Buzzell serves clients throughout the San Francisco Bay area. Our business includes motor repair, rewinding, and service, including onsite and offsite inspections, testing, and consulting. Matt Buzzell, company president, said, "Our job is chasing problems and fixing them."
Our clients are primarily industrial. We service motors in printing presses, fire pumps, chillers, elevators, ventilation blowers, and other applications. Our clients, from hospitals to high-rise offices, hotels, condominium complexes, and parking garages, are conservative in their maintenance spending and consequently many need emergency service.
You might think we take a lot of measurements on a regular basis to keep things running smoothly. Actually, we don't do much predictive maintenance or history work. We do problem call-outs instead. When things break, our clients call us.
In facilities that have a "maintain before break" philosophy, insulation resistance test results go into a database for trending. In facilities that have a "run to failure" philosophy, insulation resistance measurements serve a different purpose.
We use insulation resistance tests for troubleshooting (go/no go). Without these tests, troubleshooting would take longer and cost more. That's not good when production has completely stopped. A breakdown is always an emergency.
While any failure is problematic, motors present additional challenges - a specific motor failure or its cause isn't necessarily obvious. When a motor stops, it often looks like there's nothing wrong with the motor, at least from the outside. So, the first thing the typical owner does is try to start it back up again - not a good idea. That can result in major damage.
We ask the client to wait until we can run an insulation resistance test to see if there's a short to ground. This way, we can prevent additional damage to a unit or related equipment.
Because insulation resistance testing is a critical part of our service, I make a practice of knowing what testers are on the market and evaluating them for use in this business. It took me eight hours of researching testers to find one that fit my criteria.
The Fluke 1520 megohmmeter does almost everything I need. I was particularly impressed with the safety aspects. The meter will tell you when you have a live circuit. It won't apply power to it, either. The discharge after testing is also great. I also like the low-ohms capability on the 1520.
Pretest TasksThrough years of training and experience, we have developed standard procedures for quickly and accurately identifying motor problems.
"When a client calls and says their production isn't running, there's no time for guesswork," Buzzell said. "We have to know exactly what we are doing."
Testing involves several steps, including these pretest tasks:
Basic fact gathering. We don't just jump in there and start measuring things. We talk to the customer to find out what they've done - maybe they attempted a repair or made some operational error that led to the breakdown.
This helps us focus in the right areas and reduce downtime accordingly.
Next, we go to the motor and gather basic information. This includes collecting nameplate data and taking measurements with a digital multimeter (DMM) to understand the environment in which the motor is operating. The DMM measurements include voltage, fuse, and ground connections.
Throughout this process, we use our senses to observe what's happening. A particular smell can speak volumes if you know what it means. We have a list of things to smell, feel, and listen to.
Control contacts check. Next, our company performs three sets of tests with the insulation resistance tester. Before testing the motor itself, we check the control contacts for quality of contact:
1. Lockout and tagout the disconnect to the starter.
2. Manually engage the starter so the contacts close.
3. Set the 1520 to the low-ohms range.
4. Measure the resistance across each set of contacts.
5. The reading should be nearly 0. If it's higher than 0.1 ohms, that set of contacts needs to be replaced.
Resistance of line and load circuits to ground. Next, we measure the insulation resistance of the line and load circuits to ground.
However, "before doing any insulation resistance testing, you must isolate any electronic controls and other devices from the circuit under test," Buzzell said. "These voltages tend not to be kind to them." Then:
1. Lockout and tagout the disconnect to the starter.
2. Set the 1520 to the appropriate test voltage (250, 500, or 1,000 V).
3. Identify the resistance between these points -
To pass these tests, the line and load circuits need to show high resistance. As a general rule, to ensure safe operation, AC devices need a minimum 2 megohms to ground and DC devices need 1 megohm to ground.
Different companies have different threshold minimums for insulation resistance on used equipment, ranging from 1 to 10 megohms. Resistance on new equipment should test much higher (100 to 200 megohms).
If the load-side resistance values are acceptable, proceed to the next test. If they aren't, start tracing the problem. Is the insulation breakdown in the load side of the starter, the cables, or the motor?
We conduct the third set of tests at the motor, measuring the winding resistance phase to phase and phase to ground.
Because we're performing troubleshooting rather than predictive maintenance, we aren't storing and trending measurement data to determine whether insulation is degrading. Instead, we're taking point-in-time measurements.
There's no magic bullet test that will tell you whether your motor is good or bad.
As Buzzell says, "Every test by itself is inconclusive until you do them all. Just because you have a good reading doesn't mean something isn't still wrong, but a bad reading means something is definitely wrong."
Here's the full sequence from start to finish:
Look for a reason not to energize -
It would be irresponsible to risk a fire from a possibly shorted motor. If the supply is good, there's a motor problem. In that case, run the following tests:
If any of these elements are missing, the results are inconclusive.
Publication date: 04/11/2005