photonic juergulator revisited

[pinkjimiphoton]

built this earlier, as i'd given the original away a ways back...


one ge sounded incredible, better than all the rest. imagine my surprise when i had mis-sorted, and the "pnp germanium" ended up being an NPN germanium...

ac176 to the rescue, yo.

changes from old juerg

harmonics pot now 1 meg, raised from 100k.  input lug is wiper (pin 2) rather than pin 3 as before. big difference.

ground unused lug of perc pot. (lug 3)

output volume pot raised to 100k rather than 50k.

power supply 9.79vdc

q1 ac 176 npn
hfe 95
uf 119mv

c  2.01v
b 1.94v
e 8.42v

q2 bc550 npn
hfe 415
uf 699mv

c 9.05v
b 8.50v (drops from there all the way to zero as the 47u cap discharges.. that cap is the key thing in this )
e 8.48v

stupid pedal trick imminent of course...



i think i got this right... will try and draw a proper schematic tomorrow.

works much better than the previous version... go figure.

the layout on tagboardeffects will work.... just swap out the two pots and add the ge npn as q1.

i damn near went crazy trying transistors in this... that cap make shit nuts!! if i pulled the pnp, i got a much better signal... at first... clear and loud and harmonic... til the cap began to discharge, at which point it would de-percolate.

people with issues trying to get this circuit up, power down between transistor swaps, and short that cap or it will make you @#$%ing CRAZY.

FWIW, every 176 i tried worked. none of the pnp's i had, ge or si sounded as good. <shrugs>

a REAL high gain darlington may be worth trying in q2. q1 seems to wanna see around 100 if npn, and 50 or so if pnp.

this is the weirdest circuit on the planet. <smh>

[blackieNYC]

Took another look at the old juerg S.P.T. just the other day.  It's been on my list.  Love to see the new schemo and demo.
I get the feeling this is your favorite, no?

[pinkjimiphoton]

not my fav, (i likes fuzz faces) but this thing is pretty cool....

for the sake of keeping it all in one post, here's all the changes... the vero on tagboard will still work with almost zero modification.

hi guys,
i gave this to someone a year or so ago, and was bored today, so yep, built a NEW one.
a tagboard effects dot blogspot dot com exclusive...
ok, first, a couple subtle changes to the circuit you MAY wanna try...
first, make the input the WIPER - pin 2 - of the harmonics pot. leave harmonics pin one where it is, and put pin 3 to where pin 2 used to be
up top.
this will make a subtle difference in the way the pot responds, kinda like the old les paul wiring, it shouldn't matter but it does,
next, make that harmonic pot 1 MEG instead of 100k. the combination of the different wiring and the larger pot make a significant difference.
next up, the first connection on the upper left has to go to perc pot 2. perc pot 1 will stay on row 3 hole one.
take perc pot pin 3 and ground it. that will make that and the tweak (actually i was gonna call it focus) pot work better.

ok, now transitors.
this is the photonic juergulator mk 2 here my friends...

make 'em both npn. make q1 germanium, make q2 silicon.

will work on the board as already populated and @#$% lot easier to get it going...
i was plugging pnp's in forever, and found one that sounded REALLY good. i was amazed, the circuit came to life.

next change... output pot, take it up to 100k instead of 50k. opens it up more.

in the end, i used the following for transistors with ridiculously good results...

turns out i had mis-sorted some transistors, and what i thought was an ac125 was actually an ac176..... an npn, not a pnp.

so
q1 :
ge ac176 npn
hfe: 95
uf: 119mv
c 2.01
b 1.94
e 8.42

q2
si bc550 npn
hfe 415
uf 699mv
c 9,05
b 8.50 and dropping when my meter hit it...it's all about the biggass cap and the tied transistors
e 8.48


voltage was 9.79

that's it... i know it seems weird, but every subsequent 176 i tried worked great.

something about that cap makes the whole thing weird. power down between transistor swaps, or it will make you crazy

sorry for any hassle caused, these changes are easily reversible if you prefer the original juergulator.

finally,

make thev 47n input cap 470n instead, now them knobs can percolate. it's all about that @#$%ing electro between the transistors... it holds a charge or something, i guess thats how the harmonics percolate.... if i was swapping transistors, it would still work.... for a minute or so... then fade out. makes it damnably hard to find good transistors, ya gotta short that cap to ground after power down if ya want any predictability i think., the bigger cap lets more bass in, and pretty much every npn ge i've tried around 100hfe, as well as every PNP around FIFTY hfe as long as it's ge seems to work.
make your head spin faster than linda blair at a pea soup cannery.
i'm still tinkering, will have a new schem drawn up soon with everything. really vast improvement, way less hit-or-miss AND can use the same vero miro cooked up.
ya gotta love that.

[pinkjimiphoton]

prototype built and working.

ALL the knobs work well now, way more fuzz..

i believe the whole percolation thing comes down to that 47u cap holding a charge of "harmonics" and "percolating" them... if ya pull the ge transistor out, it will keep working and sound the same.... for a while... til the cap discharges.

so... if building a juerg or perc or jerk, my advice to save much hassle, heartache and time, is power down between each transistor swap, and discharge that electro. or expect ridiculous results.

turn knobs... watch nothing happen... then suddenly change when the cap starts discharging. can be VERY frustrating!!!!! if ya power down between swaps and discharge it tho, every time ya power up it will work right. if the cap isn't discharged? fuhgeddaboudit!!

[pinkjimiphoton]

thanks to jalmon salmon at fsb for the schem...

HEY BRO!

here's the original juerg schem



here's mk 2



thank u!
hope ya enjoy the fuzz

[GGBB]

The Mk2 schematic appears to have the perc pot backwards and doesn't tie pin3 to ground.

Very little change from the original - layout wise - so I'll be able to update the PCB I did pretty easily - stand by.

[pinkjimiphoton]

sdrawkcab? figures!
lol

here's the stupid pedal trick

[GGBB]

Here's the layout and PCB for Mk2.  I added on-board status led and resistor as well as some power supply filtering. I also added some optional power protection - the 1N4739 (over voltage protection) is optional or could be replaced by a 1N4001 (for reverse polarity protection). The 1N5718 (reverse polarity protection) is also optional - either jumper it or move the 47R resistor to that position using the inside pad. I also left the Perc pot lug 3 unconnected - you can connect it to ground or to lug 2 with a jumper in the neighboring pads - your choice. EDIT: Technically you could also drop the 47R power input resistor (jumper it) if you are not using the 1N4739 but you are using the 1N5817.

Not yet verified, so USE AT YOUR OWN RISK.

[pinkjimiphoton]

schem looks spot on... i'll look at the pcb with fres eyes tomorrow... been a lonnnnnnnng day.
thanks gord!!!!!!!!

[amptramp]

Where does the current for the Q1 and Q2 emitters come from?  C4 would have to be one hell of a leaky cap.  Note that if you add a resistor in place of C4, you get something that looks like an astable multivibrator, so you cannot just replace it with a resistor.

[pinkjimiphoton]

beats the hell out of me, bro.

but it's the only thing on the board big enough to store any kinda electricity i could think of. it makes it maddening!!

like i said, you can pull q1 while it's running... and it will still work... for a while... until that cap discharges. do non-electros hold a stored charge?

trying different transistors was maddening also, until i figured out to power down and drain that cap.

i don't understand it, but that cap has something to do with it. i mean, ultimately, a bbd chip is just a whole bunch of caps doing something that appears to be similar.

no idea why or how it works like that, but believe me, it was doing what i wrote.

[GGBB]

Where does the current for the Q1 and Q2 emitters come from?  C4 would have to be one hell of a leaky cap.  Note that if you add a resistor in place of C4, you get something that looks like an astable multivibrator, so you cannot just replace it with a resistor.

I was wondering about that too, but lacked the experience and vocabulary to express it. I was thinking that perc lug 3 should be tied to Vin, not ground, but that would probably change the circuit a lot. That's why I left it unconnected by default on my layout.

[pinkjimiphoton]

you could probably do that, but i don't know if it would work or not.
my original idea was just to make the biasing resistors between b and c on both transistors tweakable.

to me, if something is connected to ground, you'll get a certain amount of load on it that can affect the audio. a pot or a variable resistor can sounds the same or completely different in different circuits.

maybe it's the antennae factor interfering with the radio waves under my tin foil hat or something. i dunno.

i DID try grounding it, and felt it lost some. ground the focus knob and it dies completely.

i'll have to try connecting it to b+ and see what happens when i go back downstairs to finish the overdrive i'm working on.

again, i'm 10,000 monkeys here.

[amptramp]

Hate to but in with another observation here, but there seems to be no collector supply for Q1.  The emitter is floating, the collector is grounded through resistance when it is NPN, so the only effect the device seems to have is to rectify the input to provide a meagre drive to the collector at some point in the input waveform, but with no emitter resistor, it does not appear to be an amplifier stage as we know it.

Note that your freehand sketch shows Q1 as PNP and the final drawing shows it as NPN.  My comments are for NPN.

[GGBB]

If there's interest - I could update the layout to allow jumper configuration of these things and add optional emitter and source resistors. I'll probably do that anyway.

[pinkjimiphoton]

Hate to but in with another observation here, but there seems to be no collector supply for Q1.  The emitter is floating, the collector is grounded through resistance when it is NPN, so the only effect the device seems to have is to rectify the input to provide a meagre drive to the collector at some point in the input waveform, but with no emitter resistor, it does not appear to be an amplifier stage as we know it.

Note that your freehand sketch shows Q1 as PNP and the final drawing shows it as NPN.  My comments are for NPN.

if ya read the thread ron, yeah, i state you can use whatever in the circuit. the original schematic shows a pnp. i found it made zero difference as long as the gain is right and it's ge.... around 100 for npn, around 50 for pnp.

the lack of the resistors you speak of is part of the percolator. i don't even understand some of the stuff ya wrote.

that said... the only diff between this and the percolator is i based this on escobedo's jerkulator but changed the  b/c resistors to variable ones. that's it.

so if ya gotta beef with the circuit, that's cool, but take it up with the interfax guy or tim e.

if ya get a chance bro, breadboard this thing. you'll discover there's more going on than meets the eye. it's WEIRD. and when stuff stops making sense ... like pulling q1 and it still sounding the same and working the same... until the one electro discharges and it fades away... hopefully you, knowing WAYYYYYYY more than me can explain it in a way my dumb ass can understand!! please!!? lol

cuz this fuggin' thing made me CRAZY for a couple days til i figured out how to get some repeatability outta it.

that said... max gain is with grounded emitters. in this case, the emitters are "floating" thru that big ass cap.
adding resistors to that is gonna LOWER the gain at best, if the "percolation" even works like that.

but for all intents, it's just a more tweakable jerkulator.

be curious to see the results of your tinkering bro if you're up to doing it and sharing the results.
i'd love to see what this thing does thru an o-scope.


gord... updating the layout is cool, but i don't think it's gonna improve the circuit any. but ya never know!

[pinkjimiphoton]

here's escobedo's jerkulator for comparison



and the percolator for comparison to that



the weird thing is how much easier it is to get a decent sound out of the juerg than the perc.

the jerk is EASY.

the juerg sounded best with the two npn's ultimately, so that's what i went with... and THAT was an accident.

all i can tell ya is what it sounds like, or shoot video. i can't explain this shit

[GGBB]

gord... updating the layout is cool, but i don't think it's gonna improve the circuit any. but ya never know!

I was really just thinking of it as a way to let people do there own experimenting.

Those other schematics are really helpful and in looking at them I am now wondering if when you switched the PNP to NPN did you also flip it around so that the Q1 collector connects to Q2 emitter, and Q1 emitter goes to the perc control? That would make sense to my pea brain at least.

[pinkjimiphoton]

i did try that. depending on how leaky the transistor was, some worked, some didn't.

i really don't understand how the hell this can even work.

[amptramp]

OK, with an NPN and a PNP, this is starting to make sense.  The average DC current through Q1 also goes through Q2 for an obvious reason - they are in series.  The input is the power transistor, a PNP stage which feeds the upper NPN stage.  C4 floats at a voltage that is set by the relative current flow through the two transistors - more bias (or leakage) in the direction that turns a transistor on pulls the emitter voltage towards its collector rail voltage, 9 volts for the NPN or 0 volts for the PNP.  Since the base bias is taken from the collector of the respective transistors, it tends to remain stable.  Interesting circuit!