I'd posted a triple 555 timer synth/sampler project on here a little while ago which I successfully completed but it's a bit of a hassle to use when I'm wanting to chug through a song. It used two knobs for effects, one for pitch/frequency, and required a careful ear. So now I've moved onto a guitar project in which I'd like to turn the waves from the guitar's pickups into square waves.
So far I've made a humorous circuit with very limited knowledge which slightly works. I first experimented with the LM324 op amp-chip as a comparator, using a few resistors for a voltage divider to V+, and a guitar jack which resulted in a very nasty output similar to sporadic white noise. But I was able to trigger a difference by plucking the low E string which generated a tone in the mist of all the noise. So through curiosity I used an LM386 in an amplifier circuit (https://ask.audio/articles/how-to-build-your-first-amplifier) to amplify the guitar signal which would be sent to the LM324. I fed the output of the LM386 amplifier circuit directly to the V- and fed ~2 volts to the V+ instead. My idea was that the larger band/gap between voltages would help the LM324 distinguish between the guitar's frequencies and smaller irregular frequencies that may occur in the circuit. I'm not sure if that idea's correct but it did something. The result was a... success?
I'm getting a nice 8-bit sound out of it if I play with around with the potentiometers to the LM386 but I continue to have issues with the smaller strings triggering the LM324. I expect they don't cause a big enough change to the field of the pickups so the amplifier doesn't peak the signal high enough to trigger the LM324. The same issue occurs when trying to get notes to sustain for long periods of time. As the vibrations from the larger strings die down the LM324's output quickly cuts out.

So the question is:
Is there a good way to go about getting a square wave output from a guitar?

I've looked around the internet but I've only found "this schematic sounds like square waves", "it's really easy", and "here's a link that leads to unknown pages or advertisements" pages which have little to no information. For anyone who may be wondering I'm doing this all to help my friend with custom 8-bit music for his Twitch streams. And who wouldn't want a guitar that sounds like an old Atari?

 

My idea was that the larger band/gap between voltages would help the LM324 distinguish between the guitar's frequencies and smaller irregular frequencies that may occur in the circuit.

A reasonable approach.

As the vibrations from the larger strings die down the LM324's output quickly cuts out.

Overcoming that will involve a compromise, because it conflicts with the above-mentioned approach. Methinks you will need a squaring stage having an adjustable threshold.
Have you checked out conventional fuzz circuits?

 

A reasonable approach.

Overcoming that will involve a compromise, because it conflicts with the above-mentioned approach. Methinks you will need a squaring stage having an adjustable threshold.
Have you checked out conventional fuzz circuits?

I've looked at a number of them but only came across basic circuits that slightly modify the input wave; making slightly it sharper, rounded off, or even doubled up. I've found a few square wave pedals and even '8-bit' pedals but no schematics. When time frees back up I'll mess around a bit with optimizing a threshold and possibly look into some op amps that are a bit more suitable for this application. After all's said and done I'm hoping to figure out a way to feed it into a XR2206 IC to give the player a larger variety of sounds. The end project's going to be making myself a customized guitar but that's getting a bit too far ahead at the moment.

 

Seems like it could be made to work, but with a lot of tricky extra circuitry to compensate for the 'edge conditions'.

You need some kind of tracking threshold that adjusts proportionally to the incoming signal, or an AGC to keep the signal level constant, regardless of the input. A troubling problem is when the signal gets really low, the circuit will become a noisy mess.

It needs a way to gracefully suppress the output as the signal drops below some minimum level.

Ditch the LM324 which was designed more than 40 years ago in favor of a modern, higher performance opamp, my favorite is the MCP6024, everything works cleaner and better with a good contemporary opamp.

 

As he says ^. Maybe a simple squaring circuit could be fed with the output of a sustain pedal, which is essentially an AGC device.

 

Maybe I'm misunderstanding the intent here, but it seems like tons of gain and heavy clipping should get you a square wave pretty easily. Really heavy solid state distortion looks pretty darn square!

If it's harder to trigger high strings than low maybe a simple pre filter could be used. If you run your guitar signal through 3, 6, or maybe 12dB/octave filter, attenuating low frequencies more than highs, then the clipping will occur more evenly across the frequency spectrum relative to how hard you play any given note.

 

ISo far I've made a humorous circuit with very limited knowledge which slightly works. I first experimented with the LM324 op amp-chip as a comparator, using a few resistors for a voltage divider to V+, and a guitar jack which resulted in a very nasty output similar to sporadic white noise. But I was able to trigger a difference by plucking the low E string which generated a tone in the mist of all the noise. So through curiosity I used an LM386 in an amplifier circuit (https://ask.audio/articles/how-to-build-your-first-amplifier) to amplify the guitar signal which would be sent to the LM324. I fed the output of the LM386 amplifier circuit directly to the V- and fed ~2 volts to the V+ instead.

Please post your circuit schematic. Maybe all it needs is a tune up.

ak