Detection of precursor wear debris in lubrication systems

On-line health monitoring of aircraft propulsion systems may realize substantial cost savings through implementation of condition-based maintenance programs. Currently, aircraft engine and gearbox oils are monitored using chip detectors that warn the pilot of excessive wear conditions. However, they can only detect large (>200 μm) ferrous metal particles. Oil samples are also taken for laboratory spectrographic analysis; however, this procedure is time-consuming and manpower intensive. Several new technologies have emerged. Inductive sensors can now detect both ferrous and non-ferrous metallic particles in the oil, down to about 100 μm in size. Vibration monitors have also been developed to detect damage conditions. We report on an alternative method using acoustics for detecting precursor wear debris particles as small as 3 μm. By monitoring the size and generation rate of these very small particles, wear trend analysis can predict accelerated wear conditions before significant or catastrophic damage occurs. The acoustic method works by insonifying the oil with a high-frequency acoustic impulse and analyzing the reflected signals. The detection algorithm discriminates between particles and entrained air bubbles on the basis of differences in acoustic signature. Next, the algorithm estimates particle size and computes a statistical history of the particle size distribution and generation rate which is used to determine the wear status of the engine or gearbox. This paper describes the operation of an acoustic sensor and presents test data acquired in a lubrication system simulator