Clutch mystery

[Christopher smith]

Clutch action was fine for 5 years of light use, but concentric slave cylinder finally started leaking. It is a Saab 93 type slave set-up (from Bean Engineering) coupling my 1500 pre-crossflow to a pinto 4 speed using a Tilton pedal and master cylinder set-up with -4 Earl's lines in my tight clearance 1959 series 1 frame. So I had to pull the engine to get at it but not the trans, to replace the concentric slave cylinder and I replaced the master as well. The clutch works fine as I test it up and down my driveway about 5-6 starts. Then I get a frightening grinding sound trying to put it in any gear with pedal fully depressed at a stop. I have adjusted the pedal stop back and forth to no avail. Reaching in through the old side port in the bellhousing I can see the release bearing and it just touches the pressure plate fingers, but I can rotate the release bearing with my finger. The bell housing seems warm but not sure why it would work fine for years and then when replaced, work fine at first and then fail to release after a few start-stops in the driveway. If you click on the image you can see the release bearing. Any thoughts or ideas most appreciated, particularly if I do not have to pull the engine a 3rd time.

[IamScotticus]

The usual suspects?

A re-bleed?

[MV8]

Bingo! It is not been sufficiently bled. Don't drive it like that.

[IamScotticus]

I had a Saturn VUE that needed the clutch bled.  Instructions were to submerge all 4 ft of hose, master and slave in fluid, and bleed it submerged. 

I sold off the car.

Not because I didn't want to fix the clutch, but for a pesky door latch linkage clip that wouldn't stay connected.

My point is, of all the things to do on a car, it's often the bleedings that have the most gremlins.  You might have to go to extremes to get it done right.

Edited July 26 by IamScotticus

[Christopher smith]

Many thanks. I think I may not have done the bleeding thoroughly since the plumbing had to take a circuitous route in the original install and my engineering friend has passed away since. Probably not as bad as the rear engined race car he did. I made various attempts at the old 2 person method and tried vacuum bleeding from the remote bleeder and also pressure bleeding at the master cylinder. All of those methods exited at the remote bleeder so flow was long downward before 2 upward runs and I guess that trapped bubbles. I am thinking of back flow bleeding if I can set up for it. I have attached an image of the old slave cylinder that had leaked but worked fine for some years after a conventional 2 person (pedal pump) bleed, so I guess I am doing it wrong. Still wondering about the mechanism that could explain why the clutch actuation is fine for 5-7 starts and then not.releasing.Could air in the line do that on heating maybe? I should have stayed awake in my thermodynamics P chem class. Sure appreciate your guys thoughts.

[MV8]

The slave is for a Saab 900 (79-93), not a Saab 9-3. The slave should have the fittings horizontal, inlet to the center (red nut), bleeder on top. The circular line helps trap air after so many cycles above, near the elbow. I would look at modifying the adapter behind it to rotate the slave about 90 degrees ccw as viewed from your pic, then reroute the plumbing. You could post pics of the adapter and the input shaft area without it for ideas.

[Christopher smith]

MV8----Thanks for your valuable info to solve my "Saab Story". Still wondering how I had it working fine in the same exact configuration for years, albeit light usage except some really fun test laps at NJMP. I guess I will still try to get the last of the air out somehow ( reverse bleed?) before I take the next dreaded steps of pulling the engine again to get at the plumbing.

[MV8]

These tips may help:

 

[Christopher smith]

Great video! I keep hearing just how crazy the clutch hydraulic systems are. I even thought that maybe after a few starts maybe the heat in the system expanded and pressurized trapped air somewhere in the line and forced the release bearing too far so it pushing the pressure plate fingers into the clutch disc thus keeping it from releasing.

 I would think that would give the symptoms I have experienced. Not sure if that could happen. But ,in any case, a better fluid bleed sounds like it might fix it. I think I can set up a pressure feed of clutch fluid up through the bleed port and through the lines up to the master cylinder. But I will try that method in the video first. I hope something works.

[SENC]

I suspect what's happening is that you're starting with a big air pocket, and as the clutch is actuated multiple times the air is getting mixed with the fluid so you're ending up with a more broadly compressible column of fluid.

[Christopher smith]

Tried another couple of bleeding methods including an extensive reverse pressure bleed feeding a lot of fresh fluid up from the bleeder to the master cylinder. No bubbles at all. and no change in how the slave/release bearing works. It still makes the awful grinding sound trying to put it in a gear with the pedal full depressed but only after  10 starts. I can see no change in the slave cylinder or release bearing movement and it looks like it is pressing against the pressure plate fingers no different before about 10 trial starts or after them when the bellhousing is hot and the noise starts. It does seem strange that I can easily turn the release bearing with light finger pressure when clutch pedal is up and with a bit more finger pressure when pedal is down -(not expected) and it did not get hot. I am getting set to yank the engine again since the hydraulic part seems fine. Perhaps I damaged the clutch disc or pressure plate somehow earlier when I reinstalled the engine but why is the noise only when hot or after some starts?. Any thoughts are much appreciated.

[MV8]

There is a point in the pedal range of motion where the pressure plate releases the clutch. Is this point near max pedal travel/on the floor cold? Does the point change in the 10 cycles/hot? Pedal height is adjustable.

Everything is pointing to bleeding technique. I get some bubbles on every system I bleed, if only from the fluid shear at different bleeder screw positions (fine bubbles), which I change back and forth while observing flow in a clear hose that is looped above the bleeder three inches or so before going down to the jar with the vacuum port for my hand pump (which I am pumping at the same time to maintain differential pressure on the gauge), and the lid loose on the master.

You could check the hydraulics efficiency by looking for immediate movement of the slave after enough pedal movement to close the compensation port in the master (where the fountain comes from if you've ever rapidly pressed (from the pedal fully up) a clutch or brake pedal with the lid off. You can use a cable camera (about $20 on amazon) that plugs into your phone or a lap top to observe while sitting in the car, engine installed, working the pedal.

You might also post pics.   

[Christopher smith]

MV8---Thanks for the detailed analysis. My pedal has a stop that is adjustable and I have tried to set it just a bit past the point of release but also tried it set well past the point of release.No difference in stopping the grinding noise issue unfortunately. I can observe the release bearing travel and even touch it through the side port where the fork would have gone in , but is no longer there. I see the release bearing travel has been the same cold or just after a series of starts and that grinding sound with the clutch pedal against the stop, when trying to put it in gear. I guess I will try one more time with vacuum bleeding out the bleed port before I give up and yank the engine again ( a pretty tight and complex job with a 1500 in a series 1 frame and oil cooler lines etc. needing to be dealt with). perhaps I will find some damage or old oil contamination on the clutch disc that, rather than making for clutch slip might make it grab when hot? Seems to work fine for a few starts. I had overfilled the engine oil last year and had some slip past the rear main seal but not sure how that would have gotten onto the clutch disc and never had clutch slip issue.

[MV8]

Get your ten cycles in, shut it off, put it in first, press the clutch pedal, then start the car. Does it start moving or lurch?

[Christopher smith]

MV8--- Gave it a try. About 10-12 starts then grinding noise started again when trying to put it into 1, 2 or R, so shut it off. Then put it into 1st, pedal down to stop, ignition off ( and also tried 0n), hit starter- no movement at all, no grinding noise either. repeated all this again-same. Adjusted stop up a bit- same So I guess if I might need to take it all apart so I am not risking much if I trash something driving it. The tough part is hooking up all the motor mounts, electric, oil cooler and accu-sump lines, radiator hoses etc..Seems to drive ok , no noise once pedal is up and car moving

[anduril3019]

Any chance the release bearing is returning too far due to engine vibration? Not sure if it's possible with a concentric release, but I had something similar happen with a standard release bearing. Everything seemed fine after bleeding, but after running for a little while, the clutch wouldn't disengage. Turned out there was a stiff spot in the slave cylinder, the external return spring on the slave cylinder that keeps tension on the fork and pushrod seemed fine while cold, but it was actually stopping at the stiff spot in the cylinder. When hot and vibrating, the plunger would move past the stiff spot, returning farther than normal, leaving a gap between the clutch and bearing and not enough travel to disengage. Also check your upper pedal stop (whatever is stopping it, master cylinder, chassis, etc.) make sure pedal is all the way up when adjusting everything.

[MV8]

Did the engine slide right up to the bell or did you need to turn the bolts with some effort to bring it together? It sounds like the clutch system is not the problem. The trans should be inspected or at least drain and look for metal flake oil. May also have done some synchro damage early on when it wasn't quite bled enough.

Edited July 29 by MV8

[Christopher smith]

Thanks guys---I did have some trouble with the 2 new master cylinders and the pedal would not come all the way up.. The threaded push rod Tilton used in the new ones was about 5-6 mm longer than my original one. I found that it made the pedal hit the frame when in the up position so had to shorten it, otherwise trying to bleed the system was not pushing fluid into the line. I guess the piston was covering the port. But fixed that so in the pedal up position there is more than 7-8 mm clearance from the frame so the piston retract fine now.. I have an adjustable  pedal stop and tried a few distances from firewall so as not to push the slave cylinder piston out of the housing but to allow release part way to the stop, maybe 20 % of pedal travel or 50 % of travel adjusted differently ( did not help). I can visually observe the effect by how far the release bearing travels when the pedal is depressed.( I was also wary of pushing it too far and making the pressure plate fingers scrape the clutch discbut it does not look like that could be the case..Both times coupling the engine to the bellhousing, it just required minor jiggling to line up but no force once the trans input shaft found the splined center of the clutch disc. I did not undo the pressure plate but maybe should have but it had worked fine for years so saw no need. I guess I need to pull it all apart to see what is scraping.

[joe7]

Looking at the picture, I'd say the TO bearing looks a bit rusty, which could make a scraping noise as its may not be rotating smoothly. (?)

[Christopher smith]

Sorry but did not label correctly- it is the old rusty one that had a  leaking slave cylinder after 5-6 years of mild use. I was just trying to show the snake sort of route the -4 flex line takes. It worked fine and the 2 person pedal bleeding technique was easy originally but I have tried all sorts of bleeding methods with the new set-up. No bubbles seen now but still get that grinding noise after 8-10 starts , when bellhousing is hot. I hope to have the engine out this week to see what could be causing the noise. What does seem strange is that With the pedal down and clutch disengaged I can just reach a finger in there through the side port and can turn the release bearing with mild pressure. Not as easy to turn as when clutch pedal is up but I would think the pressure of it pushing against the pressure plate fingers would keep it from turning . The bearing, slave cylinder and the pressure plate are not getting hot, just the bellhousing.