If you’ve been around the business world for a while, chances are you’ve been burned at least once by used equipment. Or, you’ve tried to restart an old, mothballed system that simply no longer functioned. In the early days, this was a frequent occurrence, an accepted cost of doing business. We took our chances and hoped that the equipment would, or could be made to, operate reliability. In those days, we had little choice - there was no way to measure the equipment’s operating condition. Fortunately, things have changed. Predictive maintenance technologies provide a cost-effective means of verifying the operating condition before we buy or install.
Most plants don’t consider using predictive maintenance to determine the operating condition of used or mothballed machinery or systems. In part, this is the result of the universal belief that multiple data sets are required before predictive technologies are of value. The story line from predictive maintenance system vendors always has been that trending the change in vibration, infrared, or lube oil data is the only way to use these technologies. Fortunately, this story line isn’t true. Predictive technologies can be used as accurate, cost-effective diagnostic tools when only a single data set is available.
Throughout our careers, we relied on single data sets of vibration, process variables, infrared, and other predictive technologies as our primary troubleshooting tool. They’ve never let us down. We’ve been able to resolve 99% of these problems without the benefit of trend data. Because of this little-used power, predictive technologies are ideal tools to evaluate the used equipment’s operating condition, as well as for complex production and manufacturing systems.
There are fundamental rules that must be followed to evaluate machinery and system condition. The first and most important rule is that predictive data must be acquired when the unit is operating at its full design load. On normal production systems, this isn’t a major problem. Data acquired during normal production will be representative of the fully loaded machine. Sometimes, this isn’t possible for used or mothballed equipment. These units typically are disassembled or can’t be operated. If this is the case, there’s little that can be done to determine the unit's condition. The machine, component, or system must be operating, under load, before any of the predictive technologies can be used. There are only three options available under these circumstances: assume that the unit is in good operating condition; temporarily install and operate the unit to acquire predictive data; or try to evaluate it using visual inspection techniques.
[pullquote]Assuming that the used equipment is in good operating condition is not advisable. Unless good maintenance practices have been followed throughout the storage period, there’s a high probability that the machine won’t be reliable. In fact, it might fail to operate at all. Even stored electrical and mechanical equipment must be maintained. For example, a rotating machine or component, such as a pump or electric motor, must be rotated every thirty days to prevent damage to the bearings and other components. The same is true for electrical equipment. If the equipment and its components aren’t maintained throughout the storage period, degradation occurs. Connections oxidize or components fail. As a simple example, the old-style picture tube of a disconnected television normally will fail shortly after start-up. The tube requires constant, low voltage power to retain its normal operating condition.
The only reliable method to determine the reliability and operating condition of used, rebuilt, or mothballed equipment is to test it under normal operating conditions. When this isn’t possible, it can be temporarily installed in a test fixture, like a dynamometer, that permits accurate diagnostics. While there’s some cost involved, the value gained from this approach is worthwhile. As an example, a large, integrated steel mill in the Chicago area found that after a complete rebuild by an outside repair shop, 57% of their mill motors were in worse condition. For years, the mill relied on visual inspection and no-load testing to verify rebuild quality. When the mill insisted on testing motors under normal operating loads, they found the rebuild quality was severely lacking. The use of real-time vibration analysis to evaluate these motors improved rebuild quality. Today, 98% of the repaired motors are in "good as new" operating condition.
A word of caution
If the system has been stored or idled for an extended period, flush each liquid reservoir - water, oil and grease - before running the equipment. When a machine cools and sits idle for extended periods, deposits form. When the machine starts, these deposits plug internal passages or find their way to critical wear points, such as bearings. In most cases, this result is damage and premature failure of an otherwise good machine, component, or system.
Sometimes visual inspection is the only option available. At best, visual inspection is subjective and can’t absolutely determine reliability or condition. While it identifies obvious problems, such as broken structural members or components, it lacks the diagnostic capability of predictive maintenance tools. If you’re forced to limit evaluation to visual inspection, start with a good design review.
To gain any value from visual inspection, you must first fully understand the machine or system. As in any diagnostic situation, this understanding of the machine or system dynamics is critical. You can determine what should be inspected from the design review and whether simple manual tests can be performed.
Simple tests, such as rotating a shaft to see if the bearings or other components are damaged or purging a grease fitting to check the condition of the lubricant, can exhibit a clear indication of potential problems.
Today's predictive maintenance technologies provide a positive way to determine whether a used or mothballed machine is capable of operating reliability. Using these tools avoids many of the headaches associated with equipment startup. Perhaps more importantly, they provide the means to ensure that you’re getting you money's worth.
Email R. Keith Mobley, CMRP, MBB, principal consultant at Life Cycle Engineering, at [email protected].