Clutch is impossible to bleed and won't release

I manual swapped my 190E 2.3 and the clutch will never release. I have been battling this issue for close to 2 months now. The system looks like this (we're behind the car, looking in front:

Notice the hardline over the transmission. I believe there's an air bubble there that I can't push out. I filled the system by pushing brake fluid through the slave cylinder bleeder with an oil can. The factory bleeding procedure recommends this method, but using the right front brake caliper instead.

A bit of background:

When I first tried it after the swap, I noticed I couldn't engage 1st gear from neutral with the engine running. I tried starting the car in 1st with the clutch pushed in, the car would move a bit when the starter started turning, but at idle it would stay stationary. I could move the car and the engagement point felt normal (neither too high or too low on the pedal) but there was no way to shift. I tried bleeding about a million times, what I found to be the best method was the oil can method. Thinking that the problem was elsewhere, I pulled the transmission and replaced the master and the slave cylinder. The clutch is brand new. I measured everything and came to the conclusion that the slave cylinder pushrod was too small. I contacted LUK (from who I purchased the clutch and the slave cylinder) and the guy there told me the slave I was using was wrong. He recommended another part number which apparently had a longer pushrod. I had 2 longer rods made since I couldn't find that part anywhere. The longest rod maxes out the clutch fork. Well, today I put the transmission back in and even with the longer rod, the clutch won't release. My testing method for clutch disengagement is to jam a steel tube on the seat support to depress the pedal, engage 1st gear and try turning the engine by hand. If the wheels turn, it doesn't release.

Do you think this method makes sense? Would a new clutch and pressure plate make slight contact even with no pressure on the clutch? How should I get that air bubble out of the clutch hardline? For now, I left the reservoir cap open for the night in hopes to get the bubbles to flow upwards, but I don't think that will do anything...

Bad flex (purple bit) Acting as a one way valve. Happens all the time with brake flex lines. That would be my guess

Clutches are notoriaouly difficult to bleed.

Either use a bleeder system that forces positive pressure backwards through the system, or use a motive bleeder.

Even if you can turn the driveshaft or engine with clutch pressed on, it might not be releasing enough.   I drill a hole in the bottom of the bell housing with a 1inch hole saw so I can see the clutch and assembly releasing.

If you can't shove a ziptie between the clutch and pressure plate, it ain't released....

NOHOME said:

Bad flex (purple bit) Acting as a one way valve. Happens all the time with brake flex lines. That would be my guess

Can you elaborate on that? 

I used 3/16 brake hardline and a 1/8 brake flex line for a Jeep.The flex line is on the short side, the bend radius is pretty tight. How could it act as a check valve?

Stick a power bleeder on there. Open the slave and watch for air bubbles. 

Possibly bench bleed the slave of that doesnt work? On one car we had to unhook the slave and depress the rod while bleeding to get the air out of the slave.

Is the system using stock components and mounting locations? Is the reservoir the highest point in the system? Any chance there is a link in the line over the tranmission?

Also, wouldnt it be easier to just lift the rear end and try and rotate a wheel in neutral to test for disengagement? Good luck!

In reply to Rocambolesque :

I might have misunderstood the issue. When you say "wont release" I read that as it wont let the clutch go back to gripping the flywheel.

I see brakes lock on because of the innards of the brake flex line decomposing. 

In reply to NOHOME :

I've had that with the clutch line on the Maxton which was acting as an overeager delay valve.

To the OP - any chance you could stick something like a GoPro under the car, work the clutch pedal and see if it moves the clutch fork? Also, any chance that the fork doesn't move the release bearing far enough to actually release the pressure plate? The latter would indicate that the problem is downstream of the clutch hydraulics.

The car is on 4 jackstands. I rotate the engine with a ratchet and look at the rear left wheel. It always turns when the engine is rotated. I cannot turn the wheel by hand since the clutch is always engaged.

The setup uses the stock MC (only one part number through all 190E production), stock reservoir, stock slave, stock clutch, etc... The only things that aren't stock now are the clutch line from the MC to the slave (I have 3/16 tubes, I think OE was 1/4 but it shouldn't make a difference) and the longer rod on the slave cylinder. The rod that's in the car is 11 mm longer than the stock one, which gives me about 6 mm extra travel at the throwout bearing.

The slave is bolted to the bellhousing and the rod is inside the bellhousing. There is no way to see if the fork moves like on some cars. I can hear something moving inside the transmission though, and the fact that I could drive the car a bit tells me that it's actually doing something.

If you look at the picture I drew, yes there is a line over the transmission, and I think a bubble is hiding there...

I'm a little confused by the description of the extra pushrod length and add'l travel, but I'm guessing that's not it. I had to think about it a bit, in case it was causing the MC not to get all the way to the relief port, but I don't think that's possible, and extra pushrod length should have the opposite effect even if it was possible.

The part that sits wrong about pushrod length is the mention of extra travel. It may be a longer total length extended, but I don't think you'll see more travel, as that's dictated by the relationship between the two cylinders.

If you're right about the trapped bubble, what you need is to move fluid through there faster. Whether it's a big enough syringe to push fluid through quickly from the bottom in one continuous go, or a power bleeder on the reservoir, you need to get fluid through that hump fast enough that it doesn't do that thing where the bubble sits at the top and fluid trickles along beneath it. A smaller diameter run could perhaps help surface tension to keep things moving?

I am thinking the problem may not be the hydraulics. If the disc is sticky on the splines or the pilot bearing is siezing all the travel in the world won't make a clean release. Too much travel can also make the pressure plate fingers drag on the disc. I would pull the slave cylinder off of the trans and make up some method of trapping the piston in (which it may already have) and then bleed it. Should then have a rock hard pedal. Re-install and try it.

FWIW I replaced the clutch master in my 190 about 3 years ago and I remember NOT using the factory bleed method, but hooking up a Motive to the reservoir and bleeding it at the slave.  Worked perfectly first try and every since.

What does the motive do? Pressurize the reservoir?

The longer pushrod doesn't give me extra travel really. The total travel is the same. I took some measurements when the transmission was out:

  • Distance from PP fingers to engine-transmission mounting face (let's call that A) = 52 mm

  • Distance from throwout bearing to A if SC at max stroke = 38 mm

  • DIstance from throwout bearing to A if SC fully retracted = 56 mm

This means that the throwout bearing can only push into the clutch 14 mm. I don't know if that is enough, I should have put the PP on a press or something to measure. It also means that there is 4 mm of TB "stroke" that isn't used. The plan with the longer rod was to bring this closer to 0. The motion ratio from throwout bearing to slave cylinder is about 2:1. With that in mind, a 8 mm longer slave cylinder rod should make the distance from A to TB at full retraction close to 0. Then I would get full slave cylinder stroke to push on the clutch. I don't remember the SC stroke exactly, but I do know that the MC and SC diameters are the same and that with the pedal stop on, the MC can travel 29 mm and without the stop like it is now, it can travel 36 mm. At the throwout bearing, if the SC can move 36 mm too, I should get 18 mm of travel. If that's not enough to disengage the clutch, I don't know what will do it!

I installed the 6 mm longer rod and removed the 2 mm shim between SC and transmission. That should have given me full effective stroke. Didn't work. I installed the 9 mm longer rod + no shim, which should make the clutch always partially released and it still doesn't work.

That's why I'm thinking, either there's air in that line or the clutch is too thick!

I went back to the garage today, bled it 3 more times, no change! It still doesn't release. Now I'm questioning my test... Maybe there's always a bit of pressure on the clutch... Maybe the clutch sticks on the splines like Turner X19 said... I put new moly grease in there, it should glide freely. I need to think about this overnight.

Rocambolesque said:

What does the motive do? Pressurize the reservoir?

Exactly.

It also feeds fluid so you can put a bunch through it without refilling, though if you want to keep everything dry, you can just use it to pressurize and then disconnect and add fluid manually. I think some folks do that, but that would be awfully tedious, and of course if you're not paying attention you can run the reservoir dry and start feeding a bunch of air in...

My Saab 9000 CSE turbo wouldn't bleed unless the resivour was totally full all the way to the top.  I had never seen anything like it.  Filled it up and it bled and was perfect.  Sucked the excess out and that was it.

I'm wondering if the clutch disc is installed backwards.   The symptoms match.

I'm still thinking this is just a bleeding issue.

Motives are cheap and work well.  

APEowner said:

I'm wondering if the clutch disc is installed backwards.   The symptoms match.

No, I checked that twice and noticed you cannot install it backwards

If you're near Los Angeles you can come over and do it at my place or borrow a motive. 
 

another option would be to go flex line all the way over the trans to the slave, hang the slave low on the master cylinder side of the transmission so there are no high spots, bleed, then feed it over the top and install. 
 

Can you put a T where the "Air bubble here" aero is on the original drawing and then run a line out through the top of the  bell housing (you would have to drill a hole I am sure).  It could be a really really small line.    

Wait is that curved line inside or outside the trans?  If it is outside the bell housing then I would definitely put a T in it with a line running up from the T to future proof things.

I don't see why a good power bleeder can not overcome that though. 

Have you considered that the slave is not pushing far enough?

I just finished fixing such an issue with the molvo; the 3/4" of travel was not enough to fully disengage the clutch. I solved the problem by going a a larger master cylinder that then pushed the slave out .3" of an inch further. The penalty is that the pedal is a bit harder to push.

You can get some threaded rod and make a adjustable rod to check for this.

One thing you never want to do is make the pushrod at the slave so long that it backs the piston all the way into the bore. At that point you might be pushing on the clutch release and removing your crank end-play. That will wreck the engine.

NOHOME said:

Have you considered that the slave is not pushing far enough?

 

I just finished fixing such an issue with the molvo; the 3/4" of travel was not enough to fully disengage the clutch. I solved the problem by going a a larger master cylinder that then pushed the slave out .3" of an inch further. The penalty is that the pedal is a bit harder to push.

 

You can get some threaded rod and make a adjustable rod to check for this.

One thing you never want to do is make the pushrod at the slave so long that it backs the piston all the way into the bore. At that point you might be pushing on the clutch release and removing your crank end-play. That will wreck the engine.

Based on some tests and measurements I did, you will actually re engage the clutch by pushing the fingers into it as well.

Either way, no bueno.

I had this issue.  Decided that the 2 or 3 people telling me it was a blessing issue were wrong....   it WAS a bleeding issue.  I even destroyed a trans to find out for myself....

Bought a motive bleeder and have been awesome ever since.

OK I'm going back tonight, I'll put the stock rod back in and I'll try to remove the slave and manoeuver it across the transmission and down the bottom of the car to bleed it, then I'll re-install.

I definetly don't want to push too far.

Anyways, if that doesn't work I'll get a motive bleeder.

jfryjfry (Forum Supporter) said:

If you're near Los Angeles you can come over and do it at my place or borrow a motive. 
 

another option would be to go flex line all the way over the trans to the slave, hang the slave low on the master cylinder side of the transmission so there are no high spots, bleed, then feed it over the top and install. 
 

 

Thanks for the offer, it's really appreciated! But I'm in Quebec so it's not happening haha! I thought about the flexline thing, but the thing is there is only about 0.5" all around the transmission. It is a very tight fit in the tunnel. I just tried moving the slave around with the hardline attached but theres no way to do it.

Now I guess I'll order the Motive bleeder... One more tool...

Uae engine vacuum and a bottle with an inlet and outlet for vacuum lines. You already have a big vacuum source.