Page 2 of 4

Re: Very odd clutch problem

PostPosted: Wed Mar 18, 2015 12:57 pm
by Asolepius
Just did a bit of research. The slave is not the problem, as it's bigger than the 948 one so won't travel as far. The more I think about it the more baffled I am.

Re: Very odd clutch problem

PostPosted: Wed Mar 18, 2015 7:58 pm
by Sprite33
How much free play is there between the TO and the clutch diaphragm? It could be the boot is pushing the TO bearing to far away. I don't remember everything I did when I switched to the 1275 clutch but it seems I also used different push rod. I don't know if that would make any difference though. I did use a 1275 slave but I also used a matching master.
Are you sure the fluid doesn't have air trapped in it somewhere? Sometimes it can be difficult to get it completely out. I was taught a trick by a fellow racer. I used a syringe to pump fluid in from the slave bleeder and push it out the master. That worked well for me when I had trapped air in the clutch hydraulics.

Re: Very odd clutch problem

PostPosted: Wed Mar 18, 2015 11:58 pm
by Charlie Tolman
On one of the Internet forums, someone had a problem with a new clutch that would not release all of the way, and would not totally dis-engage the flywheel from the transmission. A problem to select first gear, obviously.

The clutch disk had a greater than normal length to its central splined hub, or the input shaft of the transmission did not have its grooves cut far enough towards the body of the transmission, so that the clutch disk could not move backwards (towards the tranny body) far enough to release the coupling between the flywheel and the tranny input shaft.

The problem was remedied by removing some material from the disk hub, reducing the length of the hub.

Charlie T.

Re: Very odd clutch problem

PostPosted: Thu Mar 19, 2015 12:29 pm
by Asolepius
Thanks guys. I now have everything out of the car again, and have made the following observations:

1. With the engine/gearbox assembled, the release fork travel seems shorter that I remember. Of course, I needed a very long lever to move the release fork, and because of the over-centre action of the diaphragm I could only check engagement at either end of the travel.

2. With the flywheel and clutch assembled on a test rig involving the trolley jack (don't ask!) the movement of the pressure plate seems very short. At some point during its compression the friction plate is released, but at full compression it is not free, ie it starts clamped, releases, and then binds.

So this clutch, a genuine Borg and Beck one apparently, has a very short throw and is exquisitely sensitive to adjustment of the slave cylinder. My next question is, what is the correct length of the slave pushrod? To reiterate, the setup is:

948 block
Thin engine backplate
1275 clutch kit
1098 gearbox
7/8" master cylinder
948 slave cylinder

Forgot to mention, before heaving everything out I checked the clutch pedal movement. It stops about 1/2" from the floor, which it didn't do with the old clutch.

Re: Very odd clutch problem

PostPosted: Fri Mar 20, 2015 12:55 am
by Charlie Tolman
On my 1963 H/P Sprite, the combination was:

948 cc engine, thin engine backplate, 1275 clutch with 1275 throw-out bearing, rib-case transmission, 1098 master cylinder ( I do not know if it is different from a 948), and a 1098 clutch slave cylinder (I believe this is the same as a 948). Quite similar to your combination.

I found that a longer than stock push rod for the slave cylinder was necessary. I do not remember how much longer, but I had fabricated three longer push rods, being 1/4", 1/2" and 3/4" longer than stock. The push rod that I used was at least 1/4" longer than stock. The stock push rod probably was from a 1098 slave cylinder.

If the clutch pedal does not depress all the way, it is possible that the piston in the slave cylinder is bumping-up against the circlip (retaining ring) inside the slave cylinder body. If so, install a longer push rod.

Victoria British lists push rod lengths of 2.5" for the 948 and 1098, and 3.0" for the 1275.

Charlie T.

Re: Very odd clutch problem

PostPosted: Fri Mar 20, 2015 5:57 am
by Asolepius
The 1098 master cylinder is smaller - 3/4 not 7/8". That would give longer pedal travel and lighter pedal. Yes the 948 and 1098 slaves are the same. I think I should have the 1098 master. The 7/8 master is pushing the slave too far. I have already sleeved the brake bore for discs and can do the same for the clutch bore. I turned up some pistons on the lathe so these are in stock. Meanwhile can I get away with adjustment? Problem is that I really need to try all this before putting the engine back in, but impossible without a hydraulic connection.

Re: Very odd clutch problem

PostPosted: Fri Mar 20, 2015 10:57 am
by Charlie Tolman
A 1275 throw-out bearing ( release bearing ? ) was not mentioned in your list, but it is a very important component in the clutch linkage.
I was not sure if a clutch kit included the T-O bearing.

Charlie T.

Re: Very odd clutch problem

PostPosted: Fri Mar 20, 2015 11:04 am
by Charlie Tolman
Hydraulic connection ?

How about several feet of braided flexible hose ( or line ) ? AN-3 or -4 .

CT

Re: Very odd clutch problem

PostPosted: Sat Mar 21, 2015 11:40 am
by Asolepius
Yes the T-O bearing was in the kit so the correct one was used. The engine is back in now after much measuring and calculation. Not quite ready to test but I'll post a full account soon.

Re: Very odd clutch problem

PostPosted: Sun Mar 22, 2015 9:36 am
by Asolepius
Well, it works. I probably went about it the wrong way, but here's what I did.

The movement of the clutch from fully clamped to fully free is 8mm. This translates to 11mm at the release fork end. I then made an adjustable spacer (nuts, bolt and big washer) to go between the slave and the fork end, to give the 11mm movement. I assembled the gearbox to the engine with 2 bolts and did them up, which pushed the fork back and released the clutch, after adjusting the spacer. I now knew that the clutch worked, and shortened the pushrod accordingly.

But after the engine was back in the car the pedal still moved too far. The clutch went through the release point and them stuck again. This seems to be because the pressure plate is released but the diaphragm hits the friction plate. I found I needed a much shorter pushrod - it's now down to 2 3/8". Even so the clutch still dragged, which has been rectified temporarily by putting a spacer under the carpet! The pedal travel is now very short, but not too heavy. I actually don't mind it like that, but I think it's clear that I need the smaller master cylinder. Still looking for suitable brass or stainless tube for that job.

Not only are 1275 clutches beset with this binding at full compression, but they seem to vary a great deal. I had no problems with the previous one, until it broke.

Re: Very odd clutch problem

PostPosted: Mon Mar 23, 2015 12:30 am
by Charlie Tolman
Innovative approach !

How did you determine the 8 mm dimension / distance ?
Did you first measure the 11 mm distance, and then use the clutch fork lever ratio,
to obtain the 8 mm value ?

Charlie T.

Re: Very odd clutch problem

PostPosted: Mon Mar 23, 2015 12:51 am
by Charlie Tolman
Can you adjust the mechanical linkage at the clutch master cylinder, to generate some
additional "free play" in the clutch pedal movement, which might help in not moving the slave cylinder piston as far as the present arrangement ?
It might obviate the need for a smaller bore in the master cylinder ?

Charlie T.

Re: Very odd clutch problem

PostPosted: Mon Mar 23, 2015 7:48 am
by Asolepius
I measured the clutch movement first on the test rig, and it was 8mm. I then multiplied this by the fork ratio giving 11mm. Yes I could adjust the master pushrod but the fact is that I have a certain small pedal movement and I either have this near the driver or further away. I prefer it nearer, by stopping the pedal before it hits the floor. Just been out for a run and it seems fine, but other drivers may want a longer pedal movement.

I am now wondering if I maligned the old clutch. It's possible that it was being pushed too far and that wore away the release pad retainer. But anyway the friction plate was largely gone as was the release bearing so it needed replacing.

I just learned that I could have kept the old pressure plate and used a ball race release bearing, available from Peter May: http://www.petermayengineering.co.uk/sp ... oducts.htm

It doesn't need the steel release pad and bears directly on the diaphragm fingers. Costs about the same as a new clutch kit though.

Still looking for some stainless steel or brass tube for the sleeving job.

Re: Very odd clutch problem

PostPosted: Mon Mar 23, 2015 10:49 am
by Charlie Tolman
After looking at the ball race clutch release bearing assembly, and understanding what surface is in contact with the clutch fingers, I would still use the standard carbon-faced "throw-out" bearing.

Charlie T.

Re: Very odd clutch problem

PostPosted: Mon Mar 23, 2015 11:11 am
by Charlie Tolman
"Onlinemetals" has various sizes of seamless brass and 316 stainless tubing, including lengths of 0.875" OD, 0.065" wall, and 0.745" ID.

Charlie T.