Guitar to square wave?

[Jaicen_solo]

Does anybody know if it's possible to shape a (mono) guitar signal to a (more or less) perfect square wave?
I know it can be done with a diode clamp circuit, but are there any other ways that will yield a more pure square, rather than that heavy distortion sound you usually get? I've tried comparators, but they have two main problems: Tracking and spluttering as the notes decay.
Also, are there any methods that fit this bill that will do pulse-width modulation??

[Mark Hammer]

Comparator IS the way to go.  The trick is to somehow assure that either:
a) the input signal level is consistent enough over time that a comparator action can occur (try a compressor up front), or
b) the comparator reference/threshold is set low enough that a square output will be generated for as much of the lifespan of the note as possible without generating false positives (some lowpass filtering before the comparator will also help), or
c) you have a means to cancel/gate or otherwise attenuate the output or input signal before you reach the critical point where sputtering begins, or
d) you do some combination of these.

Note that whether one is seeking a square wave of the same pitch as the input OR some divided-down version, the problem is identical.

Although you can do PWM, a simple comparator will, by itself introduce inconsistent pulse-width if you set the threshold just right.

[Dragonfly]

theres info in these threads :

http://www.diystompboxes.com/smfforum/index.php?topic=29343.0

http://www.diystompboxes.com/smfforum/index.php?topic=51246.0

http://www.diystompboxes.com/smfforum/index.php?topic=39254.0

[Processaurus]

Look at the fundamental extractor in the boss OC-2, it does it in such a way that there is what seems to be an even duty cycle square wave coming out of the first flip flop, the same frequency as the guitar.  it glitches and cuts out when the note decays too far though.

An easy way to that squarewave sound is heavy fuzz with the neck pickup and the tone rolled down (or better, a graphic eq with all the highs above 700hz rolled off), maybe some compression before the fuzz.

[widdly]

I built this one and it does square and saw waves and stuff in between...

http://folkurban.com/Site/SimpleSquareWaveShaperMadeSimpler-712.html

Tim also has this one that does a PWM square wave...

http://folkurban.com/Site/PWM-715.html

They both crap out as the note decays though.  Like Mark says, some kind of gate after or compression before would make these more workable.

[Paul Marossy]

The Crash Snyc will get you some square wave guitar, but it also gives you distortion...

[Jaicen_solo]

Thanks for the input guys. Unfortunately, it seems that none of those solutions will give me the glitch free decay i'm looking for without increasing the complexity a great deal.
I want to use this for hex fuzz, driven from a GK pickup, so whatever I do it needs replicating six times. Six Rocktaves seems to me to be overkill, but it would be cool I think what i'll do is use some heavy filtering to get more or less a fundamental, then a bog standard diode clamp to square things up. It won't give PWM, but it is simple!

[chemosis]

yea ive been trying to get a more real sounding squarewave my self and really like the topic. the craig anderton ultra fuzz is supposed to do a real good synth sounding squarewave but there are no PCBs available anywhere. comparator type circuit does seem like the way to go. ko amps square wave generator took the fuzz section from a schumann ppl and it gets pretty close to an authentic squarewave but what pedal could I had to take out some of the even harmonics??? im wondering if the boss oc-2 which can almost do a sine wave into a squarewave generator would be solution. I also was thinking of using a filter to take out some even harmonics and then go into a squarewave pedal. Im still going to continue the quest because I know its possible to get close to a authentic synth squarewave we just need the right pedal combination

[vigilante397]

yea ive been trying to get a more real sounding squarewave my self and really like the topic. the craig anderton ultra fuzz is supposed to do a real good synth sounding squarewave but there are no PCBs available anywhere. comparator type circuit does seem like the way to go. ko amps square wave generator took the fuzz section from a schumann ppl and it gets pretty close to an authentic squarewave but what pedal could I had to take out some of the even harmonics??? im wondering if the boss oc-2 which can almost do a sine wave into a squarewave generator would be solution. I also was thinking of using a filter to take out some even harmonics and then go into a squarewave pedal. Im still going to continue the quest because I know its possible to get close to a authentic synth squarewave we just need the right pedal combination

If you're looking for answers, posting in an 8-year old thread might not help

[chemosis]

the most authentic squarewave pedal ive heard was the K.O. amps squarewave generator but is now discountined,  so the only guy that is still making the fuzz/squarewave section off the schumann ppl is industrialectlic.I think its the same pedal btu made by industrialectric now.  I wish someone would sell pcbs for this circuit. ive never got to hear the craig anderton ultra fuzz which is a comparator squarewave circuit. theres no videos or sound samples

[dschwartz]

Schmidt triggers are very good for making square waves...

[Jaicen_solo]

the most authentic squarewave pedal ive heard was the K.O. amps squarewave generator but is now discountined,  so the only guy that is still making the fuzz/squarewave section off the schumann ppl is industrialectlic.I think its the same pedal btu made by industrialectric now.  I wish someone would sell pcbs for this circuit. ive never got to hear the craig anderton ultra fuzz which is a comparator squarewave circuit. theres no videos or sound samples

It sounds like a comparator sq wave. Tim escebedo did one that had pwm, and I think John Hollis did a square wave with a 555. The issue with the comparator is the complete lack of dynamics and abrupt cut off as notes decay. I prefer the early Roland GR approach that uses diode clamps on each individual string. This eliminates intermodulation, which is key to getting a synthy square wave sound on a polyphonic instrument.

[anotherjim]

There isn't much mention of the Schmitt trigger in this old thread. It isn't immune to decay jitter, but can cut off more surely than a simple comparator.
The Schmitt trigger can be set up to behave very much like a comparator if the upper and lower threshold levels are made close.

The guitar signal is horribly complex and uncertain. Its harmonics can be stronger than the fundamental, which makes it impossible -  not just difficult - for a circuit to determine in real-time which peaks and zero crossings belong to the fundamental with 100% reliability with just any guitar, playing technique or set-up.

The circuits employing adaptive comparators and triggers can still be fooled. In the end, whatever  method is used, you end up using the neck p/u + tone=0 ploy to get as simple a signal as possible into the effect for reliable use.

The lack of dynamics should be solvable by employing an envelope follower (off a clean input signal). Now, if you have a circuit that makes a squarewave, you can use a simple single transistor "Swing" VCA. Commonly found in early electronic drum synths, you don't see it often in guitar pedals. It's a lot easier than an OTA or LDR VCA.

[LightSoundGeometry]

Hello, I have been messing with the Hex inverter and can breadboard the preamp and schmitt trigger to an output ..thats what I got so far.

I was reading you can series stack gain stages , how does this look on the IC? do I actually put two whole preamp circuits in series sing two gates, then to the trigger then my square wave output?

any reason my amplified signal is really bassy ? decoupling caps change frequency? the preamp has C1,C2 and C3, 1 and 2 are decoupling caps and C3 is supposed to be the pass filter but its not changing on breadboard to give me more mid/treble boost..a lot of DB boost though so two gain stages probaly not needed but still want to know how to stack gain stages on the Hex inverter to get a square wave guitar.



the 47K on the low threshold is too much gating, sounds like my first tonebender farting. I got a TB to work on BB finally ..if only she moves to vero and sounds the same I will be happy