By William Kruger
Why Predictive Maintenance?
Predictive Maintenance (PdM) resulted from studies that determined that Preventive Maintenance (PM) programs are only efiec-tive for the 11 percent of machine failures that are age related. This means approximately 89 percent of machine failures are random in nature and time-based maintenance programs are inefiective on these random types of failures.
There are many different PdM technologies, but the most successful programs use multiple technologies to provide the most information and consequently the highest probability of identifying a machine with a developing problem.
Some of the most common PdM technologies are machinery vibration analysis (MVA), infrared thermography, ultrasonics, oil anaysis, motor circuit analysis (MCA) and electrical signature analysis (ESA). Common characteristic among the most successful PdM technologies are that they are easy to perform and provide non-destructive, repeatable measurements.
However, regardless of the technology being used, the most successful programs recognize that there are three phases in a successful predictive maintenance program:
Following is a brief review of the three phases of successful PdM programs.
The detection phase is the most critical phase and the basis of most successful PdM programs. The main purpose of the detection phase is to identify “bad” machines or conditions that can lead to future machine failure. “Bad” machines are machines that are in a deteriorating condition.
Accurately determining the condition of the machine or more completely defining the cause of the change in the machine’s condition is the main purpose of the analysis phase. The analysis phase involves taking additional or perhaps even different types or more in-depth data than the detection phase. This additional data may require more specialized techniques or technologies. It may require testing at different operating conditions or using completely different technologies.
The main purpose of the correction phase is to determine the correct action based on the machine’s condition change. This involves taking the action necessary to correct and eliminate the problem triggered by the change in the machine’s condition. Additionally, the correction phase should verify that the corrective action did actually fix the problem(s). Alternatively, plant operations may dictate that the best action may be to simply continue monitoring at reduced test intervals.
Selecting the machines that are going to provide the biggest payback from a PdM program seems obvious by classifying machines either by size or application. Most plants that have embraced predictive maintenance classify machines based on their application.
Plants that classify machines based on application define critical machines as machines that are very critical to the plant’s operation. In other words, if the machine shuts down, the entire process will stop. This, in many cases, results in lost production. Additionally, the product in process also may be lost. When classified by size, these machines are often the largest in the plant, usually greater than 300 HP (225 KW).
Semi-Critical Machines (Production)
Semi-critical machines are defined as machines that if shut down, will cause a partial loss of production. Losing these machines, although not causing a complete plant shutdown, may limit its output and therefore its availability. Some of these machines may have built-in spares or use two or three machines to operate at full unit capacity. These machines are medium-sized, typically 150 to 300 HP (110 to 225 KW).
Balance of Plant Machines (Non-critical)
These machines have little or no effect on plant production. These machines are usually the smallest in the plant, typically 5 to 150 HP (3½ to 110 KW) and spares are readily available. Motor System
Any motor system has two subsections:
1) The electrical subsystem consists of the power coming into the plant, the plant distribution system and the electrical section of the motor.
2) The mechanical subsystem consists of the motor shaft and coupling, the driven machine and the process itself.
A fault anywhere in the motor system can prevent it from performing the intended function. This may result in reduced or lost production, excessive maintenance, or operational expenses. These two subsystems directly affect either electrical or mechanical reliability.
Electrical power is one of the most important raw materials used in industry today. Not only must we have a continuous flow of power, it also should be clean and balanced. Yet, this important commodity is also one of the least inspected raw materials supplied to the plant.
Electricity is required in almost all areas of the plant to provide the driving force that either operates the equipment that produces the products or provides the services that the plant’s equipment were built to perform. Electricity is a unique product in itself in that it requires continuous flow, cannot be conveniently stored and is normally not inspected prior to use.
Most people believe that electrical reliability ends with the successful delivery of power to the plant. But in many cases, the quality of the power supplied to the motor system may be the cause of a breakdown or failure. The result of poor “power quality” is usually long term and often overlooked as the source or contributor to the problem.
Power is normally generated far from the point of use, the reliability of the original generation is unknown and it is combined on the grid with many other generators. Many of the generating plants are smaller and privately owned. The power is transported through several different transformers and many miles of overhead and underground cabling before arriving at the plant. Many of these electrical distribution systems are owned, managed and maintained by several different entities. Once bad or “poor quality” power is placed on the grid, it cannot be removed or even rejected by the user.
Work to standardize and regulate power quality is in progress. Many states have their own specialized standards and regulations. However, generated power does not stop at the borders of the state where it is generated.
Mechanical reliability has long been a subject of maintenance departments, and considerable improvement has been accomplished by understanding the importance and benefits of improving machinery balancing and alignment tolerances. The balance and alignment condition of a machine is measured and determined using mechanical vibration analysis (MVA). In many cases, machines exhibiting unacceptable vibration levels are removed from service and faults such as unbalance, misalignment, soft foot, mechanical looseness and other faults are corrected before mechanical failure occurs. Although MVA has proven very effective for identifying mechanical faults in the motor or the driven machine, it has proven ineffective in detecting the condition or quality of the power applied to the motor.
Additional limitations of MVA exist. First, it relies on measuring the motion of the machine’s bearings or bearing housings to identify developing faults. The force generated during the early stages of most faults is insufficient to cause measurable movement. Secondly, faults that occur at locations remote from the bearings are usually undetectable with MVA. Faults in overhead fans or vertical pumps are normally undetectable using MVA. To identify faults on the entire motor system requires making multiple measurements at each bearing location. An average machine survey varies from 7 to 10 minutes.
If a motor burns up or if a breaker trips, technicians conduct electrical and mechanical inspections on the motor and the driven machine. The subsections: motor is then rebuilt or replaced and the whole process repeats. Faults that are caused by electrical problems, such as harmonic distortion, voltage unbalance, or any other electrical faults, are undetectable using MVA.
Electrical Signature Analysis
Electrical signature analysis (ESA) is a PdM technology that uses the motor’s supply voltage operating current to identify existing and developing faults in the entire motor system. These measurements act as transducers and any disruptions in the motor system cause the motor supply current to vary or modulate. By analyzing these modulations, it is possible to identify the source of these motor system disruptions.
ESA measures all three phases of current and voltage at the motor controller while the machine is in normal operating condition. ESA performs a simultaneous capture of all three phases of voltage and current, performing a complete indication of the incoming power quality and motor power. It calculates motor efficiency and motor power factor. ESA also performs a Fast Fourier Transform (FFT) on the voltage and current waveforms.
ESA is proving to be a very effective technology for detecting faults anywhere in the motor system during the PdM process. The FFT allows ESA to identify all the mechanical faults that MVA finds in the motor, the driven machine and the process itself. It also provides better diagnostic capabilities for identifying and analyzing developing electrical faults within the motor electrical subsystem. In addition, it performs a complete power analysis to identify any power issues that can lead to premature failures in the electrical subsystem of the motor system.
The automatic analysis performed during the ESA process can be far more accurate than MVA since measuring the motor voltage and current allows for accurate determination of the running speed. This accuracy is usually within one or two RPMs. Additionally, ESA uses the motor current as its transducer and very small changes in any part of the motor system causes modulation of the motor current. This increased sensitivity allows for early detection of developing faults anywhere in the motor system. ESA has successfully detected faults in vertical pumps, overhead fans and loose bearing housing on machines driven by belts.
Successful implementation of PdM programs requires a thorough understanding of the PdM process and the efficient utilization of highly trained PdM personnel together with special and often expensive equipment. Reliability engineers agree that developing faults need to be identified as early as possible and ESA fulfills this requirement. As a detection tool, ESA usually identifies most mechanical faults in the motor systebefore mechanical methods like machinery vibration analysis (MVA). Additionally, ESA accurately identifies electrical problems in the motor system that MVA or other PdM technologies cannot identify. In the analysis phase, ESA more accurately determines the system’s rotational speed and more precisely identifies the mechanical and electrical faults that lead to reduced plant availability and uptime.
William Kruger joined ALL-TEST Pro, LLC as the Technical Manager in 2005. Since joining ATP, Bill has traveled the world teaching the Theory and Application of Motor Diagnostics, helping Fortune 500 Companies implement Predictive Maintenance Programs. With his combined work in the field as well as with ALL-TEST Pro, Bill has over 40 years of proven experience in the practical engineering and predictive maintenance field. www.alltestpro.com