Switching transistors in a Fuzz Face

[Ucho]

As many of you before me (in the meantime with me, and after me...), I'm on a fuzz trek right now... 

I'd like to build a fuzz with the option of switching various transistors with a rotary switch (two actually, one for q1 and one for q2).

I'd like to understand is if the only way to do it is with a 3-poles switch (unconnecting E B and C for each transistor) or it can be done with a 2-pole switch, leaving one between E, B, and C always connected for every transistors and switching the two remaining. Serching in the forum, If I understood correctly, it can be done with a DPxT (x for the number of throws/transistors). I've read about someone talking about Oskar Bruil and his "switch-o-matic" fuzz (or something like that), and other posts by him regarding this argument, but i couldn't find those posts nor his website...

Here's a simple and generic FF schematic. The red dots are the switching points for Q1, the blue ones are for Q2.



Q1: I think there's no problem in switching Collector and Base and leaving Emitter always connected, since it's connected directly to ground.

Now to Q2: I've positioned the "Collector switching point" so that the bias trimpot is included, so each transistor will be switched WITH its own bias resistor, so it won't be needed to adjust bias when switching trans: am I correct?
The "problem" I see about Q2 is that I'm not sure if leaving its emitter connecter will affect the circuit (I'm fearing that because the emitter is connected to the feedback resistor), or, since Collector and Base aren't connected, it wouldn't conduct, so there wouldn't be any problem in that sense.

What do you think about it? Should this switching scheme work?
Has anyone already done something similar?

[petemoore]

Now to Q2: I've positioned the "Collector switching point" so that the bias trimpot is included, so each transistor will be switched WITH its own bias resistor, so it won't be needed to adjust bias when switching trans: am I correct?
  Sounds like it has to be, if separate bias is provided.
  The "problem" I see about Q2 is that I'm not sure if leaving its emitter connecter will affect the circuit (I'm fearing that because the emitter is connected to the feedback resistor), or, since Collector and Base aren't connected, it wouldn't conduct, so there wouldn't be any problem in that sense.
  Emitter should be cool with that, if the base/collector are lifted, it'd be just the little bit of 'connector'...floating, but still possibly able to pickup something, *probably not enough to cause noise, however if this were on the base I would could be a radio  antenna.
  People are using multi FF builds, I think this is close enough to how they are doing it...I don't *see a problem with it.

[Ucho]

Thanks peter for replying.

Now to Q2: I've positioned the "Collector switching point" so that the bias trimpot is included, so each transistor will be switched WITH its own bias resistor, so it won't be needed to adjust bias when switching trans: am I correct?
  Sounds like it has to be, if separate bias is provided.

Sorry for being dumb (my english is not that good, plus I'm not a very smart guy  ), did you meant "Sounds like it has to be correct"?!
And by "separate bias" you mean the unique bias trim included with each transistor?
Again, sorry for the stupid questions 


Anyway, i think I'll make a breadboard experiment: on my breadborded fuzz I'll place a couple of transistor with only their emmiter connected to Q2 emiiter (to simulate the situation I should have with a switch when one transistor is selected and others are still connected on their emitters and unconnecter on base and collector, as in previous diagram), play a little bit while swapping them in and out of the board, and hear if it makes any difference in tone, or if some noise (apart from pops when inserting/removing them) is introduced when those two trannies are connected.

Here's a schem of what I'll do.



Q2a, Q2b, and Q2c are the selectable transistors for Q2 position (each with its own presetted bias trimpot). Q2a is the one currently selected, the remaing two are unselected, but, from previous schematic, their emmitters are tied.

Has anyone done this selectable transistor thing, with a DPxT switch?
Any other opinions?

[petemoore]

Sorry for being dumb (my english is not that good, plus I'm not a very smart guy   ), did you meant "Sounds like it has to be correct"?!
  Well I don't see a schematic and can't trace your board, from the description the emitter is effectively connected to...nothing...when that transistor is not in use, that wire is connected to a 'floating' transistor [AFAiCT].
And by "separate bias" you mean the unique bias trim included with each transistor?
  yes/necessary.
Again, sorry for the stupid questions
  Nope, never done it, sounds workable.

[Ucho]

I've made a quick test on the breadbord, placing a couple of transistors and connecting their emitter (leaving base and collector unconnected) to Q2 emitter, as explained (maybe not to well), and pictured, in my previous post.

Well, it seems to work ok: I didn't noticed any problem like "tone loss" or noises.

Anyway, I'd really appreciate if someone who has done something like that before can confirm that there are no problems in doing that this way, or suggest me a different way of doing it.