Shimming the new clutch
Looks like you got a nice truck clutch there...LOL
First thing you should be aware of are the shims on that clutch. The reason the shims are there in the first place are to level the cover. Since this clutch uses a stamped steel cover it is not very precise. So what they do is use thin shims of varying thickness to "square" the cover to the flywheel. This is critical to a twin disk since having a cover that is off will cause release issues and uneven pressure on the clutch fingers. I had a LONG conversation with Red at the 2003 SEMA show about all this. I know for a fact those shims are not to be distributed evenly. This is the number one issue that guys using this clutch face. They get the clutch. Don't dear the instructions. Unbolt the cover fro the flywheel. shims go flying everywhere. They falsely assume that they are evenly distributed throughout the cover bolts. Clutch gets installed. It acts funny. They blame McLeod.
That is the reason why you have play in two of the fingers and the third is tight. The cover is not square to the flywheel!
Since you are in the LA area, I would drive that clutch to McLeod and have them service it. They should be able to tell you where all the shims need to go.
The other issue I have with that clutch is the Long style spring arrangement. With the stock style diaphragm spring there is a certain point in the spring travel where the effort gets a little softer. The point where the spring overcenters and it becomes easy to hold the clutch released. With the long style clutch, this is not the case. See the springs all around perimeter, between the cover and the pressure plate? Those apply pressure to provide the clamping force. Those are just like little suspension or valve springs. Their rate is linear. The more they are compressed the harder it is to further compress them. As a result there is no point in the pedal travel where it gets a little easier to hold the clutch pedal.
Good talking this afternoon Andrew.
Well damn, it looks like I have some sleuthing to do. I was afraid the shims were going to bite me. At least I was able to recognize that something was off. It really would have sucked to blow up a brand new slave cylinder and start the process over.
I have a total of 18 shims, 2 each on the three straps of the floater plate and 2 each on each of the 6 mounts points for the clutch cover. I think just need to go through with a set of micrometers and measure the various stack-ups to be sure the pressure plate and floater are both dead nuts square to the flywheel. Looks doable but perhaps a pain in the @$$. If I can't get it square I'll cave on sending it back to Mcleod.
As a general FYI, Andrew had one more suggestion that I thought was solid and worth mentioning. One thing the t56 bellhousing lacks is a test port to verify clutch disk clearances to friction surfaces with the pedal on the floor. Per McLeod it's supposed to have 0.020-0.025" total air gap on each disk with the clutch disengaged. Depending on the result of measurements I may drill myself a 1.5" hole in the lower face of the bellhousing so I can measure the assembly operation with feeler gauges.
I learned a lot about clutches this weekend even if I wasn?t as productive as I hoped. Forgive me if I blend SAE and metric units here, McLeod uses SAE in their parts but it was faster to measure metric as I was using an old-school slide type caliper.
My floater plate was perfect and had no issues. 0.020-0.025? is pretty easy to verify with feeler gauges.
My perimeter shims at the six mounting points for the cover were all exactly 0.5mm (0.020? each) and I did have two each under every standoff.
The fingers on the clutch were NOT riding at the same elevation which would probably destroy the throw-out bearing in short order if I leave it as is. Measuring relative to the disc below (which is flat and of constant thickness) one finger rides 0.9mm higher than the other two.
Here's my height differential. This was touching on the high finger.
The fingers are basically levers such that static height of the finger tip is dictated by: 1)
the location of the pressure plate relative to the cover (a pivot for each is cast into the back of the PP which gets pulled outward) 2)
and the shimming of the clevis that supports the pivot of each finger (barring wear of the bushing in the lever which are typically fine). These clevis supports are shimmed at the factory and cannot be disassembled without compressing the PP to release the force on them (the fingers bottom out on the frame of the cover when not assembled).
Shims under the clevis pivot point.
Another thing to watch out for in three finger clutches is that the axis of pivots at the clevis and pressure plate (the cast mounts) can get out of parallel alignment. One of my clevis supports was rotated slightly which explains why one finger felt tight and the others had some play.
Note the two pivots for each finger
I tried running without two 0.02? shims (1 each) on either side of the high finger to intentionally warp the cover and compress the PP more in that location but the 0.5 mm drop resulted in a 2.9mm drop at the fingertip putting the previously high finger significantly too low. Bascially I need something like a 0.033? total shim under each standoff at the high finger instead of the 0.040? I have for the cover to get close. I talked to McLeod and they?re sending me some shims. One trick they do on occasion is to start with a hair too much shim and hammer out some thickness on an anvil. Supposedly 0.01? total deflection of the cover is ~100 lbs so I should be fine with varying compression of the cover springs locally.
I need to offer a big thank you to McLeod for supporting their product line even in the case of my slightly used street twin. They sent me a stack of a dozen shims in various sizes I requested at zero cost, just so I'd be able to get this back on the road without sending the whole thing in. They do offer a complete rebuild service for about $450 that come new disks, springs, and redone friction surfaces.
I ended up running +0.004" on two cover mount points and -0.004" on one other. That was enought to bring all three fingers within about 0.005-0.008" to level with one another (tough to measure). I would see more than that at the fingers if my disk varies in thickness by 0.002" due to the lever arm at the fingertip. Figured that was about as close to flat as I was going to get.