I recently replaced my clutch slave cylinder on my 72 Elan Sprint. I thought I had bled the system ok but I have a strange problem in that when I depress the clutch pedal for the first time it will be soft and difficult to engage gear. If I press the clutch a second time then the pedal is noticeably firmer and will engage gear.
This issue keeps repeating so when driving I need to depress clutch twice to make a gear change.
I am going to try re-bleeding the clutch but would welcome any suggestions of why this may be happening.
Sounds like air still in the system. Its difficult to get all the air out given the orientation of the bleeder on the slave cylinder. I find once it is pumped up so its firm then jamming the pedal down with a piece of wood to the seat and leaving it overnight helps the remaining air leak out past the slave cylinder seal.
Another less likely possibility is the shut off valve to the eesevoir in the master cylinder is not sealing reliabily and it takes a couple of goes to get it to seat
‘Another less likely possibility is the shit off valve to the eesevoir in the master cylinder is not sealing reliabily and it takes a couple of goes to get it to seat’
I have only ever found that valve opens just prior to a crash
So I’ve tried re-bleeding the clutch using a Gunson self bleeding kit and unfortunately I have the same results.
First press of the pedal is soft, but second and repeated pressing is a good firm pedal. When the pedal is left for just a few seconds without being depressed, it will go soft again for the next press - but firm again if two or more pushes.
I’ve spent hours trying to ensure all air is out of the line, including attaching a suction device to the slave to pull through fluid/air.
The only other change than the new slave cylinder I fitted is the clutch adjustment. Where the manual says 2mm between the adjusting nut,I found this was not the case with the old slave, the nut was a long way towards the clutch arm - in fact I’ve had to wind the push rod back quite a lot, about 15mm to get the 2mm gap (so moved the nut back towards the slave cylinder)
Have you fitted the correct bore of slave cylinder? There are two that look the same externally but with different bore sizes. The Elan uses a 7/8" bore slave cylinder.
A problem that can occur with bleeding comes from the geometry of the system: if a bubble (or tiny bubbles, e.g. when using DOT5 fluid) is trapped in some sort of inverted siphon or high corner, it would take bleeding very quickly to hope to create enough turbulences in the flow so as to drag the bubble with it. Some circuits are known for being hard to bleed, and require a bench bleeding if only to ensure that no bubbles get trapped out of normal bleeding reach. fwiw on my Fed elan I bleed the car with the nose up (angle 15-20°) if I’m having issues.
I would study carefully the geometry of your specific system, and consider bench bleeding either the MC or the slave if applicable.
NB: bench bleeding requires some care as one would not want to let an other bubble in the unreachable corner when reattaching the element to the circuit.
I believe the “trick” keeping the circuit under pressure overnight may work in some cases if the bubble (mostly N2 + O2) can be absorbed by the fluid under pressure, but then the circuit must be operated upon releasing the pressure to move the fluid around so that the bubble does not reappear in the same location but rather somewhere in the system where it could either escape naturally or being bleeded manually. This trick is less likely to work with the clutch since the pressure is never large.
‘Have you fitted the correct bore of slave cylinder? There are two that look the same externally but with different bore sizes. The Elan uses a 7/8" bore slave cylinder’
The bore of the slave cylinder has no baring [bearing] on if all the air has been removed, which is the problem here.
Bleeding the slave can be a pita, but for me, it is a 2 man operation. I have had no luck using a self bleed system. I open the nipple fully so that air has a clear path, no restriction, whilst a mate pumps, maybe 5 pedal presses. Then close off the nipple a little so that there is some resistance on the pedal. Then another 2 or 3 pumps, pedal down and lock off nipple.
Job done. You will obviously have to top up the master a couple of times doing it my way. Does not always give a rock hard pedal the first time, and I quite often go for a drive and have another go. Perfect in the end, good luck with it.
Although I also fitted a remote bleed to my S4, Leslie makes a very valid point about getting the assistance of a second person. Back in day, on the production line, I doubt if Chapman would have stood for clutch bleeding (and adjustment) taking any more than 10 minutes per car. Two men; 10 minutes; job done. On to the next one.
Back in the 80’s I had a couple of Lotus Cortina’s, so same hydraulics, and it was always a two man job (thanks Dad).
Is it possible you set the rod nut with the helper return spring pulling the clutch arm fully towards the front of the car? The 2mm gap should be set with the spring off and the arm pushed fully to the rear until the throwout bearing contacts the clutch fingers internally. Having to change the nut position by 15mm seems to indicate so.
The first pump of the pedal would move the piston, take up the slack in the adjuster rod, and partially dis-engage the clutch. Rapidly following with a 2nd pump would operate as normal since there wouldn’t be enough time for the helper spring to home the piston and rod. Following with a delayed pump would give you the symptoms described as the helper spring is weak and would take time to move the piston to the bottom of its bore and the fluid it displaces back to the master cylinder essentially replicating the first pump condition.
*Edited for direction clarity and added a sequence of events.
A two-person approach is helpful but I’ve had pretty good results using a bottle with some fluid in the bottom and a length of tubing. Attach one end of the tube to the bleeder and make sure the other end is below the level of fluid in the bottle. Fill up the reservoir, open the bleeder, get in the car and pump. With two people you can pump it up a couple of times and have the other one open the bleeder, looking for bubbles in the tube. No bubbles = you aren’t done yet. I suppose you could also try a phone camera on video pointed at the tube to look for bubbles, and still do it alone, but then you can’t use the pump-up-the-pressure trick.
I never had any issues with clutch (or brake bleeding for that matter) back in the day but back then I was just using regular Dot 3 or Dot 4 brake fluid. I’ve heard that Dot 5 is more difficult to bleed. You don’t need to use anything fancy in the clutch hydraulics and silicone fluids especially don’t lubricate as well as regular polyglycol fluids. That’s why OEMs tend not to use them.
Given most of my work in my garage is one person working i have had to think about a lot of methods to do jobs that are better done by 2. For bleeding hydraulics I have a method that uses a 4M long piece of tubing with is pretty much 100% reliable.
The method works by connecting the tube to the bleeding nipple at one end and then route the tube so part of it can be seen from the drivers seat and the free end goes back into the master reservoir. I then undo the bleeding nipple, sit in the drivers seat at start pumping the pedal while watching the tube. When I see only fluid and no bubbles in the tube I tighten up the bleed nipple and job done.
Sometimes it takes quite a time for all bubbles to come out of the system. The fact that the tube feeds back into the reservoir means you have a closed system so you can pump as long as you like. Once finished I tend to seal each end of the tube keeping it full of brake fluid so it is ready to go when I next need it.
The only time this has failed is with some new dual master cylinders which need to be bled at an angle and when I stupidly put the slave up the wrong way with the bleed nipple below hose connection
That sounds a logical approach to explain the behaviour. I seem to remember reading some confusion as to whether the return spring should be attached when setting the 2mm gap - I’ll give this a try as this would certainly set it back to how it was originally - thanks
Thanks to everyone who gave advice and tips on this.
Looks like I finally have a good clutch pedal. I decided to replace the original two piece clutch hose with the single piece from SJ Sportscars and in doing so I found it easy to attach the slave cylinder and hold this in a position such that all air was expelled from the line. Once the new line was full I attached the slave to the bell housing and a firm pedal was instant as a result. I did move the adjustment to get the 2mm gap without the spring (I seem to remember reading somewhere that the 2mm should be with the spring).
The result is a firm pedal and go engagement on a short drive.
