Originally Posted by
cyccommute
The ICAN article that you linked actually does give a value but says that the tension can be ±20% of their recommendations. Let’s look at the implications of that wild ass guess. There is a CN494 spoke that they use with the “ICAN standard hub”. They recommend nondrive side tension of 51 kgf and drive side tension of 96 kgf. That would be a reading of 17 and 23 on the Park TM-1 meter. But ±20% means that the tension can vary from 40kgf to 61kgf on the nondrive side and 76.8kgf to 115 kgf. I have experience with the TM-1 and know that a reading of about 14 on the nondrive side would make for spokes that are too loose.
Additionally, as someone who made a living measuring stuff, I gotta say that ±20% is a piss poor measurement. It’s just a wild ass guess. If I submitted a measurement with that kind of variance, I’d have been scolded and/or fired.
I always took that +/- 20% to be the tolerance for relative spoke tension - the difference in tension between different spokes. I agree that's a wide range, but that just means there's enough resilience in the overall wheel build to accomodate that, as long as you stick to those guidelines. I haven't seen anyone suggest that spoke tension has to be
exactly the same, and wouldn't be achievable anyway. It's one of those cases where "close enough is close enough". That spec just tells you what close enough is.
And a spec of +/-20% isn't just a wild-ass guess. There are a lot of things that can operate with that range of tolerance. Depending on where they're used in the circuit, capacitors can easily have that tolerance, for example, as well as many other electrical components. I think some of my bicycle tires might have a wider range of inflation between minimum and maximum.