Help sourcing Transistors for Twin Face.

[UtilityBeltFX]

Hi everyone! I'm new here having just gotten into pedal building recently. I've built around 5 pedals and have gotten pretty good at not only circuit diagnosing but part sourcing. However I'm about to start building a Twin Face from PPCB and realized I don't know much about finding Silicon and Germanium transistors. The pedal calls for BC108s and AC128s but after searching around both this forum and the internet I quickly came to the realization that 1) Those are pretty hard to find transistors and 2) People substitute other Si/Ge to great effect.

I see that you want to measure their HFE and power leakage for PNP1/2 and NPN1/2 but I'm not entirely sure how that's accomplished especially without buying vast quantities of both.

If anyone could give suggestions for Twin Face PNP/NPNs and the best ways of sourcing/measuring them, I would be very grateful. The infinite variety of them got overwhelming very quickly and thought I should turn to the experts here.

Thanks again!,
Ross

 

[Synchrony Pedalworks]

The easiest, but not cheapest, source is Synthcube. The have transistors sets matched for the germanium transistors and you can also pick up some BC108 transistors for the silicon section of the pedal. Another option is Amplified parts. However, they do not match the transistors for the Fuzz face. Stompbox Parts has had AC128’s , but on my last look, have sold out on the good transistor HFe’s.

 

[Laundryroom David]

I buy all my germaniums off ebay. There are some great sellers. The trick is you have to test and sort them. I agree with @aefpv that your best bet would be a matched set.

WRT the BC108s, you shouldn’t have too much trouble tracking them down. Stompboxparts.com has ‘em.

 

[Laundryroom David]

one other thing - as long as the pinout works for your pcb, you can use pretty much any Ge PNP. It doesn’t have to be an AC128. For Fuzz Face, very generally, hfe of Q1 ~ 80 and Q2 ~ 110 will do just fine. Different folks have different opinions and preferences. But that’ll get you in the ballpark. Happy hunting.

 

[Big Monk]

These are fantastic:

https://www.amplifiedparts.com/products/transistor-2n169-germanium-oval-case-npn

They’re NPN as well so you can ditch the charge pump.

 

[szukalski]

I have used 2N2218 and enjoyed them as a low-gain alternative on the Si side.

 

[GizzWizzKing]

I am a bit in the same boat here. I have chosen to purchase in bulk from ebay and acquire a TC1. This was my preference because it cost the same to get ~80 Transistors and the TC1 as it would have cost for me to buy 2 matched sets or 20 transistors from smallbear or amplified parts. though i have been eyeing those black hat transistors for a while now on amplified parts due to a big monk bulletin post a bit back.

 

[UtilityBeltFX]

Thank you all for the responses! Since I'm only building one of these to start I opted to order a matched germanium pair from SynthCube (PNP Generic) as well as some BC108B's to try. Wish me luck!

 

[Laundryroom David]

Good luck!

 

[geoffrey]

I also plan on building a Twin Face (and am a novice) and was given some AC128's that range from 60-80 hfe. I know that 70-80 is fine for Q1, but would any of these lower values work for Q2 as well? How imperative is the hfe value here compared to the type of transistor?

I guess what I'm really is asking is do I need to go find something else other than these for Q2?

Thanks everyone

 

[iamjackslackof]

I also plan on building a Twin Face (and am a novice) and was given some AC128's that range from 60-80 hfe. I know that 70-80 is fine for Q1, but would any of these lower values work for Q2 as well? How imperative is the hfe value here compared to the type of transistor?

I guess what I'm really is asking is do I need to go find something else other than these for Q2?

Thanks everyone

The answer you're always going to get for this is "try it". You may end up liking it. I frequently sub lower gain transistors and end up liking the result. And being 10 away from the target hFE is not a whole lot, so I would try it out before spending any more.

 

[giovanni]

For silicon, you can also try the modern versions of BC108 like BC548, BC549 etc. which are easy to find. If you still need some Ge PNPs send me a private message and I can hook you up.

 

[Laundryroom David]

The answer you're always going to get for this is "try it". You may end up liking it. I frequently sub lower gain transistors and end up liking the result. And being 10 away from the target hFE is not a whole lot, so I would try it out before spending any more.

Agreed. I strongly suggest socketing whatever transistors you pick and bear in mind that differences in temperature will (really) change the sound of those germs make. So if you’re swapping transistors and testing the tasty toans, be sure to give the transistors several minutes to stabilize before you make any hard decisions.

 

[geoffrey]

Thanks for the great advice, guys. I’ll try what I’ve already got before I spend on any more.

 

[geoffrey]

For silicon, you can also try the modern versions of BC108 like BC548, BC549 etc. which are easy to find. If you still need some Ge PNPs send me a private message and I can hook you up.

I’ve got some BC108s for the silicon side. Thanks so much for offering to help with the Ge. If what I have doesn’t work out, I’ll definitely reach out!

 

[geoffrey]

I can’t believe this question just came to me now, but considering I populated my BC108s without even checking their hfe… just going straight into biasing them up so I can start melting faces immediately….it’s no wonder I didn’t think of it:

The question is: does hfe matter on the silicon side similar to that of the germanium, relatively speaking? Is there a certain range for Q1 and for Q2 we’re looking for in the NPNs as well?

I’m assuming “try and see” applies here as well, but I don’t see anyone posting about swapping out NPNs to try different ones like with their germanium counterparts, so does what NPN you use and it’s hfe really matter that much?

 

[Coda]

I can’t believe this question just came to me now, but considering I populated my BC108s without even checking their hfe… just going straight into biasing them up so I can start melting faces immediately….it’s no wonder I didn’t think of it:

The question is: does hfe matter on the silicon side similar to that of the germanium, relatively speaking? Is there a certain range for Q1 and for Q2 we’re looking for in the NPNs as well?

I’m assuming “try and see” applies here as well, but I don’t see anyone posting about swapping out NPNs to try different ones like with their germanium counterparts, so does what NPN you use and it’s hfe really matter that much?

Usually Q1 is lower than Q2 in an Si Fuzz Face…

 

[giovanni]

As far as I understand silicon transistor construction is a lot more efficient and their physical properties lead to more consistent results. They also lead to much higher gain devices (hfe can be in the 400 or higher). Moreover, leakage current is not usually a problem. This is all because of the physics of silicon semiconductors. All that leads to more “ideal” devices, so the circuit design has a much bigger impact than the specific transistor properties. I’m sure there are exceptions, maybe Chuck can chip in …

 

[Big Monk]

I prefer matching gains in a Silicon FF. hFE is all that matters and parts numbers are irrelevant.

Typically, matching hFE gives smoother fuzz while asymmetrical hFE gives more characteristic square wave distortion and a more aggressive character.

The asymmetrical hFE thing as a general rule in Fuzz Faces is pure myth though with respect to Ge or Si Fuzzes. At least as it’s bandied about for matching devices or achieving “that” sound.

 

[enrosenzweig]

Typically, matching hFE gives smoother fuzz while asymmetrical hFE gives more characteristic square wave distortion and a more aggressive character.

Is there a general rule of thumb as to what constitutes matched vs asymmetrical? I’m assuming that matched doesn’t literally mean exactly the same.