I'd appreciate any thoughts people have about what I'm about to relate.

A firm, which can remain nameless for now, has a population of mostly TEFC 3 phase 480 motors, with a few 250 and 500 volt DC motors here and there. Most of the motors are below 7.5 hp and non-critical, a few are up to 100 hp and critical to the process, as are the DC motors, which range from 30 to 50 hp. The run cycle is driven by the demand for the product, which is produced on a just-in-time basis. The 2008 numbers were approximately 1000 start/run cycles, with the cycle length varying from 3 to 20 minutes. The actual hours ran for the year were 260. The locations are indoors, although much of the plant space is unheated.

The owner of this facility has a maintenance plan in effect, where local maintenance people do periodic inspections and mostly lubrication PMs. Annually, the local maintenance group is assigned snapshot megger testing (500 volts, not timed, no temperatures taken)A-B, B-C, A-C, A-gnd, B-gnd, C-gnd. This is performed on all motors, their feeders, all lighting transformers, their feeders, all motor control center buswork, and their feeders. Every calendar year also, this test scheme is performed again as a crosscheck by an electrical engineering workgroup, and every second year the E/E group hires an outside consulting engineering firm to perform the duplicated checks, which the E/E group and the local maintenance group observe. The consultant group sometimes flags noticeable abnormalities, but the results of the other inspections are not analyzed.

During the inspections, the facility is operated to make product, and amp readings are taken using three clamp-on meters, and an amperage reading is recorded at some point in a machine cycle. The readings are compiled into a report, and if taken by the consultants, illustrations are added and the report is bound and stamped with the PE seal of the lead engineer. These reports are generated for each facility within the organization, and performance of the field portion of the work takes about a week. The E/E group spends about six months of every year doing this sort of work at every facility.

So, what do people think? Does this sound like a good use of time and resources? What would you recommend?

Is this the only test done? Are they conducting IR testing or vibration testing?

What is the rationale for phase to phase megger testing?

Do these tests identify problem machines? What criteria is used? How does management respond to identified problems?

What kind of failure rate is experienced in the motor population?

The snapshot megging and amp readings are the only testing done. Everything else is just visual inspection.

The rationale for phase to phase megging appears to be that nobody understood ohm's law, and that there isn't a good way to do it on small motors.

They use the IEE 1meg action level for flagging bad megger readings. Flagging consists of noting in the report that the item should have further tests.

The usual management response will be whatever is recommended in the report, if there's not something specific, there won't be a response.

Failure rates aren't compiled, but on the basis of hours of run time, they would be substantial. The intermittent nature of the process hides this quite well. Since the motors run 250 hours a year, a failure per year is actually quite a high failure rate.

The thing that has been working in their favor is the relatively low age of the motors, most of the motors are 15 years old approximately.

Wow. The sound of silence.

I doubt that everyone was stumped, except maybe for wondering where to start.

Heaven forbid that this company is so average that the story isn't even noteworthy.

I made the following recommendations to the company:
1. Cease snapshot megging
2. Cease snapshot amp readings
3. Cease repeating these readings with outside consultants
4. Seek a reputable outside engineering firm to identify relevant testing methods and intervals.
5. Fully re-evaluate PMs.
6. Use outside consultants only where they can furnish special expertise.

A huge cultural change, but also a big financial opportunity.

Good Morning BridgeGuy

You are more or less correct in making those recommendations for a number of reasons.

1. Insulation to ground and phase insulation readings should be timed and adjusted for temperature just for the fact that they will vary from test to test. The best approach would have been to make these adjustements, plus note humidity and trend them. Inspect significant variations over time and relate all tests to IEEE Std 43.

2. They definitely need to understand their failure rate and types of failures before implementing or changing the program. Also, was there any success at all in relation to this testing methodology (I think it might have caught about 10-15% of failures if viewed historically, higher if high humidity or contaminants in atmosphere).

3. Snapshot average current readings will only identify an approximate loading (very approximate). However, not very effective for troubleshooting/pdm.

4. How were the intervals selected? With machines less than 15 years old, the actual standard calls for a minimum insulation resistance reading of 5 MegOhms for <600 Vac machines and 100 MegOhms for medium voltage and DC armatures, adjusted for temperature.

The real question is whether or not the decision makers at the facility will take the advice to change.

What type of application is it with less than 250 hours of operation per year per machine?

Sorry, it was hard for me to conceive of someone running their PM program as you describe it.

I think most of the electrical effort spent on the motors will miss a majority of failures. I have some reservations about the value of the techniques used.
• A 500v megger test on a 480v motor is less than the peak voltage the motor sees. Other than giving a value on the megger display doesn’t really stress the windings much.
• What value is there in megger testing across A-B, B-C and A-C? Those tests should be done with a DLRO or a meter to check winding resistance.
• Spot ammeter readings are pretty much a waste of time.
• I’m amazed by the follow up testing by EE and outside consultants. With the manpower costs of their PM program they could easily pay for a few motor failures, especially since all this testing will probably miss the failure when it occurs anyway.

I assume the periodic and lubrication PMs are performed annually also; is that the case? The most probable cause of failure may be over lubrication of the motor with only 250 run-hours per year. I’d recommend installing pressure relief grease zerks and/or switch to a condition-based lubrication program: vibration monitoring or ultra-sonic indication.

If the larger motors are run at the same number of starts and run hours/year, investing in VFDs might be prudent.

Seems like a good failure tracking and trending program with follow-up root cause failure analysis would serve them much better.

J-

Lubrication is maintenance's #1 program. It must be done correctly. And, I place vibration as #2. In reality, multiple programs need to run in conjunction with one another. Vibration, IR, oil analysis, prudent electrical testing.

As far as testing winding, there are automatic testers manufactured with outputs to a scada or computer that can trend motors as well as making the snapshot meg reading a thing of the past. Look into Motorguard by Megalert manufactured here in the US.

Megalert.com

Thanks to everybody who took the time to reply and offer information for this post.

Let me fill you all in on the "rest of the story"...

The company in question is the Movable Bridge department of the Washington State Department of Transportation. The "plant" is the twenty movable structures in the state, i.e. twenty drawbridges of various designs and ages. The product is bridge openings, for vessels to pass in the waterway. This explains the 250 hours of runtime per year.
Since you all pretty much concurred with what I intended to suggest to the state, I wrote up the proposal and submitted it to the state productivity board suggestion program.
It was forwarded for objective evaluation to the folks who hire the consultants every year, and unsuprisingly rejected flatly.
I appealed the rejection, and the suggestion ended up being read by executive level management, in order to decide if an appeal was justified.
Evidently it was, since approximately $2,000,000 was removed from the budget of the department hiring the consultants.
The agency EEs still come out for their inspections, although they have quit doing some of the most valueless work. When asked why they changed, they deny anything is different.
The official position of the agency is that the suggestion was not adopted.
Governments dysfunction in mysterious ways....