"softer" superfuzz

[mordechai]

I'd like to build a Univox superfuzz type circuit, but one with a little less gain than the standard circuit.  I like the general tone and octave overtone of the original devices but I'm wondering if anyone can suggest alternative transistors that might yield a slightly less massive/intense fuzz tone? 

If no substitute transistors are recommended, would additional resistors to ground on some of the standard transistors help with this?  If so, what type of resistors would be advisable?

[slacker]

A lot of the fuzz comes from the clipping diodes, you could try removing those or replace them with LEDs or try extra diodes.
Making the input cap smaller helps to clean it up as well.

[Derringer]

something I like to do with the diodes is put a 10K trim pot between them and ground. Tune the trim by ear.
Try it without the diodes by all means too.

Also ... mids. Put another 10K pot between the 0.1 uF  cap in the scoop section to ground and it will let you un-scoop the mids.
Lowering that cap to something like 0.04 - 0.05 uF helps too.
... unless of course you have the switch on your build to switch from the scooped circuit to not-scooped

[Humboldts Finest]

Something I've seen is putting a small ceramic cap between b and c of Q1. This will change the coloring of the distortion, but will soften it up too. Look at the Fuzz Factory schem.

[Tony Forestiere]

Maybe replace each clipping diode with 2 or 3 1n34a germanium diodes in series?

[petemoore]

  All goodwork.
  See AMZ labs notebook...warp control and the other articles too.

[Mark Hammer]

In my own Superfuzz, I have a 3-way toggle to select between the stock pair of GE diodes, a pair of Si diodes, and no diodes.  As you would imagine, there is a shift in output level from Ge to Si diodes, and again from diodes to no diodes.  The unit still distorts in the complete and total absence of a diode pair to clip.  It's a "warmer" distortion.

So, my advice is to first just lift the diode connection to ground and see what that does for you.

Another path to pursue to "soften" is to work with the tone control, and possibly modify it.  The stock SF has two tone positions.  One is a classic midscoop that gives the sizzling top with huge bottom.  The other is an essentially unmodified signal that is simply attenuated, by a 47k/10k network, so that the midscooped and non-scooped sounds are approximately the same volume.

It is a simple matter to turn that 47k/10k network into a "rounder" tone, by simply putting a cap in parallel with the 10k resistor to ground.  The network will now attenuate and also roll off some top end.  With the 47k fixed resistor, a 2200pf cap to ground, in parallel with the 10k resistor, will introduce a rolloff starting just above 1500hz.  It will be a shallow rolloff, so you will get some treble, but it will be less aggressive.  Larger cap values will take more treble away, smaller values let more through.  You may want to make that cap's connection to ground a toggled function so that you get more variations in tone.

[mordechai]

Thanks for those suggestions, everyone.  And Mark, you've give me some very good ideas to futz with the tone switch options on my build.  I'll report back one I've actually tried them out.

[Gila_Crisis]

In my own Superfuzz, I have a 3-way toggle to select between the stock pair of GE diodes, a pair of Si diodes, and no diodes.  As you would imagine, there is a shift in output level from Ge to Si diodes, and again from diodes to no diodes.  The unit still distorts in the complete and total absence of a diode pair to clip.  It's a "warmer" distortion.

What are the differences between Ge and Si? I have a Boss FZ2 Hyper Fuzz, which is practically a Superfuzz variant. On this the clipping diodes are Si and I was wondering how would the tone be with Ge.

[Mark Hammer]

Lowering the clipping threshold via diode substitution has two impacts:

1) The overall output volume is noticeably lower.

2) The clipping provided by the diodes last longer into the note, and is more likely to occur for more notes picked.  This is often interpreteed to be a function of the "knee" of the diode, or magical germanium juju, but I suspect it is simply because of what portion of the overall note lifespan the diode affects.

[familyortiz]

I agree with Hammer about the addition/modification of the tone control, post drive. I add this because when I studied the Red Llama circuit, there was no tone control after the gain section, so all you got was very square wave behavior and harsh tone regardless of where your guitar tone control was set. Adding a 20k pot in series with a .1uF cap to ground, right before my volume control, really added a whole range of tones.