Adding to the discussion from Pirate... with no pressure on the clutch pedal, you want the TOB to sit near but not in contact with the pressure plate fingers. This gap is the freeplay you feel when you depress the pedal, before it gets significant resistance. If you have insufficient free play, the TOB will wear prematurely.
The pressure plate clamps the clutch disk between itself and the flywheel. Lifting the pressure plate requires hydraulic or mechanical pressure, which is what you 'feel' at the clutch pedal. There's generally also smaller springs built into the pedal and slave, but most of what you 'feel' comes from the 1500 pounds or so of force from the pressure plate.
As noted above, IMHO setting this up out-of-chassis helps, especially if you build some adjustability into your brackets, linkage, and/or pivot points. You'll be able to visually see that you're not over-extending your master or slave. If you over-extend the master, it just stops moving; if you over extend the slave, it will dump the fluid.
Brewing all this custom, you may need to consider leverage and hydraulic multiplication, but that isn't that hard if you're systematic. Figure about 1500lbs at the clutch, with 2:1 from the clutch fork, 1.69:1 from slave over master, and 6:1 from the pedal. I'm guessing/recollecting, so check all of these with a tape, but that would give you about 74 pounds at the pedal, which is a nice neighborhood. If you need more or less throw that starts to get screwy with master/slave size, effective leverage, etc. (BTW, ignore what's in the Novak article about all vehicles being 1:1 -- mebbe that was once true with Jeeps, but it is not necessarily true for Isuzus, Toyotas, etc.) IMHO, cycle the parts you have through their range of motion and measure throw, then work through the math ... the details will shake themselves out and let you know which way to go.
Ideally, you'll test full engagement for slip against a running engine, but another good way to do it is to set friction plate air gap. Checking release is also easy if you can see in the bell (maybe through the fork hole). I ended up drilling a resealable hole in my bellhousing to be sure I could atch everything work, but that's a bit of overkill. Working blind sucks.
IMHO, don't modify the Dak fork unless you have to -- you should be able to do an adjustable-length pivot and/or slave rod to allow you to fine-tune fork placement. Suggested working order: tweak system to ensure sufficient pilot engagement, splined engagement and friction plate movement, and TOB slide-space on input snout. THEN work the linkage.
Be systematic, and this will come together well. Unless you know somebody who has put this EXACT swap together, there's no template. You can consider the advice that folks give you, but remember that their input may have been different length and/or spline placement, their nosecone different diameter and/or length, their linkage wholly different, and more.
Ain't it fun to break new ground? <img src="/forums/images/graemlins/evil.gif" alt="" />
Randii
