Welcome to our resource for MG Car Information.



MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Clutch release bearing worn out

I replaced the release bearing assembly about 4000 miles ago. I lost this and just opened it up, the bearing is gone and the metal is broken. Any idea what I did wrong?? Must be something, or it would not have failed. Also, any advice on how to lub this bearing or prolong its life? 4000 is about 20 times too short
Frank Baker

I know you probably know this already but it has to be asked. Do you use the clutch a lot? Like sitting with the car in gear and the clutch in at lights and so on? That will wear them out very quickly. The standard one is a carbon bearing so self lubricating. The only lubrication I put on mine was on the release fork. The clutch came with some special red grease to use on there. Could there be something wrong with the clutch pressure plate causing damage to the release bearing?

Simon Jansen

Simon is correct and the fork has to be free in the bushes. Can you push the slave cylinder piston right back to provide clearance, if not there could be a master cyl fault. You don't by any chance have a heavy duty clutch fitted ??
Iain MacKintosh

No it is a borg and beck standard clutch. The clutch "looks" good, but I am wondering about it being warped?

It could be just bad luck and a product defect, but it is a quick and dramatic failure.

Yes, I think my driving habits contributed to this, but still...
Frank Baker

Was this a standard carbon faced release bearing? If so there is no way to lub it. The carbon is the lubrication. If it is a roller style these are normally sealed types. A lot in the archives about this style some good most bad. Driving habits have a lot to do with how long the release bearing will last.

Safety Fast
RJS Smith

Yes it was a carbon bearing. I understand from Brit-tek that they are no longer made by Borg and Beck, but are out sourced. I do not want to claim that my driving habits are great, but I think it is too short by far. My guess is that I got a bad part or I mis installed it somehow.
Frank Baker

If it has worn evenly I would say it was a bad part i.e. very soft carbon, which should go back to B&B. If worn very lopsidedly then installation could well be a factor.
Paul Hunt 2

It is totally gone, and the clutch pressure plate dug into the metal ring.
Frank Baker

take your foot off the pedal while cruising down the road. Problem solved!
Jimmy P.

Frank. I purchased two "clutch kits" sold under the B&B name a couple of years ago. By then, something called "Phoenix" was doing the actual manufacture and packaging. The throwout bearings were packed in with the driven plate and the pressure plate, but poorly wrapped. One of the throw out bearings was in excellent shape and the other showed cracks due to bouncing around while only protected by oil paper. I, then, contacted Bob Ford, at Brit-Tek, who went through is stock and found that 75% of the throw our bearings showed some form of damage.

The clutch pedal should, according to MG mythology, only be depressed for no more than seven seconds at a time. Also, the mythology records that a clutch used in this manner should last 80K miles by which time the throwout bearing is worn almost to the metal and the driven plate almost to the rivets. It is difficult to say whether this is true or not, since most of us only own one, perhaps two, cars. My personal experience is that the myths are true and the cars that I have owned, even in the days of my youth when I drove less moderately than now. (When visiting with friends, my younger daughter came over. As she left, there was a squeel of tires and a roar of exhaust. The friends both looked at me and one said, "The apple did not drop far from the tree, did it?".)

Thus, you either had a bad part or your technique was wrong. To correct the latter, only use the clutch for about seven seconds at a time, shift into neutral and release the clutch pedal at stop signs/stop lights and, if possible, avoid "jack rabbit" starts.

For the former, examine the new bearing under good light and magnification. (As I get older, the $20 circular light with a magnifying glass in the middle is my friend and companion.) If you see signs of cracks or chips, do not install it, but send it back for replacement.

I have never owned an MG that I have covered 80K miles with. I did own my original 79 for 65K miles. At that time, the clutch and throwout bearing demonstrated any signs of trouble.

Les Bengtson

My philosophy has been to treat the clutch pedal as if it were red hot and get yout foot off it in the shortest possible time consistent of course with smooth driving. The carbon thrust race was never really a problem in the UK but was ultimately replaced by a roller thrust in newer designs. The carbon however is still fine in the MG. Don't hover your foot over the pedal when on the move with the excuse " but my foot's not actually touching the pedal " IT IS and that's putting pressure on the thrust face and drastically reducing its life. We've seen that all too often here and it's a very difficult habit to get out of. Good luck.
Iain MacKintosh

My briskly driven MGB Tourer is in the family for about 25 years now, we've changed the clutch at about 70K miles intervals and the carbon release bearing was worn completely to the metal and the clutch plate also to the rivets.
In the 80's my father used his MGBGT daily (probably more carefully) and had no problems in 110K miles.
Willem van der Veer

Minor scratches, chips, gouges have no effect on the bearing life - it's not like a metal bearing. Rough or scratched surfaces on the table the bearing contacts can kill it though. I've seen pressure plates that were slid across a concrete floor eat a new bearing very quickly. The life of this thing is ENTIRELY a matter of driving conditions and driver technique, mileage means nothing. When I was doing a lot of them, in a hilly city, normal life was 50-70k. Extremes were 3-150+k.
Ian is dead on regarding the footrest syndrome - I used to go out with customers who "Never rested foot on pedal" but used up clutches. After 5 minutes driving, guess where the foot was? An aid in this case is a very heavy pedal return spring. It tends to reduce the "sit at the light with clutch down" situation a lot, and the people who need a footrest generally don't put enough pressure on to overcome the spring. A switch on the pedal that causes something annoying to happen is a good training device - hook it to a static generator so they can't hear the radio!
FR Millmore

I'm just throwing this theory on the wall to see
if it'll stick...

When you changed the clutch, did you also change
any of the hydraulics? ie: the clutch slave hose?

On the front brakes, it is not uncommon for a bit of
internal rubber on old hoses to flake off, drift about,
and cause the calipers to lock up. There are plenty
of threads on this BBS about this.

If the clutch slave hose is old, I don't think that it is
beyond the imagination that a bit of rubber could
cause the slave cylinder to retain an extra amount
of hydraulic pressure when at rest.

If this happened, then an extra amount of pressure
would be applied to the T/O bearing at all times,
and after while...poof!...the T/O bearing becomes

Maybe a hose change and a complete flush of
the clutch hydraulics are in order.


On another note...depress the clutch pedal only
when switching gears. Do not press the pedal
when starting the car, or when idling at a stop.

Not only does this wear out the T/O bearing
prematurely, it also wears out the crankshaft
thrust bearing. If you think that changing clutch
parts are a chore...try changing the thrust bearing!

I remember running into a webpage that explained
that oil molecules are great at reducing friction in
"slipping" applications - not "compression"
friction applications.

When the clutch pedal is depressed, the engine
thrust bearing comes under what is considered a
combination of both rotational "slipping" and
"compression" type friction.

Oil molecules seem the have a heck of a time dealing with compression friction and within
a few seconds, you'll have metal-to-metal contact.
Not good.

Stay off of the clutch pedal unless switching gears.

Daniel Wong

As I said before, my driving technique may not be the best - I drive it hard, but the clutch has never been a foot rest and 4000 is premature in any book. My current guess is that I was not getting a release out of the slave and that it was in contact (light or hard) 100% of the time. I think a return spring will pull it off the pressure plate
Frank Baker

Frank. Please review the postings by Fletcher Millmore and Danny Wong. Unless your bearing was damaged by impact, as one of mine was, and has a series of cracks in it (which, if installed would lead to catastrophic breaks), or, as Danny suggests, the flex hose is not allowing the pressure to release properly, there should be no problem.

The face of the throwout bearing is always in very light contact with the clutch plate (pressure plate). The pressure plate is the mechanism which moves the throwout bearing backwards as the cluch pedal is released. The hydraulic pressure, as the pedal is released, is overcome by the springs on the pressure plate, moving the throwout bearing backwards which results in the clutch fork moving forwards and pushing brake fluid out of the slave cylinder and upwards into the master cylinder. Thus, a spring arrangement, as used with roller type bearings, is not necessary with this type of system. I had a documented 65K miles on my original 79 B without problems. Willem has documented 70K miles of "spirited driving" between changes and 110K miles of "my father's driving" between changes. I would consider this to be a fairly good indication total destruction of the bearing and damage to the pressure plate is not going to be corrected by jury rigging a spring in the system.

You need to find the root cause of the problem and correct it, if possible. (Bad parts cannot be corrected, only replaced with better parts.)

Les Bengtson

Danny's point on hydraulic problems is valid, Another one would be similar seal swelling or debris causing the master cylinder to maintain pressure in the system. MC pushrod maladjustment or friction in the fork pivots would be possible as well. The bottom line is that something is maintaining excessive pressure on the bearing for extended periods of time.
Use of a return spring on the slave would require an adjustable stop to limit retraction to a small clearance, or an adjustable pushrod to allow the slave to bottom, acting as said stop. Either is undesireable and unnecessary if all is right in the works,
FR Millmore

Hi, Frank - just standing back, looking at the forest here, you have never said what year your car was, so this is a long shot, at best. If you happen to have a side-by-side master cylinder, is it possible that the bleedback valve was put in the clutch side instead of the brake side? That could put a little bit of constant pressure on the bearing, though I'd think the springs in the pressure plate and slave cylinder would negate much of that effect.

Like I said, a definite longshot - - Alec
Alec Darnall

Whew! Longer shot than I thought! Upon reflection, the MGB never used the side-by-side arrangement like the MGA and the early Spridgets. Still, a plugged bleed hole (the little one) in the Clutch Master could simulate the same issue.

Hope to hear how you make out on this one - - Alec
Alec Darnall

In my first post I mentioned checking that you could push the slave cyl pushrod back into the cylinder. If you can then this eliminates the slave, flex hose and master cyl at a stroke. There is always slight pressure between the carbon race and the pressure plate and this is intentional and provided by the slave cyl internal spring for the purpose of providing automatic adjustment. Spirited starts etc have no effect whatsoever on the thrust race, this only affects the disc, flywheel and pressure plate surfaces. Provided that the pivots are free and the steel thrust pad of the pressure plate undamaged then the cure rests in your driving style. Sorry Frank but that's the bottom line. I like FRMs idea about a warning device, it could be temporary only but would be very helpful. On the subject of depressing the clutch before starting the engine, this is essential to ensure that neutral is selected for safety reasons. Advanced driving techniques recommend that it be kept depressed when starting the engine again for safety reasons in case there in a breakdown in the selector mechanisms which happens to be the case on one post on the MGA board. As for the argument between relieving the gearbox load on starting or adding additional friction due to thrust race pressure I believe the jury may be out as there is nothing much between them.
Iain MacKintosh

Whilst fitting a return spring to the slave piston should remove any pressure from the release bearing (assuming there is free space in the closed end of the cylinder for the for the piston to move into) it is not a good idea. Without a stop to limit the movement (which together with a pull-off spring some have found necessary to prevent premature failure of *roller-bearing* release bearings) you could well find that it moves the piston so far the biting point becomes too close to the floor and may even grind when selecting reverse. You should also be aware that the slave cylinder contains a spring whose purpose is precisely to push the release bearing *up against* the cover plate to remove any free play that can develop with fore and aft crankshaft movement, which otherwise will give a varying and low biting point.
Paul Hunt 2

One thought I have had is whether the pushrod adjustment for the clutch pedal could come into play.
There should be a little free play or movement in the pedal before it starts to move the clutch MC piston.
If there was no play the pedal would be stopping the Master cylinder before the piston is all the way back.
I wonder if this could cause pressure to remain on the release bearing - not totally released. The same might be true if the pedal return spring is missing or broken. That little bit of weight of the pedal may be like resting your foot lightly on it.
Oddly enough I last changed the clutch on my '73 B in about 1982. I had bought a clutch from a local automotive supplier that mainly dealt in American cars.
(I also remember standing in line at those places and they made an absolute killing on carburretors, alternators, fuel pumps, and starters for those cars, so don't let anyone cut up Lucas and SU) Anyhow the clutch kit had a roller bearing type release bearing. The original carbon bearing was still good, but I installed the roller bearing anyway. I made no modifications of any kind and it is still going strong today after about 72,000 additional miles and 23 years.
I don't know the brand of the clutch kit, but I doubt if it was Borg & Beck.

One final thought. Frank, I don't suppose that you know if the car has a heavy duty clutch in it do you? I used to put heavy duty clutches in our MGB because if standard is good, then heavy duty has to be better doesn't it? One of the problems that I had continually had was premature throwout bearing failure. I just sort of chalked it up to MG and kept throwing heavy duty clutches at it. One day it finally dawned on me that the clutch in our daughter's MGBGT has a much lighter pedal than in our roadster and the threowout bearing in it had not worn out in many years in spite of her driving the car on a daily basis. I finally put two and two together and started using standard duty clutches in our roadster. I have enjoyed a light pedal ever since then and have not had to replace a failed throwout bearing since then. I think that my putting two and two together resulted in an answer of four for us. Good luck - Dave
PS. Sorry this is so late in the thread. I just got out of the hospital after having an enlarged (football sized, 6 lb) spleen. Actually, maybe I shoud have said...after giving birth to....
David DuBois

Dave. Glad you made it through without signficant problems. (If you can walk away from a landing, it was a good one is what they taught us in pilot training.) Your article on the dual fuel pump system is posted to the website, if you should care to review it. From my perspective, I could install the pump following your directions. This should mean that "any other idiot" should be able to do it also. Glad you made it trough Okay and thanks for the work you did.

Back when we transitioned from the Minuteman I system to the Minuteman-Modernized/Command Data Buffer system, we were taught that one had to "prioritize" one's actions. To have finalized a tech article, with photos, while awaiting surgery would, to me, indicate such an ability.

Thanks, Les
Les Bengtson

Les - And a great landing is one after which you can fly the plane again, so I think that using that definition, I had a great landing. Thanks for the good words. I am moving rather slow and thinking about every ove before makin it, but in all I am feeling realy good. Cheers - Dave
David DuBois

Well, It was a standard clutch in a 1965 B with a Ford Type N 5-speed kit. I pulled the engine and the clutch pressure plate is fine with some very minor wear, but no damage. The pressure plate is fine - no real wear. It was only the throw out bearing that failed. I hate to deflect blame from myself (It is always possible that I am guilty of some stupidity, I would not want to compound it through denial, but...) but in this case - I think it is a bad part. The slave is fine - no problem. The hose is new and it is stainless braided. So if the function of all the parts is good, and the wear is minimal - It was not a "shared experience" between the part pushing and the part being pushed - bad part.

I also saw a Pantera over the weekend with a spring mounted on the slave and through the slip pin on the fork. Very close to the MG bell housing - slave mounting setup - I think I will look into that, but I think it is un necessary.

Bad part - 10+ hours of labor, I will replace the clutch ($150) and I had the car down for 3 weeks - Oh joy.
Frank Baker

5 speed conversion.. All bets are off. Ford 5 speed? Different input shaft? Different Clutch fork? MGB bellhousing machined to accept different tranny? I would look at everthing.. Is everthing square? Postion of the slave on the bellhousing? Slotted mount? Whole bunch of issues now. Least of which maybe driving habits. I have a Datsun 2000 5 speed that fits a early 18GB engine used a Datsun Disc, standard MGB plate Datsun roller bearing with pull off spring basically not a MGB setup. Whole different ball game


Ron Smith

Whilst you might now be on your own Frank and the issues might be different, I'm not so sure that it's that different, Is the dimension form the front face of the transmission to the face of the fitted thrust race the same in both forward and retracted position or at least nearly so. Is the thrust absolutely concentric on the input shaft at fork mid travel position and is the face of the thrust absolutely square with the shaft in all directions. When fitted, does the slave fit easily and can the piston still be pushed further back into the bore thus ensuring clearance. Is the centreline of the clutch fork exactly in line with the centre of the pushrod when it is in nline with the cylinder centre line or is it kinked and finally is the cylinder mounted firmly and exactloy parallel to the centreline of the transmission.

Might be enough there to keep you going for a while !!
Iain MacKintosh

Hey all,

Have a type 9 conversion as well. Clutch slipped when the engine was warm or hot.

Thrust face still has carbon on it, but is worn very thin for 10,000 miles. My suspicion is that the geometry of the conversion is not quite right, leaving the stock slave pushrod with no free play.

I am going to verify this for sure, and then shorten if required.

Shareef Hassan

Actually the clutch pressure plate, fork, slave, and throwout bearing are MGB - dead stock, The bell housing is exactly the same on the fork mount and other areas - The transmission bolts up differently. The input shaft runs in a different bearing, but is the same diameter. This bearing does not care about what is connected to the input shaft - It is down stream from the clutch.

I have a spring pulling the bearing back now - I will tell you how it works
Frank Baker

I found one bit of humor in a GERMAN service manual. It was an original VW factory manual. It was suggesting various methods to stop throwout bearing wear. It stated that if the ball bearing replacement for the carbon ring was also wearing out that the driver should be advised to find another foot rest.
This was in a Combi manual for 1957 VW service vehicles.
Sandy Sanders
Sandy Sanders

This thread was discussed between 20/09/2005 and 28/09/2005

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.