|MadSci Network: Engineering|
Jon, I can appreciate your dilemma because I have "been there, done that". By that I mean dealing with the issue of variation. Two adages come to mind: "If you cannot find anything wrong with something, you are not looking close enough" and "One cannot measure quality, only the lack thereof". In this discussion, purity and cleanliness are synonymous. The lack of cleanliness is defined as the level (distribution and size) of foreign material within the steel measured by a standard method of detection. For your application, the presence of foreign material in the steel certainly affects fatigue life and several other mechanical characteristics. However, there are other properties that can affect premature failure. Among a few: hardness, grain size and chemistry. In order to determine the cleanliness of steel, one must know the specification the steel must meet to be acceptable. For example, a bearing made for an aircraft application must comply with a higher standard of cleanliness compared to a bearing made for a packaging machine. This will involve extensive melting and re-melting operations to remove impurities and also more extensive testing. It is no surprise that a bearing for the aircraft specification will cost more. When you perform your product function deployment, economics is a factor that should be evaluated. Lets talk about testing for a moment. There are two types of tests that are performed to evaluate cleanliness: Destructive - Samples are removed from the steel bar (Normally the ends of the bar), polished and examined with a microscope. The variable involved is the magnification used. Non-Destructive - One method is to scan the bar using an ultrasonic flaw detector. The variables involved are inspection procedure, transducer frequency, scan type and scan speed. The criteria for testing are documented in the purchase contract. The standards can include SAE, AISI, ASTM, NDT, customer requirements, internal manufacturing requirements or in the case of aircraft applications NASA standards. Material that does not meet the testing requirements is diverted or scrapped. Variation in the purity of the steel is affected mostly by the process performance during melting and re-melting operations. Process performance can be measured using statistical methods. A measure of confidence can be ascertained by knowing if the steel supplier maintains registration with ISO-9000. During an ISO audit, compliance to documented standards and procedures is evaluated. Evidence of product characteristic measurements, calibration of measuring equipment, training of manufacturing personnel are just some of the areas that are evaluated. So Jon, it becomes necessary to know something about the steel supplier to rate the quality of the raw material going into the bearing. This also applies to the bearing supplier too since purity is only one factor that affects reliability.
Try the links in the MadSci Library for more information on Engineering.