![]() |
925
Industrial Drive South |
![]()
|
When
a part fails, what should you do? It might be more appropriate to ask, "what should you not do?" When in doubt, do nothing. Put another way: look but don't touch, because what may appear as "junk" to the untrained eye could well be valuable evidence to those skilled in failure analysis. |
![]() |
|
Getting
Started The greatest of care must be taken in the proper selection, preservation, and cleaning of fracture surfaces. Equal care must be given in the transportation of a fractured part for analysis. And make sure you send everything. If a bolt breaks, send both halves, the nut, and preferably the complete assembly to which the bolt was fastened. It's necessary that the entire item be examined whenever possible. When a fracture occurs, it seems natural to put the mating pieces back together. Whatever else you do, you must restrain that very natural tendency. Even in the best of circumstances, fracture surfaces are extremely delicate and can be easily damaged. It may appear to the naked eye that no damage would take place, but that's not the way a microscope would see it. Many microscopic examinations have been frustrated by careless repositioning of the parts, by careless packaging and shipping, and by inadequate protection from corrosion, including contact with fingers. The surface of the fracture must remain as uncontaminated as possible and protected from further damage. Let's assume that an on-site inspection by the failure analyst was determined to be unnecessary or impractical. Instead, you have been asked to ship the critical parts for inspection. At that point, you will undoubtedly be instructed on the specifics of protecting the components during removal and transit. You will probably be asked to wrap each component separately to assure that the damaged areas of each piece are thoroughly protected. You can imagine the damage that would occur if two metal components were simply tossed into a carton with no protective padding and allowed to bang against each other while being shipped. Corrugated cardboard, newspaper or bubble pack securely bound around larger components might be recommended - plastic wrap or a zip-lock bag for small parts. Each situation is different, however. It's far better to ask and be certain than to assume and make an error. To help
avoid damage from corrosion, handle the specimens as little as possible,
Try to keep all vital components out of the weather. A common recommendation
is to dry specimens with a jet of warm air and then pack them with a suitable
desiccant to absorb moisture. And if you suspect that the problem itself
was one of corrosion, never smear the part with grease in an attempt to
preserve it. Check with the analyst first, because appropriate, timely
action is necessary for the most accurate result in corrosion analysis.
|
|
Major
Losses
|
![]() SEM Photo of Aluminum Fracture |
![]() Note loss of detail after four light brushings |
Don't be in a hurry to clean up the mess. In the case of a major loss you can expect competing interests to be vying to affix blame. The wise route to follow would be to contact an unbiased failure analysis expert and bring her or him into the picture at the start, because the obvious answer may not be the correct answer. Was it a faulty beam that caused the collapse of that structure or could it have been a bad batch of concrete in the pier? A few snapshots aren't always going to give you the answer after the bulldozers have been through. It takes careful and thorough examination of all possibilities while the elements are still in place. Even in major cases, the principal rules still apply: look but don't touch, call an expert, and preserve the evidence. Just the Facts Please Discovering the cause of component failure is like unraveling a complex mystery. A large number of factors, often interrelated, must be understood to determine the cause of the primary failure. But as in finding the solution to any mystery, it is first necessary to get all of the facts out on the table in their original state of failure whenever possible. Knowing what not to do is just as important as knowing what to do. Failure Analysis and the SEM |
![]()