that constitute a successful “failure analysis” that accurately tells the truth. If I don’t know, I’ll say that. All this effort takes time.
It takes “quality time’ one-on-one between the part and the inquirer, me. In business, time is money. Application of resources and forensic investigation take time. Presentation of results takes time. Always tell the truth based on the available evidence. Knowledge is power. It takes time to obtain answers and present them. The pen is mightier than the sword. Writing a report takes time.
Two hunters were alone in the great north wilderness after the small float plane departed to pick-up the rest of their gear and to return in an hour. It was a sunny day and the hunters are “skipping stones” on the lake and walking barefoot at the lake’s edge. A huge, grizzly bear appears from the lakeside woods and begins to chase both of them. The first hunter runs, then, suddenly stops to  put on his boots. The second hunter yells back to forget the boots, just run from the bear. The first hunter replies that he only needs to outrun the second hunter.
Failure is a general term used to imply that a part in service (1) has become completely inoperable, (2) is still operable but is incapable of satisfactorily performing its intended function, or (3) has deteriorated seriously, to the point that it has become unreliable or unsafe for continued use.
Analysis is a term that means to break up, separation of a whole into its component parts. This refers to an examination of a complex part, its elements and their relations. It is considered a method of philosophy to resolve complex expressions into simpler or more basic ones. It is a clarification of an expression by elucidation of its use in discourse.
Design – user needs. Design is the leader in part development. When designing any part, it is important to consider addressing all the needs contained in its intended use and anticipated service life. Once the design satisfies those purposes, then other aspects are determined. The design must be reasonable and allow other crafts to realize what is expected, required and then comply with their available resources to produce the part. Design must comply with the codes and contractual commitments agreed to with the final owner of the part.
Metallurgy - is the science and technology of metals. There are thousands of metal alloys to consider. Only a few types of metals are suitable for the intended design. Only a fewer types of metals are readily available at a reasonable cost with the properties needed for the design. The metals may be heat treated to modify the grain structure. Metals change when subjected to the effects of welding temperatures. Confirming the selection of the right metal for the intended purpose is a fundamental choice in the process of analyzing a failure.
Processes - there are many welding and joining processes readily available for welding or joining metal parts. It is a fundamental consideration to select processes that are readily available, serviceable and that there are technical personnel qualified to successfully operate the process, repeatedly. Much information is available about welding and joining processes. The design must know fundamental aspects of these processes and how to calculate strength of welds. When the design specifies difficult welds on a part, the cost increases and causes many inspection problems for fabrication or manufacturing consistency. Inconsistency is a good starting place in failure analysis.
Compliance with design, sometimes called “quality” and is more likely than not consistently found in welded parts. There are many reasons for non-compliance of a part with the code or contract. When performing failure analysis, quality variations in parts are considered.
Conclusion: When I perform failure analysis, I “talk” with the parts. I let the parts communicate to me what happened and tell me what they experienced when trying to meet the intent of the designer. Failure analysis is also the realization that the process of continuous improvement may have achieved its goal, or worn-out, or that the part failed unexpectedly due to unforeseen circumstances, or perhaps prematurely, or just the part just missed the “mark”. This is what takes time to determine when I perform failure analysis. -- Dr. Jesse A. Grantham
|
