MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

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 Dragging

This might be yet another clutch bleeding thread but I hope not!

Newly re-built engine.
New clutch with an Alan Anstead release bearing.
New slave cylinder push rod (13H3655).
Master and Slave cylinders replaced along with the hose approximately 3500 miles ago.
New clevis pins.

I have a remote bleed point adjacent to the heater tap. I have done numerous bleedings with the Eezibleeder in all the configurations and combinations suggested in the archive, plus a few more: pedal held down, pedal not held down, car on the level, front end raised, off-side jacked up to bring the slave's bleed point to the high-point, parked on a slope at an angle with o/s uppermost. None of these methods have persuaded an elusive bubble to show itself. I haven't tried the glamorous assistant option as the archive suggested that pedal pumping was quite likely to just make any recalcitrant bubbles shuttle backwards and forwards.

The clutch is working and has allowed me to do a good number of test runs BUT while it slips into 1st and reverse with no fuss sometimes when it's fresh out of the garage and cold, after just a couple of miles, it grates if I try to engage 1st or reverse for a standing start. 1st goes in OK if I go via 2nd first but that doesn't work for reverse so I sometimes switch off, engage reverse and start it with the clutch depressed. Before the re-build the old clutch was working fine and going into 1st and reverse with no problems.

There is a small amount of play between the pedal lever and its clevis pin that equates to just over 1/4" at the bottom of the pedal lever. Could that be enough to be causing the problem? One post in the archive suggested that this might be the root of the issue with a new clutch that would not happen with a worn clutch.

Do you think removing the pedal to weld up and re-drill the hole will cure the problem?
C Mee

you don't mention if you've tried pushing the slave piston in along the bore and holding it there with a miniature G-clamp or a spacer between the end of the pushrod and the piston, and then bleeding. This method removes the space where the air bubbles get trapped.
David Smith

Thanks David, you saved me having to type exactly that!
GuyW

Dave, Guy

Thank you both, I've not tried that one, I thought I'd read somewhere that pushing the piston in too far can damage the seal but seeing as the method comes with two recommendations I'll give it a go tomorrow.

Many Thanks.

C
C Mee

No, it won't damage the seal. If you visualise it, the pipe in and the pipe out ( to the bleed nipple) are side by side at one end of the cylinder. So when bleeding, if the piston is down at the other end it leaves a whole cylinder pretty much out of the flow of the fluid. If there is air trapped there it doesn't get carried out to the bleed point. Pushing the piston back up the bore should eliminate this possibility.
GuyW

Just bear in mind that pattern parts can vary regarding construction and machining details. While on a FIAT the parts are very much the same as on a spridget, a guy local to me has an X1/9 and he had to replace a leaking clutch slave cylinder and initially he could not bleed the clutch, he just couldn't push fluid through the system, it was driving him mad. In the end he dismantled the slave cylinder and found the problem, the original cylinder was bored to near the end of the chamber so the piston and seal couldn't cover the fluid and bleed port, the new pattern part was fully bored and the seal covered the fluid port preventing fluid flow, a simple fix once understood as the piston needed backing off a bit so as to not cover the fluid and bleed ports.
David Billington

You don't have to have much wear in the pedal hole to effect the clutch operation
The telltale is pedal height, -if the clutch pedal sits lower than the brake pedal, there is a problem
IF you're going to weld it up, have a good measure up first so that you know where the hole was
They wear away from the master cylinder so don't get caught redrilling in the centre of the worn hole , the original hole will have been towards the master cylinder
I'd be tempted to, (if you're going to weld it up),to weld it and refit the pedal then mark the spot with the pedal sitting level with the brake pedal and even creep up the pedal 1/8" compared to where it was to get a tiddle more leverage/travel

willy
William Revit

If that hole in the pedal is worn, then the pivot bush may be too. You might want to add a grease nipple there if you have it apart. But I find it hard to understand how lost movement there would give the symptoms of a clutch that works ok when cold, but not when hot.

If there is still air in the system this will be more dense when cold, taking up less space and being dense, it is less compressible. As it heats it expands, takes up more volume but can increasingly be compressed down, so that might explain the differing operation of the clutch between hot and cold.

Other temperature controlled variables? Gearbox drag maybe? I assume you are using standard 20/50?
GuyW

Colin

An alternative to use of a G clamp is a bungy cord.

Once the car is on twin axle stands at the front so you can do the job from the front, crack open the remote bleed nipple, remove the R clip (if you have one instead of a split pin), remove the clevis pin and keep the pushrod in the slave and then put your bungy cord in the clevis pin hole in the pushrod and then push the pushrod in and attach the other end of the bungy to something suitable at the front of the car, in line with the pushrod. Then bleed the system. Do not forget to top up due to compressed slave piston once pushrod released and reconnected.

Cheers
Mike

M Wood

Thanks for all the advice.

I have tried pushing the slave's pushrod all the way in and holding it using a modified Mike Wood method via a long cable tie. The most awkward part of the exercise was removing the split pin that was too well bent back round the clevis pin; thanks for the R clip hint Mike, that makes it much easier.

Sadly no improvement! I'm repeating the exercise over night but don't expect it will be any different as I didn't get any bubbles last time.

I am fairly convinced that the problem is a combination of a new clutch with a strong, high spring plus slop due to wear in the pedal's clevis pin and pivot bolt holes. I'm going to order a new pivot bolt but the bushes are NLA. Can anyone tell me the dimensions of the bushes so I can get some from a bearing specialist. Alternatively, I've read that engine/clutch spigot bushes can be used, can anyone confirm this? I will certainly do as you suggest, Guy, and fit grease nipples.

I'll have to wait until I've got the pedals out to decide whether my welding and drilling skills are up to sorting the ovalled clevis pin hole or whether I need to give them to a machine shop to do.
C Mee

Use 2x crank spigot bushes (1A1559 iirc) in the pedals, perfet fit, one each end.
Paul Walbran

A 1275 clutch drag fault I have encountered a couple of times:
If the linings are very slighly oversize, or the hydraulic ratio wrong (too great, commonly if an 1100 slave is fitted) the diaphragm centre comes into contact with the plate at full pedal stroke, thus pushing the plate onto the firewall - hence drag.
When drag is encountered our natural reaction is to push harder on the pedal - which makes it worse.
In this scenario there is usually a short bit before full stroke where the plate is free before the diaphragm contacts it, so if dragging with full pedal test be easing off and seeing if the drag reduces or goes.
First time I found this it stumped me till i mounted it all up on the flywheel off the engine and used a press to activate while I watched what happenned.
Paul Walbran

Thanks on both counts Paul, I'll add some bushes to the order!

I'm aware of the over-centre possibility as it happens on my Morris with a mechanical clutch linkage. I haven't tested it properly on the Sprite as the bite point on the old clutch was very low. The new clutch is a Borg and Beck and, apart from wear and dirt, is identical to the old one.

I had to replace the engine back plate (My friend did it. Don't ask!) and the replacement turned out to be very slightly thicker (<1mm) than the original. I can't think that would have any effect on the clutch travel, though.
C Mee

hi colin

When I stripped the clutch out the last time there was evidence of the clutch 'overthrowing' as Paul explained. I slightly reduced the boss on the clutch plate to give it a bit of extra clearance. All was well after that.
Bob Beaumont

Colin,

I had wear in the clutch pedal see first photo.
It was elongated so I drilled it and fitted an oilite bearing and a new clevis pin, second photo.
Followed Guy's advice and fitted grease nipples to both the clutch and brake pedals.




Jeremy MkIII

Oilite bearing with new clevis pin and grease nipples on the clutch and brake pedals.




Jeremy MkIII

Thanks Jeremy. That's what I had in mind for grease nipples and the Oilite bushes look just the job. I think I might need to turn to someone with better engineering expertise to fit them for me though.
C Mee

Pivot bolt and bushes ordered.

The manuals suggest that the bolt can just be removed after disconnecting the push rods and return springs. It looks as though the angle strips with captive nuts for the pedal assembly are in the way.

Is 'simple' removal possible or do I need to disconnect the hydraulics and remove the whole pedal assembly?

There's no great rush now as I've just discovered some rust that will need cutting out and repairing before I can go out on the road again. Better I find it now than Mr MoT man or Mr Plod! Rather spoiled an otherwise glorious Easter Sunday.
C Mee

You can undo the pivot bolt far enough to disengage from the thread, but not then remove it as it would foul as you have said. One solution is to then file or grind one flat of the hex head down level to the shaft diameter, you can then just withdraw and replace the pivot bolt clear of the support angle irons. Leave the other flats as they are so an open ended will still work when you refit it.
GuyW

Thanks Guy.

After all the bleeding fuss I really didn't want to disconnect the hydraulics again.

C
C Mee

It's close, and you need to turn the bolt so the new flat aligns pretty accurately to the edge of the reinforcing piece, but it does work. Or at least it did on mine. I don't know how much variation there is between different cars. I also found the bolt quite tight to remove, even when it isn't siezed in position. You cannot easily drift it out as there isn't room to tap with a hammer because of the transmission firewall. I found the way was to get a ring spanner at an angle on the head so that it jammed in such a way that I could both rotate it back and forth and pull/ lever on it as well. Quite a struggle in the confined space of the footwell!

Hopefully, if it ever needs to be out again, my grease nipples on the pedal bushes will mean it will all come apart much more easily!
GuyW

It's been well over a month but I finally managed a run out today. I think the Sprite was as eager as me because it fired as soon as I turned the key whereas it usually takes at least three attempts.

The clutch worked perfectly and felt really positive with no slop and a much higher bite point. It can still graunch a little bit if you rush a change but if you're patient and let everything stop spinning it goes into gear smoothly as the clutch isn't dragging any more.

Spigot bushes are shorter than the originals but went in just fine. I doubt they'll wear out in my life time, especially now that I've fitted grease nipples. A note for anyoe considering this: make sure you allow for clearance between the cylinder yoke and the grease gun, I put the nipples centrally on the shaft and just got away with it! My drilling of the clevis hole wasn't as central as I would have liked but the oilite bush effectively regained most of the distance lost through wear.

Best of all, I haven't had to touch the hydraulics.

Thanks for all the advice, pictures and suggestions, you have helped to transform my Sprite's clutch action.

Colin
C Mee

15 months on and clutch drag has returned. :(

As before, when cold it goes into 1st and reverse as easily as one would expect. Once warm, however, there's enough drag that the gearbox internals never stop spinning with the clutch fully down. I've checked that it's not over-throwing as the gears are spinning however far the clutch is depressed. The issue is also causing problems with down-changes from 3rd to 2nd with it reluctant to slide into place, this caused a couple of 'moments' when hill climbing this weekend.

To recap: The master cylinder and slave cylinder were replaced not long after I bought the car. I can't find the invoice for the master (it has a large plastic filler cap) but the slave is definitely correct for the 1275, as is the shiny push rod that went in when I rebuilt the engine and fitted a complete new clutch and roller release bearing during lockdown.

I'm intending to try lengthening the push rod by adding a blob of weld on the end but haven't had time to try it yet.+

However, another thought occurred to me just now: My clutch fluid goes black very quickly following renewal whereas the brake fluid stays clean and clear all the time. Could the blackness be the rubber from the clutch hose breaking down? If so, and maybe the hose walls are thinning, could the engine heat be softening the hose so that it bulges when under pressure? The hose was replaced at the same time as the cylinders. When I first had the car I had a similar problem but that was resolved by removing the driver's carpet and sound deadening to give a bit more pedal travel!

I haven't been able to test my new theory yet but would welcome your thoughts.

Does anyone supply braided clutch hoses?
C Mee

Colin : Lengthening the slave cylinder pushrod. Before you add a blob of weld to lengthen it, try slipping a 7/16" nut onto the end of the rod. It will stay in place, nestled between the rod and the concave end of the piston. It's a quick way of checking what effect lengthening actually has on your clutch operation.

Secondly, think hard about what effect lengthening the rod has. It does NOT alter the amount of clutch thrust bearing travel. This is dependant totally on the amount of fluid displaced when you press the pedal. What then effects the thrust bearing travel is the amount of lost movement in the baroous mechanical pivots and linkages. All that lengthening the push rod does is to move the active zone of the piston movement further up the bore, but it doesn't change the distance moved. It can still be worth trying in the particular circumstance of the piston movement being too close to the end of the cylinder, especially if the cylinder has a limiting circlip in the end. But I think that is rare.

I think your concern about the Flexi hose is much more likely to be behind this fault.
GuyW

Thanks Guy, that's really helpful.

I've been going backwards and forwards in my mind regarding the logic behind lengthening the slave cylinder push rod. Your explanation is helpful, though I have read reports from folk who claimed it solved their problem.

I will try the 7/16" nut method and if I resort to the weld blob route I'll use the old push rod that I have, of course, saved in case it comes in useful.

After all that, I am, however, more convinced that it is the hose that's the cause of the problem. I fully expected Nigel to be the firs to reply with his usual comment about p*ss poor rubber products! A quick Google has given me several results for braided hoses so I know where to go if the 7/16" nut doesn't do the job.

Thanks

Colin
C Mee

ETA: typing whilst you were posting - note, I've not just pinned it all on possible piss-poor rubber as your hose may have been bought in a good period for this part, the question about now was not added in either, one trick pony, how very dare you.


Colin,
yes Goodridge do a clutch hose, PTFE inner with a stainless steel outerbraid. I've had their brake hoses on for 15 years now, one of the very first things fitted to the car as rubber parts were piss-poor back then (how are they now?).

So yes the black could be the rubber hose.

But then, the clutch fluid on mine is black despite me very thoroughly flush change it and cleaning the master reservoir 4 times since the replacement concentric slave, and changing the master cylinder last year, and not having a rubber clutch hose since 2007.

Based on stripping down my previous clutch master which was fitted in 2015 I believe these (921) modern made clutch master cylinder have the internal metal finish go black so may also be a contributing factor - or perhaps they're just getting coated in the contaminant.

https://www.goodridge.co.uk/products/car-clutch-kit-for-mg-mgb-bgt-125-midget-1962-1980?variant=39265501806657
Nigel Atkins

It's not just on Spridgets. My Ford clutch fluid turns black quickly too, but clutch action is perfect. And it's not just cars.

-------

"It's not just endemic to Ducatis guys. Hondas, Suzukis and Kawasakis do it too.

I asked why back in the 90's at the big bike dealership.

We surmised that since the clutch master cylinder displaces so much more fluid, the dirty fluid was getting dragged back up into the master cylinder reservoir. Just from the rubber seal rubbing against the bore, the fluid turns black from aluminum residue."

"Whatever the case, it seems to be true across not just different bikes but different manufacturers. My Moto Guzzi and Duc have had the same problem, as did my old Suzuki. "

https://www.ducati.ms/threads/why-does-clutch-fluid-turn-black-so-soon.67854/#:~:text=We%20surmised%20that%20since%20the,turns%20black%20from%20aluminum%20residue.


anamnesis

Good to know it's not just my Sprite then!

Interesting that it seems to be clutch only and not the brakes. Presumably it's because the clutch is used more frequently and because the shorter pipe run's volume is less so contaminated fluid from the slave cylinder is more likely to be drawn back into the master.
C Mee

Well, I've bled it and bled it and tried different push rods to no effect.

To me anything less than ten years old is New which is why I've never really given a thought to the master and slave cylinders that I replaced not long after I bought the Sprite in 2012. Maybe the problem's down to Nigel's p**s poor rubber products, so I've replaced all the rubber seals in both cylinders. After numerous bleedings with the car at all angles it's almost right but still not completely right.

It must be down to the throw of the clutch fork. It has just dawned on me that it could be down to the fact my 1275 Sprite came equipped with a Morris Minor gearbox. The clutch fork cannot be a MM one because that has a mechanical linkage, not a hydraulic one so the linkage end is different.

Checking the Moss website, I find that there are three possible clutch forks for Spridgets, two for the 948 and a different one for the 1098/1275. Could it be that my Sprite has the wrong fork and that doesn't give the same throw at the release bearing end as the 1098/1275 one?

Is it possible to identify which fork I've got with it in situ or will I only find out by taking the engine out?

Googled images suggest to me that the position of the hole for the pivot bolt is different.


At least I've got that off my mind, it was going to keep me awake otherwise!!!

Colin
C Mee

With clutch drag/suspected incomplete disengagement like this it is worth checking the stroke of the slave cyl pushrod. If less than normal the problem is befire the slave, if normal it is inside the bellhousing (suspect fork, faulty diaphragm, etc etc). That way you remive the risk of removing the engine needlessly.

The magic figure for the MGB is 11mm, with 10 as a drop-dead lower limit. The 1275 is similar but escapes me right now. Maybe someone with a nicely working clutch could measure and report.
Paul Walbran

The clutch fork arm can also bend; and, it can be bent back again. So don't throw it out if it is bent.
anamnesis

11mm rings a bell with me too. When building my concentric slave system I needed to know if the (ford) slave would give me the same movement at the diaphragm as the external Sprite one so I rigged up a bench test to check. I know I was surprised at how little movement is actually needed.
When checking the slave piston movement check too that the clutch forked lever isn't contacting the rear edge of the bell housing cut out, limiting its travel. The lever arms do bend with time and PO's big feet!

(sorry Anam, cross posting, but I think we made tge point!)
GuyW

Yep, I recall 11mm when I rebuilt the car. I was using an earlier 948/1098 slave with a 1275 engine/gearbox and had a problem with the clutch overthrowing.I measured the movement on the bench by coupling up a spare master. The larger bore 1275 slave cured the problem as well as making the clutch pedal lighter
Bob Beaumont

The fork didn't appear to be bent when I rebuilt the engine but that has to be a possibility.

I know, following Guy's explanation, that the lengthened push rod shouldn't make any difference but if the fork is bent, could that provide a temporary solution until I get the engine out again, hopefully over the coming winter?

C
C Mee

I think when they are bent, the outer end comes up against the edge of the bell housing hole, causing a premature stop, before the inner end has pivoted forwards enough to fully operate the clutch. Altering the push rod length wouldn't help.
GuyW

That is what I have found too. It is usually what causes the fork to bend. I suspect it happens due to the wrong offset in the release bearing (1098 or 948 bearing instead of 1275) requiring the fork to be further towards the cover than it should be to make up for the difference. It can also happen if the release bearing collapses.
Paul Walbran

9mm of travel measured to the slave clevis pin.

I think the next step is to have the clutch pedal out to weld up and redrill the clevis pin hole. My previous attempt at drilling it out to take an oilite bush may not have moved the pivot point back far enough from where the wear had taken it to.

Would anyone be able to give me measurement for the correct centre for the clevis pin hole?

Colin

C Mee

Clive,
you might be surprised how much slop you can have and the system still operate.


Nigel Atkins

And there I was thinking you were going to be helpful Neil!

That puts me back at square one. :(

Another bleeding session this morning freed up a few more micro-bubbles and the car is at least driveable, I just need to select a synchro gear before slipping it into 1st or reverse when manoeuvring.

My double de-clutching between 3rd and 2nd is getting better too.

C Mee

With the bleeding you're best with at least two things, time and patience, and not to let the reservoir fluid level get too low, so at least three things.

You've heard all the different methods people adopt for bleeding the clutch so you can, or have already taken your pick. I'm not much cop at any of this stuff but I've always found the one man and a jam jar method to work fine (as long as you don't forget to check the master cylinder fluid level).
Nigel Atkins

9mm of travel at the slave clevis sounds rather near to the lower limit of acceptable to me.

Do you have adjustable pushrods at the master cylinder? There may be too much 'take up' slack there. And does yours have the pedal lift springs in the footwell. If not, then they may be worth adding.
GuyW

Thanks Guy. It's a non-adjustable push rod, do you know if anyone supplies them because that would/should be a relatively easy fix.

The bite point is very low down and there does appear to be lost movement before you begin to feel resistance and yes, the pedal lift springs are in place.
C Mee

We have seen a few hard case clevises and pushrods like that. Yes they can be quite happy with the resulting reduced stroke when the clutch is old and the linings are no longer sprung apart, but when a new plate goes in and the linings do so, it requires the full stroke and the worn clevis becomes a problem.

Pedal clevis hole in the wrong place could indeed be an issue and worth checking. Try and get the top of the pedal as far forward as you can, the brings the bottom a bit further back and so more stroke before hitting the floor.
Paul Walbran

Similarly, a worn pedal pivot hole is bad news too. But I think you've checked / dealt with that?
GuyW

Colin would you like a spare pedal to play with? It might even be better than yours straight away. There's little demand for pedal boxes so I do have a box of pedals just taking up space....
David Smith

David, thank you for the offer, I'm definitely interested. Please will you send your contact details to me at meecclspa at aol dot com?

Thanks, Colin
C Mee

Colin,

my oilite bush helped considerably with the problem but the real solution was the master cylinder.

I got a NOS Lockheeds from Paul Hunt at Powertrack and that sorted the problem completely.

Had previously tried lengthening the pushrod (prior to fitting the oilite bearing) by making it adjustable (see photo) and that some effect but didn't cure the problem, especially when the engine was hot.


Jeremy MkIII

Thank you Jeremy

You've got me thinking! I still have the original master cylinder. If I remember correctly, I tried replacing the seals but, and I can't remember why, I bought and fitted the new master cylinder (plastic reservoir type).

I'll dig the original one out and take a closer look and maybe invest in another set of seals.

Cheers

Colin
C Mee

Thank you Jeremy for stirring the dormant memory cells.

Old master cylinder found, complete with un-used seal kit. It's a bit grubby externally with a bit of rust on the mounting flange but nothing that a few minutes attention won't cure and the bore looks clean and shiny though the seals are definitely past their best.

As soon as I picked it up I realised why I abandoned it, the clevis holes are completely oval, almost as bad as the ones in Nigel's photo, so I thought it wasn't worth working on. It was just my obsessive 'never throw anything away it might come in useful' trait that made me keep it safely wrapped up in its replacement's box.

There's a new pushrod on order (I didn't realise they were available 1st time round) so, just maybe, there's a glimmer of light at the end of this particular tunnel.

Colin
C Mee

Colin,

Two sets of seals were replaced as the thinking was that was the problem but resulted in little significant improvement. It is a better idea to begin with the seals though!

I also use silicone brake fluid (Automec) and that doesn't go black (or remove paintwork when accidentally spilt) and is much more resistant to absorbing water. However there are reports that it is not as easy to bleed the clutch with silicone when compared to mineral fluid. A vacuum bleeding kit makes light work of that task and is inexpensive.

Good luck, hope the seal change works for you.
Jeremy MkIII

This thread was discussed between 29/03/2021 and 30/08/2022

MG Midget and Sprite Technical index

This thread is from the archives. Join the live MG Midget and Sprite Technical BBS now