Hi guys,

After a long time again i posted something here!!By the way,there is good news and bad news!
The good news is,i got a new job in a new company..

The bad news is we have 2 DC Motors here which consumes lot of brushes..The electrical guys are saying that the brushes are breaking frequently due to vibration and i dont have an analyzer here to get a reading!!The commutator surface is clean and the spring tension is also ok.

Can anybody tell me what are the possible causes of frequent brush breakages?

Thanks in advance!!

Have a nice day!!

Are they breaking or wearing?

See problem M here
http://www.reliance.com/prodserv/motgen/h7000ch3.htm

General info on DC motor brush life here:
http://www.reliance.com/prodserv/motgen/c7090/

Jenish:

What do you mean by the commutator surface is clean? Is it pure copper color? Or is there a glaze of carbon on it about the color of straw? If it is pure copper color, then you most likely have brushes that are too hard.

You may also want to check the comm bars to make sure that none are raised and that the mica is properly undercut and beveled.

It takes a tremendous amount of vibration and low spring tension to cause brushes to break.

I have also covered DC motors and brushes in the July, August and October issues of Uptime. July and August specifically deal with brush and comm issues and contain the appropriate tables.

Howard

Hi guys,

Thanks for your quick responses!!

About the commutator, what i quote was just the comments from the electrical guys here!!I couldn't have a meeting with them further..So i assume that commutator surface clean means the film is uniform as usual!!

Thank you Pete for your valuable documents!!

By the way, what are the possible vibration signature expected when the brushes have problems..I saw 360Hz(SCR frequency) as high as 1*RPM.(By the way i got a reading from another guy.)Is it indicate a problem in SCR?

What are these brush rig frequencies and commutator frequencies mean? Sorry,itz the first time i got a headache from DC Motors!!Ha haha..

Thanks again!!

Have a nice day!!

From physical reasoning:
1 - Number of commutator segments times running speed could be prominent in vibration if one bad/damaged brush.
2 - Number of brushes times running speed could be prominent in vibration if one bad spot on the commutator (assuming brushes are equally spaced).

Look for harmonics of above frequencies (for example 2*segments*running speed) for more confidence.

360 hz is associated with ripple on the dc as you know. Faulty scr could certainly cause increase in ripple, but even a healthy motor/supply may have a high 6*LF.

Thank you pete,

we have 5 brushes/arm and i think total 4 arms in this motor...
What could be the frequencies if the motor is running at 1456rpm?

So from your experience,how could we identify a faulty SCR if it is always the prominent frequency even for a healthy motor?

I read somewhere that arcing will cause variations in the speed and we will get 1*RPM and 2*RPM peaks with side bands around 1X.In this case i saw the same in the NDE but at DE the 2X is very small.I couldn't check the sidebands as i had access to the report only.What sidebands around 1X we can expect if the brushes are arcing too much?

Have a nice day!!

Dear Jenish,

Vendor Warning!

Breaking brushes are generally caused by one of three phenomenons and are usually a combination of two of these:
1) Weak spring force – be careful of reports of “good” spring force unless it is actually measured. Pulling up on shunts to feel if the force is good or not is not very scientific and older OEM specifications always called for lighter brush pressure as it was felt that mechanical wear was more of a problem than electrical wear. In more modern times, we have all performed testing that shows electrical wear is more damaging to brush wear AND commutator wear. In other words, check brush pressure again and if a normal industrial application, target 3 to 6 PSI.
2) Heavy vibration – if you can’t measure, you have no idea what you have. Brushes do not like high vibration nor do motors so getting some meaningful data quickly will tell you if you have issues and what components of the system need attention. In heavy vibration applications, traction motors or mining equipment, we can add features to the brush to minimize high vibration conditions such as heavier shunt wires and pads on the spring end of the brush. Once breakage starts, you will have a runaway situation and need to address quickly.
3) Friction chatter – this phenomenon happens when the rotation of the commutator is into the brush toe and creates an unstable mechanical condition. This is extremely prevalent in paper mills. If you are running a radial type brush arrangement (typical for European manufactured designs), this should not be an issue unless the holders has gotten off-kilter. Usually this condition is on reaction type (angled) holders and the long side of the brush will be the first edge in the face of rotation. If you have this condition, the first tell tale sign will be fraying of the shunts on the brush as the brush is chattering in the holder a very high frequency. You will also see evidence on the thickness and width faces that looks like shiny marks.

I agree with Motor Doc on commutator conditions – commutators should have an even conductive film to allow for smooth brush operation. Be wary of terms like harder or softer brushes – hardness has little effect on brush wear and is only a measure of surface deflection. Flexural strength is more related to wear and stronger brushes will handle high vibration applications.

We are a global company and will most likely have a technical representative that is responsible for your area. I can also send you some presentations if you would like. You could email direct at rod.acklin@morganplc.com

Good luck!

Jenish,
Has any work been performed recently on the drive(s)? Has the drive been scoped? You should be able to "see" scr pulses on the output side of the drive.
Relating to the mechanics of the motor: the papers attached by epete/ along with motordoc's post are pretty good places to start. --I had written a response that echoed parts of both posts, then had sense enough to go back and read what others had written.
What size are these motors? What sort of load are they seeing? (current?) What environment do they run in? I am looking forward to any responses from your last question as well:
I would think if it was arcing bad enough to be picked up on the nde bearing with a standard accelerometer (100mv/g); you would soon have other problems on your hands. But, once again, I am always eager to hear from (&learn from) others.
Thanks, Jeff

This message has been edited. Last edited by: Jeff4,

I think my comment about number of brushes times speed was incorrect - brushes would probably not be equally spaced.

6*LF is the normal frequency of ripple coming from a three phase full bridge rectifier. The degree of filtering varies. The less filtering, the more ripple and more vibration at that 6*LF frequency on the motor. With a dc voltage time waveform, you could distinguish whether you had normal ripple (would see dips in the voltage at 360hz) or an problem with an scr (would see a bigger dip in the voltage every 6th dip).

This message has been edited. Last edited by: electricpete,

On almost all DC motors, brushes should be equally spaced with the exception of “V” type holders used on some GE designs. These have a leading angle brush and a trailing angle brush but if you measured the same orientation brush, they also should have symmetry.

Hi guys,

Thanks for all your comments!!

By the way, i could be able to collect some more information about this problem from our mechanical and electrical guys who were directly involved..
The motors which we were talking about are the Stand #8 and Stand #13 motors of our rolling mill and they are of 570KW and 748KW respectively..

R1-8:-
Make: SICME
440V(Armature),310V(Field),1398/1398/1226A(Armature),18.6/9/7.8A(Field),860/1400/1600RPM.

R1-13:-
Make:GE
750V(Armature),220V(Field),1050A(Armature),50A(Field),1050RPM.

OK. Now the Scenario!!

Electrical:-The tip of the brushes are getting chipped and sometimes the wires are getting broken.The commutator surface was thoroughly checked for raised bars and mica.We have blower on the commutator side for colling it.You know the atmosphere is not that much hot during these months(comparitively) but anyway it will be always hot in a steel plant!! Probably we will try to take an IR image of the commutator and see how hot it is!!Again it will arise another issue of what emissivity to be used!!Ha ha ha... Can anyone suggest any range of temperature for commutator? Can i assume that above 120ºC is a fairly hot commutator?

No modification had recently done on the drives..

The SCR frequency is prominent, but levels are low and could be seen in both ends.

Commutator surface were choclate brown and uniform as per them. Spring pressure is at the recommended levels as per the manufacturer!!

Mechanical:-
There was not so big vibration levels as per the reports. The overall level is around 1.5mm/s.I dont think this vibartion levels will affect the brushes!!

Thatz all for now..

Have a nice day!!

Jenish,
Where do you measure vibration? What kind of bearings do you have? At such a low speed shaft displacement would be a better indication.

Jenish,
Out of curiosity, the cooling fans have filters on them don't they?

What edge is getting chipped Is it the leading edges?
And to follow up on what Acklin wrote above, what is the direction of rotation of the motor to the angle of the brush holders?

Dave

It can sometimes be of value to check vibrations on the brush bridge using a isolating plexiglass stinger with bolton accel of "suitable" length. Great care should be exercised as opening the door introduce oxygen to the carbon dust and sparky environment and flames as long as 1-2 meters can be the result, so never stand in front of the door..... and be careful. Don´t do as the plant I worked in (nuke) where they had a fire hose to cool the carbon brushes in the summer in the 6-700MW gen with water mist until it went full waterflow into the brush fixture. Water was not noneconductive enough you could say, but it was a long time ago. Much fun stuff we did in those days would not be allowed nowadays, for some odd reason. Olov

Hi guys,

Thanks for the comments!!

About the position of the vibration measurement, unfortunately i don't have any information. But I think they took it from the outer casing near the bearing.. I will try to contact him for these details!!

Dave, of course the cooling fan have filters..
The brushes are chipped on the leading edge. About the rotation again i have to go to my electric guys!!

OLI, you made me frightened!!
When i went for taking the IR images, I was standing just in front of the door at most 1m away!! Thank God, nothing happened!! Ha ha ha...

I will come back again with more information about my IR images..

Thanks again!
Have a nice day!!

Jenish,
It´s just in the short moment when you open the door you should be very careful, when the air rush in to the brush area, after that it´s no problem. I have not heard of worse than eyebrows being removed but it is a bad experience and it´s from guys that work with these brushes and do that every day. I made a stinger like that for them and they warned me to be careful when using it, part from the electrical issues. So I only tried it on special cases. Olov

Jenish

I think the brushes are breaking at leading edge due to raised bar. (And no, you cannot see confirm raised bar just by looking at the commutator).

You mentioned 5 brushes per arm and 4 arms. Are the brushes breaking only in one position across the axis ? e.g. Only the second brushes in all the arms are breaking.

You can use an insulation rod to detect the raised bar while running. Run the motor at slow speed, place the insulation rod on top of the each brush in one arm. If any of the brushes (in one arm) gives a knocking feeling, then you have a raised bar. You can confirm the raised bar by placing the rod on the same brush in the another arm.

Jenish,

Checking for a raised bar is good advice and can be easy to do if you can isolate the motor from the roll. When down and locked out, you could dial indicate the commutator. There are also electronic devices that will measure run out and give you a polar plot but again, the machine needs to be isolated and turned mechanically. Not easy to do if hooked up to a rolling mill.

Don't discount friction chatter as it looks like you might be experiencing this phenomenon. You are running a GE and an Italian design motor. Typically, any European designed DC motor is going to be a radial brush holder design (perpendicular to the contact surface), but it looks like your SICME is a GE type design and has an angled brush holder arrangement. GE licensed this design to Ansaldo in Italy and possibly others. The “V” holder arrangement, half leading and half trailing orientation brushes, was very popular for mill motors and worked very well in a lot of applications. In general, if you have a leading edge type of holder arrangement, there is the possibility of friction chatter due to loads and / contamination.

Since you are seeing leading brush edge breakage and shunt fray or breakage, my guess is that you living through classic friction chatter. A more detailed description is a follows: the coefficient of friction between the brush and the commutator or collector is high enough to cause the brush to jump up and down, much as chalk does when squealing on a blackboard. The commutator need not be rough to cause this problem!

Carbon has higher friction when it is cooler (light loads). Some grades have higher friction than others, and certain contaminants increase friction (sulfur dioxide, for example). Some brush holders designs will tolerate higher friction than others. For instance, friction chatter is more common with leading holders, where the rotation approaches the brush towards the acute angle of the brush face.

The solution is usually found in using lower friction brush grades, eliminating contamination, using grades that work in contaminated atmospheres or by changing holder design (trailing vs. stubbing, etc.).

How many % of max load is motor used? How are motors used constant speed or varius speeds?
(dry air combination with to little currrent per quadrat inch of the brush increase bruch wear)
Are every second brush more worn?
Do you have a photo of brush arms/collector?
Br
Monitor man