Weird Hydraulic Failure

[Moseso]

Scene 1: Clutch pedal getting lower fairly quickly -- 48 hours from first noticing the phenomenon to being unable to disengage the clutch.
First thought, of course, is leaky master or slave cylinder. So I look for signs. NO signs of leakage. All dry around the master cylinder and the reservoir is full. It would be hard to locate a leak at the slave cylinder which, of course, is covered with motor oil -- but there has been NO drop in the reservoir level. Fluid is not escaping from the slave cylinder.

Scene 2: It is apparent, from simple observation that there must be air in the system. The feel of the pedal and watching the clutch lever action are consistent. The lever is barely being moved at all by a full-travel push on the pedal. So, I bleed the system. Lots of air comes out, but the reservoir level doesn't drop as much as I would expect, and a good 15 (or so) pumps of the pedal (I didn't really count, but that's close enough) doesn't restore function, even though the last three pumps looked "right" -- no more bubbles, clean fluid. It's better, but not right.

Obviously, I am going to have to dismantle the M.C. No hurry, the car is up on blocks for numerous winter updates and repairs, of which this is only one. So the question is this: Can someone posit a theory as to why a M.C. "sucks air," which is to say: develops an air bubble via an opening through which no fluid is escaping?

 

[hondo402000]

Have you thought about the clutch fork pin shearing too? Since you didnt list the car the failure happend on I assume it has the same clutch fork and pin issue as a TR6

Hondo

 

[Geo Hahn]

The MC can certainly fail without leaking, indeed that is how mine have usually died. It's that little rubber bit that is at the top end (opposite end from the piston) that leaks. It is sort of a check valve that should let fluid into the MC but then seal the inlet when the pedal is depressed.

I can't account for how air would have gotten into the system w/o fluid leaking out.

 

[Moseso]

Hondo --
I did consider the possibility of the dreaded pin-shear. But my observations seem to rule it out. The actuating lever feels fine. The hydraulic system just ain't moving it far enough to do its job.

Geo --
I haven't thought to look in the reservoir, while the pedal gets depressed. If the seal has failed in the manner you suggest, one would expect to see fluid being disturbed in the reservoir by in-rushing fluid.

Here's my best guess, but I'm just guessing... The inlet seal rod has, somehow, come out of the little keyhole that pulls it away from its seat at the end of the return stroke -- preventing any fluid from entering the MC from the reservoir. Therefor, some kind of suction on the pedal return stroke is drawing air in past the main piston seal. Does that make sense to anyone? It'll be days (maybe weeks) before I get inside there. I may have to dredge this post up out of the past by the time I finally see what's going on inside my MC.

 

[SkinnedKnuckles]

I had the spring in the slave cylinder break once. I suppose that without the spring the slave piston could somehow retract a bit farther than normal so that some of the master travel goes to returning it to position. What travel is left may not be sufficient to completely disengage the clutch. Don't know about the air in the system.

Some have also reported that the hydraulic line can weaken and balloon under pressure, I believe, which would also limit slave cylinder travel.

 

[tr8todd]

+1 on the broken internal spring in the MC. Also check to see if you have excess wear on the clutch pedal where the pin passes thru. If equipped, improperly adjusted push rod.

 

[TR3driver]

Therefor, some kind of suction on the pedal return stroke is drawing air in past the main piston seal. Does that make sense to anyone?

Yup, makes sense to me. Or perhaps the return to the tank is blocked some other way, like maybe a marble dropped into the reservoir.

The only other possibility I can think of is that perhaps, somehow, the line from the reservoir to the MC is sucking air but not allowing enough fluid out to notice.

 

[TR3driver]

I suppose that without the spring the slave piston could somehow retract a bit farther than normal so that some of the master travel goes to returning it to position.

That shouldn't apply to a TR3, which works differently than a TR6. On a '3, there is a spring that is supposed to return the slave piston to the bottom every time, and then you occasionally adjust the freeplay so the clutch starts to actuate with the piston only 1/16" or so away from the bottom.

One good thing about this system is that a broken taper pin is immediately apparent (not enough range to adjust the freeplay to spec).

 

[SkinnedKnuckles]

Yup - I'm referring to a 6. I haven't any idea about a TR3.