Well, the Krikit tension tester arrived today, and I have to say that I don't entirely trust it (or perhaps the Zero tension tester is a much more precise unit?). Maybe I'm just old-school, but I prefer the method where a certain amount of pressure is applied up and down on the belt, and the resulting distance is measured (like on a motorcycle chain, or like the early belted motorcycles...my old Honda 250 being an example).
Anyways, on the Zero, when I put just 2 or 3 pounds of pressure on the belt (up and down), I get a resulting distance of about 1/2" to 9/16". Does that sound about right?
After i got my krikit, i tested making several measurments. The tool itself seems to give the same measure if i measure repeatedly at the same spot. However belt tension seems to vary from time to time, and side to side.
It would seem that the proper thing to do is always check on both top and bottom side of the belt (without moving the wheel/belt/bike) , and do several measurments having moved the rear wheel between. I believe this might be the best practice given any measurment method/equipment.
In my opinion, the correct measurment should be (if possible) when the top and bottom reading is equal. Trying to deduct an average might also be good, however if the deviation is large, i would guess moving the wheel slightly and then trying to do it again would be the best option.
The easiest way to achieve a good readout should be with the rear wheel off ground, however if one is able to get reasonable equal readings on both sides of the belt on ground, I can't see any reason why that reading would not be good enough..
The real question about the krikit (as with any other tool) is whether each gauge measures the same...