stator coils being loose in the slots 2

[elecfrance]

Hello
Recently a motor1 1600 kW 1500 tr/min was examined because of a suspect sound : the stator coils were losen.(that's what my collegue said). It was replaced by a new one motor2.

I have to look another 1600kW 1500 tr/min motor3 (exactely the same year of installation than motor1) which may have stator coils being loose in the slots : a sound is suspect too on this motor.
I'm not on the site but will go in a few days.
Must I verify vibrations? 100 Hz? 1*RPM?
Can someone notify a vibration collector (3000 euros or dollars) with FFT good resolution?

Thanks for your help
pierre

 

[electricpete]

Diagnosing coil looseness with vib is not an ideal method. Direct inspection is of course better but not always practical.

I have seen coil looseness as evidenced by 2*lf, 4*lf, 6*lf etc. (100hz, 200hz, 300hz etc in your case). 2LF alone is different and not a looseness problem. In my case the magnitudes were very low (?0.03 ips at 4*LF) and only noticeable on a log scale. But the looseness upon inspection was severe.

I am told it can show up as ozone for high-voltage motors >6kv. Very easy to monitor for ozone in the outlet stream using Drager tube... available from the industrial safety department of most plants.

 

[elecfrance]

Thanks
Can you explain what do you do with ozone in 6 kV motors? to search what?


I will search 100 200 and 300 Hz in vib spectrum.
Do you think i must look to the current too (fan with long start time...)?

I can't do a direct inspection...

Do you know a good vib collector? microvibe SKF? another?
Thanks

 

[electricpete]

Ozone comes from partial discharge which will be expected if severe winding looseness persists in a >6kv motor. It is a phenomonon called slot discharge. Vibration can erode the coil semiconductive coating. The looseness and the damaged coating leak to arcing in the slot.. which in turn can lead to further erosion of the coating. Good motor should be much less than 1ppm in exit airtstream.

In case you're not familiar with ozone.. it's the small like aerosol spray that you smell during an ac hi-pot test. It occurs as a result of partial discharge in air.

We use high-end Entek Datapak data collectors and I believe the cost is about $10k. We are very happy with them. Maybe someone else can suggest a good one that would meet your needs.

If you look at current you would most likely be able to check for rotor bar problems which are most likely in high-inertia, high-speed, high horsepower applications. In that case you need to look at the current spectrum for pole pass sidebands around 60hz. Should be <1% of the 60hz.

Other than that I'm not sure what you're hoping to see in the current.

 

[edison123]

elecfrance,

Direct inspection of the stator (however difficult it may be) is the best option to check for coil looseness.

 

[testtech]

elecfrance-

Does the noise occur during start-up or continuously. If at start-up you may have a rotor bar issue.

Vibration may pick up a winding or electrical source vibration. However, this is not necesarily useful information, since some motors inherently produce field or rotor bar related vibration and will do so for years. This &quot;noise&quot; may be a normal characteristic of the motor.

When you get the vibration analyzer, you may want to track the vibration for several months to try to determine whether deterioration is occuring. Also, with the spectrum analyzer, use a microphone to obtain a sound spectrum. Sound spectrum will reveal rotor bar frequencies (# of bars X shaft speed) during start-up as well as field related vibration during normal operation. As noted by electripete, look for multiple harmonics of field frequencies for loose windings.

 

[elecfrance]

testtech,
we have current transformer in the electrical room and i will check rotor bars wtith current method ( gig fan 1.6 MW tl =20 s) .
The noise of the motor3 is growing (a man look every day) and the other motors of the same power are &quot;for the moment&quot; correct.

edison123,
we don't want to inspect stator (money...) but monitore it until the level of vibrations is high.

I will check radial vibrations 100,200,300 Hz

What do you think about CSI Motorstatus which can monitore vibrations during long time?
Thanks

 

[edison123]

elecfrance,

The point I was trying to make was that loose stator coils is a serious condition. Loose coils not only damage the semiconducting coat and will also chafe away the coil insulation leading to earth fault with possible core damages. In which case, the repair work becomes more expensive. By the time you allow the vibrations to become large enough (btw, how will you determine what is the allowable vibration level pertaining to loose stator coils, I dont know of any standard that specifies vibration levels due to loose coils), it may be the case of too little too late.

 

[elecfrance]

For the first motor, it was decided (not by me...) to change the motor because the reparations are too expensive. It's a global impregnation (5.5 kV) , so hard to repare

For this motor, i think we have another one in another site.

I agree with you, but i have not the final decision.
(vibration > 5 mm/s)

Have you experiences in reparing stator with global impregnation?

 

[edison123]

elecfrance,

With global impregnation, it is not possible to have loose stator coils. After all, one of basis of global impregnation is fill all the voids with the resin thereby preventing any coil movement. So, you (or your colleague)may have to re-evaluate the diagnosis of loose stator coils.

I agree with you that rewinding of globally impregnated winding is a costly and time consuming proposition. In my repair shop, we carry out rewinding of many impregnated machines and I can tell you that its costs more than to rewind coils with resin rich insulation system.

 

[testtech]

Here is a quick and easy analysis tool:

Go to uesystems.com and then accessories. They have spectrum analysis software that will reside on you PC. Install it on a notebook that has a built in microphone. Use the notebook next to the noisy machine to obtain the sound spectrum.

I have walked up to noisy machines with nothing more than a notebook computer and confirmed bearing and other faults in this manner.

 

[jag27]

what role do the bearings play in generating noise
if ou are experiencing noise issues with a motor
say a BLDC motor, how can u pin point if the noise is coming from the bearing,rotor or the windings

thanks

 

[mvcjr]

I think testtech means that each type of noise has its own unique sound pattern. Using a computer to capture these noises and comparing it with a database of identified patterns of noise from various sources you can quickly determine what is causing the noise.

Am I right?

God bless!

 

[jbartos]

Suggestion: The power supply to the motor might be examined for its quality. Harmonics and interharmonics might cause motor vibrations and might structurally affect the motor winding and rotor cage. Supposing that you have the existing motor overhauled, the same motor conditions might happen again. Considering the size of the motor, a reasonably good power quality analyzer might be in place. It will pay for itself handsomely.

 

[electricpete]

Testech - That looks like some great software. I can't count the number of times I have wanted to record a noise and analyse it. A great assortment of bearing and other noises that operators like to ask about and I usually have very little answer. I see there is a trial version which looks like it will work with any mic-equipped laptop (although upgraded mic might help). Do you know the cost to purchase??

 

[Highspeed]

To monitor vibration, I also use the soundblaster in whatever computer... 16 bits acquisition system for that price is hard to beat...

If you have a copy of Labview, National instrument provides drivers and samples allowing all kind of spectrum analysis using the Sound card.

CoolEdit(recently purchased by ADOBE) is also another software to be considered as well as SpectraScope 2.8 and Spectrum Analyser Pro (

http://www.pas-products.com)

I do not use the mike input but a preamplified accelerometer connected to the line input of my sound card(monitran >>

http://www.monitran.com)

Those are a lot cheaper than B&K.
Good luck

 

[elecfrance]

jbartos,
i will not check power quality because it 's a power plant... I think it's not necessary. All the other motors have not the problem.

testech and electricpete
If my PC has a micro, i can try...

edison123,
I was the first surprised when my collegue said it was a problem of looseness in stator coils. I hope that photos exist ! But i m sure that it is a global impregnation (i found a document of 1971 for this motor )

Thanks

 

[jbartos]

Comment on elecfrance (Electrical) Oct 28, 2003 marked ///\\jbartos,
i will not check power quality because it 's a power plant...
///Is there a power struggle or something?\\ I think it's not necessary. All the other motors have not the problem.
///This does not necessarily mean that one motor may be affected by a nearby nonlinear load generating excessive harmonic pollution, e.g. AC-DC converter.\\\

 

[elecfrance]

highspeed,
can you tell me what reference of Monitran have you taken?
preampli + sensor? How do you calibrate?

jbartos,
it's a fossil power plant (and no rectifier), so i think power quality is good.

 

[electricpete]

I agree that looseness in the slots is expected for b-stage coils and probably not for vpi coils.

Is it possible that the previous problem was movement of endwindings? Maybe loosenss of the endwinding blocking material?