Ferrography is a state-of-the-art predictive maintenance technique based on wear debris analysis. It provides a comprehensive non-intrusive evaluation of the health of lubricated components while the equipment is in running state. The wear debris (ferrous as well as non-ferrous), present in the lubricant sample are precipitated on a glass substrate which can be viewed through a powerful bichromatic microscope. Under a specified condition of magnification and lighting filters used for transmitted and reflected light the shape, size, concentration and colour of wear debris are observed. The four major findings from ferrography are wear mode, rate, severity and location. The identified wear modes include abrasion, impact, fatigue, erosion, corrosion, scuffing, severe sliding, etc. The concentration of wear debris indicates rate of wear. Size of debris indicates the severity of wear. The colour of the particles identifies the type of material which pinpoints the affected component(s).

Specialist skills are required for these complex observations and their interpretation to provide meaningful assessment of the health of the equipment and suggest corrective measures. These skills are encapsulated into an Expert System developed in-house and made available to the engineers/technicians doing the analysis. The user-friendly expert system guides the analyst through a detailed step-by-step procedure and records all the observations regarding the presence of different types of wear particles. The system then offers a diagnosis as to the cause of abnormality and also offers a set of corrective actions that should be taken, along with the time in which the actions should be taken to bring the equipment to a healthy state as soon as possible. Thus, one of the key benefits of the Expert System is that consistent expert decisions can be taken even by non- experts.

At Tata Steel, the Expert System has been very successfully used for monitoring the health of critical equipment like turbines, blowers, compressors, mill drives, gear boxes, diesel engines, etc. The paper also highlights the effectiveness of Expert System through a case study.