Unstable Carbon Brush Temperature

Currently we are facing some problems about carbon brush temperature. It's not stable and the MVAR also increased.
Our machine specification GE 9FA and the generator is hydrogen cooled generator. specification

There are 48 carbon brushes in the generator collector end (The temperature and current of the brushes are unstable and MVAR also increases). I wanted to know proper reason behind this.

I think the question is not good enough to elaborate my query.

Thanks in advance
Can you relate incident(s) to:
o Sync-Gen Load changes?
o Atmospheric changes?
o Ambient Temp changes?
o Operational changes?
o Time of day, wk, month, shift?
o Coolant Changes?
o Scheduled maintenance change?
o What shows up on instrumentation?
o Is a similar machine in operation?
Regards, Phil Corso

The brushes "transmit" (conduct) the current--supplied by the excitation system (the "AVR")--to the generator rotor windings by their contact with the slip rings.

If the current is unstable, then it's likely the output of the excitation system is unstable.

More current means more heat--regardless of the type of conductor. More amperes flowing in/through/around a conductor causes more heat in the conductor. Carbon brushes are a form of conductor.

Many times carbon brushes can be affected by humidity, as well as oil residue/vapors. This causes the surface which is in contact with the slip rings to glaze over, which can reduce the amount of current being transmitted, causing the excitation system to try to "force" more current to make the generator terminal voltage setpoint (or MVAr setpoint if the unit is being operated in VAr control mode, or Power Factor setpoint if the unit is being operated in Power Factor Control mode) equal to the setpoint.

You have not said if the brushes were recently replaced--or if any of the brushes have been replaced during the troubleshooting of this problem. In fact, you haven't said what troubleshooting has been done, or more importantly, what the results of the troubleshooting were.

You have not said when this problem started; suddenly, or has it increased over time, or after a maintenance outage.

There is an old, but very effective, method of "polishing" the slip rings--called "burnishing" the slip rings. It involves taking heavy cotton canvas (such as from an old, cotton canvas fire hose) and attaching it to the end of a piece of wood (something like a length of 2"x4") and then pressing it against the slip rings while the generator rotor is spinning for a period of time. This removes build-up from the slip rings and improves the ability of the brushes to conduct (transmit) current to the slip rings. (This was also done to DC commutators when DC motors and generators were still common. In fact, I think that's where the practice began.) It used to be done when the generator was on-line--but in today's safety conscious world that would probably be a great, big, giant NO-NO. But, the unit could be started and run up to FSNL and excitation removed so that no current was being applied to the brushes from the excitation system and the burnishing tool could be used much more safely. I'm sure you can find details of this tool and methods on the World Wide Web using your preferred World Wide Web search engine.

When new brushes are installed, they should be properly prepared. In the "old days," new brushes were rectangular and had flat ends as shipped from the manufacturers. A "mandrel" or "jig" was made that was the same shape as the slip rings, and a piece of sand paper was attached to the mandrel/jig. The ends of the brushes which would be in contact with the slip rings were then slowly rubbed back and forth over the sand paper to make the end have the same shape as the slip rings--to better transmit current by having fuller contact between the end of the brush and the slip ring. (Nowadays, many of the more well-known manufacturers of slip rings pre-form the ends to a slight curve--but unless they are purchased from the generator OEM or for a particular generator they probably just have a curve, and one that's not exactly matched to the curvature of the slip rings.)

Another condition which can occur over time on a synchronous generator with slip rings and brushes is called "brush bounce." This results in the brushes "bouncing" up and down in the brush holder as the slip rings are rotating beneath them, and this results in poor contact between the brushes and the slip rings. This is best determined using a strobe light as it's very difficult to see with the naked (or safety glasses-protected!) eye. I imagine some high-speed video cameras may also be able to capture the phenomenon. It usually results from improperly maintained slip-rings, and once it starts it can accelerate the problem. Some times brush bounce can be made worse by increased generator rotor shaft vibrations. Has the generator rotor shaft experienced any change in vibration level(s) recently, before or since this problem started?

Lots of generators experience oil leaks along the generator shafts. Either the hydrogen seals are worn and not working properly, or the generator bearing "sealing air" system isn't working properly. These can cause oil to leak out of the end shields along the shaft and either directly contaminate the slip rings and brushes, or vaporize and indirectly contaminate the slip rings and brushes.

And, as was mentioned, an increase in humidity (or entrained moisture in the cooling air passing over the slip rings) can cause problems.

Brushes and slip rings are very much and very often thought to be a "set it and forget it" thing--meaning that once they are replaced they don't take any maintenance. That's very far from the truth, as can be seen from the above few more likely problems with brushes and slip rings. This is why it's important for operators to periodically pass by the collector and look carefully at the brushes and slip rings. Usually, there is a light source in the housing over the collector (slip rings and brush assembly) with a switch. That switch should be turned off, and the operator should look for sparks--which SHOULD NOT be present. This is in addition to monitoring the air temperature of the air entering AND leaving the collector housing (usually by one or more RTDs in the collector cooling air supply and discharge air paths connected to the Mark*). And, looking for oil leaks and contamination, as well as any water leaks.

So, if you require more assistance, you first need to tell us what troubleshooting you have done--AND what the results of the troubleshooting was/were.

Then you need to tell us when the last time the brushes were changed (some or all of them).

Then you need to tell us if the collector area is contaminated with lubricating/seal oil.

Then you need to tell us when the problem started.

Hope this helps! Please write back with more details, and to let us know how you progress in resolving the problem. If you want more help, you will need to provide more information. There isn't one possible cause--there are several, including problems with the operation of the excitation system (which was not discussed in any detail above).

You conspicuously did NOT tell us what troubleshooting has been done at site regarding this "problem" and what the results of the troubleshooting were.

There are usually RTDs (Resistance Temperature Devices) mounted in the collector housing, monitoring the temperature of the air entering the housing and the temperature of the air leaving the housing. Since you have a 9FA, it was most likely built by GE-Belfort (France) or to specifications supplied by GE-Belfort (France). Unfortunately, GE-Belfort has decided that the typical things GE usually provided for decades were not always necessary (in GE-Belfort's esteemed opinion only) and has changed tried-and-true standard equipment, usually for more modern and complicated devices and programming. And, they have also been changing what were established signal names, so, what might have been DTGCHnn (where "nn" is a whole number that represents the RTD number from the generator manufacturer) is now who knows what name, if it even exists. But, I would venture this is not something they would change--but when it comes to GE-Belfort, all bets are off and they can, and quite often do, change things just because they can.

DT would stand for Detector Temperature; GC would stand for Generator Collector; and H would stand for Hot (as is discharge air; DTGCCnn would be the Generator Collector inlet air temperature).

If your machine has these RTDs, or some similar device, you would mostly likely find them on the Generator display on the HMI, where the other generator temperatures (stator temperatures) are displayed.

You said the problem started a few months after an HGPI (Hot Gas Path Inspection) was performed. Technically, an HGPI is procedure that only involves the turbine, not the generator. Many times, generator work is also performed when a scheduled maintenance outage is being performed. Was any work done on the generator at that time, and if so, what work was performed on the generator?

I don't think I said MVAr increases in relation to carbon brush temperature, so I'm not exactly able to understand what you are saying. If MVAr increases (I am presuming by increasing MVAr you are referring to a lagging MVAr increase), presuming the grid voltage is relatively stable at the time the MVAr is increases, that means that more current is being applied to the generator field through the brushes and slip rings, and increased current flowing in any conductor means the conductor temperature is going to increase.

I didn't ask how you are measuring the temperature of the brushes.... Can you please tell us how the temperature is being measured?

AND, how long has the site been monitoring the brush temperature? Since commissioning, or just since a new hand-held temperature detector was purchased and made available to the outside operators making rounds?

What "mode" is the generator being operated in? By generator mode, I mean is the generator being operated in automatic Power Factor control mode, or automatic VAr control mode, or simple AUTO Voltage Regulator mode--or, possibly MANUAL regulator mode?

How stable is the grid voltage?

Are there any alarms associated with the exciter ("AVR")?

Is this a constant problem, or just occasional?

Again, no response to the question of how the brush temperatures are being measured.... Specifically, has the site been tracking brush temperatures for any period of time, other than just since the recent outage, or just before the recent outage.

The generator OEM is the best resource for brush loading calculations.

It is possible that the brush composition has recently changed, or that brushes from different manufacturers (the OEM builds almost nothing anymore, outsourcing most things, with very little quality assurance required in their bid specifications). It's simply possible that this "batch" of brushes is some physically different from past batches of brushes.

Alignment of brush holder rigging is also important.

You really haven't provided much actionable information/data. You say the temperature is unstable, but you won't say how you measure the temperature(s) or what the temperature range is, or for what period you have brush temperature data. It's interesting that "so much" work was done on the slip rings.... Was it normal, planned maintenance? Was it something the OEM suggested doing and the site agreed? Was it something that was prompted by some other reason--brush temperatures, a high usage/wear rate, something else???

Best of luck with your issue(s). Please write back to let us know what you find.

I can suggest another website/forum for help with this problem: IGTC has a forum that is free and gets a LOT of generator-related questions, and a LOT of respected people who post replies and comments and even anecdotal information. The URL is:


They also have an archive of past threads, so you might start by researching them using their 'Search' feature. Unfortunately, the website is not very user-friendly, so you will have to click around to get registered (free) and find things, but for generator-related issues and questions it's quite good.