Just found this - Elektor 1988 Guitar Fuzz Unit

[lion]

Going through some Elektor archives, looking for something else, I came across this circuit. I don't think I've seen it before. The description says it features controlable sym/asym distortion with individual controls for the posite and negative portions of the input sine waves.

I get the idea, but it's not immediately clear to me how it functions, maybe it's just drawn in a odd way

Anyone seen it before - or even build it?

Erik

[duck_arse]

using pots as the bias dividers - never seen that before.

[Mark Hammer]

That's the "Variable Fuzz Box", if I remember correctly, whose functioning and promises have intrigued me for a long time.  The claim was that it could be used to continuously adjust the symmetry/asymmetry of clipping.  Or am I confusing it with an earlier circuit?

[teemuk]

I see NPN and PNP transistors basically configured as shunt diode clippers with base voltage adjusting DC reference for the emitter-base diodes.
Each reference voltage is adjustable from rail voltage to half rail voltage. Yep, clever move to Integrate the control to bias circuit.

[mth5044]

Subscribing for interest! Seems like a low parts count for high functionality! Input cap looks pretty low. - Gracias, Padre

[PRR]

> Input cap looks pretty low.

17Hz

[Aph]

Looks like this is the first version? :
https://www.diystompboxes.com/pedals/Elektor_variable_fuzz.PDF

[Ice-9]

I would definitely change the power connections though, having the bypass switch connect and disconnect the battery is a bad idea and will likely cause a loud pop when switched.

[Rob Strand]

I know I've seen that schem before but that's about it.

Edit: Teemuk already said it.

The biasing of the transistors is opposite to what you would normally do for linear operation;in that T1's bias voltage is low and T2's bias voltage is high.   The transistors are off in the middle region.  The transistors are actually functioning as programmable clippers.  When the output at the R4 (100k) swings low T1 turns on and stops the voltage drop decreasing further.  Similarly for T2 on the positive swing.   R4 is the series resistor that feeds the clipper.  The resistor R6 (1k) might add asymmetry but it could be there as paranoid protection against T1 and T2 both being on.

[Mark Hammer]

Looks like this is the first version? :
https://www.diystompboxes.com/pedals/Elektor_variable_fuzz.PDF

That's the one I thought it was, but the two circuits have a number of differences, not the least of which is the absence of diodes in the first one.

[Aph]

Ok, found the page. No year of publication. Does have a very short description:

http://www.luciferstrip.com/fuzz/elektor-guitarfuzzunit.jpg

[garyg]

Going through some Elektor archives, looking for something else, I came across this circuit. I don't think I've seen it before.  ...  Anyone seen it before - or even build it?

Oh wow, I built one of those back in the day (really '88? wow...) recall it sounding kind of crappy to be honest and probably one of the reasons I didn't bother building any pedals for a couple of decades...

Maybe my build was dodgy but I seem to recall it being really spluttery, kind of all or nothing. Interesting circuit though looking back.

[Mark Hammer]

Thanks for the link to the full write-up.  DO NOT follow their suggestion to use a 3PDT to kill the power in bypass mode.

[Keppy]

I seem to recall it being really spluttery, kind of all or nothing.

That's not surprising, looking at the circuit. There's no tone shaping, no volume control, and a linear gain pot that crams most of the audible differences to the left side of the gain control. Half of the asymmetry control will probably seem redundant as well, since it can create asymmetry toward either side of the waveform and these are unlikely to sound very different.

The idea is interesting, but this thing is CRYING out to be modded.

That said, I love stuff like this, and I can't wait for someone to frankenstein this thing and post their schem.

[MaxPower]

I believe I've tried this and a similar one.  Doesn't live up to expectations if I remember correctly.

[lion]

I seem to recall it being really spluttery, kind of all or nothing.

That's not surprising, looking at the circuit. There's no tone shaping, no volume control, and a linear gain pot that crams most of the audible differences to the left side of the gain control. Half of the asymmetry control will probably seem redundant as well, since it can create asymmetry toward either side of the waveform and these are unlikely to sound very different.

The idea is interesting, but this thing is CRYING out to be modded.

That said, I love stuff like this, and I can't wait for someone to frankenstein this thing and post their schem.

I'm thinking:
Reduce C1 to taste to control any bass content issues.
Add a tone control at the end - BMP, Crunchbox style, etc - and a vol control.
Small (pF) caps in the opamps FB path, and/or simple LP and/or HP filtering at strategic points.
If usable noise comes out of it the pot control issues/tapers could be optimized I'd guess.

Anyone considering to have a go?

Erik

[garyg]

That's not surprising, looking at the circuit. ... That said, I love stuff like this, and I can't wait for someone to frankenstein this thing and post their schem.

I'm hardly an expert in this stuff now but back then I would have been pretty clueless. Sounded interesting so I built it, no idea how it should work.

But it is an interesting idea, I do like seeing this kind of take on things and, with duest respect to everyone, not another TS/op-amp with clipping diodes.

[teemuk]

In my opinion, most of those Elektor articles / circuits exist mostly to demonstrate novel circuit ideas or applications. Take this one for example; I'm not remarkably eager to build just another fuzz, but I do admit that the circuit packs a handful of useful ideas for which I will likely find more interesting applications later.

So, some neat concepts there. Besides variable clipping references and that neat bias circuit the circuit architecture is pretty much that of dozens of other "distortion" effect designs, like MXR Distortion+, for example. Basically, a high-gain non-inverting opamp gain stage (that implements a gain control) has its output signal clipped by a shunt diode clipping circuit. In this case, the main idea of the circuit is introduced as user-adjustable "diode" circuitl instead of just using a pair of generic shunt diodes that are referenced to a fixed, non-adjustable voltage (usually 0V or Vb).

Want to see some neat applications of that circuit... Google patent search for Dirk Baldringer (Circuit assembly for distributing an input signal) - I suppose this idea is employed in "Dual Drive" pedals - and David Tarnowski (Distortion contrik techniques and and configurations) - which would be ADA's "D-Torsion Core".

Edit: Also wanted to add, conceptually it's a great idea to vary characteristics of harmonic distortion by controlling symmetry vs. asymmetry of the distortion. What I don't understand is why the variable control is implemented for both halfwaves. Implementing that variability to just one of the half waves already does the trick (though I still like that biasing scheme) so it's wasted effort for the second one. An arbitrary signal doesn't really "care" which polarity you clip stronger than the other. Also, there's already a gain control for tuning preferable input sensitivity.

[RickL]

What I don't understand is why the variable control is implemented for both halfwaves

I agree that there is some redundancy  but wouldn't a little bit of clipping on one wave and a lot on the other sound different than no clipping on one wave and a lot on the other? Isn't that the point of using back-to-back Si/Ge diodes in a similar circuit instead of just a single Si diode?

[Rob Strand]

Isn't that the point of using back-to-back Si/Ge diodes in a similar circuit instead of just a single Si diode?

There's a few differences.   I guess teemuk's beef is why offer "two diodes up and one diode down" *and*  "two diodes down and one diode up".

Most circuit have a cap between the opamp and the clipper.   A single diode (or a different numbers of diodes) charges the cap so a DC voltage develops.    The above circuit is DC coupled so it doesn't have that effect.