Excessive brush wear on DC motor.

Greetings! We have a field excitation DC motor on our SAMP machine. We are experiencing 1 brush on the DC motor to be wearing out a lot faster than the other 3. (My understanding is that the brush that carries the least amount of current gets worn out the fastest)
The specs of the motor are as follow. 500VDC , 49 full load Amps. Supply: 460V 3 phase.
Running load 20Amps.
The brush contacts are jumped. i.e 1-3 and 2-4 are paired so that equal current passes through each brush set. The behavior of brush number 3 is unexplainable: wearing out almost 40% faster (replaced one on Thursday and down 40% in 3 days approx 40 hrs of machine runtime). We have 3 DC motors that are all showing the same problem.
The commutator on 2 of those motors were replaced but we are still experiencing the same characteristics.
We contacted SAMP (Italian manufacturer) and they have advised us to buy their brushes($33 a pop) which is about 3 times more than what we are currently paying. I don't think that their brushes will somehow magically fix the problem we are experiencing. The brushes that they use are from the same manufacturer (helwig) that we are currently buying them from.
I am not able to figure out what the problem is: I connected a microscope and the signal out of the drive is as clean as it gets. No problems there.
Any insight would be greatly appreciated. Best regards

Is there much dust from the worn brush? DC brush wear is the result of mechanical friction and electrical erosion. Friction produces carbon dust; the result of electrical erosion is the vaporization of carbon with little physical residue.
I am not able to figure out what the problem is: I connected a microscope and the signal out of the drive is as clean as it gets. No problems there.
What 'drive'? Please explain.

Dawoodi Bohra wrote:
  Quoted Text Here  
You are supplying DC to this motor or AC?

snipped-for-privacy@gmail.com wrote:
  Quoted Text Here  
Supplying 460VAC.

Dawoodi Bohra wrote:
  Quoted Text Here  
Am I missing something here?
1. You have a DC motor, with commutator and brushes, rated at 500v DC, 2. You are feeding it with 460v AC 3. ????
I must be missing something. Has it got a built in rectifier?

wrote:
  Quoted Text Here  
HUH! I assume that you have some sort of rectifier in between so that the motor is actually getting DC and that you have limited the voltage to be sure that the motor rating is not exceeded.
I suspect that this is what Mike wants to know.

Don Kelly wrote:
  Quoted Text Here  
Sorry for the late reply. The machine has a digital DC drive that converts 460VAc to DC. We are currently running the machine at 75% load and therefore pulling 345VDC.
We have replaced the springs on the brushes and also confirmed using the Oscope the waveform at the armature. (clean();)
Also connected a AMP-probe to find out that the brush set that is wearing faster is pulling 10amps while the other is pulling 9 amps. I don't know whats going on.

remove the X to answer ----------------------------

  Quoted Text Here  
Ah, brush wear questions....
Rapid brush wear in general can be caused by a lot of things (off-neutral, atmosphere contaminants, light load, poor commutator surface, the list is long...).
But you only have trouble with one position. Has any other maintenance been done? (for example, remove/replace brush-holders, disconnect/reconnect any wiring).
What's the appearance of the brush face when you remove it? You can tell a lot from this. Parallel grooves imply rough commutator surface. Burning/pitting on leading or trailing edge imply commutation problems (spacing, interpole). Chip/flaking imply 'chatter' vibration (bad springs, high-mica, un-even commutator). Two perfectly polished surfaces, but at different angles indicate the brush is cocking at different angles in the holder.
Perfectly smooth, polished face but rapid wear mean 1) no/low current density 2) atmospheric contaminant (unlikely to affect just one) 3)wrong grade.
Some thoughts about *one* brush wearing differently. Check the spacing of all brushholders. We used to do this by wrapping cash-register tape around commutator, set down all the brushes, draw a line on the paper along the leading edge of each brush (pay *attention* to rotation and leading edge), remove paper and measure the spacing between marks (you have to also mark where the tape overlaps). If it's off even a little bit, this can cause some brushes to spark/arc much more than others.
Spring tension off. If the one 'bad guy' is in a difficult place, your workman may have 'cheated' a bit and not checked this. Some types of holders have 'constant pressure' springs, while others have adjustable tension.
Not the same current flow. Wiring of the brush 'pigtail' wrong. Check to see where it goes, and is it securely connected. With the machine shutdown, use an ohmmeter and contact the back side of the brush (not the holder, not the pigtail, not the bus bar, the *brush*) and measure back to the common bus-bar. Compare to readings for other brushes.
An interpole is bad. The commutation of armature segments as they pass under the brush is aided by 'commutating' or 'interpoles' between the main magnetic poles of the machine. If one is not carrying current, the brush associated with it will 'arc/sparc' like crazy. That would burn through the brush pretty darn fast. But it would have to be some wiring work that was done. These poles are normally all connected in series, so for *one* to not carry the same current as others means 1) maintenance reconnected it wrong 2) the pole winding developed a short. Before you jump on this one (it can be an expensive repair, and if it's not the problem you waste a lot of money/downtime), check the brush face for burning/pitting (bad commutation will leave this tell-tale sign). Simple resistance measurements of the interpole are not always conclusive since they are very low resistance normally.
Any inspection windows? Can you see any excessive sparking at this holder when running?
Are the replacements the right grade/hardness?
Oh, check the gap between holder and commutator. Too large and the brush will cock forward/back in the holder causing chipping/cracking.
Hope this helps...
daestrom

daestrom wrote:
  Quoted Text Here  
Thanks a lot for the great insight. I will keep you posted on my findings. Best regards S.