MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

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 Midget and Sprite Technical - Clutch problems

1275 midget, competition clutch with roller bearing.
Car had been running well for a couple of years with no clutch problems.
Week before last the slave cylinder gave up the ghost. Managed to drive home (30 miles) mostly motorway.
Changed slave for one the same as came out. Unable to get a pedal afterwards despite bleeding both ways. Noticed fluid in master cylinder was dirty so have changed that (cylinder)as well. Now have a good pedal BUT clutch does not fully disengage. Lots more bleeding with no air apparent (have extended bleed line). Slave pushrod has been hacked about at some stage and appears a bit longer than standard .
Any ideas why it does not fully disengage now?
David k Brenchley

Isbit the exact same make and size as the one you replaced... you never know with a PO and a modified push rod

What about the clutch peddle ... was the holes and the clivis pin worn ? Thats fairly common issue and can make the clutch not fully disengage ... my thought is it could have been on the edge of becoming a problem and then once you distrubed it, then it became what it is...once that clvis pin starts wearing egg shaped then it puts pressure on the sheet metal holes it runs thur and they become egg shaped and then they both just grind on each other...and it will go down hill quick

Also id Try switching the push rod from the old to the new slave... just a thought there is a reason that has worked that mod worked for you up until now. No point in reinventing a modifed wheel

Prop
1 Paper

David,
The clutch fork arms have a habit of progresively bending until you get to the point where they contact the edge of the hole in the bell housing before the clutch has fully disengaged. This could have happened if you had to press overly hard on the pedal of a failing clutch, when driving those 30 miles home.

The check is to get an AA to sit in the car and press the clutch to the floor, whilst you lok under the car and check the clearance on the pedal. If it is touching the bell housing edge of the hole then it has bent. :-(
GuyW

Prop, The push rod is the same one as came out (slaves don't come with pushrods over here). As I said replacements where the same as came out. No obvious wear on peddle or pins.
David k Brenchley

Guy, when depressed there is a clearance between fork and bell housing.
David k Brenchley

That's good news David, as if it were bent it would be an engine out job!

The piston travel, and therefore the disengagement is purely a function of the fluid displaced at a single pedal stroke. So if no wear in the various mechanical points it should work, assuming all of the air is out. (did you push and wedge the slave piston right back up inside the bore whilst bleeding?)

I believe some cars have an adjustable pushrod at the master cylinder. That could take out some slack. Is that set right?

There are also 2 diameters of slave. If you have used the fatter one then the travel will be less. It may not be immediately obvious.

If the slave piston is operating too far down the bore one trick is to insert a 5/16" nut into the inverted cup shape of the slave cylinder so that the pushrod nestles into it and is extended a little. The shapes will keep it in position pretty well, but its only a temporary fix. Bear in mind that the slave piston only moves around 1/2", whilst the bore is nearer 2". This allows for the self adjustment as the clutch wears, with the piston moving further down the bore as the clutch wears, whilst retaining the same actual stroke.
GuyW

Guy, seems i need to wedge and bleed tomorrow, bugger, it is a 1" slave which i think is correct for a 1275?
Will order a 7/8 slave if no better tomorrow.
David k Brenchley

David, off hand I forget the sizes. I know the early one has the bleed nipple at an oblique angle whereas the later ones it is a right angles to the body. I may be leading you off at a tangent anyway as I think the early one is the smaller diameter version with a longer stroke.

One slightly troubling thought is if a PO had apparently lengthened the push rod, or at least "doctored" it, perhaps they had already had a similar problem and had attempted to botch a repair which has now failed.

Incidentally, wedging the slave is just a suggestion. I find that it gets over the potential prolem of bleeding through the system, with fluid flowing into the slave and immediately out of the adjacent bleed nipple, inconveniently leaving stagnant unflushed fluid - and air - sitting undisturbed within the body of the cylinder!
GuyW

****One slightly troubling thought is if a PO had apparently lengthened the push rod, or at least "doctored" it, perhaps they had already had a similar problem and had attempted to botch a repair which has now failed. *****

Thats what drew me in as well...Mods are great as long as they are your own

Its tough to eyeball the differance between 7/8th and an inch

Would jacking up the cars front end make a differanse for bleeding the slave? ...i cant remember were the bleed nipple is located on the orgianl slave its been so long

Prop


1 Paper

Prop - its cast on the side of the cylinder. Most people jack up the rear of the car although i think jacking up the front is good as it raises the cylinder bleed so the air will travel up hill?
David k Brenchley

David, in this photo, I've sliced a clutch master cylinder right through the bleed hole and the orientation is correct for a midget - inclined at ca 47 degrees. The bleed hole is not shown, but it is near the bottom of the cylinder. This shows that bleeding from below may not expel all air easily - a bubble will be left at the top. I believe that tilting the car may help - but not just any old tilt. To get the bleed nipple more VERTICAL, the front of the car should be jacked up as high as possible - but only ONE side - the same side as the clutch slave cylinder (the driver's side for a RHD drive car, the passenger's side for a LHD car). Do you see what I mean? Have I explained this properly? I'm investigating this problem right now but I have not yet tried the experiment. Please do tell me what happens, if you are able to attempt it. Thanks. My theory is that this is a real design flaw - the bleed nipple should, of course, be VERTICAL - but it isn't anywhere near that!! Pressure bleeding is often necessary to force the air out by swirling things around - but my hope is that a proper tilt might do the trick easily! John


J.E. Davies

John, if that is sliced through the non-vertical bleed hole, I guess the second sentence "The bleed hole is not shown" should say that the "Feed hole is not shown". A typo?

If the bleed hole is at an angle, leaving a crescent shaped area above the bleed hole that might contain a bubble of air, then pushing the piston back up the bore as far as possible as I suggested to David, will reduce the volume of that cavity in the roof of the cylinder
GuyW

AFAIK the slave cylinder is on the right on a left hand drive car.
Dave O'Neill 2

Yes David; as John says, that is the pasenger side on a LHD car. ;-)
GuyW

Sorry all, GuyW is right - I meant the feeder hole is not showing in my 1st photo (it's in the bit that was sliced off). In this second photo, the clutch slave cylinder in shown attached to the gearbox - is this clear now? The arrow points to the possible (probable) air bubble. John.

J.E. Davies

Ah yes, confusing.

With regard to the port angles, I believe that the inlet port is angled on the early (pre-1275) cylinder, as it is the same item as used on the Mini, whereas the inlet on the 1275 cylinder is straight, although offset.

http://www.moss-europe.co.uk/slave-cylinder-clutch-gsy110.html
Dave O'Neill 2

Dave's photo highlights extra confusion in this area. The 'feeder' pipe can be attached to either hole, some photos show the bleeder valve attached to one hole, some to the other. I can certify that one can bleed the clutch successfully no matter which hole the feeder is attached to - but the feed should surely(?) be the lower hole, the bleed valve at the top. Just like brake bleed valves.
J.E. Davies

Good morning everyone . Well I pushed the slave in and locked it and then jacked the car up so the bleed was as near vertical as possible ( thank god for hydraulic ramps) and had another bleed. Tiny air bubble removed. Now waiting for a proper pushrod in the post.
I'm beginning to think that the slave leaking was a red herring.
As I said I have the Peter May uprated clutch cover with the finger plate removed and using his roller bearing. You cannot see much inside the casing and no bits seem to be rolling around.
Has anyone who has fitted these bearings had to lengthen the pushrod?
Has anyone had the bearing wear out - About 10000 since fitted but it does have a hard life doing regularity rallies.


David k Brenchley

The other thing to consider if you get clutch misbehaviour (can be slip or drag) is whether the plate could have got contaminated with fluid.

AdrianR

Yes david

Ive been sort of down that road

I have the datsun 5 speed kit from riverside

In that kit there is a roller bearing to replace the carbon block that rubs directly on the fingers we have to remove the rubbing block from the fingers also

The kit also comes with an (adjustable) push rod

On my situation i also have tilton 3 masters set up that is also completely adjustable.. so there was a big learning curve on how to adjust everything correctly from scratch and i did some good wear on my throw out bearing in only a couple thousand miles ... and didnt notice untill i blew my engine and had to rebuld it, but the premature worn out roller bearing would have showed up fairly soon had i not blown the engine

It dosnt take much contact between the roller bearing and the fingers to wear them out fast

At the time i was also notourios for riding the clutch peddle at stop lights...A HUGE NO NO for these types of roller bearings ... so watch that also

Note... the adjustable push rod from river gate is a piece of threaded 8 grade rod the same size dia as the push rod and 3 nuts ... 2 to postion and secure around both sides of the fork..1 nut adjusted agianst the slave to ride and block agianst


Prop

1 Paper

Well it looks like the roller bearing has failed so out with the engine ASAP.
David k Brenchley

Sometimes its easier to just do it and be sure. 45 minutes is my record, from deciding it needed to be done to engine + box sitting on the garage floor.

One piece removable front helps with access, and that doesn't include picking up all the tools.

AdrianR

Well its all in bits now, the roller bearing is complete and still rotates, but has worn the fingers almost through in some places ( seems to have been engaging at one point first all the time. The clutch finger plate outer is cracked and the clutch plate is worn and torn up in places.
Ho hum, wonder how long the new set up with an australian release bearing ( that is noticeably lighter than the uk one) will last. All seems to engage much more smoothly and at the right angle.
David k Brenchley

David,
Do check the condition of the clutch fork arm pivot. If it is worn or loose, with the extra weight of a roller release the end "nods" down and then operates eccentrically to the clutch diaphragm springs and this causes them to wear badly. Its worth checking that it isn't worn, and go further by adding washers as shims to take out as much free lateral movement whilst still of course allowing it to pivot freely.
GuyW

Well that sucks

I'd build it back and recheck it on the bench thru the access holes and make sure it was within margins before reinstalling

I guess you could rig up a longer line off the master to operate it

Or maybe ??? Use a vacume hand pump to actuate the slave ... ( im not sure about that idea)

Prop
1 Paper

The further the clutch bearing is from the pressure plate the greater the angle and sweeping action across the plate. For example if a flywheel has been lightened across the clutch face than the bearing starts off at a greater amount of off centre than with a non-lightened flywheel.

The angle of engagement exists because the clutch arm is on a pivot. The movement of
the bearing at the end of the arm removes the angle but consequently creates a small or large amount of off-centre engagement. The more the flywheel has been skimmed off the clutch face side and the more the plate itself wears the greater then sweeping action of the engagement.

Or at least that's my understanding. It's one reason why the co-ax/concentric slave is a such a dream to use on a type 9 box. With the standard gearbox car your choices are more limited.


Daniel

Guy - new arm fitted. prop, done that.
Daniel , yes the cover plate looks like that but also cracked on the spring plates behind. New bearing gives original throw so appears to have less “droop” when operating.
David k Brenchley

Watch also for over throw of the clutch release mechanism. That has dire consequences for the clutch, and especially for the cover fingers and for the driven plate which eventually fractures. Ask Bill SDGPM about that !! That said, it seems unlikely unless you have a long standing mis match between the master cylinder and the slave cylinder.

Daniel's photo suggest to me that a flat faced roller release bearing has been used. There is a very definite edged wear groove on the fingers where they have pivoted over the angular edge of a release bearing. The Peter May bearing has a rounded face where the fingers make contact. The rule of thumb used to be to only use a flat faced release bearing if there was a corresponding flat pad mounted on the fingers for it to match to, but to remove the pad if a round faced release bearing was used.
GuyW

My bearing was round edged. The one other thing that effects bearing engagement is the base section it's built on. Alan Anstead is probably the expert on this whole issue.
Daniel

That's surprising Daniel. The photo does make it look as if the wear has produced quite a distinct sharp edged step.
GuyW

My pressure plate looks just like Daniels - also using a round edged bearing.
David k Brenchley

That's what I have is a round edge bearing

It really sounds to me that your over throwing the push rod pin

Is so, the easiest way to correct would be to find the spot of disengagement on the peddle then install a bolt and nut under the clutch peddle as a stop

What I did .. was Jack the back end up had someone turn the back wheel while I depressed the clutch peddle and watch peddle position as it relates to where the tire just started to rotate freely then adjusted that but nut and bolt stop o that position plus an extra small amount for good measure...1/2 inch ?



That worked great until I finally figured out how
To adjust the clutch peddle position on my tilton set up
1 Paper

The photo and the results are deceptive because what you are the clutch fingers with no load on them. When the fingers are being pressed in by the bearing, the 'cut' shape is different.
Daniel

Daniel
I am no expert on clutches: far from it. The release bearings that I make are a direct replacement for the carbon thrust type and all are expected to contact with the clutch cover pressure pad hence I cannot comment on the problem David is experiencing.
Alan

Alan Anstead

Alan - you may not consider yourself an expert but I bet you have all the data on the differing bearing carriers for carbon bearings and so could tell what original carrier was used for mounting a roller bearing? So when any well known supplier retails a roller bearing for both 1275 and 1098 engine/gearboxes you would be able to tell which carrier it uses (one or the other).
Daniel

From memory the Peter May bearing for 1275 sits on a 14mm deep Mk 1 Sprite carbon thrust bearing casting. That is the type with a roll pin that everyone complained about the carbons fracturing.
The 14mm deep casting will, usually, with either a bit of boring or sometimes alternatively sleeving, accept a deep groove ball bearing onto which I fit a mild steel faceplate to be used on the Mk1 948..Shown is the casting type with bearing fitted but minus the faceplate.The roll pin hole is visible to the left hand side.
Alan


Alan Anstead

This thread was discussed between 28/10/2017 and 19/11/2017

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.