pulse conditioning for 555 pulse fuzz (fixed/it was better when it wasn't fixed)

[iainpunk]

im currently working on a  protototype of a 555 based glitch fuzz, which is basically the 2nd half of a Atari Punk Console, driven by a schmitt trigger.



the problem im running into is that as long as the schmitt trigger is High the 555 timers output is also high, this results in that the 555 only dictates the pulse length if the pulse which comes in is shorter than the set pulse length.

Basically it is only determining the minimum frequency, while this is a useful and fun tone to use, its not what i originally intended the circuit to do.

how do i limit the duration of the pulse that comes out of the schmitt trigger?

thanks in advance, Iain

[Rob Strand]

I've played with a lot of 555 trigger circuits.  The key thing to note is while
the trigger pin is low the output stays high.   What you need is a pulse shaper
so the 555 only see narrow pulses.   You can play with the pulse width in your
case but you can get very reliable results with pulses in the order of 1us to 10us.
Pushing it you can get down to maybe 300ns.

As a generic trigger circuit this one is probably the most reliable.
See figure 4.5.5,
https://learnabout-electronics.org/Oscillators/osc45.php

You will see similar variations like this one, which aren't quite as good,

http://www.circuitstoday.com/scmitt-trigger-circuit-using-ne-555

They are OK for < 5V.  They don't reset the trigger as quickly.

how do i limit the duration of the pulse that comes out of the schmitt trigger?

You would need to reset the first 555 early with another monostable.  Still might not do what you want since it could trigger immediately after.   I think the real issue is the triggering since one mono should be adjustable to the maximum pulse width.

As for better triggering with the guitar that comes down to filtering and pulse shaping *before* the trigger circuit.

[iainpunk]

turns out that having the pulse coming out of the schmitt trigger shortened to make the monostable dictate the overall pulse width instead of the minimum pulse width sounds worse than i expected...

quite unusable, both rhythm and lead sound really really bad.

when striking a chord, it automatically sounds like thin noise, and the playable range for lead is to small, about one and a half octave sound good lower sound really tin, quiet and has the tendency to randomly generate octaves and other overtones, and the octave above playable range is an octave down, which makes it the same as the octave before it, but way glitchier, resulting in too little of a playable range.

so, ill keep it the way it is, just setting a minimum pulse length, which sounds quite cool, broken and computerey when playing power cords or major chords, also lead playing works more than one and a half octave.

[Rob Strand]

quite unusable, both rhythm and lead sound really really bad.

It was always going to be an angry sound.  I guess it comes down to what you expected.

Imagine if you could extract the fundamental frequency, which isn't that easy anyway.  Then the output of the pedal would at best be a single note with a synth tone (an angry synth tone as there is no filtering or envelope shaping to follow the envelope of the input signal).

So now think of a complex signal like a chord, many fundamentals at once.
That's an even more difficult problem.  It's not going to do well at all.

[iainpunk]

yes, i found out the hard way.

i wasn't really expecting anything to be honest, but the problem i thought i ran in to wasn't a problem at all, but rather a better version of my original plan.

i really like the tone when the first octave down comes in at the 12th fret of the D string. Playing barre chords on humbuckers gives it a bit of computer glitchiness combined with the Jesus and Mary chain noisiness, with out of phase pickups, it goes completely haywire, while still remaining some resemblance of the chord progression you intended to play, very nice this way

i also found out that most op amps, when they have single sided power, they don't like it when you want to use them as a comparator to ground

[Rob Strand]

i also found out that most op amps, when they have single sided power, they don't like it when you want to use them as a comparator to ground

If you add a bit of hysteresis to the comparator, effectively making it a Schmitt trigger, it will behave a lot better.  The MXR bluw box does this.  See IC 1B.  The larger you make R6 the smaller the hysteresis window,
http://www.tonepad.com/project.asp?id=33

You can also play with adding high and low-pass filters before the comparator.    That lets you roughly focus which notes cause the triggering.   The Blue box has a bit of that going on as well.

[iainpunk]

i also found out that most op amps, when they have single sided power, they don't like it when you want to use them as a comparator to ground

If you add a bit of hysteresis to the comparator, effectively making it a Schmitt trigger, it will behave a lot better.  The MXR bluw box does this.  See IC 1B.  The larger you make R6 the smaller the hysteresis window,
http://www.tonepad.com/project.asp?id=33

You can also play with adding high and low-pass filters before the comparator.    That lets you roughly focus which notes cause the triggering.   The Blue box has a bit of that going on as well.

hye, yeah, i have experimented with different amounts of hysteresis and filtering, but i liked going in raw, unfiltered comparing with the bias sounded the best, so i realised that if i just use a comparator opamp for the input, i could just as well ditch the op amp and use the trigger input, which is a comparator as well, instead of the first opamp, and use a bias trim to bias the input at 3 volts and turn this into a single chip and really low component count little glitchy fuzz monster

[Rob Strand]

hye, yeah, i have experimented with different amounts of hysteresis and filtering, but i liked going in raw, unfiltered comparing with the bias sounded the best, s

There's no rules for "Angry" pedals like this.    The best sound comes from playing about for sure.  The technical stuff just guides your options.