MatthewMiller

Okay guys, I need help diagnosing a clutch issue. My car is a 96 LT4 (396cid now) ZF6, with a Fidanza aluminum flywheel. Last May I had a new SPEC Stage 2 kit installed after the prior disk came apart (it had been in a long time). It developed serious chatter, but otherwise seemed normal. Over the last couple months or so, the engagement point of the clutch has been getting higher. Three weeks ago at an autocross it began to slip. This was only after maybe 2000-3000 miles on the SPEC clutch kit. The engagement point was now within the last 1" of travel.

I took it all apart. The disk, plate, and flywheel friction insert were all barely worn, so it does not seem like the high pedal engagement was due to a stacking issue. There may have been air in the hydraulics, as the reservoir was near empty. But I don't see how that could cause too high an engagement point (I can see how it would cause too low an engagement point or failure to disengage completely).

I obtained a NOS Valeo race clutch from Jim at Powertorquesystems, along with a NOS throwout bearing and oillite pilot bearing, and I got a new Centerforce Dual Friction disk per Jim's recommendation. I also installed a new steel insert for the friction surface of the flywheel. I also installed a new slave cylinder, because I wasn't sure if the old one was leaking or not (reference prior low fluid level).

Driving it after putting it all back together shows that the clutch engagement is still high. I didn't try an all-out test for slip, because I'm trying to obey Centerforce's instructions to break the disk in first. But now I'm afraid to drive it around and break it in because the throwout bearing may be constantly under load!

During assembly, I observed the following.

• With the bellhousing on but the transmission still off, the fork was not under load from the diaphragm fingers. I take this to mean that stacking height of flywheel, pressure plate, and disk are fine and the fork correctly engages the throwout bearing. That is, all of these parts are okay and not causing the problem. Unless I'm missing something, that leaves only the hydraulics.
• I was able to compress the slave cylinder piston by hand (pedal not depressed of course). The slave had to be compressed to go all the way down on its mounting studs, but again this could be done by hand. Is that normal?

About the only thing that hasn't been replaced is the master cylinder. Again, I don't see how failing seals or air in the hydraulics can cause too high of an engagement point. But I saw in another thread that Tom400CFI said something about the replenishing port not opening if the pedal doesn't come back all the way, causing pressure to build up in the line. It feels like my pedal comes back all the way, but I'll have to check. Is there a way that the master cylinder can malfunction or get a clogged replenishing port, in a way that would prevent it from properly releasing the line pressure when the pedal comes up? I can replace the m/c I guess, since it's cheap (I hope it's not too hard a job - it seems buried in there!).

What am I missing?

Tom400CFI

I think that the only thing you're missing is a test of the new clutch!


That is normal and in fact I was going to recommend that you do exactly that (depress the slave by hand) to see if you could. That you can push it in, is proof that the TOB isn't under any abnormal load. No more than what the slave spring is applying anyway.



It's possible, but not likely. I also agree with you about the MC failing, not being a cause. Failing hydraulics can't very well, "work better".
Anyway, your "test", compressing the slave proves that the port in the MC is free and clear; if not, you wouldn't have been able to push the slave in. The fluid displaced by the slave piston, travels up the line into the master bore, then through the port and into the reservoir. So that is all working. Your pedal (or at least the MC piston) is retracting back far enough or that port would be covered and compressing the slave would be a no-go. I think at this point, your hydraulics are working fine.

I'd go "get on it" (prepared to "get off it" quickly if necessary), but to see if it will hold. "Break in" for a new clutch is (I'm sure you know this) the disk friction material mating with the FW/PP surfaces, much like brake pads on a rotor. If it's going to slip now, it'll slip later and isn't going to work....and a quick slip isn't going to ruin it anyway.


.

MatthewMiller

Tom, thanks for the feedback. I will double-check that I can manually compress the slave cylinder this evening, and that the fork is still unloaded when the slave is removed. If the answer is still yes, then I will go drive it as you suggested and report back.

MatthewMiller

Well, so now I'm not so sure. I pulled the slave cylinder and it didn't immediately compress. After a lot of force it did finally move, but it acted like it was "stuck." Then I remembered that when I installed it before (when I last tried to compress it by hand) it wasn't really properly bled at first. I basically did the "ranger method) of bleeding with the pedal and it firmed up. So I wonder if before I was mostly compressing air. I have the old on the bench for comparison, and it is clearly easier to compress.

FWIW, the end of the fork where the slave rod contacts it has plenty of freedom: probably 1/2." So the fork is not compressing the diaphragm at all by itself. I guess there are only two possibilities:

1. The pedal/hydraulics really are somehow holding the clutch partially open with the pedal released; or
2. Somehow something is screwed up and the pressure plate doesn't have full clamping force even when the pedal is released. Not sure how that could possibly be the case for two different clutch sets plus a new friction insert on the flywheel, though.

So the plan now is to just replace the master cylinder. I'm too lazy to reinstall the slave and take the thing of jackstands to test drive it some more. I have a new m/c coming in the morning, and it's a cheap and fairly easy thing to try. Given how much time and money I've put into this ordeal, it seems silly to not just eliminate that variable. Then I'll drive it and see.

Tom400CFI

I agree with your bullet pointed possibilities; it's either the hydraulic system or the PP.

Sounds like you have a good plan. keep us posted.

mtwoolford

One of the few things that hasn't happened to me personally but which I have read about is that the inner lining of the hose between the clutch master cylinder to slave cylinder will collapse in on itself. Sure when you press down on the pedal with your foot there is enough pressure to move fluid down to the slave cylinder, but not necessarily enough pressure in the opposite direction when the pedal is released to allow the clutch to fully disengage.

On a similar note, I did get a new master cylinder for a Toyota once, that for whatever reason the piston would not return far enough to uncover the brake line port and allow the brakes to release fully; I'd first notice a little brake dragging, but eventually the brakes would lock up. And that was with a factory new master cylinder.

MatthewMiller

Well that wasn't quite as easy a job as I expected (replacing the master cylinder), but it's done. Driving it this evening, I think it's improved just enough to feel like it did before this debacle got started. That is to say, I think it's back to normal. The engagement is still on the high side, but I think it kind of always was. It's just that when it started slipping it was really high, and that's also what it was like on Thursday. Now it feels okay, I think. Also, it held full power through the torque peak, without question. So that's a good thing.

I plan to drive around a bunch tomorrow in town (to break in the clutch), and then I'll hopefully know for sure. Strange stuff, this!

mtwoolford

thinning or using a thinner spacer between the master cylinder and the firewall increases the effective stroke of the master cylinder piston; that's what I had to do with a spec lite weight billet steel flywheel and spec stage 3 plus clutch; worked out very well.

MatthewMiller

If I understand everything correctly though, I think for my issue I would want to try the opposite: extra spacing between the m/c and the firewall. Keep in mind most hydraulic clutch issues seem to be a low pedal with failure to fully disengage, but my issue is the opposite of that (engagement high up in the pedal travel). I did scrape off the old gaskets on either side of the spacer and add new ones (I cut them out of thin gaskent material).

MatthewMiller

Just an update here. After replacing the m/c, things seem to be working well. The engagement is still a bit on the high side in the pedal travel, but to my recollection it is now about where it used to be. It is also gripping well. I autocrossed the car yesterday and it was fine.

Two (or maybe just one) very strange thing is that as soon as I ran the car after all of this, I found a strange problem at idle: it sometimes doesn't want to idle a a high enough speed, and when I let off the throttle and clutch (like when coming to a stop) it will almost die. It's as if the IAC isn't active at these times, but then at other times it runs fine. I would expect if there were a problem in the harness, then the problem would happen all the time. Ditto if were related to the below problem. I wonder if the IAC can have an intermittent failure mode? They are not cheap to replace, so I am hesitant to just throw a part at it.

Another issue that may or may not be related is that about 10mi after I started driving the car, the speedometer stopped displaying and the odometer stopped counting (no cruise control either, of course). I realize that the engine does certain things about drop-throttle and idle based on the VSS signal, so this could be related to the above problem. But keep in mind that the idle problems occurred before the speedo problem. The VSS is outputting a good signal. So now I need to test the continuity of the VSS circuit to the PCM. I see no evidence that the harness got damaged during the clutch job, but who knows? I am very frustrated by this!

ETA: Rockauto has a Delphi replacement IAC for $68. So not as expensive as I thought. Current plan is to diagnose and fix the VSS first, then address the idle/drop-throttle issue if it still exists after that.

MatthewMiller

Apparently they can. I ended up replacing the IAC that seems to have fixed the idle.

Still haven't solved the speedometer though. I believe the PCM is getting the signal from the VSS, because otherwise there would still be some issues with the car not running right on drop-throttle while rolling, and I would expect a DTC code for the lack of VSS signal. My best bet at the moment is that I screwed up something in the harness going from the PCM to the CCM, so the speedo signal isn't reaching the CCM. That harness is right around the clutch master cylinder, and the speedo stopped working when I replaced the m/c. So I need to pull the PCM and battery and really look at all the wiring down there.

Tom400CFI

Indeed. I had mine take a dump recently; it wouldn't stay idling when cold, but was fine when fully warmed up. Problem came and went for a week or two, finally seemed like it was a goner. I bought a new IAC, and the day I brought it home, the old one started working again...still working fine. So now I've got a spare.

I think you're on the right track w/the speedo diagnostic. Have you checked all the modules for codes to see if you have a communication error w/the ECM>CCM or CCM to cluster/

MatthewMiller

I ran the codes through the dash display, and got the following:The 81 code in the CCM is due to there being a 94-95 OBD1 PCM in place right now, but this has always been there since I've owned the car and everything always worked before (SYS light flashes intermittently on the dash, of course). The two ABS codes seem to have been a one-time error right after I started driving the car after the master cylinder change. It set lights on the dash, but cleared itself right away on the next startup. I can't see any problems with the cables at this point, but it seems a bit suspicious that the speed is also out...like there could be a relationship. The airbag error is expected since the car has a non-airbag Momo steering wheel.

I got no display at all for module 4, the PCM. It didn't even show a 12 code or the three-dash line. It just skipped past that module without ever even showing "4" at the bottom of the display. So I don't know if that is normal when there are no DTCs set for one of the modules, or if it means there is no communication between the PCM and CCM. What do you think?

Tom400CFI

I'm not sure. Any comments that I could make at this point would be WAG's. I would hit the FSM at this point, either starting w/"Speedo not working", or the C81 just to see where that took me.

Another option would be to PM Hooked On Vettes. He is a master at the CCM/PCM diagnostics, IMO.

Sorry that I can't offer a clear-cut diagnostic path.

MatthewMiller

The C81 doesn't worry me at all. I am sure that is a result of the previous owner installing an OBD1 PCM. One thing I plan to do in the near future is attempt to move it back to a proper OBD2 PCM (the original one developed problems).

The FSM implies that I should be getting a P1652 code if the VSS Output Circuit (the speedo signal from PCM to CCM) is not reaching the CCM. But I'm not getting that code. And again, I wonder if something larger is wrong at the PCM side (although the car now runs perfectly well). But in any case, the general idea it says is to test that circuit, and the PCM grounds, too (the VSS output is created by the PCM grounding the incoming VSS signal, so if no ground then no signal). If that doesn't turn up any solutions, then I will definitely PM Hooked on Vettes. Thanks for that tip and for being my sounding board!

MatthewMiller

I solved the speedo problem. It was the VSS all along. Somehow it was making a/c voltage, but I hadn't realized I should check for continuity across its two terminals. That should be around 400ohms. On my old one it was infinite - i.e., total discontinuity. D'oh! So I replaced it and now everything works perfectly.

So the lesson for today is this: when testing your VSS, check it for a/c voltage and also check it for continuity between its two terminals!

Tom400CFI

Nice work. It's always good to fully explore and find "proof".

VikingTrad3r

but tom and matt are alive and well on here.

in regards to the very high engagement point on these zf’s....i see you solved yours by simply getting a new master?

You had both slipping clutch and high pedal engagement point.

Do we know why these clutch systems end up with high engagement ponts as the hydraulics fail? Seems the complete opposite of what it should.

Further, ive read in a few places on the interwebz that “high clutch pedal engagement means zf clutch has worn”.

How is this possible?

I am diagnosing a high pedal engagement with a non slipping clutch 90 zf6

i also have a normal engagement with a slipping clutch on another 89 zf6.

MatthewMiller

My slipping clutch was only with the Spec kit and seemed to be caused by a shitty throwout bearing that puked grease onto the disk. There are no good bearings made anymore for the ZF6. The high engagement did seem to get better with the new m/c. Go figure. However, high engagement seems endemic to ZF6 replacement sets. I wonder if it's literally the disk just not being quite as thick as stock: that would do it. You can adjust that also by slightly machining the six "legs" or standoffs where the pressure plate mates to the flywheel, I suppose. There's a spec for the height of those standoffs relative to the p/p's friction surface, but it doesn't take into account the disk thickness. I don't know how to account for that with an installation trial and error. But yeah, as the disk gets thinner (wear), the p/p surface has to extend further toward the flywheel to get the same amount of engagement, which means the pedal will be higher at that point of engagement. No way to avoid that.

pedricd

I bet the variance is largely the disk thickness + flywheel (if SMF) + PP…. Maybe slave rod length or stroke of slave varies too??? EDIT: Although thinking about it again the entire system should self adjust…so that leaves master/spacer/rod length, stroke ratios and fork/pivot?

I know you are tired of hearing it… but RAM conversion you get a quality TO bearing and I think engagement point is a bit better…