Clutch master cylinder misalignment

Hi guys,

My LS1 conversion has 3 track days and about 200 street miles since finishing it a few months ago.

There have been many of the normal sorting issues, and I'm still tinkering with brake bias, alignment and a few little things - like the motor barfing oil after I finally overfilled it after many tries.

Those things aside, it's a blast. It's well balanced and you can definitely do stuff mid-corner with a 944 that you couldn't get away with in a 911. Love the NASCAR soundtrack as it pulls up the straights. It needs stiffer springs and bars, more tire and probably a wing would help especially cresting hills in 3rd gear. But, overall it's a LOT of car.

Here's a problem that's developed.

I'm running the .88" Wilwood MC and a custom pushrod. Stock Clutch pedal with stock spring, travel and geometry. Not that it matters, but a Stage I clutch and aluminum flywheel.

Until last Thursday I've not had a DNF at a track day with any car I've run. This was a Porsche dealer sponsored event, so the little grey 944 didn't get a second look from anyone (but the corner workers who usually know better) until it started eating up all the modern stuff...but that's probably no surprise to you guys.

At the end of the previous track day, the clutch MC started leaking a tiny bit...just enough to add some fluid. I freaked out a bit thinking about the implications of pulling the bellhousing and all the rear suspension...blah blah. The slave cyl was dry when I checked after that day, phew. So, I thought it was leaking from the reservoir (a motorcycle part I adapted).

I didn't worry about the leak too much and ran a full session, then during my 2nd session the clutch pedal started to go, so I came in after about 5 laps.

I couldn't get it bled and couldn't pull fluid through with a vac. Then the pedal went straight to the floor and stayed there. It wasn't until I looked into the pedal cluster that I noticed it was leaking a LOT from the main cylinder bore. The carpet was wet from fluid. So, we pushed the car on the trailer for a short track day. It was a big bummer to DNF with only about 30 minutes track time, and at 10am. Oh well.

Just tonight I got the MC out and the piston is toast. It looks as though the shaft is out of alignment, which caused the soft aluminum piston to grind itself away on the top of the cylinder bore. It's the end of the piston nearest the shaft. Obviously it is way out of alignment. I hadn't worried about clutch alignment because the pedal ratio and relation to the MC is unchanged from stock - but obviously this is a problem.

See pic below - if you have a strong stomach.

Has anyone seen the same issue?

I have a rebuild kit and will hone the bore, so the MC can be saved. But, it doesn't make sense to put it back together like this. Any suggestions?

Thanks as always.
Dan.
[You must be registered and logged in to see this image.]

Could be possible that you left some of the old paper MC gasket attached to the firewall and it is cocking the new MC at an odd angle.

Good idea, but no. I'm using the firewall plate from Sloan and the cyl is mounted flush.

I compared the Wilwood with the OEM cyl and the difference is that the OEM part - the OEM is longer and would have a different angle with the shaft. See below.

[You must be registered and logged in to see this image.]

I'm not sure if that matters though.

Messy huh?

Dan.

The problem is that the firewall plate needs to have that clutch MC firewall angle designed into it as well. So you have a SEVERELY cocked MC now, that's the problem. I bet your worn MC is worn at 12 o'clock where the convex pushrod interfaces with the Concave piston area?

Rob,
I'm also using the stock mc with Sloan's firewall plate.
What do you suggest to adjust the mount angle?
Mike

Hi Rob,

I hadn't heard anything discussed in the past about this. I'm using the 3/4" wilwood and clutch pushrod you fabbed for me.

Eric and I worked out an angled brake MC mount for the brake cyl reinforcement plate he made. That was just as much to get the MC up and away from the cylinder head as to improve the rod angle. Until now I wasn't aware there was also a need to do the same with the clutch. On my car, the studs coming out the firewall are pretty short - so it's going to be a big deal to change that angle significantly.

The clutch MC is located in the same hole as stock and pedal geometry is also unchanged. So, the alignment of the pedal mount and MC bore is unchanged. I'm not sure how the difference in length of pushrod matters but obviously is is a problem.

And yes, the MC is scored across the top. I have a track day next weekend so I'll probably end up honing it, then doing a rebuild and putting it back on the car in order to use the car next weekend.

I haven't read about anyone having this issue before - and the Wilwood 3/4" MC seems to be a popular setup.

Thanks for the input.
Dan.

There is a reason why Porsche stamped an angle into the firewall sheetmetal at the Clutch MC opening, just as they added the angle to the Brake Booster "cereal" bowl. I always wondered if the firewall plate would cause a problem because it sets the Clutch MC parallel to ground when it should be angled up. The 12 O'clock MC piston wear/damage seals the deal in my mind. The fact that the Wilwood MC piston concave interface sits farther away from the pedal pivot point than the factory means that it would be less prone to off-angle pressure than the factory MC. I have MIC'd both and have the drawings. The pushrod adapter only makes up the difference between the factory pushrod length and the Wilwood pushrod length and also allows for adjustability. My larger 7/8" Wilwood MC is mounted to the factory firewall and have ZERO problems.

I just checked with my magnetic angle finder and there is exactly 9* on the pushrod angle between my factory pedal pivot location and Wilwood MC that is mounted at factory angle on the firewall. My suggestion would be to make a 9* adapter between the firewall plate and whatever MC you are using.

I have the exact same problem. Just with my master brake cylinder from wilwood. I still have the cereal bowl from the LT conversion so I'll have to measure it.

But on the Brake MC it can set parallel if you drill the pedal pivot/clevis up 1". To that guy who modded his clutch pedal for more throw by lowering the pivot/clevis point.....take warning.

xschop wrote:But on the Brake MC it can set parallel if you drill the pedal pivot/clevis up 1".

Already did that. Requires an angle. I can hear/feel the piston scraping.

What does your clevis adapter look like? I would make an offset adapter if I was running a single MC like the tandem. I went thru the trouble of machining the angle into the Dual MC housings, but the fact they also have a heim joint on the pushrod makes up for any small angle irregularities.

For the brake MC using a dual Wilwood Eric made an adapter plate that angles the MC up and away from the cylinder head. It was more for clearing they cyl head, but an added advantage was a better angle with the pushrod. BTW, I'm not running a booster and ended up drilling the brake pedal up 1". At .75" the pedal feel was too firm with my caliper combo.

Notes here: [You must be registered and logged in to see this link.]

For the clutch cyl I'm trying something else. I removed some material from the mounting face, so the cyl is now at an angle pointing up. From looking at the shaft angle from inside the car, it looks to still be somewhat out of alignment, but hopefully not enough to cause issues. The clutch is pretty stiff, so there's a lot of force being transmitted up through that shaft and onto the piston/bore.

In the long run, I think another solution is needed so the shaft is in perfect alignment - 9 deg up?

Again, I'm really surprised this hasn't come up more often.

Thanks guys,
Dan.

Dan J wrote:Again, I'm really surprised this hasn't come up more often.

I dont think its going to either. Theres not alot of people running aftermarket clutch and brake MCs. I havent run my car yet and so didnt think too much about my brake MC yet but did notice the scraping sound/feel when bleeding. I just thought I got a bad MC from wilwood and would deal with it later (lot of other things to do yet). But after you posted this, I thought .... duh. With my LT conversion, I was using the cereal bowl which angled the MC up. But then, after installing the LS, I bought one of Gregs firewall plates and didnt even think of cereal bowls purpose.

Good thing you posted this!

This is crappy news... I'm just about to bleed my brakes / willwood tandem master cylinder. I drilled the clutch pedal pivot up 1" too. I'll see if I have the same problems as 944-LT1 and let everyone know.

spence wrote:This is crappy news... I'm just about to bleed my brakes / willwood tandem master cylinder. I drilled the clutch pedal pivot up 1" too. I'll see if I have the same problems as 944-LT1 and let everyone know.

Great. Let us know!

Spence - how does your MC clear the cylinder head/valve cover? Do you have much clearance?

Dan.

There's lots of room with the LT1 using the willwood tandem... not sure about the LS

I just bled my brakes. I've got most of the air out; the pedal is still a little spongy. I'll finish it up tomorrow. I did not hear or feel any scraping noises. Everything seemed OK

I too have the Sloan reinforcement plate. Would it just be better to cut that section out? It's already bonded to the firewall.

I don't even see how I could add 9 degrees anyway. The master cylinder is too long and is already close to the wheel well part of the engine bay. I take it a hammer will fix that problem?

Another idea...clutch MC

The studs on the firewall are only just long enough for bolting the Wilwood MC with the sloan reinforcement plate. One idea I'm going to try on the Wilwood MC is to machine the 9deg angle (or something close) on the mounting surface of the MC.

The optimal way to go - a lot more fabrication required, but may be worth thinking about for those with the budget and skills - is to use a full Tilton pedal cluster mounted to the floor. Something like this:

[You must be registered and logged in to see this link.]

Dan.

**1986 Porsche 944 Turbo with LS3***
Wow.  I saw the picture in this post as I was ordering another Wilwood clutch master cylinder.  I took mine apart and thought "What the hell?"  It was just like your picture as you can see in my pictures.  This losing fluid and getting air in the line has haunted me, now I have the problem identified.  

Problem identified:
1.  People have been using a large firewall plate which changes the angle of the clutch master.  I have been using a firewall brace only for the clutch and it has created the same problem. If you pull the brace off, you can see Porsche stamped an angle in the firewall to bolt the master to.  The firewall plates are changing this angle which as I read in the past posts is about 9 degrees.  
2.  Wilwood tolerances only allow about 5 degrees of angle that I have read about which make me about double out of Wilwood tolerances.  

My fix for now:
1.  Remove the firewall plate.
2.  Find a different way to brace the firewall after everything is mounted stock as Porsche intended.  

Has anyone solved this?  Please post the fix if you have it.
[You must be registered and logged in to see this image.]

For whatever reason the Tilton master cylinders aren't doing this. Only seen it from the Wilwoods

I solved the Wilwood clutch master by first narrowing the mounting holes as we all know needs to be done, but also shaving down the outside mounting corners at an angle just enough so it does not contact the firewall. Before shaving these corners, I found that the Wilwood is slightly wider than the Porsche firewall stamp. When you test the fit, it would teeter up or teeter down since the outside corners were contacting the flat side of the firewall and not truly mating to the 9 degree surface stamped in the firewall by Porsche. After that I modified my brace to install after the master was installed. This made the master flush with the mounting surface and the brace still functions. I hope this helps someone. My master is now working with outstanding results. I have the Wilwood 13/16.