September, 2007
by Vlad Bacalu, ATS Product Manager
Predictive Maintenance Technologies

Predictive maintenance (PdM) techniques help determine the condition of equipment in order to predict when and what type of maintenance should be performed. This approach offers cost savings over routine or time-based preventive maintenance because tasks are performed only when needed. There are many predictive maintenance technologies, and some have become standard in many industries. Those standard technologies include: thermography assessments, vibration monitoring, oil analysis combined with wear particle analysis, ultrasound measurements, and motor current analysis.


Early Detection
Predictive technologies enable the gathering of equipment health data much like a doctor does a physical, blood analysis, and X ray on a human. These actions do not fix the problems, but can find them in a very early stage, not easily detected by visual or other basic inspection of the equipment. In short, monitoring the equipment health and making maintenance decisions based on these conditions can prevent catastrophic failures and extensive downtime.

Thermography or thermal imaging creates a visual picture of temperature. This tool allows a technician to find temperature differences, usually abnormal hot spots, typically associated with problems due to high electrical resistance, or excessive mechanical friction. By using thermography assessments technicians perform maintenance on only the components that need attention, and rectify the problems before more costly failures occur.

Ultrasonic measurements are used to find process leaks (air, gas, steam, and vacuum), detect low speed bearing defects, find electrical arching problems (Corona effect), and to determine the proper amount of lubricant applied to high-speed bearings. The application of this tool in leak detection is based on the fact that most leakage problems produce a range of sound. The sound, when properly detected and measured, provides data on the location and severity of the leak.


Know Your Oil
Oil analysis is one of the key components of predictive maintenance and includes the following: lubricating and hydraulic oil analysis, spectrochemical analysis, and wear particle analysis. Oil analysis determines the condition of a lubricant. Spectrochemical and wear particle analysis determine the condition of equipment based on concentration and composition of wear particles.

Oil analysis is a proven method to predict and prevent certain maintenance problems before they occur. It also helps to insure that lubricants are performing as expected. As a predictive maintenance tool, oil and spectrochemical analysis are used to schedule oil change intervals based on the actual condition of the oil, or the analysis data can be used for maintenance decisions. For example, monitoring the amount of trace metals in successive oil samples will indicate wear patterns of parts in the equipment and will provide an indication of impending equipment failure.

Vibration analysis is used to find problems such as misalignment, out of balance conditions, and bearings defects by taking and recording periodic readings. Maintenance personnel can compare these readings to a baseline or trend and make appropriate decisions. When wear reaches a certain level, the bearing is scheduled for replacement before it fails. This reduces the amount of reactive maintenance and insures that replacement occurs with minimum impact on the production schedule.

Motor Current Analysis is a technique for diagnosing problems in electrical motors by monitoring input electrical signal. This technique can diagnose problems such as broken rotor bars, abnormal air gaps, and shorted turns in low voltage stator windings. Quite often, motor current analysis is used in combination with vibration analysis.


Finding Failure
In conclusion, regardless of which technique is used, the ultimate goal is to identify imminent failures and make a prognosis of the remaining useful life of analyzed component. This enables corrective maintenance action to be undertaken on the identified failing component at a convenient time allowing better planning of maintenance work. Many predictive techniques can be used to check for the same problem. For example, a problem indicated by an oil sample could also be checked by vibration monitoring, or thermography. Knowing how to use the tools and where to apply them is important to a successful program. The impact on the business is what creates the return on investment, so choosing the right predictive technique is critical.