Can anything be eliminated from my octave fuzz creation?

[Ben Lyman]

I started with a Green Ringer and kept adding things from other pedals until it looked like this. Maybe some things can be taken out? I even have the pull down resistor in there somewhere and some possibly unnecessary voltage dividers. Also, since I turned the first stage booster into a variation of Steve's Ursa Minor, I have some weird stuff going on. If you could change anything to consolidate this circuit into something simpler, where would you start?

Here's what it looks like on the BB right now:


and what it sounds like:

[Mark Hammer]

Not eliminated, but R9 doesn't have to be any larger than 500R to *do* something.  Keep in mind Q1 gain will be increased to the extent that AC has an easier path through the pot and cap than through R7.

Nice-sounding circuit, BTW.

[Ben Lyman]

Thanks Mark. So R9 is my gain pot but of course it's parallel to the 390r (R7) that was left over from when it was an LPB-1. If I take out R7 it goes silent, but maybe I can take out R7 if I replace R9 with a 1k pot. Then I'm afraid something else will go wonky on me but I think that will be my next step anyway. I have some C1k pots that I need to use for something.
I also just noticed that I forgot to show R23 as a 1k in the schematic, fixed now.

[Ben Lyman]

I was able to pull a few things out without changing the sound. I also tried a lot of other experiments but each time I did, it just threw everything out of whack.
I'm still very happy with the sound, happier about the parts count but I wish I could lose R12 and R13 somehow. They seem to be important to the bias at Q2b but I wonder if there is another way. Right now Q2 voltage readings are right on for a Green Ringer Q1, which is correct for my thing. Here's an updated schematic:

[Kipper4]

Have you tried scaling down R12, R13? It won't help with your parts count and it will futz with the RC filter formed by C6,R13.
Probably best left as is. Since it sounds good mate.
Even if you looked into biasing Q2 with a Zener or something again you'd lose the filters effect and parts count may go down by one resistor but cost would go up with a Zener.
I dunno.
Your limited somewhat by the number of pots in use as to what size enclosure you use.
I hope you find a happy balance.
Rich

[amz-fx]

Shouldn't the junction of R6/R8 be connected to the base of Q1 to bias the transistor? In which case you don't need R3, but if removed, you would change the value of R8 to 430k (470k will do if that's all you have).   If you don't remove R3, the bias of Q1 will be incorrect when R6/R8 is wired to the transistor.

Alternately, you could delete R8 and reduce R3 to 680k with the R3/R6 junction connected to the base. I like that a little better actually.

regards, Jack

[Ben Lyman]

Ya kipper, that's what I was thinking. The 160k is actually a 150k plus a 10k in series cuz that's all I have so that's not great. Oh well, it does sound good. I still have to play it at a regular volume at some point today just to make sure it sounds the same after last nights experiments.

Jack- that free schematic thing I use is tricky but ya, that is a connection even though it's missing the little dot. Thanks a bunch for the tips, I will try that next for sure.

[ashcat_lt]

Is C6 actually having any impact in the audio band?  I haven't done the math, but my eyeball tends to think it's got to be a pretty low cutoff.  If you can live without it for tone purposes you might be able to DC couple those stages and save an R or 2 along the way. 

What's the DC voltage on either side of that cap?

[Ben Lyman]

What's the DC voltage on either side of that cap?

5.42v on Q1c and 1.81v on Q2b 

Thanks ashcat, definitely going to look into that too. My gut feeling as I tinkered with it last night was that something weird and magical is happening from that point on... the way each transistor interacts with diodes and the final tranny as a buffered output stage seems to be very important but who knows... I'll mess with it later. As it stands, everything from C6 to C9 is a stock Green Ringer except my addition of the Ge diode and B500k pot, and all the voltage readings are pretty much right on according to the build doc.

[ashcat_lt]

5.42v on Q1c and 1.81v on Q2b 

Thanks ashcat, definitely going to look into that too. My gut feeling as I tinkered with it last night was that something weird and magical is happening from that point on... the way each transistor interacts with diodes and the final tranny as a buffered output stage seems to be very important but who knows... I'll mess with it later. As it stands, everything from C6 to C9 is a stock Green Ringer except my addition of the Ge diode and B500k pot, and all the voltage readings are pretty much right on according to the build doc.

Well, that's a pretty big difference.  If you remove C6, R12 and either R11 or R13, you could drop that voltage by messing with R10 versus whichever of the others is left, but that's going to be serious attenuation of the desired signal as well.  Might be able to fix that with more gain in Q2, but...

Another thought being that Q2 is biased kind of based on the difference between its base and its emitter.  If the emitter sat higher, the base wouldn't have to go so low.  Course then you'll be messing with gain again...  It's probably easier to just leave it as is. 

[Ben Lyman]

Ya, probably best leave it alone, I think it sounds really good. I did pull R11 last night and didn't notice any difference yet, so that's something. I think that last schematic might represent the final incarnation, I'll test it later with my real amp when I get a chance.

[Fast Pistoleros]

that is definitely sweeter sounding than a green ringer . good work ben. post a stripboard layout when you get her done and I will verify it for you   

[Mark Hammer]

I'm guessing that the clipping stage on the input adds a bit of compression/sustain that makes the octave more than a quick blip.  It may or may not have been designed with that in mind, but that doesn't stop it from being a good idea. 

[Ben Lyman]

that is definitely sweeter sounding than a green ringer . good work ben. post a stripboard layout when you get her done and I will verify it for you   

Aw man, I don't know how to do strip board! I guess now is as good a time as any to try and learn 

I'm guessing that the clipping stage on the input adds a bit of compression/sustain that makes the octave more than a quick blip.  It may or may not have been designed with that in mind, but that doesn't stop it from being a good idea. 

That is probably true Mark, I just tried a bunch of different combos of OD and booster pedals with my breadboard to decide what I wanted, I like the different sounds it's getting right now. I know very little about circuits, all my ideas are coming from spending many years stacking pedals together to get the sounds I like and always sorta wishing some of the combos were included in one box. That's why I think there is probably more components than necessary in this one but I will probably leave it the way it is.

[PRR]

> I wish I could lose R12 and R13 somehow.

You need gain-of-3 and two outputs. I don't see anything significantly cheaper than 2-Q 6-R. Maybe if I burnt $10 of brain-cell I could sneak-out one $0.12 R.... but it's not my project.

I do not see a for-sure reason you need R10, but overdrives can be fussy.

> The 160k is actually a 150k

My thumb-count says 150K may be a hair better. Sure you could jumper the 10K and find out faster than I could think.

> Shouldn't the junction of R6/R8 be connected to the base of Q1 to bias the transistor?

It has to be, or no-go. So this is poor draftsmanship. (And I was not allowed to blame my pencil for missing dots.)

> Is C6 actually having any impact in the audio band?

20Hz; but we "need" a cap there so Q2 isn't hanging-on Q1's bias (which may be wacko in overdrive).

[Ben Lyman]

Thanks PRR, I added the all important dot  that free schematic.com thing is fussier than an overdrive 
Anyway, I think you are right about R10, it goes at the light of dawn when I can crank it up a bit louder than tonight.
For tonight I lowered it to 1k just to see what happened and I got more volume and no nastiness. Maybe tomorrow I will also give R13 a much needed appendectomy.

Jack nailed it on the Q1b bias, no more R8 and R3 is now 680k, perfect bias and all is well.
Here's the latest schematic, maybe the final, pending tomorrows experiments:

[samhay]

What is R23 doing?

[Ben Lyman]

What is R23 doing?

Good catch, thanks, I might be able to pull that one. The original GR shows no resistor right there but I put it there to hold my Q4c voltage at a perfect 8.8v like it says in the build doc. I'll check it out later, thanks!

[ashcat_lt]

Um...  You need R13.  It's part of the voltage divider biasing Q2.  Get rid of it and the base pulls up to 9V through R12.  That won't sound good.

R23 helps set both gain and bias of that stage.  Removing it might not change much, but...

Is R22 a trimmer?  Must be, right?

[Ben Lyman]

Um...  You need R13.  It's part of the voltage divider biasing Q2.  Get rid of it and the base pulls up to 9V through R12.  That won't sound good.

R23 helps set both gain and bias of that stage.  Removing it might not change much, but...

Is R22 a trimmer?  Must be, right?

By "appendectomy" I just meant removing the (un-seen) 10k from the 150k that is making R13 into a 160k. It will probably change it a little but I think PRR is right in that it won't be enough change to worry about and it will eliminate one part.
As for R23, ya the voltage in the doc says 8.8v right there with no resistor at all. I had about about 9v with no resistor so I put the 1k and got 8.8v so I left it that way.
R22 is not a trimmer but it could be if that's how someone wanted it. I call it the "Octave" knob and plan to put it on my pedal. It goes across the two diodes' cathodes and the wiper goes to Q4b, then I can blend the diodes, which are a mis-matched pair, one 1n4148 Si and one 1n34A Ge. In my vid I tried to show the effect of blending them, far right (clockwise) gets the Si at Q3c and more pronounced octave, turning it back far left (CCW) gets the Ge at Q3e for less octave. If you use matched diodes, the most pronounced octave is center of the dial and CW or CCW just blends the octave with an out-of-phase tone or in-phase tone.