Q2 Darlington Fuzz Face

[chromesphere]

I posted this on my facebook page, was wondering if anyone here had tried it.

The story is, bought some germanium transistors, they were SFT307.  All consistently measured under 30hFE which is odd compared to my past experience with the SFT307.  So i was disappointed and was considering taking it up with the seller on ebay.  But then i thought, maybe ill try a darlington pair test in a fuzz face.  A paired 2 together and used for Q2, they were both approximately 25hfe and had less then 20 leakage, i cant remember exactly, but the result was 480 hfe (lol) and leakage approximately 200-250ua.  Difficult to tell at this point until i can test it out in day time and crank it up but from my low volume test tonight, didn't sound half bad...dear i say, good...it did however require tweaking of the bias trimmer on collector of Q2 to get it sounding less splatty and more fuzz face-y :S

Thoughts?

[alanp]

Steve Daniels has commented on Darlington germaniums -- I think he called his project "Oh My Darling".

[chromesphere]

Ive read that article some time ago, ill read it again thanks Alan.

I think the transistors you can use for this sort of thing are very limited, they have to be very low gain () and low leakage, both have to be under 20 maximum (20x20=400ua of leakage current and 20x20=400 hFE, both are getting up there, leakage probably being the more critical of the two).  The challenge in the past has been finding that combination. I have wanted to build a bazz fuss with NPN's but that combination is even harder to find.  Anyway, this combination has worked well, I'm going to have to do a sound comparison of this one, for my own sake and to show you guys what it sounds like incase you have never tried before.

[culturejam]

I have wanted to build a bazz fuss with NPN's but that combination is even harder to find. 

I've built a few Bazz Fusses with a pair of 2N4401 and it worked fine. The "standard" setup with an integrated darlington has a super-high gain anyway, to the point where it doesn't really much matter what the gain is on the tranny, as long as it's greater than, say, 2000-ish.

[chromesphere]

The gain was not the concern as such, but with germanium npn's with that sort of high gain the leakage will be through the roof.   A sziklai pair might be a good compromise, silicon + ge pair, and of course you can reverse the effective polarity with an NPN as the first transistor in the pair.  You could also get high gains with a silicon with decent gain in the pair, say 2n5088 coupled with a low gain germanium.  What does this achieve?  don't know...maybe a slight change in tone with that ge in the set.  Bit of fun all the same, going to have to mess around with it.