DIY Electric Car Forums banner

1 - 20 of 23 Posts

·
Registered
Joined
·
224 Posts
Discussion Starter #1
Brushes on my 11-inch Kostov wore down to the point of failure at 34,600 miles. One of the brush holders melted due to the heat generated when the brush springs bottomed out and the brushes began to arc against the commutator. The vehicle has only been in service for three years.

I was quite surprised by this premature brush failure. The brushes looked good last inspection at the 22k mile mark (one year ago). At that point the brushes appeared to be wearing evenly and there was about 1/16” of brush material extending above the brush holder (approx. 15/16” total length). At that point I figured they should have been good for at least another 30k miles. But over the last 12k miles most of the brushes lost over 0.75” of material!

I was racking my brain trying to think of what might be the root cause of this accelerated brush wear. One thing that came to mind was the fact that I moved and now have to drive on a couple miles of dirt road every day. I moved two years ago. Another thought was the occasional drive through flooded streets where the motor has likely taken in water. This has happened a few times over the past three years. I then did some research on DC motor brush wear and found several papers that mentioned the presence silicone as a possible cause. Here is one such statement:

“It is known that the presence of silicone will cause very rapid wear of carbon brushes. The wear rate can be many times the rate that would occur without the silicone present. The concentration necessary to cause the rapid wear is extremely small. It is less than a part per billion. The exact mechanism of the wear is not completely understood but the effects can be quite damaging.”

After reading this I immediately remembered, with a cold chill, something I had done. A while ago, a little over a year, the motor started to develop brush squeak (slip-stick) condition that was very noticeable under low rpm no load situations (coast). I blew out the brush dust with compressed air but that didn’t seem to help. So I applied some dry silicone lubricant to the brushes. Apparently that was a bad move! The silicone took care of the squeak but I’m willing to bet it was the kiss of death for the brushes.

Hopefully others can learn from this and not make the same mistake. I had no idea of how delicate the brush-commutator relationship was and how susceptible it can be to chemical contamination. Another point that was mentioned the papers I read was the use of silicone RTV in air ducts for motor cooling having a similar effect on brush wear.








 

·
Registered
Joined
·
7,791 Posts
Brushes on my 11-inch Kostov wore down to the point of failure at 34,600 miles.
....A while ago, .......So I applied some dry silicone lubricant to the brushes
Hi Darren,

Thanks for sharing. It may well have been the silicone lubricant. Any stuff put on the brushes and comm is bad, IMO. Dirt and water can be tolerated. It's best kept clean, but many motors work in dirty/wet conditions for decades.

Typically a small amounts of silicone gassing will not be a problem in well ventilated motors. With enclosed or poorly ventilated motor, yes, rapid brush wear. I can't say I ever heard of a case like yours, but my bet would be the lube.

Sorry for your loss :(

major
 

·
Registered
Joined
·
224 Posts
Discussion Starter #4
I got my Kostov motor repaired and the EJETTA is back on the road! I ended up getting new brushes custom made by a motor shop in Indiana: Flanders Electric Motor Service www.flanderselectric.com . I sent them a sample of the original Kostov brush for them to analyze and duplicate. A set of 8 brushes was only $115 + $10 shipping. The only difference that I can see in the construction is the shunt pigtails in the new brush are slightly longer and stiffer than the ones on the OEM brushes, other than that they appear to be identical. Good news for Kostov owners; Flanders has added this brush type to their catalog (Item No. 90003 - ML1285 Carbon). So those of you that are patiently waiting for Kostov K10/K11 brushes from Bulgaria you now have another option…

In addition to the brushes, I also had to replace the brush holder yoke assembly. Fortunately I was able to get one used from the good guys at Rebirth Auto http://rebirthauto.com. Kudos to Rebirth for coming through on that! The used brush holder came with a set of brush springs. I compared the spring tension of the two sets of springs and the originals seemed a little weak at 0.9 lb (2.4 psi) vs. the used springs at 1.7 lb (4.5 psi). From the material that I found on DC machines it seemed that the accepted range of brush pressures for this application and brush type was in the 3 – 6 psi range. Based on this information I decided to use the stiffer springs (4.5 psi brush pressure).

The damage to the commutator turned out to be not that bad. It cleaned up nicely with a little light sanding (no emery cloth or metal oxide sandpaper). I spun the armature at around 500 rpm and started with 100 grit and worked my up way to 1200 grit. I also verified the runout was acceptable and undercut was good. Before final polishing with the fine paper I deburred all the comm bar edges with a jewelers file. That was time consuming.

Getting closer to final assembly I shaped the brushes with 220 grit paper wrapped around the commutator (abrasive side out). I cleaned out all the abrasive dust and carbon with a shop vac and compressed air and ran the motor unloaded for 5 hours with a 12V battery charger to start the seating process.

Before final assembly I set the brush timing to the neutral position by using the AC induction method. This was accomplished by energizing the field windings (terminals S1 & S2) with a 30A 60Hz current source and adjusting the brush yoke assembly while looking at the AC voltage induced across the armature (A1 & A2). The voltage will be at a minimum (essentially zero) when the brushes are in the neutral position. I read that the accuracy of this method can be questionable if the brushes are not fully seated so I repeated this process several times with the armature in various positions. All runs yielded the same result which had the brush holder inline with the main field pole bolts as one would expect.

More on brush timing; I didn’t notice when I first took the motor apart but the factory position of the brushes was set at around 4 retarded. I noticed this while setting up the new brushes and saw the match marks I had put on the old holder assembly. Basically they had the brush yoke rotated almost all the way to the end of the adjustment slot in the direction of the motor rotation (CW facing comm end). I never thought to check brush timing when I initially purchased the motor since it was brand new and I assumed was set up correctly from the factory. Perhaps this could have been one of the contributing factors in the short brush life I experienced…


Comparison of Kostov K11 brushes: OEM brush on left & Flanders brushes on right.


End bell with new brush holder assembly. The factory jumpers connecting the two opposing sets of brushes are pretty anemic (8 AWG). I decided to beef this up a bit by adding another set of 8 AWG high temp wires in parallel.


Running motor unloaded @ 12V to start the seating process.


Final assembly of motor with brushes in neutral position.
 

·
Registered
Joined
·
3,930 Posts
Nice brushes. I think I may go with Helwig split brushes next time. I put two extra jumper wires on each set of brushes instead of the weenie single jumpers per set. What kind of amperage were you running through your motor?
 

Attachments

·
Registered
Joined
·
224 Posts
Discussion Starter #6
….
What kind of amperage were you running through your motor?
I’m assuming you’re asking about the current at the time of failure.

I ultimately became aware of the brushes arcing when I was driving the parking garage at work. Although I imagine it had been happening for a while prior to that judging by the melted brush holder. The motor current in the parking garage was minimal – probably less than 100 A…
 

·
Registered
Joined
·
3,930 Posts
No, I was more asking your amperage in general. Some folks have a pretty high amperage setting and actually drive quite aggressively so over all pump high currents through the motor. The Kostov is not the best for super high currents for long periods or lots of heavy quick stop and go's.

It was an issue that built up to the final failure.
 

·
Registered
Joined
·
1,700 Posts
Nice brushes. I think I may go with Helwig split brushes next time. I put two extra jumper wires on each set of brushes instead of the weenie single jumpers per set. What kind of amperage were you running through your motor?
What do extra jumpers improve or what do they help avoid?
 

·
Registered
Joined
·
3,930 Posts
DC motors have jumpers going to the opposing brushes. The jumper wire on the Kostov motor is pretty weenie compared to other motors like the Warp9 or my GE 9" motors. It helps reduce restriction and allows current to flow better and helps to reduce heat. The Kostov was designed for forklifts and not designed for the currents and voltages we drive through them.

That is why I put mine in. The stock wire just seems way too small in diameter to handle all that current. More than likely it really does nothing that is of any real benefit. I just think it needs more. If the others have more, why not.
 

·
Registered
Joined
·
362 Posts
Wow, great thread!

I've been fighting a problem with my Kostov 10 inch and I think it is related to my brushes and at least one brush holder that wants to hold onto the brushes too tightly.

Thanks for posting the warning about silicon lubricants. The thought of lubricating the brush holder ran through my mind earlier this evening as I was reinserting the brushes.... sure glad I haven't done anything about that yet!

REALLY glad you found a new source for brushes. When my problem first surfaced and I decided it might be the brushes, I ordered a set from ReBirthAuto. I've not installed them because I've only used roughly 1/8 inch in a little over 8000 miles. But nice to have a backup set so I don't have to way so long for them to come all the way from the factory.

What sort of tach sensor are you using in your Kostov? I have the Kostov version that isn't working very well... ok, not a all... and they even admitted the design wasn't the best. So I'm thinking about using a different sensor that will function with my Soliton Jr. while I have the motor out of my EV.

Thanks,
Peter H.
 

·
Registered
Joined
·
3,930 Posts
I have been under the impression that graphite is an excellent lubricant. I was also under the impression that silicon issues only presented with the new Helwig Split brushes with the new soft top and that the silicon actually damaged that soft top.

I don't believe that the silicon damaged your brushes. I believe it was just over heated. I have couple old brushes that had the terminal wires pop out of the brush. No silicon was used.

 

·
Registered
Joined
·
224 Posts
Discussion Starter #12

What sort of tach sensor are you using in your Kostov? ….
Peter H.
I have two rpm sensors on the motor. One for the tachometer uses the factory crank position sensor and looks at a 60-2 tooth reluctor wheel on the flywheel.

For the controller rpm I was using an automotive ABS type sensor which is a magnetic style two-wire device looking at a couple bolt heads mounted on the fan. It worked ok for the Soliton controller but would not register motor speeds below 200 rpm or so. This turned out to be problematic when I upgraded the controller to v1.5.2 software and the check engine light would now flash when I crept along at a traffic light or in a parking situation. This was a result of the new SW that warns the driver when the controller senses ≥50A of motor current and no tach pulse. It was more of a nuisance than a problem really but prompted me to upgrade to an inductive prox sensor when I had the motor apart. You can get one for $26 from Automation Direct http://www.automationdirect.com part No. AM1-AP-3A or $89 from Prick Jackhard’s online store…
 

·
Registered
Joined
·
362 Posts
Thanks for the feedback. I ordered one of the sensors from automationdirect yesterday before heading to work. Paid the extra for 2 day delivery in hopes it would be here on Saturday so I can put my rig back together. Planning to mount it in place of the sensor the Kostov factory sells, mounted to the inside face of the drive-shaft end of the motor. Just not sure of the appropriate gap.

I have a sensor in there, but not sure it is working. So rather than take a chance, I just ordered another that I'm more confident is in working order. I don't want to have to remove the motor again.

I hope to connect an o-scope to it before reinstalling the motor, to test the signal output. The old sensor was so improperly gapped it was generating a negative 1 volt pulse rathar than a positive 5 volt pulse which was clearly not adequate.

Thanks,
Pete
 

·
Registered
Joined
·
2,045 Posts
I was also under the impression that silicon issues only presented with the new Helwig Split brushes with the new soft top and that the silicon actually damaged that soft top.
Pete,

What I believe Tom Brunka said about silicon was that they thought it combined with the carbon and formed silicon carbide and it was this that caused the rapid brush wear.
 

·
Registered
Joined
·
3,218 Posts
Pete,

What I believe Tom Brunka said about silicon was that they thought it combined with the carbon and formed silicon carbide and it was this that caused the rapid brush wear.
Yes. This.

However, I've seen at least one motor manufacturer claim it is the acetic acid produced when certain types of silicone sealant/caulk cure that is the culprit. I'm not sure I buy that explanation, though, since acetic acid is a fairly weak organic acid and thus won't really react with carbon or copper (or the copper oxide in the commutator film). It *is* possible to make silicon carbide by exposing silicon and carbon to an electric arc, as described in this scientific paper:
http://jes.ecsdl.org/content/110/4/298.abstract
 

·
Registered
Joined
·
3,930 Posts
Pete,

What I believe Tom Brunka said about silicon was that they thought it combined with the carbon and formed silicon carbide and it was this that caused the rapid brush wear.
I sent an email and he sent me a PDF that I will post. My assumption was wrong.
 

·
Registered
Joined
·
224 Posts
Discussion Starter #17 (Edited)
1200+ miles on the new brushes, I’d say their fully seated by now. No noticeable brush wear at this point. One observation however is that these brushes are much louder than the originals. Not the slipstick noise as before (nails on chalkboard at low speed & unloaded) but more of a commutator whine (motor rpm*number of comm bars at all speeds loaded and unloaded). The noise has decreased substantially since the initial installation and is not really noticeable on the open road but can be heard when driving next to buildings or in parking garages.

Another observation is the intermittent wear pattern on the commutator - the leading and trailing edge of each comm bar appears not to be in contact w/ the brush. The burnishing marks on the copper are still visible on the first and last 1/16” of each segment. Notice the hour glass pattern on the edge of the film indicating a high spot on the center of each bar – see pic. I did check the commutator surface with a dial indicator and didn’t see any radial runnout. However my dial indicator only has a resolution of 0.001” and may not be able to detect a slight variance on the individual comm bars. I’m thinking this condition was caused by the sanding technique I used to restore the commutator. And perhaps this is the root cause of the commutator whine mentioned above…




One mistake after another; looks like I need to break out the comm. stones and try again…
 

·
Registered
Joined
·
1,560 Posts
Mine whines also. I just learned to accept the noise since it seems nothing serious is going on. Wakes up the shoppers in the Walmart parking lot since it kinda sounds like a turbine.
 

·
Registered
Joined
·
224 Posts
Discussion Starter #19
Thought I’d post an update. Shortly after my last post I ordered a set of commutator stones from http://www.carbonbrush.com/; med (90 grit), fine (120 grit) and polish (220 grit) for $12.00 each. I ran the motor unloaded at around 1200 rpm and setup the shop vac hose in the area where the stoning was to happen. After the first pass across the commutator with the medium stone I stopped the motor and took a look. Sure enough, only the center most portion of each comm segment was shiny which confirmed my supposition of low spots on the edges of each bar. I continued stoning with the medium stone until the entire commutator surface was shiny and true ~ approximately one minute. Then repeated with the fine and polish grade stones. I vacuumed out the brush area of the motor with the shop vac and followed up with a compressed air cleaning.

First test drive after the stoning I immediately noticed the brush commutator whine I was experiencing before was completely GONE! The motor was now as quite as when it was brand new. I’ve now put about 1200 miles on the car since the stoning and the latest motor inspection showed proper filming of the commutator and brush length virtually unchanged (5/16” of brush material above each holder).

This was another one of those DC machine learning moments for me and the art of proper commutator care. I don’t know what I did when I ‘touched up’ the commutator before with sandpaper to cause the high spots in the center of each bar. But I learned the hard way that there is no substitute for using the proper tools ie. comm stones instead of sandpaper…
 

·
Registered
Joined
·
181 Posts
To be frank, I think 33K miles is pretty good for soft pure carbon brushes.
Reason for accelerated wear in the last 12K miles is probably the fact that as the brush shortens with wear, its spring unwinds and starts exering less pressure which worsens commutation and leads to your result.
Not sure if your new brush type is the proper grade but if you send me 8 brushes we can test. In all cases an 8-set of original brushes is 50$ MSRP...of course this is Ex Works Bulgaria so add courier and 20% bank fee. I will speak with RebirthAuto so they start stocking some brushes.




Brushes on my 11-inch Kostov wore down to the point of failure at 34,600 miles. One of the brush holders melted due to the heat generated when the brush springs bottomed out and the brushes began to arc against the commutator. The vehicle has only been in service for three years.

I was quite surprised by this premature brush failure. The brushes looked good last inspection at the 22k mile mark (one year ago). At that point the brushes appeared to be wearing evenly and there was about 1/16” of brush material extending above the brush holder (approx. 15/16” total length). At that point I figured they should have been good for at least another 30k miles. But over the last 12k miles most of the brushes lost over 0.75” of material!

I was racking my brain trying to think of what might be the root cause of this accelerated brush wear. One thing that came to mind was the fact that I moved and now have to drive on a couple miles of dirt road every day. I moved two years ago. Another thought was the occasional drive through flooded streets where the motor has likely taken in water. This has happened a few times over the past three years. I then did some research on DC motor brush wear and found several papers that mentioned the presence silicone as a possible cause. Here is one such statement:

“It is known that the presence of silicone will cause very rapid wear of carbon brushes. The wear rate can be many times the rate that would occur without the silicone present. The concentration necessary to cause the rapid wear is extremely small. It is less than a part per billion. The exact mechanism of the wear is not completely understood but the effects can be quite damaging.”

After reading this I immediately remembered, with a cold chill, something I had done. A while ago, a little over a year, the motor started to develop brush squeak (slip-stick) condition that was very noticeable under low rpm no load situations (coast). I blew out the brush dust with compressed air but that didn’t seem to help. So I applied some dry silicone lubricant to the brushes. Apparently that was a bad move! The silicone took care of the squeak but I’m willing to bet it was the kiss of death for the brushes.

Hopefully others can learn from this and not make the same mistake. I had no idea of how delicate the brush-commutator relationship was and how susceptible it can be to chemical contamination. Another point that was mentioned the papers I read was the use of silicone RTV in air ducts for motor cooling having a similar effect on brush wear.







 
1 - 20 of 23 Posts
Top