Generator Collector Rings 4

[electrageek]

We have an older GE Generator (70MW) that is being put back into service. The collector rings have some rust on them but it can be clean off easily. My question is at what point should we consider replacing the collector rings. It is very expensive and I would like to know what the criteria should be. Any Generator experts out there.

 

[rmw]

I thought collector rings were a special grade of brass. Rust? I am not a generator expert (nor any type of expert for that matter) but I would think that the criterion that would be the most important is the surface finish you arrive at when you get the rust cleaned off. If the surface is too rough it will eat the carbon brushes alive. You should expect almost a mirror like finish.

rmw

 

[waross]

I would look at the amount of material left in the thickness of the rings. There are grades of brushes that will dress a commutator if run for a few hours or days. I am not current but something similar may be available for slip rings.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[burnt2x]

I would rather refer to any document from the OEM, if there are some. IMO, slip rings are made to last long enough. Verdigris (rust) are just superficial and correctable without any noticeable effect of slip ring performance. Burnishing for a few minutes could the job.
Hope this helps.

 

[edison123]

electrageek

Is this a two pole machine (3000 or 3600 rpm) ? That would account for the steel rings.

Mechanical checks:
1. Surface finish (smooth, no ridges/cuts)
2. Connection points (loosen, clean the joints, check for stud cracks)
3. Material left for service (OEM manual provides the minimum service diameter)

Electrical checks:
1. Megger both rings to ground and ring to ring
2. AC HV test at 3 KV minimum
3. Measure contact resistance of each ring to lead

 

[ScottyUK]

A 3kV test on a 70MW machine seems a little high, especially for a service run machine. I'd be surprised if the rotor is rated for much more than 500V in normal service. Is the risk of provoking a fault worth the cost of a rotor rewind? I'd probably settle for a 1kV Megger test on a small machine.


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

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

 

[electrageek]

edison123

It is a 2 pole machine 3600 rpm, and the rings are steel alloy. The field is rated at 250Vdc 780. There are four rings two per phase (+ and -). Thanks for the info everyone.

 

[edison123]

Scotty

For new rotor windings, the ac hv test is 10 times the rated voltage with a minimum of 1.5 KV as per stds. For overhauled machines, the test level is 80% of the new winding test voltage.

In this case of 250 V, I would recommend 2 KV AC.

 

[wolf39]

Mirror like finish would result in squeeking carbon brushes and considerable wear and tear. Just clean the collector ring surfaces carefully but don't use sand paper. This would contaminate the surfaces with carborundum which in turn also leads to excessive brush wear.

edison123 is right: I also would apply 2 kV AC for one minute. But first clean the collector ring assembly and field winding (were accessible) as well as possible. Before hi-pot testing I'd recommend to perform a dry-out procedure.

Wolf

http://www.hydropower-consult.com

 

[LionelHutz]

10 times rated sounds high. The NEMA MG-1 recommended test voltage is 1000 + 2 x rated voltage. If it is adjustable supply it is 1000 + 2 x AC voltage of the supply. So, say this is using 208VAC rectified to get 250VDC the test voltage would be 1400V. If you are using DC then you go to the peak voltage which would be 2kV.

 

[edison123]

Lionel

2 U + 1 is for stators/ 10 X U is for rotors per stds. (IEC/BSS et. al.)

The reason is that during faults, rotor experiences much higher surges than stator.

 

[ScottyUK]

Hi Edison,

I am cautious of provoking a fault on a rotor which is threaded into the stator - the costs of a stripdown are very substantial on a big machine. If I had the rotor in the shop then I totally agree with testing to the standards but I am less inclined to undertake a test which has scope for initiating a fault which might otherwise lie dormant for years in a machine which is built up. An in-service failure is bad news, but a failure initiated by a test will have me answering some real awkward questions from some very unhappy plant owners. Sometimes being right isn't worth the consequences! The outcome would be the same in terms of an unplanned outage whether a test takes it out or it fails in service. The most likely failure is a single earth fault which is unlikely to be terminal for the rotor, although a second one might well be. In this case where the machine is possibly not committed to commercial operation then you are right, a proper test is probably justified.



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

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

 

[wolf39]

Scotty:

This is a no-win situation, i.e. whatever you do, it may turn out wrong. However, after a long stand-still period it would be wise to do some inspection and testing before the generator is put back into operation. You'll have a severe problem if you don't hipot test the rotor and the unit fails in service. This is called an unscheduled outage and if the operator is committed to deliver a guaranteed amount of megawatthours, he has to purchase this energy at high costs from somewhere else. If the hipot test fails before re-commissioning, you have fewer problems to face and you have followed a standard procedure for which nobody can blame you. To be on the safe side, the power plant insurance company can be consulted for comment.

As edison123 pointed out, the minimum test voltage for new field windings is 1.5 kV and 80% of a standard test voltage should be applied to an overhauled winding. Before the AC test voltage is applied, however, a dry-out procedure has to be done, followed by a megger test. A dry-out can be achieved by running the unit at rated speed for a couple of hours with reduced cooling water for reaching a cooling medium temperature (air) of about 40 degrees centigrade.

Regards,

Wolf

http://www.hydropower-consult.com

 

[wolf39]

I forgot to mention that for the dry-out procedure the unit should run in an unexcited mode.

Wolf

http://www.hydropower-consult.com

 

[edison123]

Scotty

The OP asked for a reliability check for bringing back a moth balled machine and hence my suggestion. Of course, the tests should be done only when the conditions conducive to conduct (how is that for alliteration!) them are met.

Any test, theoretically, has a 50% failure probability. As wolf rightly points out, not only an in-service failure would result in unplanned outage, but there are also high odds of collateral damages like core iron damages, fire etc. Of course, the "management" must be brought on "board" about the possibility of destructive testing.

May be my strong views on stringent testing norms are biased by my repair/service experience .



wolf

I agree that dry-out is a must for any hi-pot and I generally use Generator short circuit method to bring out an early and simultaneous IR improvement in both stator and rotor. I go up to 80 to 90 deg C in a span of 3 to 12 hours depending on the machine size.

(BTW, I visited your website and saw names like Mario Lopez/Jurgen Groth. Are you connected with VSHK anyway? Awesome photos of Itaipu - I am sure people were holding their breath till that 1800 ton monster was put in its place. If you are interested, my url

http://www.edison.co.in.)

 

[ScottyUK]

Wolf, Edison,

I think we agree!

there has been some good discussion in this thread - thansk to all.



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

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

 

[rmw]

As a comment, I was involved in a situation recently with a 2 pole ~150MW machine and it had brass collector rings. (60 hz, naturally). The OEM couldn't deliver in time and the shop ended up machining them themselves.

rmw

 

[slavag]

Hi.
Sorry for another Q, but I wouldn't open new thread.
Our client have now damages generator and have send it to repair. As far, as we know some, de-magnitazing ( don't know exactly) procedures needed.
Could you please attached some link on the issue and short explanation, what is it and what procedure.
Thanks in advance.
best Regards.
Slava

 

[electrageek]

Hi Slava,

I was the original poster and it was fun to see the direction and information that came out of a simple question. However I believe you should start a new thread with your question otherwise you will not get the attention you could. It is not the same as mine and involves a very different problem.

By the way, on the original question, the answer appears to be to replace the collector rings based on a minimum diameter specified by the manufacturer.

Thanks to all
electrageek

 

[byrdj]

I'm not da expert on Collector Rings But I am interested

was the call to replace base on minimun diameter an actual mic diameter or was it based on the depth of a spiral groove in the rings. I have cut that groove deeper then took a grind pass to resurface

I've seen these folks do some good work on rings and brushes

http://www.cutsforth.com

If your call to replace was based on minimun diameter (book value) is there any suport engineering to support.

would it be stress due to reduce ring OR I think it might be more to due with brush geometry. I know when you take a grind, the brush holder gap is adjusted IF the rigging will allow.

Now If you have any questions related to stroking the CVs (something I might know about)