i highly doubt that the gain pot on the ua5 is nonlinear.
Perhaps I'm using the wrong terminology. Let me explain it this way: the gain knobs seem more sensitive as I turn them up. For example, turning the knob one clock-hour, from 09:00 to 10:00 seems to increase the gain less than turning the knob one clock-hour from 04:00 to 05:00. I've not measured this, but that's the best way I can think of explaining the increased sensitivity I sense at higher gain.
brian;
i understand what you're saying and you're 100% correct. i'm quite sure that gain pots in the ua5 is linear; however, the actual resultant GAIN is not.
hopefully the following 2 graphs will help to explain.
(this is based on Analog Devices SSM2017/2019 Gain equation).
i plotted 100 resistor points representing an infinite gain pot (10000 ohm > 100 ohm). this is column A.
column B is the gain based on the gain resistor (Rg): G=1+(10000/Rg)
column C is the gain in dB: A=20*log(G)
x axis (linear) is Rg, y axis (linear) is dB:
you can see that as you turn the gain pot linearly, gain in dB goes up non-linearly (logorithmically, in fact).
if you change the x axis to log scale, then it staightens out.
marc