AC Audio, Switches, and Potentiometers. Oh my.

[twjtwj]

I have the following circuit:

Path 1: Audio in -> audio switch 1 -> pot 1 -> audio out
Path 2: Audio in -> audio switch 2 -> pot 2 -> audio out

I've also tried the pot / switch order reversed:

Path 1: Audio in -> pot 1 -> audio switch 1 -> audio out
Path 2: Audio in -> pot 2 -> audio switch 2 -> audio out

Keep in mind only one switch is open at a time. Although the switch I use does have a max of 10ohm leakage. Also the other audio wires are all grounded together.

What I've noticed with either setup above is that if I use the pot for the open channel (audio switch not allowing ac through), it effects the volume for the other (closed, lets AC through) channel. It's not quite linear, but if I turn it all the way down it shuts off audio completely. 

What I think is happening is that because audio is AC, it still travels back to the other pot from the other side, even if it's channel is open. So in the end what I did is this setup that worked:

Path 1: Audio in -> audio switch 1 -> pot 1 -> audio switch 2 -> audio out
Path 2: Audio in -> audio switch 3 -> pot 2 -> audio switch 4 -> audio out

Now when Path 1 is open (won't let AC through) it will open switch 1 and 2, thus the AC can't sneak in and be effected by the pot's resistance. Voila.

Question:  Is this thought logic correct? The reason I'm asking is because I'd really like to get away with only using 1 switch per path.. I'm thinking 2 switches per path if overkill, or is that what's really needed here?

Thanks a bunch!
Todd

[David_AVD]

Maybe draw a schematic and show what you've tried?  Doesn't have to be CAD.  A hand drawing will do.

[twjtwj]

Here you go. The top 2 are the non-working situations. I think these would work fine in a DC situation, but since I'm using audio/AC, I need to use the bottom setup. Correct? Any other way around it without using two switches?

[David_AVD]

The top-right circuit should work.  If not, you have not wired it as per the diagram shown.

[AndyC_772]

Agreed, the top right one is OK.

Forget any idea that ac and dc will behave differently in these circuits, because that's not the case - at least, not at audio frequencies. In your first circuit (top left), you've just not considered all the paths that exist for current to flow, and that's why both pots still have an effect on the signal level. It's nothing to do with ac behaving differently from dc, guaranteed.

[twjtwj]

Let's add to the mix.  I'm not using actual hardware switches, but IC audio switches (MAX320). This consists of 2 SPST switches that are NO controlled by HIGH/LOW on the switch pin.  I also have pull downs for the high/low switch inputs (IN1 / IN2) (tried with and without).

So in the case of using this IC would it add an issue to the above right schematic? Perhaps the IC open switch acts differently than a hardware open switch?

Thanks for all of your help. I'm just trying to cover all of the possibilities/thoughts here to avoid adding things to the mix I don't need to solve/debug for.

[madires]

Here you go. The top 2 are the non-working situations. I think these would work fine in a DC situation, but since I'm using audio/AC, I need to use the bottom setup. Correct? Any other way around it without using two switches?

Top left: If the input signal is switched to pot 1, the right part of pot 2 is in parallel with the right part of pot 1. And vice versa.

Top right:  Correct circuit, but double the resistance of the pots. Since both are in parallel between signal and ground you halve the input resistance.

Bottom: Overkill :-)

[twjtwj]

So you're saying if I want two 500K ohm pots in the case for the top-right chart, I should instead use 2 1Meg ohm pots to get the desired circuit? Since the bottom schematic separates the ground and input, is that why they work as planned?

[madires]

So you're saying if I want two 500K ohm pots in the case for the top-right chart, I should instead use 2 1Meg ohm pots to get the desired circuit?

Yes, to have a proper input resistance matching your audio signal. I don't know the output resistance of your audio source, but 500k is quite high. For a standard line-level signal you would use 10k. Please check this! It might be the cause of your trouble.

Since the bottom schematic separates the ground and input, is that why they work as planned?

Both work because the outputs of both paths are separated.

[twjtwj]

It's for a guitar signal. The standards are either a 500K ohm pot or 250k ohm pot depending on what pickups you use.

[madires]

It's for a guitar signal. The standards are either a 500K ohm pot or 250k ohm pot depending on what pickups you use.

I see. If you still got problems with the top right circuit (assuming you wired everything correctly) you're facing crosstalk. In that case the bottom circuit adds enough crosstalk attenuation by the second switch stage (left one). If you like to experiment you could use some opamp to boost the guitar signal up to line level (around 1V, 10k ohms) and then use the top right circuit. If your guitar amp doesn't got a line-in stay with the bottom circuit.

[twjtwj]

Awesome. Thanks for the advice.

This would be used in a standard pedal setup for normal input. Although I was thinking of using an audio amp for a slight boost down the road (not in this circuit currently). 

In the end, if it costs more to add another good audio switch and it does at least benefit the setup (I did see crosstalk mentioned on the datasheet for that switch), then I'm okay with it. Plus I can interchange standard guitar pots with it.

Thanks for your help. It's good to know the "real" reasons I'm getting the results I see (crosstalk, etc.) instead of made up ones like the difference between how DC and AC acts in the circuit. I've learned a lot in the last year.

[David_AVD]

Let's add to the mix.  I'm not using actual hardware switches, but IC audio switches (MAX320).

Ummm... this would have been nice to know in the first place.   

Using an IC switch instead of a conventional switch brings in a whole new set of constraints.

If you need help with a circuit, please describe what you really have done.  Otherwise people waste time on diagnosing problems on the wrong circuit.

[twjtwj]

I did mention it in the first post that I was using audio switches and there was 10 ohm leakage. I just posted the quick schematic that was requested using the switch picture I had.