It is essential when beginning a root-cause failure analysis involving metallic materials to collect and maintain the as-found physical evidence properly prior to detailed examinations. This has two objectives. In all cases collecting and protecting the evidence in its undisturbed condition provides the best opportunity to derive meaningful good results from the subsequent analyses. Secondly, in a legal proceeding lack of attention to these matters can produce the accusation of spoilation of evidence. As a minimum this will cause presentation problems for the offending party in the suit. At worst such deficiencies by that party may cause the case to be thrown out.

Following are suggested steps for this process. It is understood that these guidelines assume idealized conditions and less desirable alternatives may be the unfortunate reality. However, best results can be obtained when these actions can be applied.

• The analyst should go to the scene of the failure as soon after the incident as feasible. While there he or she should thoroughly document all aspects of the failure site with multiple photographs. It is valuable to first record the failure in the overall context of other equipment and components showing relative locations and sizes. These macro photos can later provide evidence of possible interactions not considered initially. Next, take “close-up” photos of the specific failure and physical conditions in its immediate surrounding area. If the locations of the small-area photographs are not obvious the analyst should make and retain clear notes of what is shown in each photo by its number. Using a ruler or pen in a photo will indicate relative sizes. Always make more photos than you think you might need. Certain shots can be valuable later to confirm potentially important factors that may have been overlooked during the field exam.
• If the failed component of a larger equipment system is not easily separated in order to conveniently transport it for further examination, precautions should be followed in the removal process. There is a danger that heat input from a saw cut, or even worse by a cutting torch, too close to the failure will alter metallurgical properties of the metal. Removal by saw cutting is the preferred method. What is “too close” will be a judgment call dependent on several factors. However, in general make the removed section as large as possible and still permit convenient packaging and shipment.
• Never attempt to fit-together fractured parts when examining them. This can mechanically alter the condition of the features on the surfaces and thus destroy evidence critical to defining causes of the failure. In addition the fracture surfaces should never be touched with bare hands. Residual perspiration (that contains salt ions) or oils on the skin can alter the chemistry of naturally occurring deposits on these surfaces. Original deposits on fracture surfaces need to be retained there to determine if they played a role in the failure. These should not be contaminated by anything else.
• All collected components or sections should be wrapped in soft, clean cloths, e.g., dry cotton rags, to prevent damage during their transport. Fracture surfaces should be especially protected so they will not be damaged. When possible, it is also good practice to put the wrapped parts in zip-locked plastic bags to prevent exposure to moisture or various ambient gases during transport. Each sealed bag or wrapped bundle should be labeled and a written list retained by the analyst. This is for easy communications with the laboratory personnel when they receive the evidence.
• It is always most desirable to obtain an exemplar of the failed part. This is a part, which ideally is the same material, model or design as the failed part, that has or could have experienced the same service conditions but yet it did not fail. Various parallel analyses of the exemplar and the failed part may indicate deficiencies in the latter’s metal composition, strength, dimensions, fabrication details, etc. that caused or contributed to the failure. These comparisons can provide validation of conclusions later obtained.
• If corrosion is suspected to have had a role in the failure it is vital to know the chemistry of the liquid to which the failed part was exposed. Obviously such a liquid can have many variations. It may be fresh, brackish or seawater or it may be a chemical mix as is often found in chemical or electric power plants or oil processing facilities. The analyst needs to establish the actual composition of the liquid that was at the failure site not what it was assumed to be. This is especially important for complex mixtures subject to process upsets. Generally it is desirable to obtain one or more samples of the liquid directly at or slightly upstream of the failure site for later chemical analysis. The sample(s) should be stored in tightly sealed glass containers and protected against breakage for transport. Plastic containers should not be used for this purpose because ambient gases can diffuse through certain plastics. This could alter the pH of the liquid and provide misleading results.
• If one of the forms of mechanical wear is suspected to have played a role in the given failure and a lubricant was used obtaining a sample of that lube may be very useful in the analysis. This sample should be taken directly at the failure site or immediately downstream of that location. Specialized laboratory analyses can be completed to check for the performance and efficiency of the lube by determining if it contained water (which would greatly lower its effectiveness) or to find the type and number of metal particles present.

The competent failure analyst will consider the results of all appropriate examinations of the physical evidence plus all related information obtained from multiple sources. Proper initial collection of the available physical evidence and maintaining it for subsequent laboratory analyses are the first essential criteria for reaching meaningful conclusions.

Posted in: Uncategorized