Bleeding the Z3M clutch slave

[Z3MROADY]

I was replacing the brake fluid in the brake lines and clutch slave a couple days ago and I'm afraid that I got some air in the clutch line. I tried and tried over and over to bleed it out but no luck. The engagement point is now at the floor. Bummer. Before I started this project the clutch was OK but probably not as high as it should be. I should have left well enough alone. Here's my procedure. Perhaps one of you guys can tell me what I'm doing wrong and steer me in the right direction. I've read the various DIYs on this in the stickies and elsewhere so I think I was doing it correctly. 1. Check to see that brake fluid reservoir is full. Top off as needed during the procedure. 2. Remove the slave and let it hang from the hose. 3. Connect the special tool that holds the plunger in. Push the plunger into the slave until it seats lightly. The tip of the plunger is now sticking out of the slave about ½". 4. Have a buddy pump the clutch a couple times and hold the pedal to the floor. 5. Bleed the slave while holding the plunger side down and the bleeder screw up (about a 45 deg. angle). Repeat about 5 times to purge fluid and air. (Note that my pedal would not spring back up into position. I had to grab it and pull it back up each time). If I just cracked the blended so the pedal would only go partway down it would then spring up. Next step.... 6. Remove tool from slave. (Plunger moves out to an extended position). 7. Reconnect the slave and bleed the clutch line as buddy pumps and hold pedal down as in #4 above. (Again the pedal would not spring back up into position by itself and had to be pulled back up). After bleeding the line a few times the pedal finally was at the normal position and felt firm. I then started the car and checked the engagement. Worse than when I started. Clutch still rubbing a little with the pedal at the floor and hard to shift into any gear. Note: The clutch line is new so I know it's not ballooning. Clutch pedal and bushing are OK too. Can you guys see anything I'm doing incorrectly or in the wrong order? Please feel free to critique this procedure using the slave plunger tool. I ran about 6 oz. of fluid through the line and slave several times so there's something I'm not doing right to remove the bubbles. I'm stumped. Thanks! Happy Fourth of July to All!

[Muzz258]

The first time I bled the clutch I ended up in the same situation as you. resolved the problem by reverse bleeding the clutch.

1) Make sure you have lots of spare capacity in the reservoir.

2) Put about 1 lite of fluid into a hand pump with a exit hose the correct size to fit snugly onto the slave cylinder bled screw.

3) With someone watching the reservoir to make sure it does not over flow, open the bleed screw and start pumping the fluid back up into the reservoir. Don't stop until told to or you have the likelihood of getting an air slug into the system. Close the bleed screw. and try the clutch.

4) After a day or so, bleed the slave cylinder with the tool or the "push the rod against the bell housing" method to remove any residual air that has found its way into the slave cylinder.

Hope this helps.

[Z3MROADY]

I've read that some are able to successfully bleed the system without the reverse method but I may have to give that a shot as well.
What do you mean when you say "don't stop until told"? Is that to say don't stop until you reach the maximum fill line on the reservoir? If you stop before that line or before the reservoir overfills how does that add air to the line?

[Muzz258]

Stop when the reservoir hits the max level, not when it is over full.

If your bleed hose is not a very tight fit on the bled screw, a slug of air will drawn into the tube at the bleed screw when the flow (pressures) changes direction e.g. pumping, very slight drain when you stop, then pumping again.

[gmushial]

I was successful with a similar method: I started with a full slave from the bench - was careful to orient it so that any air would come out; allowed some fluid to drain down out of the res via the flex SS hose while I made the connection to the slave; and then using the reverse method - pumped my oilcan full of brake fluid until juice was coming out, made the connection to the bleed nipple, opened it, and then pumped fluid upward into the res, until I could see the new clear fluid coming up into the bottom of it. The net result was a high engagement/disengagement point on the pedal, and having never had to go back and bleed it again.

[rf900rkw]

The oft quoted "hold the slave so the bleeder is up" is pretty much guaranteed to trap air. The system does not give a crap about the orientation of the tip of the bleeder. It cares about where the bleeder port intersects the working cylinder. As designed, the slave mounted with the bleeder tip poiinting down and back is correct. This puts the end of the drilling at the top of the bore.

Also keep in mind there are many different slaves in these cars at this point. Not all of them will benefit from that clamp tool. If you still have the stock slave, that tool is correct. When done bleeding with the tool still installed, the pedal will be solid. Solid as in doesn't move a bit. Like you were pushing against stone. Not "only a 1/4" of squish solid. Zero give. If not, there is still air.


It's not the master. Before you even go there, it's NOT the master.

[fmcfad01]

If you read the bentley manual, it says that after you bleed the slave, you have to take it and depress the plunger completely a few times by pressing it against the trans bell housing. It exlains that doing so forces any air that's inside it back up to the reservoir. I never got mine to bleed correctly simply by bleeding. It always takes this extra step.

[Dakar Ole]

[Z3MROADY]

The oft quoted "hold the slave so the bleeder is up" is pretty much guaranteed to trap air. The system does not give a crap about the orientation of the tip of the bleeder. It cares about where the bleeder port intersects the working cylinder. As designed, the slave mounted with the bleeder tip poiinting down and back is correct. This puts the end of the drilling at the top of the bore. Also keep in mind there are many different slaves in these cars at this point. Not all of them will benefit from that clamp tool. If you still have the stock slave, that tool is correct. When done bleeding with the tool still installed, the pedal will be solid. Solid as in doesn't move a bit. Like you were pushing against stone. Not "only a 1/4" of squish solid. Zero give. If not, there is still air. It's not the master. Before you even go there, it's NOT the master.

OK. So the slave bleeder port should be at the lowest point and pointing toward the rear of the vehicle. Does it matter which way the slave is turned to have the hose oriented up or at the side? After I bleed the slave (with it held by the tool and still connected to the hose) the pedal is rock hard. But when I disconnect the tool the plunger moves out and then reinstall the slave the pedal goes soft again. Perhaps when the plunger moves out it is sucking in some residual air from the hose. I think I'm just pushing some air back and forth to the slave and to the hose. I'm still confused why the pedal has to be pulled back up to the top repeatedly. Note: stock M slave

[Z3MROADY]

Looking at the photo of the slave it seems to me that pointing the plunger straight down would put the air passage to the bleeder at the very top of the cylinder chamber allowing air to go up and out the bleeder. If the bleeder is pointed down (and back) wouldn't that trap air in the chamber at the cylinder or send it back up into the hose?
I think the tool is compressing the plunger back a little further than the photo to about the point that the plunger is at or about the orifice to the hose.

[gmushial]

Randy's comment is quite cogent in terms of internal "plumbing" of the slave and the appropriate orientation: so - if one has pulled the slave to replace the soft line, by holding the cylinder in its as installed horizontal orientation, and by pushing on the plunger, just a mm or two, one should end up with the fluid as shown in blue; and by rocking it back and forth maybe 5 or 10deg at most, just in case there is a trapped bubble one should get it come out, leaving again, the situation in the image. And if one lost the visible meniscus then one might want to push the rod back in another mm or two. But if one is starting with a dry cylinder, then one can preload it by using the oilcan method connected to the bleed port, and again by holding the cyl in the as installed orientation, one should be able to fill the cyl and get to the situation in the image. But as long as the cyl is entirely full and there is that meniscus in the connection to the softline, then the only air one has to get out is going to be in the soft/hard line and that's trivial: simply using the oilcan pump enough fluid into the slave to displace any air up through the reservoir.

[rf900rkw]

Awesome! Thankyou. This illustrates basically whats up.

Where the drilling for the bleeder intersects the working bore needs to be on top. There is a lot of latitude in the exact angle, just so long as the slave is clocked so the point is the highest.

However (TM), this is but one of the slave designs. And not the most common in the Z's that's I've seen. The common one has the fluid feed, the line port, coming in from the side at an angle, the the bleeder drilling going into the fluid feed drilling. It doesn't intesect the working cylinder at all. On these, the tool is essential. The tools' job is to compress the piston to the bottom of the bore, making the volume in the working cylinder zero. You then bleed the line and port, orientation doesn't matter.

Then there are the aftermarket ones that are different still.

[Z3MROADY]

My slave is still hooked up to the clutch hose with the tool attached and the rod compressed all the way in to the bore so the plunger is bottomed out. Plunger straight down. The air that is causing the problem must all be in the hose. Perhaps when the tool is removed the plunger is sucking air back into the bore when it extends out during reinstall. I guess that I really need to put the slave back in the tool, compress it all the way and then back bleed the hose pushing the air up to the reservoir. I'll try plunger up, bleeder down, plunger down, etc. I'll try all the suggestions ASAP and report back. Anyone care to comment on why the clutch pedal will sometimes stay down and has to be pulled back up manually? Anyone experience that?
Thanks everyone! Have a great 4rth!!

[gmushial]

I guess I'm going to disagree a bit here: the case of the plunger all the way in has the risk of a) closing off the port, making it impossible to get any air out, and b: when one pushes the clutch pedal the first time the plunger is forced out it misses its docking location on the release arm, causing all types of other problems. ... ie, if the cylinder was wet when it was removed, then just push the plunger in a mm or two until one has brake fluid in the port for the softline... and this is after rocking the cyl 5-10 deg to make sure air bubbles are hanging along the top of the cyl have escaped (or rocking and tapping) - though I do appreciate Randy's point in that smaller the volume of the cyl, the less space bubbles can hang out. Though the notion of plunger up, that is guaranteed to trap air. ... w/re clutch pedal: isn't it returned by a) a spring on the pedal and b) the pressure place arms?

[Z3MROADY]

The slave was not completely removed. It was just unbolted from the clutch housing but still attached to the one year old clutch line. So it is wet. Doesn't seem like one would have to completely remove it to bleed it but I may be wrong. My pedal has a good return spring but it seems to be stuck down at the end of travel when bleeding. If I manually pull it up depress the pedal with slight pressure, and then crack the bleeder open just for a millisecond (so the pedal does not go all the way down to the floor) it returns just fine. Odd. Sorry about all the minute details but I just want to describe the symptoms as completely as possible. I'll make a spacer for the tool that keeps the plunger from going all the way in. Say 1/4" away from full depress.

[Z3MROADY]

[IMG][/IMG]

BTW- This is the tool that I made with info from this forum and is being used on this project. The pipe spacer that presses against the slave to keep the seal from popping out under pressure is about 1-1/4" long. It keeps the plunger from bottoming out completely in the bore. The 1" wood dowel in the picture worked good to hold the slave unit in the clutch housing with the plunger depressed while threading the bottom nut onto the stud. Once the bottom nut was on to hold the slave I was able to start the top nut and remove the dowel. The dowel was cut to 2" long.

[rf900rkw]

Bottom the piston out. IF it blocks the port, you'll know because nothing will come out. Once bled, there should be no air in the system to be sucked in.

Greg, the piston will fully self-extend as soon as the tool is removed; spring inside. If you misss the clutch arm socket, it's because you did it on installation.

[misterTT]

I'm having a very similar situation as the OP in this thread, so rather than start a new one, I thought I would piggy-back onto this conversation.

For me, the big picture is this:
1. My OEM slave cylinder failed about a month ago.
2. I replaced the failed OEM with an FTE slave cylinder (and UUC SS clutch line), and bled the new slave while fully installed (i.e. NOT pushing the plunger in while bleeding). Note that I did not 'bench bleed' the new slave, although in hindsight I probably should have figured out how to do so.
3. The bleeding seemed to do alright, ended up going through almost a liter of new fluid with no visible bubbles at the end. Pedal pressure was good, until I tried to drive the car, at which point it was obvious that the clutch engagement was on or slightly beneath the floor. Very hard to get into gear...

4a. I then did some more digging and tried a variant of the method tested above (bleeder valve pointed up, opening the valve with plunger depressed, closing the valve before letting the plunger back out). No visible air bubbles emerged from this after going through another .5 liter of brake fluid.
4b. Reinstalled into transmission: no change in clutch operation - engagement point still on the floor.
4c. One thing I noticed in the above attempt is that the new slave cylinder has an audible "pop" when depressing the plunger by hand. There's no leaking fluid from the rubber boot, but I am concerned that this is either a defective new slave, or I somehow damaged it along the way. Basically, if I hold it in one hand and depress the plunger with my thumb (i.e. not a ton of pressure), it'll retract a few mm and then "pop", and continue depressing with consistent thumb pressure to its stopping point.

SO - here's my question: Since I don't have a known working slave with me, I don't know how much pressure should be necessary to depress the plunger all the way, either before or after bleeding. In looking at the above images of the homemade press tools, I would assume there's a fair amount of force being applied to depress the plunger in a correctly-functioning unit. Since I can depress the plunger all the way with my thumb only, is that indicative of a defective part? How about the 'popping' noise?

I'm a bit stumped!

[rf900rkw]

There are many potential issues here, possibly being compounded. Let's start at the top. Slave cylinder failures are extremely rare in these cars. What made you think yours had failed?

[djb2]

You absolutely can bleed a dry slave cylinder by pumping the pedal about a hundred times. The first 95 presses will be mostly compressing air, getting a tiny bit of fluid in each time. It won't feel as if you are making progress, and you'll need to pull the pedal back each time. Only the last few will feel productive

The clutch will clear air bubbles in normal use, although it should be air-tight.

Unlike the brake system, he clutch hydraulic system operates with high volumes of fluid at low pressure in a short line. Unlike the brake system, there is no "slug" of stale fluid that wiggles back and forth in the line. With each press fluid from the clutch master cylinder goes through the line and ends up in the slave. With each release, any bubble in the slave is forced out through the line and ends up in the master cylinder. From there it will eventually work its way up to the reservoir.

(In reality the transmission is tilted a little bit, so a little bubble might stick around until you drive downhill.)

[rf900rkw]

It only works that way when the feed line enters the working cylinder at the top like the illustration above shows. Unfortunately most of the Z3 slaves do not, they come into the side.

- - - Updated - - -

OP, in reading back, I see where you said your tool keeps from bottoming the piston. The piston HAS to be bottomed, or you're wasting your time with the tool.

[MrDangles]

Is anyone else super confused? I though you just let the slave hang, break the valve open, and force the fluid upwards with a pump?

[rf900rkw]

Air 14
Humans 3

[misterTT]

There are many potential issues here, possibly being compounded. Let's start at the top. Slave cylinder failures are extremely rare in these cars. What made you think yours had failed?

The failed slave cylinder came about while driving around locally, when at a stop sign the clutch went to the floor and simply stayed there. I was able to push the car to the side of the road, and pull the clutch back out by hand, pump it a bunch of times, and it began working again. Doing a bit of research suggested that there was either air in the lines, or a failed slave or master cylinder. I decided to try bleeding the slave cylinder once before doing anything else in case it was just an air bubble. Doing so helped tremendously for a couple days (the clutch engagement has always been at the floor since I purchased the car, without a clutch stop at all). The clutch engagement after bleeding was much higher, to the point where I could install a clutch stop for the first time since purchasing the car. However, within a couple of days the brake fluid light came on suggesting I was leaking fluid, and sure enough before I could get more fluid in there, the clutch fell to the floor again, never to return!

At that point, I inspected the master cylinder (no visible leakage), and then removed the slave cylinder. The slave's plunger was super oily, and fell out of the (thoroughly cracked) boot along with brake fluid as soon as I got it out of the car. So, there's the slave cylinder failure. At that point I ordered and installed the new slave and UUC clutch line previously ordered.

The thing that bugs me most is that for a few precious days, the clutch engagement was perfect, and not on the floor for the first time ever. Now I can't get it back off the floor, and getting into gear is pretty tough as a result. I find myself rev-matching and shifting at very precise RPMs to help get into gear, so maybe that's a plus side to all this: training myself to be less reliant on the clutch

[rf900rkw]

That was the answer I wanted to hear. There is only two ways a hydraulic cylinder can fail. It can seize solid, or it can leak. We don't have the seizing problem. Therefore the only cylinder failure is leaking. The slave is easy to diagnose; it can only leak to the outside. The Master is more complicated as it can leak externally, or it can leak internally pushing fluid back into the reservoir unseen.


You got the air out of it once. It's time to do it again.