Differential Fuzz Design

[mdcmdcmdc]

LOL, that's what scope creep gets you I guess?
It does actually have this nice quality to it when you dime the distortion of almost sounding like power amp distortion; I'm loathe to start using a bunch of subjective nonsense words but yeah it gets DOOMY with that kind of wobbly quality to the low end as if your amp is about to blow up.

I'm still tinkering with the LPF filtering and attenuation between stages; it's quite challenging trying to keep it from sounding overly smoothed out without it starting to squeal uncontrollably or the string wiping having a kind of glassy/bell-like resonance to it.

I think it would be useful to clamp down the highs a bit more either right before or after the input buffer, but I'm still trying to figure out the best way to accomplish that.

[antonis]

I think it would be useful to clamp down the highs a bit more either right before or after the input buffer, but I'm still trying to figure out the best way to accomplish that.

Try to enlarge Miller caps..
(like C8, C12 & C16)

[mdcmdcmdc]

I had already been playing around with those miller cap values, so I'm glad to see I was on the right track (or at least *a* right track?). Bless you patient souls for taking this winding journey with me! 

Small revisions:
- increased miller cap values across the board
- removed resistor before DIST control
- adjusted LPFs between stages
- added a treble bleed cap to the DIST pot
- increased R11 and R18 for a bit less gain in those stages
- decreased R31 for a bit less output overall
- put a HPF (~175Hz) and mid notch (~666Hz) on an ON/OFF/ON switch at output for some added EQ control

[a bunch of mistakes on this schem, fixed it below]

[stonerbox]

So when's a sexy demo inbound? 

[mdcmdcmdc]

Ha, maybe soon - there's still some work to be done getting rid of some squealing when everything is maxed.
Also, I definitely wasn't thinking clearly about that switching situation; I need to either change it to a 3-position switch, or figure out a common resistor between the two circuits that I can make work jumper'd across the OFF position, or nix the HPF.

[mdcmdcmdc]

Some goofs on that last schematic so an updated one here. Got rid of the squealing, but still want to mess with the caps some more...

[ThermionicScott]

Pulled the breadboard for this off the shelf, ripped it up and started again...

Musical interlude first, schematic second...

I followed that track to YouTube and learned that was a young Edwyn Collins.  Mind blown!

[iainpunk]

you could swap out the expensive switch for a cheaper pot like in this schematic, which also makes the mid scoop depth controllable.



cheers

[mdcmdcmdc]

Some more small tweaks, a quick demo of where it's at:
https://imgur.com/a/43Q8Dqz

[iainpunk]

that sounds amazing! such a sexy circuit and such great tone!
does this already have the mid scoop circuit in it or not? it sounds quite 'rich' so i'd bet on that its not there yet.

cheers

[mdcmdcmdc]

TYSM!
That's the sound with no mid scoop. A couple of changes from the last posted schematic in the demo—figured out that putting a big ass 200K resistor in front of the distortion pot worked way better than trying to do smaller bits of attenuation between stages. I was able to pull the LPF after the first differential stage entirely and lower the miller cap values in a few places (also it dawned on me that Miller caps on BOTH of the Qs of the diff amps helps a lot). I think this is fairly accurate as of today:



There's some hum in there that's a bit frustrating but again it's at the point where I'm not sure what's just coming from having a high-ish gain circuit on an exposed breadboard and what's by dint of the circuit itself. The Beavis hum reducer fragment that's hanging out in the corner is on the breadboard currently but it doesn't seem to be doing much, might be helping a bit with the low end noise? I'm using a voodoo labs PP2 which should already be keeping the 9V fairly 'clean'...

[antonis]

I should implement below configuration.. 

[mdcmdcmdc]

So (for example), 82R resistor and 100uF cap for ~60Hz? I'm guessing that smaller resistor values are preferable to not mess with the voltage too much?

[antonis]

So (for example), 820R resistor and 100uF cap for ~60Hz? I'm guessing that smaller resistor values are preferable to not mess with the voltage too much?

Quite right...

Resistors values depend on affordable (added) voltage drop between stages..
(I shouldn't go for higher than 100R..)

P.S.
Filters orders are added resulting into steeper slope (i.e. 24dB/octave for the last one) so you can deal with their particular R/C values..

[mdcmdcmdc]

Makes sense - that 820R above was a typo, it should've been 82R (82Ω * .0001F = .0082).

[mdcmdcmdc]

Tried that out, no real change - i'm pretty sure it's hum from the circuit/breadboard/cables/computer/lights/fan/life. Hopefully just RF that'll take care of itself once it's happy in a box...

[antonis]

Can't say for sure 'cause present day/night is determinant for Third quarter to New Moon phase transition.. 

[mdcmdcmdc]

So I'm currently trying to wrap my head around Eagle CAD and turning this into a PCB... seems to be going ok, but I have a small mental stumbling block about where/how to connect things to the +9V plane on the board.

Given Antonis' suggestion of adding simple filtering between gain stages rather than a more complicated/Beavis style arrangement at the beginning, +9V comes in through D1. For the filtering to be effective does this need to be a linear path from end-to-beginning of the circuit, or are there other spots where I can tap into a 9V plane without mucking up the filtering? R2 R5 and R6 for example?

[antonis]

Bump..!!

Just to declare Michael's PCB works FINE..!!
Made some mods (altering HP/LP filters corner frequecies, deleting some of them and interposing a 5 band EQ in lots - the last proved pain in the @ss  ) and resulted into a very pizazzy effect..

Thax again for the PCB, Michael..!!
(I should never try to build it on strip/perf by my own..) 

[mdcmdcmdc]

That's so great to hear! I've kept on tinkering with it a bit as well and am quite happy with where it's landed:



Just some general notes:

- I adjusted the mid-notch section a bit; it's now cutting ~10dB @510Hz and seems a little more 'musical' perhaps. Also, less pronounced volume drop when you switch it on. 
- I was trying out some different clipping diodes and eventually realised that the signal was so distorted by the time it got to the clipping LEDs that they honestly had no audible effect. So, I omitted them altogether.
- C3 and C24 are fussy.
- I tried Antonis' suggestion of a dual-gang pot for gain; the effect was quite subtle, but that may've just been mis-implementation on my end. I went back to the single pot but reduced the limiting resistor (R8) and adjusted R10 to suit.

Once again, thank you so much to everyone who helped out with this. It was a very satisfying pandemic project.

I'm going with 'hexenszene' after an Unwound song that I happened to be listening to the other day.