Idea for the GEOFEX "Juggler"

earthtonesaudio · April 22, 2009, 02:08:25 PM

If you took the CMOS version of the Juggler from GEOFEX, but made the "switch" signal into a 0-100% duty cycle PWM signal, you'd be able to blend smoothly (or choppily) from "first A then B" to "first B then A."

...At first, that doesn't sound exciting at all. Except- at the midpoint between those two extremes, you have 50% "first A then B" AND 50% "first B then A" simultaneously. I can't even imagine what that could sound like (imagine envelope effects!), and it could be any two effects (or chains of effects), so seriously, this is my next project.

Note that I didn't mention the rate of that PWM signal. More possibilities there...

trixdropd · April 22, 2009, 02:10:03 PM

So it would be possible to crossfade between your clean sound and your dirty sound?

earthtonesaudio · April 22, 2009, 02:29:45 PM

In theory, at least.

Taylor · April 22, 2009, 03:46:14 PM

Hmm, this is pretty intriguing. In particular, I'm imagining what would happen using a delay with another effect. Say a ring modulator into a delay. First, the delay would record the ring mod guitar sound, then it would output it back into the ring mod, which would get recorded into the delay, and so on. Probably a big mess in the case of those particular effects, actually, but potential.

When I first read your post, I was thinking you were planning to control the crossfade with an audio rate oscillator. Now there's a weird idea.

earthtonesaudio · April 22, 2009, 05:05:29 PM

Quote from: Taylor on April 22, 2009, 03:46:14 PM
Hmm, this is pretty intriguing. In particular, I'm imagining what would happen using a delay with another effect. Say a ring modulator into a delay. First, the delay would record the ring mod guitar sound, then it would output it back into the ring mod, which would get recorded into the delay, and so on. Probably a big mess in the case of those particular effects, actually, but potential.

When I first read your post, I was thinking you were planning to control the crossfade with an audio rate oscillator. Now there's a weird idea.

That's one of the things (crossfade) I was thinking of too. Honestly, if this works the possibilities are limitless. Envelope controlled crossfades, pseudorandom LFO-controlled FX order, who knows what else.

That's the beauty of the little widgets that don't actually do anything themselves. You gotta come up with uses for them.

earthtonesaudio · April 23, 2009, 08:23:08 AM

Here's an idea for the oscillator. The idea is flexibility.

If the Schmitt trigger is CMOS type, varying supply voltage from 3-15V should give 50:1 speed ratio. Duty cycle variable from 0-100%, switch allows selection of high (above audio) or low frequency oscillation.

Just one idea.

earthtonesaudio · April 23, 2009, 01:41:29 PM

Scratch that oscillator. I don't think it would work as drawn.

slacker · April 23, 2009, 01:48:57 PM

Just stick your rate pot wired as a variable resistor between the in and out of the schmitt inverter.
Or for something more elaborate use the clock from the MXR Envelope filter, feed it a CV and that gives you a PWM output. Feed it the output of an LFO or an envelope follower and you can sweep the pulse width.

earthtonesaudio · April 23, 2009, 02:09:56 PM

Yeah, voltage/envelope control is always nice to have.

earthtonesaudio · May 21, 2009, 01:18:11 PM

Here's a better/simpler idea for an oscillator to use. Op-amp is ideally a high speed, rail-rail type, but general purpose types will probably work fine.

Oh right, some suggested values... Resistors=100k, nonpolar cap=1n, polar cap=10u There should be a smallish limiting resistor (1k?) in series with the rate control too.

I haven't tried it yet, but I believe that you could add control voltage (over duty ratio) through another resistor going to *either* the (+) input or (-) input.
With this I think you could also add a DPDT toggle to turn the CMOS Juggler into an LFO'd A/B panner. ...Or leave one send/return open and it's a tremolo! Hah!

earthtonesaudio · July 09, 2009, 01:03:32 PM

I finally got around to attempting this last night. I'd say it was an almost complete failure, but mainly because I was in a rush and didn't lay things out very well. To make matters worse it's all on the breadboard so who knows what sort of gremlins are due to that...

Anyway, the next chance I get I'll clean up my layout and use a more proven oscillator (rather than the experimental one I just happened to already have built) to rule out those sources of error.

The part I'm still not sure about is what two effects should I use to test it? I need two circuits that are physically small (i.e. one transistor max or I run out of breadboard), sound different from one another, and especially sound different when their order is reversed.

Any ideas? I was thinking a Bazz Fuss and something else.

R.G. · July 09, 2009, 04:42:22 PM

Don Lancaster
"The CMOS Cookbook"

earthtonesaudio · July 10, 2009, 07:20:01 AM

I'm gonna have to request that one from my library.

earthtonesaudio · July 17, 2009, 07:51:46 AM

Some progress: on my breadboard now are two LPB-1 type circuits, one with a gain of 10, the other with more gain and diodes to ground on the output, limiting the output voltage to ~1.4Vp-p. Also on the 'board is the 4053-based GEOFEX juggler, and last night I was flipping the order of the two circuits with ease.

With these two circuits its easy to tell which is first and which is last. If the one with diodes is first, the output is semi-distorted and loud. With the diodes last, the output is more distorted and quieter. Simple.

I don't have a proper switch hooked up to the control pins on the 4053 (right now I'm just using a voltage divider between V+ and GND! eek!), so I can't really make any judgments about the switching behavior near the transitions. Hopefully this weekend I'll have time to test it with a real switch, and finally put it under PWM control.

earthtonesaudio · July 17, 2009, 11:01:12 PM

I just fired it up with PWM control, and it mostly works. "Mostly" because even though the PWM oscillates at over 100kHz, I can still hear a fizzy whine that varies a little when I mess with the controls or put my hands near the breadboard. I'm a little suspicious of how I've hooked up the 4053, because when I switched at low speeds, I also heard a little fizzy tick with each transition.
Anyway, I feel like this is a minor issue that could possibly be sorted out just with a well-laid-out circuit board. The bigger picture for me is hearing, for the first time, what it sounds like to pan smoothly between "A then B" and "B then A."

News:

Idea for the GEOFEX "Juggler"