I built a NPN Ge and an NPN Si version.
NPN GE: went looking through my stash and found some 2SK352,, AC175s and other GEs, but irrespective of the transistor choice I got great fuzz with incredible sustain, but a lot of noise . Didn't take the time to modify the pcb with RF caps or resistors. The transistors are probably too leaky
NPN SI: in this setup as well as in my other configs, I used a 100K pot for the bias so I could always dial in the 'correct gain'. I tried 2N2222, 2N4401, MPSA13, 2N3904 and 2N5088/9. Finally settled for Q1 = 2N2222; Q2 = 2n4401; Q3 = 2N5088. Very low gain silicons didn't work - say hfe < 100
After biasing at 4.5V, it seemed that nothing was lost tone wise raising the bias to 7V - but you win in SNR, treble content and some definition in the sound. This requires only 1k on the collector of Q3 in my setup. When you hit your guitar hard, the bias momentarily drops to 5.8V or something.
Was only trying out the GE's out of curiosity, and because it has a history of being a problematic build.
I was also wondering, is that original Silicon Tonebender by Gus still available?
"The whole point of diy is diy. It's not dsoiyathodtr - do some of it yourself and then have others do the rest" (paulc)