-------------
For timing changes due to shaving the head, my back of the envelope math is:
- cam gear is 34 teeth, or ~10 cam degrees per tooth
- the gear and belt pitch is 0.375" per tooth
- shaving the head by 0.1" needs ~1/4 tooth of backwards cam gear rotation to take up the slack, which gives a change of ~2.5 cam degrees (counter clockwise / retard) per 0.1" of head/gasket reduction
- the tensioner needs to take up 2x the shaved distance, or a shorter belt is needed.
Reading this and thinking about it has made me revisit the math I've thought I had locked in for like a decade. As it turns out, I was wrong in stating that every 0.010" you mill retards the cam 1 degree.
The one thing you missed here is that yes, there are 38 teeth, but also 38 valleys, for a total of 76 "positions" on the gear. I just went out and measured one, it's 4.48" across the teeth, 4.31" across the valleys, let's call it 4.40" mean diameter. That makes the circumference 13.8226". When you divide that by 720 (the number of crank degrees in 1 combustion cycle, how a cam is measured), you get 0.01919, or about 0.020". So every 0.020" you mill the head, it retards the camshaft 1 degree.
My error that I've held on to and argued I had compared to results from a degree wheel (which I remember doing but I've gotta say, perhaps mistakes were made) is that the belt gets shorter on both sides of the cam gear, which it does. Let's say you trim 0.040" off the head, well in order for all the pulleys to stay in exactly the same place the belt would need to be 0.080" shorter, which is correct, but it would not require that the cam gear rotates anti-clockwise 0.080", retarding it 4 degrees, it rotates anti-clockwise 0.040", retarding it 2.
I really apologize for the goof and the number of times I made this argument. I'm also going to go back and look at the math on the cam gears we made to see where our cam timing actually was.
My brain hurts.
BIG EDIT
OK, I couldn't let this go. I chucked the K cam in the race car and measured it up. Volvo says that with 0.5mm valve clearance the intake valve should open at 22.6* BTDC. We found that in our application, it opened at 12* BTDC, or 10.6* retarded. The head is milled 0.109", and we use a 0.030" head gasket, which, to my knowledge, is 0.017" shorter than the factory gasket, compressed. THUS: 0.126" removed between head and block nets 10.6* retard, or 0.0118" per degree. Bear in mind our dial indicator is only accurate to 0.001", so there is a margin for error here, but in general, you cut 0.010", you retard 1*.
PHEW.