Motor problems 1

[tito103]

I have run into a problem with one of the buildings i do preventive maintnance testing in. They have motors faulting to ground. They range from 25-125 hp , 480 volts. They are controlled with softstarters, VFD's with output reactors, and across the line starters. Each application varies. The building is only 3 years old! No other electrical problems have been found. If this info may help I had one of the faulty motors 100hp ,and megger tested the windings 1kv. I ran the test for 1 minute and tested approx. 300 megohms. Tried to energize the motor using VFD control, it faulted to ground after 20 seconds. Tested it again .23 megohms. Any ideas? HARMONICS

 

[edison123]

The 2nd megger (0.23 megohm). Was it with or without the cables from the VFD ?


* Basically, I would like a full-time job on part-time basis *

 

[waross]

Hello tito103
Is the whole building on the same transformer or do you have a higher voltage system (480) for the motors and dry type transformers for the 120/208 volt loads?
yours

 

[waross]

Hello again tito103
Are the motors VFD rated?
Is the system grounded?
yours

 

[tito103]

As far as the megger check the cables were disconnected from the motor. Yes the motors are VFD rated and the ground testing was normal. One thing I did not mention is all the mechanical equipment has a sepperate utility transformer 13.8-480volts 5000 amps. And the rest of the building has a second transformer same name plate data.

 

[waross]

I have never seen a discontinuous ground fault but a couple of my instructors had experienced the problem first hand and it was included in a couple of my classes. This could be something similar. A different cause but the same result. Try putting an oscilloscope on the 480 volt bus and look for a superimposed high frequency, high voltage.
The high voltage will typically be present when the equipment causing it is energised. The high frequency is isolated by the high impedance that your motors and transformers exhibit at high frequencies. It typically causes an insulation breakdown in the first few turns of the motor windings.
Get a 'scope and cycle your various loads until you find the source.
yours

 

[ScottyUK]

Hi waross,

I've seen a failure which fits the description: On one particular design of induction motor - a 415V 37kW Brook Crompton type of 1991 vintage - we frequently see ground faults which are present during starting / running but when the motor is isolated for testing everything is ok. On the occasions that I have been involved we have Ductered the windings and Meggered to ground and between phases. Everything was ok at 500V to ground, 1000V between phases. This is not confined to any one motor, but common among the sixteen on site.

Our theory is that the winding design allows some flexure of the conductors under high starting current and as the pump accelerates, and this movement causes the winding to short to the stator core. This is supported by the damage site. I'm curious if anyone else sees this problem and if our theory is a reasonable one. Between ourselves and the motor shop this has been our most plausible conclusion.

----------------------------------

I don't suffer from insanity. I enjoy it...

 

[jraef]

ScottyUK,
I've seen flexion in windings on large motors, but shorting to the rotor core would be rare unless there was a bad bearing or bowed shaft. You would still need to have an insulation breakdown of some sort, but if you saw physical evidence, that would lean me towards that conclusion for sure.

tito103,
The fact that this is occurring in motors started in any way is particularly troubling. Do you also have physical evidence of shorting to ground, or are your drives or protection relays just tripping on GF? The reason I ask is that there is a theory out there that concentric windings, used in the construction of small mass produced motors like the ones you describe, can exhibit minor current imbalances that are exascerbated during startup. Residual GF detection can pick up this imbalance and interpret it as a ground fault. That's why residual GF relays tend to have longer trip delays than Zero Sequence relays. Many VFDs and solid state OL relays use residual current GF detection because it does not need the balanced core CT used in Zero Sequence detection. Nothing wrong with it, but you have to be aware of the consequences of setting the trip delay too short.

If you do have physical evidence of ground faults, that's another story. High voltage spikes such as can occur from a large motor in the plant being started with Y-Delta can do damage over time, but that wouldn't be the case with the VFD fed motors because the VFD would act as a filter of sorts and not allow that yo get to the motor, plus you would likely also see SCR failures in the soft starters before you would see motor winding damage. The VFD filtering issue would aslo be true of harmonics in the incoming line causing extra heating in the windings, the VFD has its own harmonic issues, but that would be separate from problems in the incoming line which would be necessary to explain the problems with the soft starter and X-Line started motors.

Unless you have more than one set of problems...

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[waross]

Hello tito103
Have you been able to examine any of the motors? I would expect that flexure would tend to cause damage where the windings emerged from the stator slots. Over-voltage failures are typically near the line end of the winding. And yes, I realize that in some cases a voltage failure near the line end of a winding may also be at the point were the windings emerge from the stator slots. It is a good starting point for determining the cause of the problem. If it is an over-voltage problem, there is always the scope on the line as each motor starts.
respectfully

 

[ScottyUK]

Hi jraef,

Failure was to the stator core near the slot end, not the rotor. I'm pretty sure I said stator in my post, but I'm terrible at spotting my own mistakes. Thanks for lending some support to our theory - some that I work with believe the mighty Brooks are above suspicion, yet to me it screams 'type fault'.


----------------------------------

I don't suffer from insanity. I enjoy it...

 

[jraef]

D'oh! I see that now, sorry to doubt you Scotty! It's my eyes that need adjusting, not yours.



Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[tito103]

Hey Guys, I have more info concerning this post. I had the motors sent out to be examined, and they found pin holes in the stator. If this is an over voltage problem, how could this be corrected?

 

[motorspert]

tito
By pin holes i assume you mean small holes in the copper wire covering or the slot liner. If in the liner, then get the repair shop to use a thicker liner or two thicknesses if he can get them in the slot. If in the enamel wire, check the grade of enamel used.

If the pin holes are caused by high voltage spikes they will normally only be seen in the line end coils - the repair shop should be able to confirm that (the inductance of the coils prevents the spike getting much further through the winding). Fitting some surge capacitors will take the sharp edge off the spike and may solve the problem. I can't understand this spike affecting all motors VFD and fixed speeds.

Have you had a power system quality survey done? Might be worth leaving a recorder on the system for a week and seeing if it picks up any transients

 

[jraef]

Just a warning, no surge caps on the soft start and VFD fed motors, they can damage the controllers.

I still find it odd that this is happening in both types of applications. When they did the analysis, were they looking at one motor from each starting methodology? The reason I ask is that an end-turn damage phenomemon is well known for VFD fed motors that were not designed as "inverter duty". If, however, you have a line voltage spike problem (as it would take to explain the fixed speed motor damage), then the motors fed by the VFDs would be filtered against it by the VFD (although the VFD front end may suffer).

There is of course a remote possibility that you have a line voltage spike problem AND the VFD fed motors were not designed for inverter duty of course.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[Marke]

I had experience with a large fan motor that would run for a while and then blow the semiconductor fuses on a soft starter. No megger tests etc would show a fault, but it consistently blew the semi conductor fuses.
We put the whole motor in to an oven at a brick works and heated it up and the problem appeared. Once the motor was rewound, no problem. There was a short from winding to earth but only when the windings were pretty hot. If the motor had been DOL, the fault current would have blown a decent sized hole and it would have been obvious, but because the fault current was energy limited, the fault was hidden.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[aolalde]

The ground and turn to turn insulations need improvement. VPI impregnation can reduce the air pockets into the insulation which develop Partial Discharges and subsequent insulation failure. Depending on the VFD type and line length, continuous high frequency surges up to twice the peak voltage bombard the winding at the VFD carrier frequency.