1

I study and record bat calls using consumer audio devices. I've been using the Fireface USB audio interface series (802, UC, UCX) along with their other 4 channel preamp devices.

The gain knob has 3 positions at which the gain is defined - 6, 30 and 60 dB, at around 7, 12 and 5 o'clock. See cropped knob image here: enter image description here

I called customer support to ask for the gain values in between, but it didn't lead to any new info.

Edit round2: Here are my measurements (feeding a 50 mVpp, 50 kHz sinewave from an oscilloscope), the recorded audio level is reported in dB rms for each knob position). All angles are reported relative to the 6 dB starting position in clock-wise direction. X-ticks are not to scale

enter image description here

I know the gain values increase non-linearly based on this plot. The thing is that I've done a lot of recordings at intermediate positions, and am now wondering how to model this nonlinear change in knob gain.

As you can see from the plot, I don't really have a direct measurement of the gain to say if the 6 dB position is actually 6 dB, but I think it's safe to assume so.

Assuming so, my calculations tell me the gain at the knob positions are (see plot below).

enter image description here

However, does anyone have any ideas of what the typical trend of increase for such knobs is (square, exponential, log..etc?) - in relation to standard amp circuits or any kind of electronic designs?

Raw data for the plot
# degrees, 6dB position is 0 degrees
angles = [0, 36, 76, 112, 152, 192, 230, 268, 302] 
# calculated using relative gain with 6 dB position as reference.
measured_gain = [6, 7, 10, 21, 29, 35, 41, 46, 59] dB 
(Admittedly there might be a +/- 1 dB error in my measurements due to their manual nature.)
9
  • Just measure it. Shouldn't be a problem if you have the interface in your hands and you understand the math of decibels. Being able to make a custom attenuator (resistors, cable) for comparison would be useful.
    – user35252
    Jul 14, 2022 at 18:45
  • Hi @user287001, I have measured the gain at various positions, but would like to have a 'model' to compare with.
    – Thejasvi
    Jul 14, 2022 at 20:19
  • Publish the list of clock positions and measured gains. Then we may see (or not) which preamp gain control circuit fits. A good enough gain vs position formula also may be possible to generate without guessing anything of the circuit.
    – user35252
    Jul 14, 2022 at 20:32
  • (continued) In modern high cost gear the gain can be controlled in the actual circuit digitally (=switching a resistor network). The knob can be an encoder. Clever timing of the changes prevent the zipper noise.
    – user35252
    Jul 14, 2022 at 20:47
  • In a simple device, a logarithmic gain pot will give an apparent linear volume change.
    – Tetsujin
    Jul 15, 2022 at 8:22

1 Answer 1

2

If you look at where the stated values fall compared to the dial positions, you can see something that approaches a straight line.

enter image description here

I couldn't find a sensible function that maps to this extremely shallow curve without using contorted mathematics or a ridiculous number of decimal places and exponents. I have to conclude from this that the values stated on the dial are not entirely accurate.

The '6' and the '60' might be accurate but the '30' is probably rounded from something less aesthetically pleasing.

For a perfectly straight line, each of the dial markings would indicate an increase of 6.75dB. This would mean that the '30' marker is actually '33'.

However, as this is not laboratory-grade equipment, I would suggest that the markings are a rough guide only and that you should use the meters in your software to dial in the values you want.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.