Generator Collector Ring arcing and pitting problem

[electrageek]

We have a 30 year old GE 20MW 13.8kV generator runinng at 10MW with a Power Factor or 1.0. The field is rated at 250VDC at 199Adc maximum. The excitation is provided by a Basler Static Full wave unit. We run at approximately 90Vdc at 100Adc most of the time. We consistently have arcing on the outside collector rings. It is not there all the time it seems to come and got but it is generally related to the amount of field current. There is never any arcing on the inside rings.

We have tried:

1. Comparing the hardness between the inside and outside rings. They are the same.
2. Increase the brush pressure with new brush holders.
3. There is no vibration or out of roundness.
4. Swapped the polarity of the power going to the rings.
5. Checked and reworked the connection from each embeded copper cable coming from each brush to the ring.
6. Cleaned and resurfaced the ring to remove the pitting.

I am now thinking it might be a metallurgy problem. Something has happened to the outside ring metal that has increased it's resistance. The photo I attached shows an interesting thing. It almost appears that there is a pitted area about the shape of the brush itself. Could it be an pulsation from the excitation system happening at the same time each revolution? But why only on the outside ring?

I am looking for some new ideas and possible tests to run.

 

[electrageek]

Thanks wolf39,

They have changed all the brushes and put in new constant pressure brush holders. I was there last night and it was just barely arcing on one of the brushes. We have a chart recorder hooked up for Field Volts, Field Current, Pos to Ground, and Neg to Gnd. It still seems strange to get pitting in the shape of the brush usless the arcing is synchronized to something that is once per revolution. It only ocuurs on the outside brushes and they have tried swapping polarity.

 

[wolf39]

Good to hear that the problems eased.

I may have chosen a wrong term by using the word "polish". If a slipring is perfectly polished, brushes tend to squeak and rattle. I rather meant to improve the surface quality by removing the brush marks. Be careful to use grinding paper or similar because particles remaining on the slip ring surface will deteriorate brush wear.

Wolf

http://www.hydropower-consult.com

 

[ScottyUK]

The term I'm familiar with for sliprings and commutators is 'burnish' rather than 'polish'.

I will probably be corrected by the materials and mechanical purists, but I understand the difference to be that polishing is an abrasive process while burnishing is a forming process where the material is displaced on a microscopic level rather than removed completely.


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

If we learn from our mistakes I'm getting a great education!

 

[electrageek]

Thanks wolf39 and ScottyUK,

They use the recomended stone for cleaning the Slip Rings. The process removes the pitting caused by the Arcing. The pictures show the discoloration and once the light stoning is complete the arcing seems to go away for a period of time but it always comes back.

They are going to take the end apart and we are going to inspect the connections from the Slips Rings to the conductors that pass through the rotor into the Main Generator Field.

 

[ScottyUK]

Did you check the shaft earthing brush? What is the condition of it and the shaft it rides on?


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

If we learn from our mistakes I'm getting a great education!

 

[electrageek]

I have been told that it is installed and in good condition on the shaft between the Generator and the Turbine. I will be checking that tomorrow myself.

 

[ScottyUK]

"I will be checking that tomorrow myself."

Very wise - your eyes are harder to fool than your ears!


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

If we learn from our mistakes I'm getting a great education!

 

[rmw]

Sad but true. I don't know where you are from 'geek, but in the USA we have a saying "I am from Missouri regarding that" based on our State of Missouri's moniker of being the "show me" state. I have found way more times in my professional career than I would have liked to have been the case that what I was told wasn't what was actually the case with whatever situation was at hand when I went to take a look for myself. I tend to be "from Missouri" most of the time now.

rmw

 

[electrageek]

Thanks rmw,

True but can we make a case that "not having the Turbine End of the rotor shaft grounded" could cause arcing and pitting on the outside slip ring only.

 

[rasevskii]

for electrageek:

On your last visit to site did you actually check that the new excitation system has NO ground anywhere else in the circuitry? There must be an excitation transformer which was supplied new as part of it. Are you absolutely sure that there is not a neutral or star point ground in that supply somewhere? Why not post us a dwg of the new system?

There can be unpleasant surprises when someone has modified or retrofitted something old.

There is still a possibility of a ground fault in the rotor that only shows up during running of the unit. That together with another solid ground elsewhere could cause such a problem.

rasevskii

 

[wolf39]

Just a wild guess:

What if the outside slip ring was improperly heated by welding torches during the heating/shrinking process of the slip rings to the generator shaft? If this was the case the mechanical/metallurgical properties of the surface may have been affected unequally thus generating these spread surface marks.

Also, the sliprings are having spiral grooves. These grooves should be chamfered.

Wolf

http://www.hydropower-consult.com

 

[electrageek]

Thanks rasevskii and wolf39,

I had thought about the Slip Rings having high resistance of some sort but that doesn't really explain the pitting looking like a picture of the brush. You would think it would be more generally spread out over the ring. We are going to take a DLRO and measure the resistance of the various connections. The spiral groves look good and there are no raised edges.I am not sure when the Slip Rings were installed. I don't believe they were disassembled prior to installation at this site.

 

[rasevskii]

The burn marks on the ring would be typical if excitation voltage had been applied with the unit at standstill, however both rings would have been burned. This would have been possible if, during testing of the the new static exciter, current was applied to the field by mistake. This assumes that the excitation supply comes from a source in the plant rather than from the stator voltage...

Unless you have a double ground fault: one in the rotor, and one solid ground somewhere in the excitation supply: result: a hundred amps (or so) flowing in the loop..also via the shaft grounding brush.. A similar scenario has happened at times during commissioning in other places, where, however, both rings were more or less ruined by current applied at standstill.

rasevskii

 

[electrageek]

The excitation source is a PPT (Power Potential Transformer) provided by Basler Electric specifically for this application and it is an unground Y on the side toward the converter so it shouldn't be an ground from that side. The PPT is connected to the Generator Output and produces 3 phase 320Vac. So if the Generator is not running there is no voltage available. The pitting is occuring while the Generator is runnning not while it is offline. Also it should be noted that we appear to not have any rotor ground protection as far as I can tell. They have a Schweitzer 300G relay for Generator protection but forgot to provide any rotor grounded protection. We are going to fix that too.

 

[electrageek]

Sorry I meant an ungrounded delta on the low voltage side.

 

[pwrengrds]

I was interested on comments on the ringing on the current when the voltage turns on.

 

[ScottyUK]

pwrengrds,

My first reaction is that the current trace is not truly showing what is happening, and that the ringing is a result of rectifier commutation interfering with the transducer. There is simply too much inductance in the field for such rapid changes in current to occur without a very powerful source. The ringing occurs at too high a frequency to be influenced by the AVR itself. The logical conclusion is that the ringing is a measurement artifact.

What are you using to measure the current? In my experience many of the LEM Hall-effect type transducers are pretty good but some are susceptible to inteference caused by high dv/dt on the conductor passing through the transducer. I normally fit a copper foil electrostatic shield in the bore of the transducer and connected to frame earth which really supresses the capacitively coupled interference. It's a cheap and easy fix - worth a try if you find yourself questioning your instruments.

Most AVR's I have worked on use a shunt to measure field current and you could tap in to this but BE CAREFUL if you connect directly to the shunt, and use an isolating probe or a Scopemeter. Your voltages are well within the capability of the Scopemeter type instruments, so that would be my first choice in this instance.


This is developing into an interesting problem!


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

If we learn from our mistakes I'm getting a great education!

 

[rasevskii]

OK, so there is (correctly) no other ground on the excitation system...there goes my double ground theory out the window...

Probably Basler can supply a rotor ground fault relay. ANSI Device reference 64 if I recall.

rasevskii

 

[electrageek]

Thanks rasevskii,

Yes exactly what we going to propose.

Thanks ScottyUK,

This was my thought also. We are using a DATAQ 730HV digital 8 channel recorder/oscilloscope. It seems to show up at very high data rates 37,500S/S.

Today is the day they are going to take the emd covers off so we can see the field connections.

 

[ScottyUK]

Dataq 730HV: that's a nice looking toy - is it any good?


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

If we learn from our mistakes I'm getting a great education!