Simple Square Wave Shaper Made Simpler

[liquids]

So I've bread boarded this circuit up (http://folkurban.com/Site/SimpleSquareWaveShaperMadeSimpler-712.html) and was in love with the sound right away. I've had it around on and off for the last few weeks, messed with it a bit, etc as it seemed like it could be tweaked, in theory...

But while I'm trying to make sure it's not just working but functioning correctly, I'm simultaneously  trying to understand it stage by stage, and then with some understanding, tweak it.

I've searched and haven't found much of the "how it works" for the circuit outside of the stuff Tim offers on the page.  Changing values here and there unfortunately hasn't shed too much light on what does what like with other builds...and I guess I'm a little impatient, of course since I like the sound so much and want to narrow down what tweaks, or mods, I'm going to settle on so I can get the box sized, laid out, and drilled.

The Gating: audio probing straight out of the collector of the 2N5088 (without the 500k resistor)--contrary to my assumptions and from what Tim says, I got no 'gating' no matter how I set the trim on the emitter, which was surprising. I'd assumed there was some intentional transistor mis-biasing that causes the gating.  Likewise I tried resistors lower and higher than the 10k (6.8 and 15k) in its place, for example and there's a tonal difference, but again, no gating at this point in the circuit.  Not totally clear what the trimmer does, though I can hear subtle differences in how it effects the clipping, which is fuzzy but not bombastic.  After you hold a note, it fades back out of fuzz, which seems to correspond to the time the effect would gate when the rest of the circuit was present, maybe.

Along those lines, I can't quite understand what the 500K, 100K and .1uf from base to collector 'do.'  I can swap parts in and out, and hear drastic differences, but it's just blind guesswork without understanding.

Putting the circuit back to schematic up to the output of the first IC stage, again using an audio probe at this point in the circuit.  Here the gating is present.  Suddenly the trimmer einfluences the gating.   My new hypothesis was that the gating was due to the 500k resistor between transistor's collector and inverted input of U1, but I was having trouble even pinning that down clearly.   I also don't completely understand what the 100K from non-inverted to inverted input does, nor why the positive input is connected to Vr for that matter.  Also, at this point I started wondering if a 'preamp' with more fuzz into the IC would help with sustain as per the above, where the note would be fuzzier for longer, possibly...

So basically, it seems I understand very little of even this half of the 'simple' circuit!  Can anyone shed some light on any of this?  Is Tim here?   

[johnnysomersett]

I was looking at this the other day...

what is the purpose of the 'Vr' section on the bottom right hand side?

[armstrom]

That's a pretty standard circuit segment. It's used to create a "bias" or "reference" (hence  Vr) voltage for the op amps. Op amps are designed to run with a bi-polar power supply but they can be run from a single supply (+9V in this case) if you reference the non-inverting input to half the supply voltage. The two 100K resistors are a simple voltage divider that causes Vr to be half of the supply voltage. All you do then is tie together every location on the schematic where you see Vr. In this case, the non-inverting input of U1.

[johnnysomersett]

I seeeeeeee, cheers for that. i thought it might be that but wanted to be sure. I wanna make this one once ive succesfully done some simpler ones.

The to 500K pots after U1....would they be user adjustable ones? Also, do U1 & U2 come in many shapes/forms? wouldnt wanna order completely the wrong things LOL

[Mark Hammer]

I seeeeeeee, cheers for that. i thought it might be that but wanted to be sure. I wanna make this one once ive succesfully done some simpler ones.

The to 500K pots after U1....would they be user adjustable ones? Also, do U1 & U2 come in many shapes/forms? wouldnt wanna order completely the wrong things LOL

U1, as noted, produces a square wave output.  That square wave goers above and below the circuit's "zero point", which is Vr.  It has two ways to get to the outside world.  One is by passing through one diode (which only one half of the square wave can do), and the other is by passing through the other diode (which only the other half of the square wave can do).  The two 500k pots and the .02uf (realistically, .022uf) cap form a lowpass filter, such that when either of the pots are set to max resistance of 500k, that half of the square wave takes about 16msec to fully charge up the .02uf cap.  This means that instead of going from nothing to full peak instantly, there will be a "ramp-up time" of as much as 16msec.  At this point, I can't think of a reason why log would be preferable to lin for the pots, but your tastes might dictate otherwise.  Bi-FET op-amps, like a TL072, are probably a good idea.

That time will not be audible as delay, per se, but will be heard as a "triangulation" of what started out as a square wave.  Of course, to produce a true triangle out of a square, you would need to set both pots on max resistance.  Setting one for max and the other for min, lets one side ramp up, but lets the other side switch instantly.  Also, note that the controls only impose a defined ramp-up time.  They do not impose the same amount of "triangulation" across all frequencies.  For any pot setting, there will be more "triangulation" imposed the higher the note is.

Since I imagine the output level is fairly hot, it might be fruitful to replace the 100k output volume pot with a 47k resistor going into a 50k log pot.  The junction between the fixed 47k resistor and the pot then becomes available for adding a treble-cutting cap or two, going to ground.

[liquids]

Can anyone try to explain the transistor stage by itself, particularly the function of the resistors and cap from base to collector?    I do notice similar setup from base to collector in the Jinx (which uses a darlington, though)....

Or, can anyone explain what the 100k connected between U1's inverting to non inverting inputs purpose and functionality is? 

Anyone?

[rnfr]

i've also wanted to find out what was going on behind this circuit ever since i heard the samples. i've been wondering as well if the diode configuration could be used with other discrete circuits to take advantage of the slew rate limiting in order to do some tonal shaping. 
---
if we are going to look into this circuit we should also look into the designer-
here is a great interview with bobby beausoleil in 1999.  his story is really an interesting one, and he gets into most of the circuit design stuff towards the end of the interview.
http://www.beausoleil.net/mminterview.html

[Boogdish]

i've also wanted to find out what was going on behind this circuit ever since i heard the samples. i've been wondering as well if the diode configuration could be used with other discrete circuits

I added the backend of the simple square wave shaper onto the output of my slacktave and I really liked it.  That pedal had an insane output so the mod was passive, the mod made it my favorite pedal.

http://www.diystompboxes.com/smfforum/index.php?topic=55644.msg567966#msg567966

[rnfr]

i've also wanted to find out what was going on behind this circuit ever since i heard the samples. i've been wondering as well if the diode configuration could be used with other discrete circuits

I added the backend of the simple square wave shaper onto the output of my slacktave and I really liked it.  That pedal had an insane output so the mod was passive, the mod made it my favorite pedal.

http://www.diystompboxes.com/smfforum/index.php?topic=55644.msg567966#msg567966

cool! i'll have to try it!

[Mark Hammer]

Basically, anything that puts out a square-ish wave is fair game for that circuit.  You will find something similar in a bunch of  LFO-based circuits (including the EHX Pulsar) that use the standard 2-opamp LFO circuit where the one half puts out a square wave and the second op-amp slews it produce a triangle output.  Often found in synth circuits.  This particular implementation by Beausoleil (and check out HIS rig!!....not to mention his Forrest Gump-like life that found him in the company of a surprising number of 60's and 70's cultural icons) is aimed at shaping square waves in the audio range, rather than slower control waves in the sub-audio range.

[Solidhex]

Yo

  Are there any audio samples floating around of this circuit?

--Brad

[ianmgull]

About halfway down:

http://www.geocities.com/tpe123/folkurban/fuzz/snippets.html

[rnfr]

this circuit makes a nice square output.  you could put the filter after it followed buy a simple boosting stage and probably get good results with a smaller parts count.  matter of fact you could do the above circuit and the output stage with a dual op amp.