Oil analysis has been used for a number of years to determine the condition of lubricated machinery, and to predict failures from excessive wear of lubricated parts of the machine. The principal technique in use up to the present time has been spectrometric oil analysis. This technique is based on the concept that as a part wears, minute quantities of wear material enter the oil and are dispersed throughout the whole lubricating system. A sudden increase in the amount of a metal in the oil indicates that some part made of that metal has started to wear.
The technique has been successful in several major applications, principally in the railroad industry on diesel locomotives and in connection with the monitoring of aircraft engines. The technique is widely used and has provided warnings of failures before they have resulted in catastrophes. Often the cause of the failure can be corrected before serious permanent damage is done to the machine. As an example, the ingestion of sand, usually caused by a failing air filter, will cause rapid wear in a machine, but if this condition is detected soon enough the air filter can be repaired and the machine saved.
Ferrographic oil analysis is a new technique which goes much further in that it provides information about the size distribution of the particles and also permits the direct analysis of the types of particles, so that the nature of the failure can be determined. For example, it may be determined that the problem is with rolling bearings, or that most likely a gearing system is involved. Further, innocuous causes of high metal content in the oil may be detected and corrected without the necessity of tearing down a complex piece of machinery. The technique is aimed at answering two questions: “Is the machine sick?”, and, if so, “What is the matter with the machine?”.
Studies over the last two years have shown that wear particles belong to unique classes and certain types of wear particles are associated with specific modes of wear. An analysis of the wear particle size distribution and the types of wear particles present in the lubricant permits the prediction of failure of a machine and the determination of the nature of the failure.