Volume Control and Sound Quality

[mosorensen]

This may be an easy question, but I haven't found a good answer anywhere.

I want to A/B test a couple of dacs and headphone amps, and I found a Niles SV-5 selector box to do the switching. The Niles box is essentially a passive pre-amp, with a five-way input selector and a volume pot. My question is whether this volume pot will have any effect on the sound, if it is left at max volume? (my guess is "no", but I cannot figure out how to think about it)

Some technical details: (1) According to the Niles schematic, the volume pot is 100K, so leaving it a max volume effectively places a 100K resistor in parallel over the signal. (2) The dacs have output impedances around 30 ohm (3) The headphone amp has input impedance around 50K ohm. (4) All interconnects are short (around 1 foot).

I am wondering whether the volume control (at max) in the selector box affects the sound, if I am running either:

A: dac > selector box > headphone amp > headphone

B: dac > headphone amp > selector box > headphone

(Setup A is for A/B testing the dacs, and setup B is for A/B testing headphone amps).

Thanks

 

[jan.didden]

I cannot imagine any effect, although there will surely poeple who will come up with some senseless 'explanations', especially if they don't like your results. 😉 And then there will be those that will state that the switches will destroy the sound. Hard to win this one.
But you can always open up the box and temporarily disconnect the top of the pot from the signal.

Anyway, if you leave the control at max, how are you going to make sure that the levels in the various situations are matched (to a fraction of a dB)?

Jan

 

[PMA]

The Niles box is essentially a passive pre-amp, with a five-way input selector and a volume pot. My question is whether this volume pot will have any effect on the sound, if it is left at max volume? (my guess is "no", but I cannot figure out how to think about it)

Without high resolution measurement it is impossible to tell. Depending on design and wiring, even the passive box may add unwanted power line related spectral components (multiples of 50Hz or 60Hz) that may be coupled by EM field or may depend on wiring of the whole audio system. As usually, there is no simple answer for simple question, though from simple engineering point of view your question might have appeared simple.

 

[mosorensen]

Thanks, this is very useful. I will just leave the pot at max and ignore it for now.

Jay: About volume matching, both dacs have calibrated outs plus their own volume controls, so that should to it.

I think I got confused by trying to square two different thoughts: (1) Logically, as the resistance goes to infinity, the effect of the pot should vanish (having no pot, is effectively like having an infinite resistance in parallel over the source). (2) The threads I have seen about passive preamps suggest that the pot should have a resistance that is less than (about half) the input impedance of the following amp, which is this case is 50K. So I was left thinking that the resistance should either be as large as possible or around 25K, and one of those cannot be right.

Anyways, thanks for your help.

 

[DF96]

The pot resistance does not go to infinity, it goes to its maximum value. At this position, the main issue is the output impedance of the source and its linearity not the input impedance of the load.

The input impedance of the load becomes most important at around -6dB attenuation, which is just below maximum volume for a log pot.