Fuzz Factory build: I want to try alternatives to AC128 - what's interesting?

[Philtre]

So, I built a ZVex Fuzz Factory from the layout at tagboard effects (http://tagboardeffects.blogspot.co.uk/2012/10/zvex-fuzz-factory-compact-layout.html) using two AC128s from Fuzzdog and it's a lot of crazy fun. And now I want to build another one, but this time with different PNPs. What have you guys tried? What do you think is interesting or fun?

[somnif]

Funny enough, I enjoy mine with 2N3906's.

Yep, silicon.

Different flavor from germanium, but with the knobs its not hard to rein in and get it to play nicely.

[Philtre]

Funny enough, I enjoy mine with 2N3906's.

Yep, silicon.

Different flavor from germanium, but with the knobs its not hard to reign in and get it to play nicely.

Interesting. Cheap as chips, too. ;-) Does it still oscillate like a crazy thing?

[somnif]

Funny enough, I enjoy mine with 2N3906's.

Yep, silicon.

Different flavor from germanium, but with the knobs its not hard to reign in and get it to play nicely.

Interesting. Cheap as chips, too. ;-) Does it still oscillate like a crazy thing?

Oh sure, its touchy as hell and will quite happily screech like a banshee. I have the 8 knob Aion version, so I can lower the gain a little on Q1, and bump the bass response up a bit and it tames the silicon a bit. Still, very different tone from the germs, but it has its place!

(one of these days I will get around to building a toggle-able version for Si or Ge at either tranny, but I'm remarkably lazy, so for now I just swap them in sockets as I wish)

[thesmokingman]

I've built one with ac122s as a kit from musikding in germany (first pedal I ever built in fact), did one with ac128 because I thought I was somehow missing something, and 2n404. low gain pnp silicon would be neat for a different flavor ... there's also the spam can transistor version of the fuzz factory ... I'll get around to doing one of those eventually.

[ahiddentableau]

When I built my FF last year I went a bit crazy with germaniums.  I bought 7 or 8 different types, then built R.G. Keen's test circuit and tested them all.  After a long A/B process, I'm (oddly) disappointed to say that I think the type of germ matters hardly at all.  Two pairs with similar gain and leakage will give you the same sound and response.

So my advice is don't bother focusing on type.  Focus on gain and leakage.  That's what matters.

Good luck!

[Philtre]

When I built my FF last year I went a bit crazy with germaniums.  I bought 7 or 8 different types, then built R.G. Keen's test circuit and tested them all.  After a long A/B process, I'm (oddly) disappointed to say that I think the type of germ matters hardly at all.  Two pairs with similar gain and leakage will give you the same sound and response.

So my advice is don't bother focusing on type.  Focus on gain and leakage.  That's what matters.

Good luck!

Which types did you try? I guess I'm looking for a cheaper option than an AC128..  ;-)

[somnif]

Which types did you try? I guess I'm looking for a cheaper option than an AC128..  ;-)

You can occasionally find large lots of Russian surplus germs pretty cheap on ebay and spend a few long weekends testing every one. For example:

https://www.ebay.com/itm/50x-MP41A-41A-RUSSIAN-MILITARY-GERMANIUM-PNP-TRANSISTOR/253414889625?hash=item3b00b45899:g:uyoAAOSwCtJaU0Wf 

Just scan around, see what you can find. Theres all sorts of random crap out there to find  .

note: always check to make sure a transistor is one actually useful for our purposes. Gains in the 50-200 range, make note of pnp vs npn, that sorta thing. Google is your friend, the fuzz face (et al.) is so ubiquitous that I bet just about every transistor on the planet has been tested in it, and someone will have an opinion (which you are welcome to ignore, of course!)

[ahiddentableau]

With regard to cost, I guess it depends on how you're looking at things.

If you're just doing one FF and have no future plans to make other fuzzes, then I'd say bite the bullet and just pay Small Bear or another reputable seller for a set you like the look of.  Sure, they're expensive, but you know you're getting the good stuff and you don't have to expend all kinds of time and effort sorting things yourself.

If you're going to do multiple builds, then it can make sense to source and sort your own supply.  This is more involved than you might like or immediately recognize.  It means you have to:

1) Source your transistors
2) Build a test circuit
3) Use the test circuit to sort your raw stock so you can use it intelligently

This is kind of involved.  It's not difficult, but it is time consuming.  So if you're not in it for the love, it's almost certaintly not worth it.

If you're going to do it for the love, then some comments regarding those three considerations:

1) Sourcing: If you look on eBay, for instance, there are sellers who sell raw stock.  The thing you have to accept is that you are, in fact, buying raw stock!  This means you're paying for 10 or 20 or 50 transistors at a time, and at the end of the day only 20-30% are going to be useable.  The level of variability is high.  You might get entire lots that are basically junk.  You might luck out and get a lot that are mostly keepers.  But the burn rate is high.  Understanding this in advance is important.

2) R.G. Keen has a simple circuit for testing Ge transistors.  It's on GEOFEX and is worth looking at in any case.  Lots of great info on Ge devices (thank you Mr. Keen).  But if you're going to build it you have to buy a specialized resistor (I think it's 2.49K 1% but don't hold me to that) to make the math work out.  This is easy to get from Mouser, but it takes time to get.  Another headache buying a set avoids.

3)  Then you have to actually measure all the devices.  This is also tedious.  Another headache.  It also gives you a good sense of just how temperature dependant germaniums are.  You'll have the device in the sockets and just the lightest touch will send the readings all over creation.  What I'm saying is, the testing is itself tempermental and seems to drive some people crazy.  Are you one of those people?  If so: buy a set!

All that said, I don't regret sourcing and testing my own supply.  I like cork sniffing, it turns out, even in a situation like this where the result is that I find out I don't think it ends up making a difference.  And now I have a healthy supply of tested devices for future builds.  If that appeals to you, then do it!

tl;dr version: it's a pain, so buy a set if you're not prepared to invest more time and money than you think you'll need to invest.

As far as devices go, I've had good luck with 2N404s and AC170s.  The 2N404s I sourced were especially consistent.  But YMMV.  It will vary.  Almost certaintly!

[selfdestroyer]

I have had fun with 2SA72 in my PUDGII and AC128 in my Aion Flare. Both had great results to my ear.

Cody