what can be confusing is from worn jets bending the tab changes the angle of the float sometimes to a pretty large degree. What used to "level" or parallel to the bowl gasket is now an angle pointing up, as is the front of the floats are doing a wheelie. The further the tang is bent the higher the floats will be for when the float pressure stops the flow from entering. So finding the correct measurement at any position or place along the float might hinge on luck. I tried unsuccessfully for a month, 20 times, bench testing the fuel levels with all the floats, jets and pins in their original place, using the measuring methods of "float height" 21mm. The results were either fuel pouring out the over flow, or no fuel at all getting into the bowls of random bowls. I think I blacked out. I then dumped all disassembled parts into a pile, then randomly picked a part to assemble the 40 year old rack of 4 Keihins rather hastily, and re adjusted each float to be even with each other somewhere around level or just before level when the tang just touched the spring loaded pin. And in the 21 clear fuel line test and after 2 or 3 attempts of carb installations, the 21st haphazardly, careless approach worked, both on the bench and after installation on the bike. I kind of knew they would, because in the middle of the 21 attempts, right in front of the carb rack, I would only "appear" to make an adjustment like a placebo effect, but didn't actually make an adjustment, and the subject bowl would react to the adjustment as if one was actually made, going the entire opposite way.
Syncing the carbs the same thing happened. Following the spec measuring directions, didn't work. I guessed at it looking at daylight coming through being the same, they worked. One of the carbs came off of another rack that were separated in use by 30,000 miles.
