the SWinGeR bOx!!! schem/verified layout

[pinkjimiphoton]

ok, here ya go, stupid pedal trick...

this is before the modifications to the e of q1 i'm gonna try. i have enough room on the vero to pull it off.

i really like the sound of this thing, especially with the les paul. very thick and creamy.

[LucifersTrip]

more diverse than expected...the eastern stuff (and tone) you played at the end is great

[pinkjimiphoton]

was a hell of a swan song. i tried modding it as gus reccomended.. no dice.
now it won't even run. arrrrrrrrgh.

tried it with a 1 k trimmer in parallel with a 22u cap. no dice. tried 100r.

the circuit is getting power, but i can't get it to bias... i can barely get audio thru it now. fudged.

gotta try and debug it now. the initial vero/schem however seem to work great.

this one may be a lost cause...i've found once vero gets heated up all bets are off. sucks.

[pinkjimiphoton]

fuzzes don't work so well when ya solder the center lug of the output volume directly to ground. <epic facepalm>

ok. it's working. used a 100r resistor and a 1k trimmer wired as a variable resistor to ground from e of q1. also bypassed the resistor with a 22u electro cap.

well, at 1k, the fuzz loses all it's balls. all the way up (just the 100r) it sounds almost the same as it did, so i leave it there. 100r is probably plenty.

the 22u makes no discernable difference i can hear. so it's in there, i'll leave it. sometimes you can only fix it so many times when it ain't broke.


all that said, battery voltage

battery 9.87v
q1
c  8.21
b  .58
e 0.01

q2
c 6.72
b  .58
e  0.0

put the wart on it? collectors shoot up to 8.70 or so on q1, tone goes out the window.

thinking i may try a 7808 in the circuit to clamp it to 8 volts, i think i could compromise there.

but the problem is then if the battery falls below about 9 volts, the reg will melt down.

my advice is screw it, stick to batteries. i hate to do that, cuz i'm totally into everything on one power supply.

any other ideas are welcome!!!

anyways... onwards n upwards..

[Mark Hammer]

Didn't listen to all of the demo, but the first bit sounded nice.  Has the grind of a Leslie to my ears.  Bet it would sound terrific with a flanger set for slow Leslie sounds, or some chorus pedals with a shorter delay range.

[pinkjimiphoton]

yah mark, it seems to sound a lot better with a bit of reverb. can't wait to try it on a gig this weekend.

if ya skip ahead to about 10 minutes in or so, you can see how it reacts with feedback...it's pretty sick!

i have a feeling it would be right up your alley.. 

[LucifersTrip]

put the wart on it? collectors shoot up to 8.70 or so on q1, tone goes out the window.

my advice is screw it, stick to batteries. i hate to do that, cuz i'm totally into everything on one power supply.

remember the 100 ohm resistor trick?  "bring the power in via a 100 ohm resistor" (PRR).  there's a ton of vintage circuits that don't sound right until that's done.

[pinkjimiphoton]

first thing the power hits on the vero i layed out is a 100r resistor dave. i have it labeled r00 on the layout.

[LucifersTrip]

sorry...didn't remember that...then, you're results are more surprising. is it possible that resistor needs to be larger? I would at least try multiple wall warts...

[pinkjimiphoton]

good point, but i'm figuring if it does it with one, it probably will with others.

i'll try it tho.. in or a dollar..

i'm wondering if it is something inadvertantly dumping some signal to ground.. maybe a defective power jack?
it didn't seem to do this when i had it out of the box.
but then, i didn't try a battery until then.

dave... stay tuned.. i gotta real motha fuzz bomb coming in the next day or two, me and my friend russ are working up
the audio designs fuzzound. i built a working npn ge version tonite, it's insane... like a superfuzz that cleans up like a fuzzface.

sick.

[Mark Hammer]

So I put one together on perf last night.  At least I think I did, because it is not currently working at all.  No sound.  Nothing.  Nada.  Garnicht.  Rien.  Could be a bad pot.  I'll see after work.

I have both 3391 and 3392 transistors on hand so I "went stock".  Q1 has an hfe around 275 and Q2 is around 175.  I used a 100nf, instead of 47nf on the output.  I also figured that it was pointless being able to turn the input sensitivity ALL the way down, so I split the 22k into a 12k on the input lug of the pot, and a 10k added on the ground side, such that it still provides the same resistance but covers a different portion of it.

On the output (more in a moment), I split the 1meg into a pair of 470k in series.  I stuck a 180pf to ground from their junction (little bit of LPF), and a 5817 and 4148 to ground at the junction of the 2nd resistor and output pot.

I'd love to be able to tell you how it sounds, but ça marche pas!  I was attentive to the pinout of the transistors, so I don't know what's up.

That 1M on the output is awful big.  In tandem with the 100k volume pot it must bring the level down something serious.  I'm wondering if it couldn't be reduced a bit, and maybe complemented by an increased in volume pot value.  So, make the resistor 100k and the pot 1M, or the resistor 510k and the pot 500k.

[pinkjimiphoton]

hmmmmm!!! sorry to hear it didn't work... did you breadboard it or vero it or?

i'm betting the key is bias, if you went stock. did you try it on a wart, or with battery?

did you use trimmer for the c to + on q1?

i can't really imagine the mods you make having any effect on this circuit whatsoever in terms of stability.

are you getting voltages? if so, what?

the other day i built another "new" fuzz. couldn't get it to work right. i was like... wtf...

turned out i had reversed the wiring between ring and tip on the input jack. (facepalm)

i really wanna help ya get this thing running, i think you'll love it once it's dialed in. i'm betting the original transistors
won't work well, i think you need fairly high gain to get it to work.

let's squash this bug, and hopefully we can get it running... and see if maybe you can help me figure out why i get such a huge tonal change between battery and wart. i tried resistance (and bypassed resistance) from e of q1 to ground as gus suggested, but no dice there...at least not noticeable.

one thing i did notice also was lower gain transistors barely fuzzed. try some 5089's or something in there, maybe even a darlington?

[pinkjimiphoton]

oh dear...
just found a mistake on the vero. needs track cuts at c3,d3, AND e3. didn't mark e3 on the vero... will repair and replace..that may be why it's not working mark!!!

sorry!!

edit:

[Mark Hammer]

Immune to vero errors, but perf is my kryptonite.

Immune to ring-tip reversals here, too.  It's a pair of two-lug mono jacks for testing.  Used a battery. No trimmers, just the 6.8k and 10k that Herb shows.  With a battery that reads 9,55V, I'm measuring about 8.5V on the collector.

[pinkjimiphoton]

like gus said, the biasing goes crazy on the transistor at very small changes.
my guess is that the voltage is too high for the circuit to run at, and perhaps the transistors are at cut off?

maybe splice in a pot until ya find the sweet spots? i think the original transistors may have been significantly lower gain, too.

sorry you're having all this hassle!!

in a way it's good, cuz i caught the bug in the first vero.

keep at it, you'll get it. any chance the pinout is backwards from what you're expecting? i know, dumb questions...

for the life of me, i can't ever get stuff to come out well on perf. it just don't stick!!
i always seem to melt plastic into it and it gets AFU.

[Mark Hammer]

No problems.  I whipped it up between 10:30 and 11:30 last night.  Serves me right for trying to make something just before bedtime!

The boardis pretty small - 5 holes by 17 holes - so there isn't much room for trimmers!

[pinkjimiphoton]

that's pretty freekin' small!! i await your findings eagerly! 

[Gus]

Jimi

I simmed a circuit.   Note the ratio of the supply voltage to the collector voltage.
I would make the bias control R3 if you use one.  


EDIT R6 was removed
I forgot to change the notes in the screenshot R6 was in series with C1
Also the emitter voltage changes with changes in power supply voltage

[pinkjimiphoton]

far out... at 9v it's exactly what i get, same at 10v. so... i'm a dunce. i gotta look at this tomorrow when not full of cactus squeezin's

[pinkjimiphoton]

flash forward years.... the original one of these went to my bud ed years ago, he never tried it til the pandemic thing... now he loves it. so much he asked for another as backup... so i did it all over again for him. i did make a couple changes. the input fuzz pot is now just a vr instead of lug 1 being grounded. i ditched the 22k series resistor with the input.
ended up adding a weird way to vary the input capacitance on mine, so its not on the vid attached, which is the resurrected circuit minus the two changes mentioned earlier. sounds pretty much exactly like the original still. vid is before spending a bit of time dialing in the trimmers. both 5089's were within about 20 points of each other, somewhere around 520hFE. higher gain in q1 like the original.

trimmers are a must. didn't measure the voltage, i always just bias by ear. full tilt almost seems to be the sweet spot. i set both trimmers half way, then bring up q1 til it sounds good and q2 til its loud. that's it.

figured i'd re-add the graphics for posterity sans the botosucket crap.

if ya build it, enjoy.  if not... fine by me
if ya use the components i reccomend and that layout, it will work. i've built a few of these (the old ones, not the new variant)