mixing silicon transistors in a fuzz face

[MordechaiBenZev]

So I've got four nice silicon transistors: two BC183s, measuring about 200 and 400 Hfe...and two BC109s, measuring 250 and 500 Hfe.  I love the Gilmour tone with the BC109s, but also the "warmer" tone the 183's give.  So I've been thinking about combining them in the circuit.  If I was aiming to find a happy medium between the two leaning SLIGHTLY more in favor of the Gilmour tone, would you recommend doing a Q1=BC109/Q2=BC183, or vice-versa?

I know I can socket them and use my own ears, and I WILL...but I'm curious about your opinions.

Thanks and happy new year.

[FSFX]

So I've got four nice silicon transistors: two BC183s, measuring about 200 and 400 Hfe...and two BC109s, measuring 250 and 500 Hfe.  I love the Gilmour tone with the BC109s, but also the "warmer" tone the 183's give.  So I've been thinking about combining them in the circuit.  If I was aiming to find a happy medium between the two leaning SLIGHTLY more in favor of the Gilmour tone, would you recommend doing a Q1=BC109/Q2=BC183, or vice-versa?

I know I can socket them and use my own ears, and I WILL...but I'm curious about your opinions.

Thanks and happy new year.

With a silicon FuzzFace, the transistors themselves are never likely to have any significant effect on tone. If you can hear it then it is the placebo effect or the bias is just a bit off so it is distorting slightly differently due to different hFE values. But just changing the signal level is going to affect that and as everyone knows, the FuzzFace has a very low input impedance and so whatever you drive it from will have a really significant effect on the distortion and the tone. That is why it works best driven from a high impedance source like direct from a guitar rather than from a buffered source. The thing that will really have a significant affect on tone is the values of the capacitors that you use throughout the circuit.

It is worth reading some of the excellent analysis documents about what is really going on in a FuzzFace circuit to understand what really affects the sound.

https://www.gitec-forum-eng.de/wp-content/uploads/2019/02/poteg-10-08-05-5-fuzz-face.pdf

[Steben]

Caps, gain and corresponding bias.
Silicon usually has higher gain than germanium, which results in slightly higher input impedance etc etc... the volume roll off on guitar gives more easily clear cleans.
Experiments will tell you more. And trimpots...

[antonis]

As many wise guys said:
(and prooved..)

"Any circuit based on particular BJT beta should be considered a VERY BAD circuit,,"..!!

[pacealot]

As many wise guys said:

"You want that I should whack him, boss?"

"Nah, just significantly lower his input impedance, that oughta teach him a lesson!"

[Gus]

this might help
https://www.diystompboxes.com/smfforum/index.php?topic=114153.msg1058953#msg1058953
Note how things change between the 2N2222 and the 2N5089 model.
That thread should help you understand how things can be adjusted in a FF like circuit

[Lino22]

Recently, i have done myself a favour and built two identical fuzz faces. I picked B183s and BC109s with identical gain, so one was BC183 and the other was BC109. I adjusted the C1 and C2 collector voltage to the same values.Then i plugged them into A/B box. I used my LP and my Strat to compare them.

They did sound absolutely identical.

At that very moment my life has become easier. You may prefer a manufactured FF with BC109 over a manufactured FF with BC183 because the voltage levels inside are set differently, mainly because their gains. Also their Volume Pot values could be different and that also has an audible impact on the sound, even if maxed.

[FSFX]

They did sound absolutely identical.

Welcome to the world of reality rather than that of myths and mojo.

[Knobby]

As many wise guys said:
(and prooved..)

"Any circuit based on particular BJT beta should be considered a VERY BAD circuit,,"..!!

So, is it possible to redesign the fuzz face circuit, or any other classic 60s fuzz for that matter, to use any transistor with an HFE greater than a specific value, such as a cheap modern high gain device? Presumably setting the gain required with the appropriate other components? It's thirty odd years since I studied transistors and I was never sure I quite got the detail of it all. People's experiments in the past with piggy-backing and finding low gain alternatives would suggest not, but I've not dug into it all that far to find out.

[antonis]

So, is it possible to redesign the fuzz face circuit, or any other classic 60s fuzz for that matter, to use any transistor with an HFE greater than a specific value, such as a cheap modern high gain device?

Of course..

[pinkjimiphoton]

i cry bullshit.
not all transistors sound identical. if they do, its cuz of where you chose to bias it.
take the same stock circuit.
no trimmers.
socket.
audition a shit load of transistors.
you will find some sound great, some not so great, some like ass.
this is the effect of different gain on the circuit, and the effect of different transistors tonally.
if you bias it to half supply, yeah, they may sound pretty similar, if the ratio of gain is good between q 1 and 2...
but half voltage isn't necessarily the sweet spot to make it sound good.
god.
i must have built a thousand sili fuzzfaces over the years.
yes, i can make them sound pretty much identical, but that was more in tuning than anything else.

[FSFX]

not all transistors sound identical

Of course not in a circuit like a Fuzz Face that has no emitter degeneration for Q1 and with the fuzz control and its emitter decoupling capacitor for Q2.

Doing anything like that in a design makes it dependent on the beta or hfe as well as the leakage and junction characteristics.

The Fuzz Face uses a circuit design from the early 1950s when they were just learning how to use these new fangled devices invented less than a decade previously. So the design is not optimised to eliminate variance in transistor characteristics.

That is why most modern audio circuit designs are not done that way now.   

[pinkjimiphoton]

of course they aren't. thank god for that, cuz then we wouldn't have this glorious circuit

[Steben]

As many wise guys said:
(and prooved..)

"Any circuit based on particular BJT beta should be considered a VERY BAD circuit,,"..!!

So, is it possible to redesign the fuzz face circuit, or any other classic 60s fuzz for that matter, to use any transistor with an HFE greater than a specific value, such as a cheap modern high gain device? Presumably setting the gain required with the appropriate other components? It's thirty odd years since I studied transistors and I was never sure I quite got the detail of it all. People's experiments in the past with piggy-backing and finding low gain alternatives would suggest not, but I've not dug into it all that far to find out.

Yes, but beware that the larger the gain, things change a bit. The input impedance changes for example. The higher the hfe, the brighter a fuzz face wil sound.
Mind you, I for one never really liked the low hfe germanium fuzz faces for that reason with single coils. They are slightly too dull. And as soon as you get into silicon ballpark with certain hfe's you simply do not go for germanium because of all the advantages of silicon. I find low gain GE faces a bit better for P90/humbuckers guitars.