That would suggest that there is no stretch or wear during the life of the belt.
Maybe that's so?
I have not found that to be true.
Actually - if you get the C/C right and measure the belt twist, why can't that be the gauge?
I didn't explain clearly enough (slugging coffee, launching for work).
"Center to center" is a (rough) value used to calculate how much deflection (think a short chalk line versus a long one) is appropriate for the belt in THAT setup.
NOT to measure "center to center" when adjusting the tension
"
In order to have a MEANINGFUL SPECIFICATION, it must read like: "3/16" deflection using 18 lbs. force."
To read/listen "half specifications" repeated around the campfire is painful.
I completely agree with your statement but most specs I have ever seen for chains/belts simply state some amount of deflection without any force spec.
I agree, but think about the variation to that. It kinda assumes a seasoned mechanic knows how much force to use to twist or deflect.
For example, from a Lycoming diesel service manual: "cylinder head torque: one man on a 6' wrench" in that case, 1000 to 2000 foot pounds will work depending on the size of the gorilla.
Same thing with belts.
The Commando belt drive is a compromise, to be sure. That is, to set up the belt to a tension the BELT would last the longest and be most efficient, would be WAY TOO TIGHT for what the clutch, overhung on the mainshaft, supported with sleeve gear bushings can tolerate.
So really, it comes back to as tight as the rest of the stuff not designed to have a belt & sprockets on it will tolerate. So, "as snug as decent clutch (therefore gearbox operation) operation will accept. "