Switching between inverting and not inverting a small signal

Question

I want to be able to choose whether an audio signal being fed into an amplifier is inverted or not via a TTL signal. What's the simplest, cheapest, best way to achieve this? My goal is to replace the mechanical DPDT switch in this guitar sustainer circuit with something silent, controlled electrically, and can be assembled with through hole components.

Here is what I am currently thinking. So you can invert a signal with DPDT switch pretty easily. Perhaps a DPDT solid state relay exists? I found one on digikey, but it was $170. I read I could make a DPDT switch out of an arrangement of two SPST-NO and two SPST-NC switches. SPST-NO switches are cheap, however SPST-NC switches are $20-$30. I can buy four SPST NO switches and one not gate so that I can flip the logic for two of them. Below would be the result. Here's the link to the circuit diagram: https://crcit.net/c/de4ea0a434f9430594dfb446a37e6f1b

The circuit diagram image for SPST SSRs makes them seem easy to replace with MOSFETS, but I wouldn't know which ones to use. Is this an avenue I should explore?

What I have is cheap and simple enough, but I don't know what effect it will have on the signal. I know the SSRs add 6 Ohms of resistance each, but no idea how it might end up filtering the signal. If you know a better way to do this, I am all ears!

Answer 1

My goal is to replace the mechanical DPDT switch in this guitar sustainer circuit with something silent, controlled electrically, and can be assembled with through hole components.

Consider the analogue switch (also called a transmission gate) and sometimes (on more esoteric circuits) it's called a multiplexer. You can get them as 2:1 (SPDT) and they route signals bi-directionally so you can make your signal inversion using two 2:1 devices and create the equivalent of a mechanical DPDT switch. On-resistance is a factor but I'm sure a few tens of ohms isn't going to be a showstopper.

There will be many available through hole devices to hand.

Or, you can make a 2-qudrant analogue multiplier. It's basically an invert/non-invert linear circuit that with the appropriate control voltage (logical 1 and 0) inverts your signal. You can also do this with analogue switches using a high speed control signal (with variable duty cycle) and get a smooth changeover from non-inverted signal to inverted signal.

Answer 2

My goal is to replace the mechanical DPDT switch in this guitar sustainer circuit with something silent, controlled electrically, and can be assembled with through hole components.

There's no such thing.

I've installed hundreds of true bypass systems in guitar pedals over the years - the same applies to your guitar. DPDT switch is the only reliable way. You'd need to install an extra tiny DPDT switch on the guitar.

If you insist on the electronic solution, RG Keen has a series of articles on quiet pedal switching which you can adapt into an active circuit for your guitar. But all solutions will likely involve FETs in a flip flop circuit which will insert a few solid state components in your audio path. This will go against your requirement of it being silent (while it might not go pop when you switch, it will necessarily load your signal). Also JFETs won't be through hole unless you use a breakout board.

Answer 3

Also, one other possibility... if you were to use a single-bobbin humbucker for your sustainiac input, and if you had access to the coil tap, would it not provide you with a 180d signal by default?

If I'm remembering my stuff correctly, you could then use transistors in an either/or config to route one feed or the other to ground. Run each feed through its own coupling capacitor (to ensure the ground doesn't eliminate BOTH signals through the confluence point,) and sum them together. You may still need some anti-click capacitors somewhere, but this would get you the phase control using only a single GPIO output.

(Single-space humbuckers have rotten tone when the coils are split... but you don't care too much in this case as you aren't listening to the output... you're using it to propagate a magnetic field, which shouldn't be terribly sensitive to 'tone,' as the inductor coils themselves will have awful frequency response output by their very construction.)