Clutch pedal free play

First of all, the term “hydrostatic” just adds black magic to a stupid simple self adjusting system. With the self adjusting version, the slave piston when at rest has 3/4" of free movement to the front of the car. Because of that, it exerts essentially no pressure on the actuating lever unless the clutch pedal is depressed. All the return force is supplied by the clutch itself: the clutch springs or diaphagm push back on the release surface. Travel is limited by the mechanical limits of the clutch mechanism. At this neutral point, the release bearing is just touching the release surface. It doesn’t turn, because it’s a graphite donut, rather than a true bearing. With this system, the release bearing can wear since it’s always in contact with the release surface. And mechanics, confused by the various procedures and parts supercessions can set it up incorrectly and cause premature failure.

Your setup isn’t self adjusting, at least, it wasn’t intended to be. The clutch pushes the beariing back to a neutral point, as with the self adjusting system. But the external spring which is in your photo pulls the arm slightly forward of the neutral point, to lift the bearing off the contact surface. As the clutch wears, the release bearing is pushed further back, and so you have to adjust it from time to time to restore the gap. To check whether you have it it adjusted correctly, your best bet is to sight the release bearing through the port in the bell housing. The exact gap is less important than whether there is a gap…no gap means wear is likely.

1 Like

I’ve been following this thread closely as I’m going through the same. I’m curious as to why yours has a spring and bracket. Mine is 1967 and it has no spring. I replaced the clutch with a 10" disk, finger type pressure plate and according to Terrys website it should have the shorter cylinder. I used the one that it came with and the clutch worked as it should. I have only put 10000 miles on the car and rebuilt this cylinder halfway through the mileage. So I crawl underneath it today and it’s just dripping with oils and brake fluid blowing back on the undercoat. I peel back the rubber and was covered in deluge of rust. So how does water get into the insides. It must get past the rubber, maybe sliding down the pushrod. This is on a car that was never driven in the rain, parked all winter and very low mileage.The last time I did this I remarked to myself that a stainless rod would be the best thing. Mine was just rotten inside and definitely wore the hole bigger.You have more threads than I do. I would think that it would wear the hole in the rubber quicker than a smooth shaft.

I went through my spares this week and discovered a new slave, TRW/Lucas, with the part number you mentioned and it says right on it, all years 61 -74. As you can see in the pic the bleeder and intake are on opposite sides, not a problem, but they are threaded 3/8 instead of 7/16 UNF as my old one. Why do things have to be so complicated? I’m going to attempt to rethread it the right size. Apparently I didn’t have mine adjusted properly as I only have 1/8 " of carbon left. I kept the old one and it has a good 3/16" left. I guess this slave adjustment is just one more thing we’ll have to keep an eye on. I might add that I"ve owned the car 35 years, and it’s always been in a garage and the last time I drove it
was 2006. I’m attempting to get it back on the road and everything I touch needs work, the carbs are next. The cylinder I purchased in 2009.

Why indeed. There was a succession of slave mechanisms used in E-Types, enough to cause endless discussion, not to mention the occasional heartbreaking clutch failure. The two styles were self adjusting and manual adjusting, with some variations in implementation depending on year. Jaguar refers to the self adjusting style as “hydrostatic”, for no particular reason. The self adjusting cylinder has no external spring, the manual adjusting clutch has the spring. The self adjusting clutch is adjustable, just to make things more confusing. A lot of folks switch to the manual adjusting style, because it can provide longer release bearing life at the expense of an occasional adjustment. The original parts were made by Lockheed, not Lucas.

The short vs long cylinders is a bit more obscure: the clutch arm needs about 3/4" of rearward travel to release the clutch. If the slave is self adjusting, it also needs about the same in forward free travel to adjust for wear over time. So a 2.5" cylinder may allow just enough piston travel if everything is set up just right. The 3.5" cylinder allows some head space. Oddly, it’s associated with the manual adjusting set up, although there’s minimal need for forward travel in that case. If you’re converting, you should pay close attention to the length of the actuating rod, and don’t hesitate to make up a new rod if the travel isn’t appropriate.

As for the dust boot, I used the Isuzu piston (discarding the internal spring) and dust boot:

The Isuzu piston has a different type of seal, which may be better, but the big news is that it comes with an accordion boot that has a much more positive seal than the original. The Isuzu boot clings to the shaft, so that it doesn’t wear on the rubber:

The squeek you hear in the video is because the rod end is worn. one more thing on the to do list. It highlights another problem: Jaguar specified that free play be measured at the actuating rod, but doing so includes free play introduced by wear in the release arm mechanism. So you can adjust correctly according to the FSM, and still end up with not enough motion to allow the bearing to fully release, especially with the manual adjusting version. This is why I keep saying that you have to sight free play through the peep hole in the bell housing before you can call this a done job.

Guys I might have a WONDERFUL solution for a driver…
Its on the way as others have said, my set up is new and bled to death BUT , there could be something MUCH BETTER.
There is a combo issue here…

  1. The repo throw out bearing is OK if you know how to adjust the rod for the right pickup without wiping the bearing. BUT its not exactly the right size.
  2. The repro clutch slave , same…there are upgrades which others have mentioned.
  3. The U.S. set up is not great, so I might have a simple , upgrade solution along with the Bob Beers kits.
    Not expensive, give me a week.
    And for some reason the actual master clutch cylinder is 1 size fits all, therod was different from my original the lenght was off a hair…more to come

Wow, I adjusted mine per the manual and check it from time to time. I have 25,000 miles on the TOB and see minimal wear.

Lots of voodoo and hand-wringing over something that seems quite simple.


I just don’t understand how the slave cylinder part # 29801 is supposed to work. I thought that as there is supposed to be a spring inside it then there would be a circlip at the end to stop the piston coming all the way out. But this is not the case. When I removed the push rod I was just lucky enough to stop the piston coming right out hence there is obviously no retaining circlip. So the piston under the pressure of the internal spring aided by the hydrostatic pressure head from the clutch reservoir to the slave cylinder will always move out to take up any play between the clutch operating lever and the push rod. Hence the external spring has to be precisely set to balance out the internal spring and hydrostatic pressure so that there is 1/16 ins of play. The longer the push rod the greater the force exerted by the internal spring hence requiring more tension on the external spring.
It just seems a stupid set up. If the clutch had a release bearing similar to, say, Triumph cars there would be now need for any play. Plus how on earth when lying on your back under the car can you accurately set 1/6 ins ?
I also suspect that the bore of 29801 is too large so that there is not enough travel to get the clutch to clear.


C.29801 is NOT the slave cylinder that Jaguar used in the “hydrostatic” (auto-adjusting) configuration on the later S1 4.2s before they reverted to the manual adjustment. It is the same length (3-3/4") as those cylinders (C.21479), but probably has some differences. The Jaguar Parts Lists show that the original 3" slave used on the early S1 3.8s (C.16989) and the 3-3/4" “hydrostatic” slaves (C.21479) did have a circlip to retain the “innards”. The Jaguar parts list for the S2s does not contain a breakdown of the components that make up C.29801, so I don’t know if it was originally specified with a circlip or not. Maybe just another way of saving a penny :smiley:

I think you are misunderstanding this. In the manually adjusted configuration (ie non-“hydrostatic”) which use the external spring, that spring is much stronger than the internal spring, and when the clutch pedal is released, the external spring pulls the cylinder piston all the way back into the cylinder against the hard stop of the back of the cylinder. There is no fine balance of pressures involved. The 1/16" (not 1/6") of free play required does not need to be precise. It just needs to be anything more than zero. The less you make it, the more frequently you will need to check it to make sure that the balance of wear of the clutch disk and throw-out bearing (one increases the play, the other decreases it) hasn’t taken up all the play. In the non-hydrostatic configuration the function of the internal spring is to keep the separate piston and cup seal together, as without it the seal may “cock” in the bore.

In the hydrostatic (self adjusting) mode, there is no external spring. The internal spring applies “light” pressure on the fork to keep it in “light” contact with the carbon throwout bearing. It relies on the pressure being light enough to avoid excessive wear on the TOB throughout the life of the clutch.

Please explain - I’m not sure I understand what you are saying here.

I think if you guys use the willwood
Even though it’s not stock
You will be surprised by the difference

There should be a circlip to retain the piston. Yours is missing.

The word hydrostatic should be banned from jag-lovers. All hydraulic systems are hydrostatic, and hydrostatic pressure or “head” plays no part in the self-adjusting system. It also leads to the term “non-hydrostatic”, which is just wrong. The two types of cylinder are “self adjusting” and “manual adjusting”.

The self-adjusting system (formerly known as hydrostatic) relies on the springs or diaphragm in the clutch itself to push the release bearing back… The bearing rests on the release flange, but there’s no pressure holding it there, so friction is low. To support this, the piston has to rest about mid-cylinder, which is what the internal spring is for. You adjust its position so that it has about 3/4" of forward travel. There is NO external spring. As the clutch wears, the piston is pushed more and more forward by the clutch diaphragm, and is poised between clutch pressure and internal spring pressure. The only significance of hydrostatic pressure is that it keeps the forward part of the cylinder filled with fluid, but it has essentially nothing to do with the self-adjustment process…

The manual adjusting system (formerly misnomered “non-hydrostatic”) works differently. It expects the external spring to hold the release bearing about 1/16" off the release flange. 1/16" is Lockheed’s generic specification for all clutch systems. The Jaguar procedure is idiotic. What the FSM tells you to do is to allow the release lever to be pushed forward by the clutch as with the self-adjusting version. Then you manually pull the lever forward another 1/16" and adjust the rod until the piston bottoms. The external spring holds this forward position when the adjustment is complete, and so the release bearing is off the plate. As the clutch wears, the freeplay will close up and you will need to readjust.

Why do I say this procedure is idiotic? First of all, if your car is actually driven, the release rod end and the fork pivot accumulate wear, so there’s no telling how much of that 1/16" is just lost motion in the release mechanism. But the more fundamental problem is this:

clutch fork

The fork ratio isn’t 1:1, so that the arc of the outer portion of the arm traverses more linear distance than the inside. In other words, between sloppy joints and an unequal rod, 1/16" freeplay at the rod translates to a much smaller amount play at the release bearing. There are those who will defend this as Jaguar engineering at work, I think it was a careless error. The only way to adjust the bearing properly is to sight it through the inspection hole. The good news here is that any gap at all will minimize wear.

When you depress the clutch pedal the bore diameter of the clutch master cylinder and the travel of the piston inside determine the volume of fluid sent to the slave cylinder. The amount of travel of the rod inside the slave cylinder is governed by the internal bore of the slave cylinder. A slave cylinder with a smaller bore will give more travel at the expense of a slightly higher clutch pedal pressure.
After replacing my slave cylinder with a new one and setting the free play as best I could, with the engine running and the clutch pedal fully depressed when I engaged reverse gear it would grate. So to stop this I have to first engage any forward gear and allow the synchromesh ring to stop the gearbox rotating then quickly engage reverse. So I am surmising that if the slave cylinder gave more travel this grating would not occur. This what I mean by using a smaller bore.
Taking your explanation of how the slave cylinder works as correct, which I’m sure it is, it seems like I need a longer push rod so I can set the external spring pretty tight so it forces the slave cylinder piston right back to the stop. At the moment if I set the external spring tight then with the current push rod the slave piston moves fully back and then they there is a lot of free slave cylinder piston movement before the clutch level arm starts to move. If I screw the push rod out of the threaded joint to eliminate this loss motion it gets to a point where the push rod is barely screwed in and that seems no good to me.
I will have another go at this when I have completed a few other jobs on the car. At the moment I am awaiting the arrival of a new camshaft oil feed so I can’t start the engine.


Thank you for your explanation - I now understand what you are saying. It is important that when you set the free play, the slave piston is fully pushed back or you will run into problems. You also need to adjust the tab that the external return spring attaches to (mounted off the slave cylinder mounting bolt) such that the spring is tight enough to pull the piston fully home. You may find that your push rod is not long enough to satisfy both of these criteria, in which case you can fabricate a longer one by simply cutting off the head of a suitable bolt. If your slave was once set up for the “hydrostatic” self-adjusting mode you will definitely find that the push rod is too short, and it will have to be replaced. Provided there is no air in the system and it is adjusted as described above, and the throwout bearing is the appropriate one for your clutch pressure plate, there is enough motion available in the system to fully engage the clutch and avoid grinding when engaging reverse. There have been several threads on this very topic in the past month or so, and I believe that in all cases the problem has been overcome by (a) bleeding out all the air, and (b) properly adjusting the free play with an appropriate length push rod.

As David mentions, I had to make a longer pushrod from a decapitated bolt. I also had to tweak the tab to put more tension on the spring:

The paint mark was just to see if it moved later (it didn’t).

When mounted this way I suppose one could make a revised tab such that a short extension on the tab wraps over the side of the slave to prevent any rotation of said tab, once installed.