Square Wave Tone Pedal Question

[amptramp]

If you are using this circuit across the full range of bass and guitar, you should beware of the "wolf note" you get when the second harmonic exceeds the level of the fundamental and you suddenly go up an octave.  Tracking a frequency has always presented that problem and some people use lowpass filtering to  reduce the probability of it happening, but usually this just changes the frequency at which it happens.

Go ahead and experiment with it but be aware that you may get this problem.

[FiveseveN]

This crude drawing should hopefully illustrate the effect I'm after!

OK but effects are AC-coupled. Meaning "-1" and "+1" can just as well be 2V and 7V so you don't have to deal with the trinary business. I'd say start with Anderton's fuzz, it doesn't get any simpler than that, and see if that's what you're looking for.

[Mark Hammer]

The 220nf/100k pair on the input of the Ultra-Fuzz have a bass rolloff starting around 7hz, which is fine for bass.  However, the calculated bass rolloff for the 100nf/10k pair linking the first and second op-amp stages is 159hz, which may be a bit too high for a bass guitar.  Consider increasing that 100nf cap to 220nf.  Indeed, simply swap the locations of the 100nf and 220nf caps, so you don't have to buy anything.  A bass rolloff on the input of 16hz will not impair anything.

[PRR]

Find him a Schmidt with adjustable hysteresis, zero to large. I'm curious too.

[sinthmart]

What you drew is called a meander. Apparently this is the most accurate name for such a waveform.

(I am withdrawing my questions about the Schmidt trigger. I tried a near symbiosis of the proposed scheme (for the theremin) and the result was disappointing. The sound is scratchy, dry, not musical.)

[ElectricDruid]

To those asking specifically what the aim is -

This crude drawing should hopefully illustrate the effect I'm after!

Yep, that looks like a comparator output!

Thanks for putting me onto the Schmitt Trigger, however I have a couple of questions:

I would like to build this pedal so I can utilise all frequencies from both my bass guitar (possibly in drop tunings) all the way to the top end of my six string electric guitar. Would a Schmitt trigger still make this possible? How would you recommend the best way of going about this is?

The schmitt trigger doesn't really care about *frequency*, but rather about *level*. It's working in the "time domain", not the "frequency domain", to use the jargon. What that means is you can draw a waveform graph like you've done and you have all the information you need (a waveform graph like that is "time domain").
Of course, any practical device has *some* frequency limits because of things like the slew rate, the time taken to get from the lowest level back up to the highest level, but the highest note on a guitar is not going to be so high that that should be a limiting factor.

Would using two solid state relays possibly be a better option?

No. Relay switching will be power-hungry, noisy, and too slow. Stay solid-state!

HTH

[ElectricDruid]

Here's my example from playing with it in LTspice:




The "Threshold" control sets the hysteresis. You can see from the waveform plot that the output doesn't go high until the input is a long way negative, and vice versa. I had the threshold very high for this plot. This provides a large amount of noise immunity, but will gate off the end of your note as it dies away.

The basic plan in use would be to turn the guitar up loud and then turn the threshold knob until the spurious triggering stops. And then rock the fuzz out!

There's probably no need at all for the threshold to have such a ridiculously wide range, so putting another fixed resistor above the 100K pot would be a good idea - say 330K or something. Oh, and you'd probably want 100n for the input cap for Bass. I used 10n because it minimises the time taken for the HPF to settle around the 4.5V Vref, which makes the sim run quicker.

HTH

[soggybag]

Ugly Face is a great example of a square wave from a 555, works great.

http://www.super-freq.com/pcb-projects/ugly-face/

Mockman by RunOffGroove is another. This uses two op-amps to boost the input by x 10,000 which makes a pretty strong square wave out of anything at the input.

http://runoffgroove.com/mm2.html

[ElectricDruid]

Mockman by RunOffGroove is another. This uses two op-amps to boost the input by x 10,000 which makes a pretty strong square wave out of anything at the input.

The trouble with this approach is that noise at the input gets the same amount of gain. x10,000 is enough to turn 0.1mV of noise into 1V, so I'm slightly surprised that noise isn't mentioned anywhere. It can hardly be a quiet pedal! You could add "coring diodes" like Mark suggested earlier in the thread, I suppose.

[R.G.]

I simm-ed up a noise gate by mocking/mucking up a comparator with mild hysteresis from one half of a TL072. When I got it to making good square waves and squelching with hysteresis down in the millivolts, I rectified/filtered the output to create a gate drive signal for a J176 to mute a signal.

I was about to use the second half of the 072 when I thought - hey, wait a minute. The second half will just  be  making the same square waves as the comparator for the noise gate. I tapped off the comparator output from the gate and used it as an output signal muted by the noise gate. Seemed to work fine.