It's approx 200miles since my new engine went in with new exedy clutch etc and it's been fine until now but the other night the clutch went nasty.

Pedal is very soft and whilst the clutch still kind of functions the useable range has gone right down so the instant the pedal is moved away from being tight against the floor it bites. It's barely driveable.

My homework leads me to believe in most cases this would be a knackered/leaking slave cylinder, but I don't think that's the case here.

Basically the clutch fork is totally loose to pull forwards and backwards by hand (ie far enough forwards to disengage from the slave cylinder, so beyond the normal stopping point).

From memory the clutch fork should be firmly in contact with the slave on account of the clutch working pressure (the springs on the pressure plate), since I know you have to 'lock' it into place like that after mating engine & box from when I've done that before.

I have a horrible suspicion that the only way the fork can be free to pivot back & forth like this, is if the big circlip thing inside the clutch (think it holds the bearing to the pressure plate?) has come undone somehow.

Is that right? Or is there another thing that could cause the fork to be 'free'? Like I say, I'm sure it's not the slave cylinder since it wouldn't make the fork go loose just because a seal had gone etc.

And assuming this bloody clip has come loose (how?!), does it mean I have to take the engine out again to fix it? I don't have access to a ramp so I don't think it's practical to attempt to remove the box from below rather than lifting out the motor?

Thanks chaps.
Jim

 

Hi,

the clutch forks on these do tend to be loose, and there is a sping that hooks on to the fork to hold it forward.
If the circlip had failed you would have no clutch at all as these are a pull type clutch.
It is likely that your slave or master cylinder is failing. To check the slave, pull the rubber boot back at the fork end of the slave cylinder, if this has any fluid in it then it will require replacing.
Also while your checking this also make sure the slave cylinder mounting bolts have not come loose, as when i replaced my clutch i had to pull the cylinder forward before tightening the bolts otherwise it did not contact the fork correctly.

Stu

 

Thanks for the reply mate. I've yet to check that (hopefully tonight) but the fork really is far too loose and there is no resistance to it at all.

I checked the fork after the last time I drove it so it might be that there now is no functionality as such. Like I say the fork can move forward past the point where it would normally stop (ie well past where the slave output bit falls out the back of the fork!), and this couldn't be influenced by a fault with the slave from what I can figure out.
Normally when once the circlip is engaged the fork has something like an inch or so of play but it can be moved fully forwards & backwards while feeling like nothing is really attached to it.

So yeah I'll be checking the slave behaviour etc as you suggested, but I'm not hopeful.
Assuming the clip has somehow come undone (is that at all common/how could that happen with a new one?), is there any way of re-engaging it without dropping the box or lifting the motor out?

I've tried pushing the fork backwards at the top and knocking it firmly, but no joy.

 

Hi,

Mine also has alot of movement, it may well be worth loosening the slave and pulling it as far forward as possible and re-tightening to see if that helps.

The clutch on the Impreza differs from conventional clutch systems as the release bearing pulls on the pressure plate instead of pushing on it to disengage the clutch, hence the reason for the clip to fix the bearing to the pressure plate.

If the clip had come away you would not be able to operate the clutch at all, as when you push the clutch pedal down the clutch fork pulls the release bearing to disengage the clutch, if the clip had failed then the bearing would pull away with no effect to the pressure plate.

To engage the release bearing with the preasure plate you just push the fork back as you have already done (I thought mine had not engaged due to the movement when I replaced my clutch, but it had, they just have alot of freeplay).

It is possible that the release bearing has disengaged from the pressure plate, but this would probably be down to component failure as the way it clips together is very secure (When you take the gearbox off, you have to remove the clutch fork first as it is not possible to disengage the bearing from the pressure plate in situe).

If there is a fault in the hydraulic system, it could cause the pin on the slave to retract slightly, so checking the condition of the hydraulic components should be the first port of call.

Stu

 

Well, even though I was quite convinced that it was the spring clip that had come loose despite fully concurring with your argument Stu, I've (fortunately!!) turned out to be wrong. I think.

I was going to post up that I agreed with your comment about "It couldn't be disconnected if there is still some capability to use the clutch", but I thought it had since progressed to having no clutch action at all.

Last night I got the chance to check it out so I thought I'd confirm first that the clutch was indeed not now functioning at all. Pressed the pedal and it went to the floor, this time with no bit of useful clutch at the end of the pedal travel. I pulled the pedal back up (as I had done several times already) and suddenly it was fine again. Full pedal has come back.

Now as I say your argument makes perfect sense and I agree with that theory entirely, but I'm still amazed that the fork could have been so loose and feel like it was not attached to anything if the release bearing was still correctly clipped in place in the pressure plate. It doesn't make sense to me since whilst there'd obviously be some play/movement there would also be resistance due to this connection with the plate.

Am wondering if it's possible that the clip came undone, and has now just snapped back in place (like it does when you first engage it after mating engine & box, since I've had the 'pleasure' of doing that job a few times now over the years!).

It shouldn't since it was a new clip with the rest of the clutch, but now it's working again there's a lot of resistance to the movement of the (top of the) fork which was the big alarm bell for me when that was missing, since it suggested it was not connected to the pressure plate any more. The fork's ability to be moved by hand is totally different now that the pedal is working again.

But then if that was the case, it would not have had the residual clutch operation like you've said. So I'm still a bit confused about what was the cause.

Naturally I'll be bleeding it as a precaution now, and if air comes out then it's pretty much proof the slave is all it was, but if there's no air in the sytem (assuming the master cylinder is also ok) wouldn't that indicate that it was indeed the snap ring for the pressure plate at fault?

Cheers
Jim

 

Hi,

It is good to here that it is working again.
I am assuming that the excessive fork movement that you were experiencing was with the slave still installed?

The reason I ask is that when I replaced my clutch I had a similar problem in that when I pushed the pedal down it stayed on the floor and I had to manually pull it up, which happened each time I operated it. At first Like you I thought it was due to the clip not being engaged in the pressure plate as there was a large amount of movement in the fork, but it turned out that the slave just needed to be pulled closer to the fork (there is a small amount of adjustment in the mounting holes of the slave cylinder).

I don't think your problem Was down to slave position, as it worked previously in that position and does again now (assuming you have not moved it), but I am wondering if there is a fault in the hydraulic system that caused the operating pin on the slave to retract, and now it has pushed back out to contact the fork.

This could also explain why the movement in the fork is now reduced, as if the slave is now contacting the fork correctly it would restrict the amount of freeplay.

Out of interest, did you pull back the boot on the slave to see if there was any leaking fluid?

I think bleeding the system would be a good idear, perhaps even change the fluid if this has not been done for a few years.

Stu

 

Hi Stu

Yes that movement was with the slave still installed. I haven't touched that yet. I'm definitely going to bleed it as soon as I get a chance just to make sure it doesn't have a load of air in it, so I'll check the positioning of it at the same time.

That's actually the opposite of what 'appeared' to have happened! The pin seemed to be extended really far forward from the slave, to the extent that it was kind of flopping around, almost as if it was actually a cable coming out of the slave!

Nah. I was sorely tempted to, but with that being one of the ver few components of the car that I've not previously had in pieces over the years I got the better of curiosity. It's taken many years of bad ideas to accomplish that though!

I've owned the car for approx 10 years of its 21 year age, and certainly never touched the clutch hydraulics so I'm sure it's due a little bit of TLC.

Appreciate your input by the way. I'm still doing this though in terms of quite what was going on...

We have two theories, both of which seem to be logical and fit the symptoms/circumstances, but are mutually exclusive, so it doesn't make a lot of sense.

 

Hi Jim,

By pulling back the boot, it will not cause the slave to come apart, as it is just a dust cover (and also keeps the pin in line). I recomend checking this as it easy to do and will help to rule out a leaking cylinder.

Let me Know how you get on.

Stu

 

Will do. Not sure when I can do this but hopefully over the weekend if I don't get nagged too much....