DC Motor Rapid Brush wear

[Drivesrock]

Hi all.

I followed with great interest the previous threads with experiences on selective brush wear on offshore drilling installations - now I’ve been called to something similar! But in this case, all the brushes wear out very quickly. Last incident; in a space of 5 hours the brushes went from ok to gone (literally) and the motor was lost!

Motors: BBC GMW 400 M32, 746kW, 720Vdc. 7 brushes per arm x 4 arms. Brushes were/are EG236S.
Top mounted water/air heat exchanger with closed loop internal air circuit. Purge air for keeping internal air pressure positive.
Application: mud pump. Two motors mechanically coupled with SCRs in master-slave configuration.
Floating neutral supply to the SCRs.

First incident: Routine inspection, well within the expected life expectancy of the brushes and both motors found to have brushes dangerously low just a few millimetres left. All brushes changed.
Second incident: Month later and Motor 1a all brushes worn down, motor 1b OK
Last incident: Another month gone by, Motor 1b all brushes worn down, motor 1a OK.
Period: All happened within last 3 months but actual running time much less.
Supposedly motor 1a SCR is always the master.

OK, we’re going through all the classic reasons now and looking for recent changes to anything and everything but I’d like to ask you all a few questions in the meantime.

1. Anyone else had this kind of problem where all the brushes wear out ‘suddenly’?
2. What grade brushes are used elsewhere for this application on these motors?
3. The maintenance guys say they’ve always had a problem with low insulation readings to ground with all of their mud pump dc motors. Now, the cooled air is blown down at the commutator end, through the motor, up and through the filters and then through the heat exchanger. Is this normal elsewhere? Seems like they are blowing carbon dust into the motors.

I look forward to the replies.

 

[waross]

Has anyone recently used silicon to repair the air duct system?

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

 

[Skogsgurra]

Right Bill!

Silicone fumes create very abrasive chemical products that kill brushes like emeraud paper.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

Now that I did not know! Thanks guys.


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

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

 

[Drivesrock]

Thanks so far.

The motors are cooled by on top mounted water/air heat exchangers and there is a purge air line from a compressor that injects air into the terminal boxes making them and the rest of the motor positive pressure. The air purge line feeds 4 dc motors in parallel of the 2 mud pumps but mud pump 2 motors a & b have not apparently suffered any rapid brush wear.

 

[edison123]

Was there a good conductive film on the commutator ? If not, I would suspect improper brush grade. Was the brush manufacturer contacted ?

 

[MotorVib]

Take a look at how much water is in the air. I have seen brushs wear fast when there is litte water in the are. (I think it was 2 grain of water per ft³.)

Chris

"In this house, we obey the laws of thermodynamics." Homer Simpson

 

[Compositepro]

Water in compressed air lines is so common that I would look there first. There should always be a self draining water separator at the end of the line before the motor if you don't want water in the motor.

 

[flexoprinting]

Yes moist and dirty air is a possibility but he stated two motors out of four do show this condition.

 

[Compositepro]

Water is only one possibility, but it is very likely that the amount of water would vary from line to line wthin a system. The water condenses in the piping and drains to the lowest point. Water vapor often gets past the driers during flow surges where some user needs a lot of air for a short time. Other branches in the pipe don't get the wet air. Water will also collect more in cooler pipes that are shaded from the sun or in conditioned spaces.

 

[oftenlost]

Have the drives been tinkered with? Sometimes, motors operated below base speed with less than full field amps can take a toll on the brush wear.

 

[MotorVib]

Back when I work on some brush generators I got a book from National that ran down differnt problems and how to fix them. you could try and get a copy. It was called National Electrical Carbon Corporation Brush Digest. I did some looking on the web but I did not find it.

Chris

"In this house, we obey the laws of thermodynamics." Homer Simpson

 

[burnt2x]

Could you please take a photo of the commutator and let us see? A nice and even film on the commutator surfaces should be desirable. Try another grade if nothing else works!EG224 is a bit harder than EG236S, lower contact drop; with the same current density capability.
Please see attached.

 

[UKpete]

Hello Drivesrock.

My dc motors background is in traction motors - these are similarly rated to yours (but not with so many brushes - commutators don't like running with brushes at low current density, the surface film gets worn away). Ventilation is either forced, by an external blower, or self (by an armature mounted centrifugal fan). Air in is always at the commutator end. On the self-ventilated motors the air is drawn in under the train and frequently draws water droplets, dust and leaves through but this doesn't result in abnormal commutator conditions. The usual victim is the winding insulation.

Is your commutator mechanically sound? You should run a clock around it, it shouldn't exceed about 0.2mm TIR, ideally 0.1mm, and 0.01mm bar-to-bar. Sudden steps in commutator profile is a brush killer, the sparking set up there also gives selective wear of the commutator at that point and gets progressively and irreversibly worse until the commutator is re-turned in a lathe. If you measure on and off the brush track, you can determine if you have electrical wear, or mechanical wear respectively.

I dare say that your worn brushes already show evidence of sparking on their trailing edges (soot), but if you look at their sides for wear marks from sliding in the brush boxes, it is a quick indication of whether you have brush/commutator instability.

 

[Drivesrock]

Hi all, thanks for the replies.

A complete as possible 'shake-down' of the drives and motors was performed first. A history check also to establish recent changes. And then testing and waveforms checked with everything looking as I would expect. Of possible significance; There was a possible loose connection of the phase-back signal on the back of the regulator rack - these are analogue drives with wire-wrap connections between the cards - and it was either already disconnected or was certainly so when the wire was disturbed a part of the checks. This connection is on the output of the speed regulator; An AC capacitor - part of the snubber circuit - had a very slight sign of a leak in the past and was swapped out but it checked out OK with a meter on the bench.
What was left of the brushes is to be sent for analysis. The guys onboard correctly do not use silicon 'glue' to seal junction boxes covers etc.
My last report from the vessel is that after 5 days everything is still OK with commutators looking good.