Fuzztainer simple distortion

[WaveshapeIllusions]

I swear, it took me longer to figure out the name than it did to put it together!



Here we have a basic distortion stage with feedback loop diode clippers. The first stage gain is varied by a pot. The second stage has an LDR in the feedback loop rather than a pot or a standard resistor. The clipping diodes are LEDs.

Right now it's just a basic design, there is some refining to do. I left the tonestack (and the makeup gain/output) up in the air for now; use whatever you'd prefer. It actually doesn't sound too bad without one though.

Oh, right, the sound. You'd probably want to hear about that. It's very saturated. At minimum gain it's a heavy fuzz. At max it's huge, square wave-like. The LED/LDR combo gives it some pretty long sustain. Depending on how hard you play, they end up fighting each other and you get a bit of a tremolo effect too. It alternates between saturation levels.

The one pot for the gain does get to zero. You can add a resistor for minimum gain in, but with the pot at zero it's basically a unity gain buffer. I suggest socketing the LEDs to find the best set for you. I used some bright red ones.

I hope you find this project interesting.

[psychedelicfish]

I hope you find this project interesting.

I certainly do!

[psychedelicfish]

Redrawn the schematic:
[img]https://docs.google.com/open?id=0BwrTwddNc0sjZncxODh5a2JhbE0[img]
Call me OCD, but your schematic was too messy for me. Forgot to add in the diodes in the  power supply, CBF doing that now

[psychedelicfish]

will try again with the image

[John Lyons]

Name's a bit close to this though. 

[psychedelicfish]

NVM that, looks like this website doesn't like google drive then. Anyway, one mod I thought of:
A 1M pot (may need a different value depending on LDR) in parallel with the LDR, when set to full resistance this would turn gain to max and increase the effectiveness of the LDR, and when set to minimum resistance this would turn into an ordinary distortion pedal. Also, while it is nice  and slightly simpler to do the voltage divider thing to get a "dual supply", for stompboxes I wouldn't recommend this, because some people like to power multiple pedals with the same adapter, and the "dual supply" would cause problems.

[Quackzed]

  thats messed up! the ldr resistance is high when the leds are dark (no signal), lots of gain! as the signal gets big enough to turn on the leds it clips AND turns down the gain of the stage! so then the leds get dimmer and as a result the gain goes up, which lights the leds and turns down the gain, and on and on... thats effed up! so the leds do light up? thats pretty crazy, i've never seen that before, now why couldnt i have thought of that!?!
wait, the leds are in parallel with the ldr, so when the leds are lit the resistance goes down and allows MORE current through the ldr so they stay lit longer but the gain(volume) goes down and as it does the leds clip harder till the signal is too low and then as the leds are lit less the resistance of the ldr goes higher and makes it ... ok i'm getting a headache... WTF?!?
you should call it the ... the ... what DOES that? a tug of war isn't quite whats happening, thats too simple... its more like ... ok i give up. 
thats a brain teaser alright.... it makes me think of some of those crazy perpetual motion machines... thats as close as i can get, ideologically. 

what is really happening there? anyone?

bump

[Mark Hammer]

Bill Berardi had a more elaborate design called the Fuzztain in EM in the mid-80's, that also used optoisolators in double duty for compression and clipping.  Google around and you might find it.  There were some interesting bits in it that you might want to implement.

[pinkjimiphoton]

speaking ofc which, i was just looking at that mark, so....

[Quackzed]

this whole ldr in the fb loop of an opamp opens up a lot of possibilities, never really thought much about it.
also i never built an opto compressor, so maybee thats why this idea seems so cool to me.. 
but i still want to know what is going on with the ldr in parallel with those leds... 
in the fuzzstain schem, i think the leds are clipping/and being lit up in second opamp and this is affecting the gain of the first opamp (with the ldr)
so when they clip/light up it reduces the gain of the first stage, i'd guess its set up so the leds stay close-ish to their conduction voltage and never get slammed and square off the signal, before they can/ as soon as they start to clip the gain of the first opamp stage gets turned down via the ldr...
so it seems like this setup would prevent the leds from 'super hard clipping' as well as keeping the signal just hot enough to clip.
must add to the sustain as well as keep it from becoming squared off and too buzzy... probably a good topology for keeping within the conduction curve
of a l.e.d. clipper... sort of like automatically hitting the brakes whenever the clipping gets too hot in the leds...
but the first circuit is still eluding me. 

[WaveshapeIllusions]

Psyc, I'll be updating the schematic with your suggestions in a bit. I agree with the single supply being preferred for stompboxes. I like the parallel max gain setting pot. I have a idea to set the range (min/max) that I'll test out when I get home. I also switched the second stageto inverting so we can get attenuation as well.

Quackzed, it is an odd little function. Right now the effect isn't as extreme as I'd like, probably due to the LDR (physical) position. The LEDs do light up, though not as much as I'd hope. Diodes to ground may work a bit better, it may need a higher supply voltage or brighter LEDs. For extra craziness, try the LDR as the feedback divider resistor. The gain will snowball wildly.

Thanks for the suggestion Mark, and thanks for the schematic Jimi. That's quite a bit more fleshed out than what I've got.

I'm still trying to get the physical isolator set up right. The univibe method with a little reflection chamber may help. Perhaps more parallel LEDs?

I think LEDs to a Vref may work better. Bias it so the difference between both ends is close to their forward voltage so they light up more?

[pinkjimiphoton]

you're quite welcome, i look forward to seeing what ya come up with!!

[WaveshapeIllusions]

Here's an updated schematic. Inverting second stage, with a +9V supply. Still no tonestack. Raised the feedback divider resistors to make the gain less crazy.



I went with a 500k pot. Decided against the min/max range pot. What I had was the one on the left here:


Kind of like RG's tapering resistor, but with the LDR. The idea was you could vary how much of the pot was in series/parallel, thus setting the min/max range. Then I realized that at one end it'd short the LDR ends together. So I went with the one on the right.

I couldn't figure out a good value for it. If it was too low, the LDR wouldn't do much until it got near the same value. Then I figured minimum range was less important anyways, I wanted it to be able to attenuate a bit. So I stuck with the 500k in parallel for max.

[WaveshapeIllusions]

After looking at that Fuzzstain for a bit, I gotta say that's an interesting circuit. The set up with the diodes is interesting, aside from the use of the optocouplers.

As far as I can tell, the resistors in series with the diodes are also in the feedback loop. Which is an interesting take on feedback loop clippers, since most are straight diodes that give 100% NFB above the diode drop. That gives ne an interesting idea... What if you used a larger resistor in series with the diodes? Anything above the drop would get more gain?

The switch set up is interesting, as is the clean blend. I think his way of modifying the gain of a previous stage may be a bit more effective. What's weird, though, is that the LDRs are in the loop of the opamp that drives the clean blend too. So not only does it affect the clipping, it affects clean volume too, even blended full clean. If the goal is to approximate the compression and clipping of a tube amp, I'd think it does it pretty well.

Going to need to come up with a new name for my circuit though, if it gets off the ground.

[Quackzed]

a name...  ok here are some suggestions based on my idea of whats happening ...

my take is, and i may be wrong here, that every time the signal is big enough to light the leds and clip, it causes the ldr resistance to drop and allow current around the clippers and stops them clipping, like dangling a carrot in front of a donkey, it never gets close enough to heavily clip...

the carrot overdrive.
the teaserdrive.
the fly (ever try to catch a fly?)

i had a thought, if you put a resistor in series with the ldr as drawn, then there would be a larger resistance(and more gain) when the leds are lit so it would distort more, i think. and the leds would light more... say a resistor of (whatever the dark resistance of the ldr is) would double the dist.
thats assuming i'm getting whats happening there... but im pretty sure it would work that way... it'd also be in parallel with that 500k pot, so it'd be less active at low gain pot settings and more active at high gain pot settings... due to having less/more effect on total resistance in the feedback loop respectively...  right?  i think thats right.

[Gus]

another circuit schematic with photo control of gain is the Bixonic Expandora

[deadastronaut]

OT,  bu this is weird..it happens now and then, just with Gil   

[Quackzed]

 
i remeber you mentioning that before, but i wasn't exactly sure what you meant, but there it is... weird!
i suspect its some sort of higher dimentional crosslinking of our sub ether-net interconciousnesses due to the Gathering.  

the sensation you are feeling is the quickening.
Who am I?
We are the same, Deadastronaut! We are brothers!

... awkward silence due to inexplicable personalized highlander reference noted....

[pinkjimiphoton]

all these years later, i was looking for the fuzztain, found this, remembered i wanted to try it and never did...

so i whooped up a vero layout... better late than never i guess. haven't built it yet, but seems to be good to the last schematic.

added polarity protection and a snubber cap to the power supply, and a pulldown for the input. should probably move c5 over 11 holes
ooops. lazy.

fwiw

[pinkjimiphoton]

sorry, don't use the one above, i think there's some mistakes.
this one should be good, and is laid out a little better.
i don't like doing all the connections at the end of the board all the time. i DO like having an extra pad or two in case a wire breaks.
i added a small cap for rfi noise, and a polarity protection diode, a 1m pulldown resistor and an optional 100k volume pot.