Schumann Two Face.

[digi2t]

I've tried for months now to find one to poke into, but no luck. I found one person, over at DAM, but he flat refused my request for a peek.

Anyway, knowing that the finest minds of the trade reside here, I was wondering how some of you wiz kids would draw up a circuit that would resemble this description;

Two germanian transistor fuzz-face style distortions linked together at the ground rail.
One distortion controls the positive half of the wave. The other distortion controls the negative half of the wave. This lets you create very odd asymmetrical wave forms.
The TWO FACE FUZZ has two outputs which lets you have one fuzz come out of each output or you can create a mix.
A boost foot switch lets you adjust the positive and negative side to adjust how much you are going to boost.

Here is one version;



But I also found another version, with two additional switches;



No clue on what the addition switches do. Maybe gain boost switches (one for each section), a la Gemini III Dual Fuzz? Who knows.

Basically, I'm looking for something to breadboard. Ideas anyone? Any takers?

[nocentelli]

http://www.diystompboxes.com/smfforum/index.php?topic=73802.0

http://electriciantraining.tpub.com/14180/css/14180_40.htm

BJT phase splitter with diodes from both the out of phase outputs to two separate fuzze faces: another bjt on the end to re-mix the signals, alongside separate outs? Totally guessing here, btw.

[Gus]

Maybe something like this I GUESSED +- 9 supplies and simple mixing. I used a FF like circuit I like for this example/guess. The -9 rail got cut off in the screen shot


Not Ge, a fast sim to show a basic idea of NPN and PNP Si FF like circuits with a common ground.  The output of each FF like might come off the 2nd stage collector
Don't know from the description if the fuzz inputs have a input volume control or series gain setting resistor
As always adjust cap values
I would not adjust the 2nd stage collector resistor value much over 10k

This is posted as a simple idea from the description.  The inputs of each FF like circuit might have issues without resistors or an input volume control.

EDIT just found this link +-12VDC supply
http://www.schumannelectronics.com/two_face.html

Another input blend idea The input could go to the wiper of a potentiometer and the outside lugs could go to each FF like circuit this would work with the feedback bias resistor and change the gain at the first transistor stage of the fuzz

[digi2t]

Huh, first time I've seen that. Interesting. Here's another theory that eatyourguitar offered over at ilovefuzz,



although it doesn't address the boosting, or splitting part, but I'm planning on trying it.

Total guesses are what I'm expecting here. Another thing that strikes me is the "Boost", "Drive", and "Clean" controls. These are the same controls found on the Lion;



which I traced out a while back;



Now, I'm wondering, could the Two Face be a Lion, with phase splitting going on? An 072 for the clean signal, a 4049 for the boosting, but where/how do the NPN/PNP fuzz circuits fit into the equation?

This sounds like a job for Super Gus.


EDIT:
OMG!!!! I hit preview, AND GUS HAS REPLIED!!!! This is like TOO CRAZY!!

Thanks for the quick sketch Gus. Something to chew on, for sure. Personally, I would prefer to keep this germanium though, since I have shitloads of Russian transistors that I can use for a project like this. I'm clearing the Iron Butterfly off the breadboard today.

[Gus]

I posted a guess
If it has a dual opamp one stage could be a input booster
Drive as the opamp gain control
Clean as a volume control after the gain stage
If it uses a opamp input buffer at +12VDC  -12VDC it should stay clean at lower gain settings

Sensitivity could/would be a pre-gain control for the FF like part series resistance or a volume control
Level standard volume control maybe a 100K
Gain FF like gain control

A switch to disconnect the mixing of the outputs the output could be a passive or active mixer

So maybe
input opamp gain stage with a volume control going to the clean out (drive, clean controls)

Booster spilt to both FF like circuits one + NPN one - PNP

FF stage pre-gain control(sensitivity)

FF volume(level)

FF gain (drive)

Switch to combine or split the outputs

Then you adjust the mix with both level controls

Boost might be adding more gain to the input opamp stage or adding gain to the active mixing output if it has a active mixing output circuit

Maybe I will sim and post the complete idea.

[digi2t]

Maybe I will sim and post the complete idea.

If you do (I hope you do), I'm ready with the breadboard.

What you're guesstimating would/should be awesome. As for the 1/N/2 switches, does it make sense to you if they were gain switches for each fuzz section? You know, like the Gemini III dual fuzz?



EDIT; Yes, from what I understand, and have seen, the original does run on a +/-12v scheme. It seems to be a standard that all Schumann pedals were build to.

[Gus]

I used stock parts from LT Spice
The two Si fuzz circuits are the type I like to build.  If you you search you might note it is based on the same type circuit I started posting in the 90's

You can substitute a Ge NPN and PNP fuzz of your choosing, HOWEVER you want to have a known output resistance of the 2nd transistor stage of the FF like circuit, total collector resistor value about 10K should be good

for fuzz I like a 220k input resistance when using an opamp input

sim works well at +-9VD with a bias adjust (R4, R14)

I did not show a power supply circuit +- 9VDC batteries should be fun to start with

Not built a sim guess.

As always adjusts parts to your liking but the posted sim screen shot might be a good starting point

Any questions?

[digi2t]

Gus, it looks brilliant!

Gonna start plugging parts into the breadboard tomorrow. This is going to be fun!!

[LucifersTrip]

you'll prob be able to get some idea what does what from the youtube vid...another sputtery glitch machine.

[digi2t]

Thanks Luce. At least I'll have something to compare to sound wise.

[Gus]

Looked at the youtube link

Is the clean for bypass or does it mix in with the outputs?

Sounds like minimum or no power supply decoupling caps for parts of the circuit, this idea is from the oscillations OR maybe the FF like circuits run at about +-9VDC but are powered via series resistors selected to drop the +- 12VDC to a lower voltage with no bypass caps

The two switches between the controls in the second picture could be some kind of bias and/or feedback settings for the FF like circuits could be something like changing the 2nd emitter to first base feedback/bias resistor value.
EDIT or maybe it could be different voltage dropping resistors to the FF like circuits

I might guess from looking at the video the output volume wipers have a series resistor that goes to the output jack and the link switch shorts/connects the outputs together at the resistor side not connected to the wiper mixing the signals together.  

EDIT found this site http://bassfuzz.com/schumann-pedals/schumann-two-face-fuzz/

[digi2t]

Is the clean for bypass or does it mix in with the outputs?

If I were to take a cue from the Lion, I would hazard to say that it's mixed with the outputs.

The two switches between the controls in the second picture could be some kind of bias and/or feedback settings for the FF like circuits could be something like changing the 2nd emitter to first base feedback/bias resistor value.
EDIT or maybe it could be different voltage dropping resistors to the FF like circuits

I'm not as interested in those at the moment. I'll wait until I see what the breadboard starts producing first. Voltage dropping would seem feasible though, for added sputtery.

I might guess from looking at the video the output volume wipers have a series resistor that goes to the output jack and the link switch shorts/connects the outputs together at the resistor side not connected to the wiper mixing the signals together

Makes sense.

I'm going to start with your original scheme first, and see how it sounds/performs. BTW, if I go with Ge transistors, what would you recommend as resistor adjustments? I realize that it's tougher to call, due to the leakage angle, but ideally, do I stick to the plan, or do I tweak some values?

[Gus]

still thinking about the clean control
I think Ge's should work with a the bias control
might need some minimum value caps at the FF supplies if you use the dropping resistors

[digi2t]

A better demo;

Looks to me like the CLEAN control overrides everything.

Split or Link control seems to join or divide the two circuits.

[Gus]

OK for the clean maybe have two 100K resistors off the wiper of the clean control and have one go the + out and one go to the - out.  This will mix the clean with the fuzz split or linked.
Split or linked is shown in the last screenshot as "connect to mix outputs"

Use what ever Ge ff like circuit you like but try to keep the Q2 output collector resistors about 10K total.

I posted Si Fuzz sections for more possible consistency.

[Gus]

Anyone have an another idea of what this might be? 
Mine are guesses as to what the circuit might be like.

[digi2t]

I have your idea on the breadboard now Gus, but it's awfully noisy. Specifically the pre-boost section. I've been doing some reading, including trying to conjure the spirit of John Schumann, and I'm going to try something a bit different. I'm keeping your twin fuzz section, but deleting the boost in front. Instead, I'm going to try a 4049 type post overdrive for the boost section (Craig Anderton Tube Sound Fuzz). Instead of the 100K/10M switch, I'll use a pot to vary the boost, with a footswitch to kick it in/out.

The fuzz section is nice though, with Russian transistors filling in quite well. I will need to tweak the resistor values though, just to fine tune the fuzz vis-à-vis the transistors I use. Sensetivity control works well, but I'm still up in the air as to the 3 position switches. I'm still thinking that it might be some kind of gain control, like the Gemini Dual fuzz uses.

I'm redrawing what I want to try right now, I might have some time tonight to play with it some more.

Stay tuned......

[digi2t]

OK, sudden turn of events. I decided to go in a different direction (I'm havin' a Jimi Photon moment  ).

I had an unexpected, yet pleasant, surprise. Here is what I have on the board now,




I dropped the preamp front end, choosing to go direct to the fuzz'zz, then pumping each fuzz into a 4049 overdrive. I'm only using one chip, since I have enough room to carry both fuzz circuits, in and out. The 4049 boost pot is a 1M dual gang. I'm using Russian NPN and PNP germanium transistors for the fuzz circuits, with trimmers to bring the collector voltages even at 4.5v. I also opted for a 2K pot on each fuzz circuit for fuzz control. The 100K sensitivity pots are there.

Now.... (Gus, get your sim cap on here)

The surprise is this; for some bizarre reason, when I'm using an output separate from the other, the 2K gain control of the other fuzz section affects the one in use. As a matter of fact, if the gain of the unused fuzz is cranked, and you dial the gain of the fuzz being used to 0, instead of losing the fuzz completely, I get a beautiful octave fuzz.  I mean, it's giving a Superfuzz a serious run for the money folks! Even the sensitivity control of unused fuzz affects the fuzz being used. If the unused gain is dialed down to 0, then the gain of the fuzz being used reacts normally, full fuzz at 10, no sound at 0, and the sensitivity pot of the unused fuzz does nothing.

What the hell is going on here? 

The 4049 boost can really push everything hard. I think a 2.2M pot would be the max I would use here. As for the fuzz'zz being linked, playing with the controls can yield some very interesting textures, although the great octave fuzz is unavailable.

As for the clean blend, I can keep the originally drawn pre section, and use a dual gang pot on the back end to swing between clean and dirty.

But how I'm getting that great octave... that one is really blowin' my mind.

Sorry I massacred your schematic Gus. I just didn't feel like redrawing everything.

[Gus]

could be feedback to the input of the FF like circuit causing changes or power supply interactions

The fist stage of a FF is kind of like an inverting opamp stage the 47K from the Q2 E to the Q1 B is the feedback and part of the bias resistor(think pinch knob)
If Q1 had gain like an opamp it the input node would be close to 0 ohms with feedback

Why trimmers in the collector legs?

Have you tried the passive output section like in the last screen-shot I posted?

Something else to try use two caps maybe 100uf at the 2.2K voltage dropping resistors to FF like nodes to ground to test if the interaction changes, this would be a power supply coupling test