Fuzz face topology - alternative gain controls

[storsvedian]

After years of doing mods I'm finally getting into building my own stuff from scratch. As per usual, one of my first stops is the dreaded classic-sounding-yet-highly-tweakable fuzz face kind of things (though I'm probably leaning towards more of a MK1.5/Vox Tone bender type sound) and one of the things I want to address is the gain control. Although we all love what a fuzz face type circuit does on full blast, I've never liked the idea of having to turn down the guitar volume to make it clean up (IMHO the stock gain control is of little use for lower gain sounds). I've been toying with various ways to control the gain in a more amp-like way and I thought it'd be interesting to hear your thoughts on each option while waiting for parts to arrive.

My idea is to use a fixed 1k resistor to set the gain of Q2 on full, while adding a gain control at another point in the circuit. These are the three main alternatives that I've come to think of so far:

A) A pot wired as a voltage divider at input, effectively emulating a guitar volume pot. Seen in the Heathkit TA-28 (which I own) and tried it in my Jen Double Sound SI fuzz face-circuit.
Pros: Great gain response.
Cons: Leaves the rest of the circuit at full blast, giving a real bad S/N-ratio at lower gain settings while also affecting input impedance. Can get a bit thin sounding at low gain settings (though I haven't tried playing with the pot's value to combat that).

B) A pot wired as variable series resistance at input, as seen in the clean trimpot of the Analogman MK1.5/Sun Face as well as the impede knob of the Ramble FX Twin Bender Unprofessional MKII/1.5. Only heard it via YouTube-demos.
Pros: Seems to give a good general cleanup response. Leaves the circuit stock when put at max (zero resistance).
Cons: Presumably the same as A.

C) A pot wired as variable limiter resistor between Q1C and Q2B, as seen in the MKIII Tone bender (of which I own an original unit).
Pros: The MKIII gain pot has a really nice and even response (although it's range is somewhat limited due to it's 100K value, really interested to see how it would perform with a 250K/500K pot). Leaves the circuit stock when put at max (zero resistance).
Cons: Seems to darken the sound on low gain settings. Probably affects bias if no coupling cap is added (which I guess is the function of the 0.22uF cap before it in the MKIII).

With the goal to minimize noise and to give an even gain response, C seems like the most reasonable way to do it. How would you pursue implementing it in a fuzz face type circuit? Do you have any other favourite alternative way of controlling the gain of a fuzz face?

Also please forgive me for being quite novice with electronics. I may have gotten a lot of stuff wrong, but I'm here to learn

[luix]

You can try this solution, but the input impedance will also vary.

[nooneknows]

what about a plain FET , common source buffer feeding a 6.8K in a 250K pot to ground, center to the 2.2uF input cap? that way you can put the pedal anywhere in the chain. the R could simulate the pup impedance, the 250K the guitar volume pot. I think I even tried it somewhere in the past and it didn't sound bad

[luix]

what about a plain FET , common source buffer feeding a 6.8K in a 250K pot to ground, center to the 2.2uF input cap? that way you can put the pedal anywhere in the chain. the R could simulate the pup impedance, the 250K the guitar volume pot. I think I even tried it somewhere in the past and it didn't sound bad

The guitar pickup is not only a resistive impedance but also inductive and capacitive combined, the fuzz face greatly interact with those element so it sound different on every pickup, you should simulate also the L and C to get a realistic sound.

Maybe your solution will sound good but usually guitarists love the way fuzz face catfight with their guitars

[plush]

Check this great article on achieving pretty plausible pickup emulator by using small signal transformer w/ switchable wingings. Can emulate both single and humbucker with it.
Put a buffer in front of it and, BAM! you can use your FF with low impedance input signals (place it almost everywhere in your signal path).

http://www.muzique.com/lab/pickups.htm

[nooneknows]

The guitar pickup is not only a resistive impedance but also inductive and capacitive combined, the fuzz face greatly interact with those element so it sound different on every pickup, you should simulate also the L and C to get a realistic sound.

I know, but the simpler solution I proposed (btw I tried it in the past) is something very close to what MXR implemented in the M-173 Classic 108 Fuzz for example, where they used a BJT instead of a FET, with a 10K on the emitter output, and there it seems to work fine.
The aim is to reduce the gain of the first stage when a very low impedance load is connected to the fuzz: the gain is very variable, due to its very low input impedance, so a resistor could be enough (a solution implemented in some way in early fulltone fuzzes too).

[luix]

Check this great article on achieving pretty plausible pickup emulator by using small signal transformer w/ switchable wingings. Can emulate both single and humbucker with it.
Put a buffer in front of it and, BAM! you can use your FF with low impedance input signals (place it almost everywhere in your signal path).

http://www.muzique.com/lab/pickups.htm

I'm arrived to the same solution without knowing this article

Measuring the TM019 primary inductance gives me about 4H and 7H on the two taps so this is a good approximation for a SC and a HB pickups.

I'm using it to record some samples of my fx and it works very well!