Harmonic Percolator - make one!

[pinkjimiphoton]

all germanium percolator... the germonic percolator.

2n2273 ge pnp, hfe 87, and a ac176 ge npn, hfe 100 ish (would change of course depending on temperature.

i'se sick...so bare with me. but you may like it. it's pretty fat when it's loud.

universal old school little less than unity fuzz.  does all the hp tricks, but a little "warmer" for such an obscene thing.

[garcho]

@Mustachio

Or should I go with silicon/germanium.

from Small Bear, regarding their 2N404a:
Tested in the George Giblet circuit, gains will run between 55-200 Hfe, leakage under 300 microamps, and low noise floor.

I've only built the Harmonic Percolator straight off the Giblet schematic (on page 3 of this thread), but it worked like a charm, no fiddling necessary, at least, to my ears. I used 2N404a (165 Hfe), 2N3565 (300hfe), and [EDIT] 1N695 for the semiconductors. Also, did ye olde 100Ohm series resistor and cap PS filter, with reverse polarity protection, and ended up with about 12V from my wall wart. Haven't measured much because it sounded fantastic right off the bat. Regarding the style of distortion, Albini said it perfectly: "More of the good, less of the crap".


@Jimi

drink some OJ, eat some soup, and get some sleep dammit!

[pinkjimiphoton]

lol.... oj only with tequila

soup til i'm drowning

what, it don't look like i just fell outta bed? 

i made this stupid video, then decided it needed more balls than it needed that nice tone.
went back to whatever npn i'd used originally.

beauty!

[Jazznoise]

Hello fellas, so I finally built mine but the voltages seem funny.

Q2 and Q1's Rc are both 100K. Q2 = 2N3904, Q1 is a fairly low hfe AC128 (about 30-40, repeated measuring drifts considerably). Power supply is a One Spot at 9.5V.

Q2
E 2.55
B 2
C 2.1

Q1
E 0.17
B 0.25
C 2.1

Is this just because R4 is too large? I haven't plugged it in simply because I can't see a base voltage that low giving me anything. Hoping I don't have to pay for 2n404's, as I can get AC128's for free within small quantities.

[pinkjimiphoton]

q2's e seems kinda high to me... gonna wait for someone with some intelligence to speak tho.

cuz...i ain't it!!

[Jazznoise]

Your modesty is as compelling as always, Jimi!   

I'm willing to speculate on the Emitter resistor on Q1 being too high or maybe the 220K resistor for Q1's Rbe being too small, but I aint sure!

[pinkjimiphoton]

 

what i do lately if i can't figure it out is replace the suspect resistors with pots... or in some cases, use roach clips on wires to put pots in parallel with the suspect resistor,
and twist the pot til it works.

measure the resistance, go with the next standard value that's close.

or, if ya do the trimmer thing, say ya gotta 100k resistor.. if you parallel that with a bigger pot, and use this parallel resistance calculator gizmotron to see what your load is, you can also dial it in. i used it on my toneblaster...100k resistor with 100k trim in parallel.. can dial it in to whatever ya need.

i don't think i'm being clear, sorry, i need coffee bad!!

here's the link...saves my arse daily!!

http://www.sengpielaudio.com/calculator-paralresist.htm

[Jazznoise]

Bumping this as I've still had no luck in getting the thing working. Figured out my 2N3904 is reversed and now my voltages look WORSE!  I used a 100K for both R3 and R4 and I seem to drop about 8 V just over that 100k. I get that that gives me about 0.08mA of current flow, so should I be trying to lower R3 to allow more current flow? If I parallel a 100K the emitter on Q2 pulls up to about 1.3 volts, about 2 volts with a pair of them. But Q1 still has a huge voltage drop between Vc and Vb

My transistors are the same. Here are the new voltages:

Q2 =  2n3904
E 0.9V
B 0.3V
C 0.3V

Q1= AC128
E 0.18
B 0.3
C 0.3

I've swapped the AC128 for everything, even a 2n3906 and one thing I've discovered is my feedback resistor R2 is showing up as about 6K resistance. Am I just an idiot and is this due Rbe is so small or could this be the cause of the transistor being pulled so low?

[duck_arse]

jazznoise, how does yours sound?

I breadboarded to the giblet circuit, and couldn't find any transistors that got near the posted voltages. pulled it apart, and then rebuilt it with the geminiIII values, with both Ge. then pulled out the top Ge and put in the oldest, crapest 2n3565 I own. still no voltages right, the top trimmer only seems to make it slightly louder, and the collector bias resistor at 150k makes for shorter sustain, so I'm putting a header there with the 680k.

now I'm going to solder it, and be done with it.

[Jazznoise]

I'll put it this way, it goes a great rendition of 4' 33''

[duck_arse]

..... and no-one likes modern music ....

so you get no output at all?

looking at yr voltages, you should have the same on BOTH emitters, whatever it is, because they are connected.

they are connected, aren't they?

[Jazznoise]

Actually I'm really into some of that stuff. Cage's Piano Sonatas are pretty cool, and I like alot of the Minimalist stuff (though less the phase based music). Peices like I'm Sitting In A Room even are very engaging, once you stop thinking in terms of just harmonic movement.

But this is wildly off topic.

No output. I can touch the Emitter of Q2 and get it to "brrr", but it's actually quieter than when I do it with no power. Weird, eh? The AC path to ground through the circuit is easier than going to my amp.

I'm pretty sure it's to do with the nominal current flowing through the transistor stages and the series feedback on Q1 dragging it down. Not sure what to do about it, though.

[pinkjimiphoton]

those voltages are way off.
i'll crack mine open when i get a chance and give ya the voltages i have in mine.

looks to me like you're not getting any juice to the circuit.  if memory serves, should be around 5v at the collector.

check your caps, first thing i can think of. somewhere you have an open circuit. is this on a breadboard, or a circuit board jazz?

[Jazznoise]

I've checked and double checked for shorts, I'd say if I run the scribe again I might snap the thing.   Once again I can pull the voltage up by lowering the resistor from Vs to Q1, but the Ge transistor just never seems to pull up. That said, I never tried to get it upto 5! Might parallel the 100k with a 10k or something and see if I can't ballpark 5v. Though that might cause other problems...

That'd be great, Jimi! If you could give me your values for R3 and R4 that'd be great, too. There's some trick to biasing this that my brain just doesn't get.

It's soldered onto veroboard. My caps should all be fine. It's a 47uF Axial Electrolytic for the ground cap for Q1 and Q2, ceramics for C1+C3 and a pair of white (Possibly mylar, little square plastic things) caps for C2 and C6. I've no cap meter, but the values all translate correctly from Cap Code jibberish.

[garcho]

the only thing between both emitters and ground is a single 47uF cap, so they can NOT be at different voltages. unless, of course, something isn't soldered in correctly.    Did you make your own layout?

[Jazznoise]

Why must I always confuse the emitter and collector? Seriously considering getting it tatoo'd somewhere on my forearm..

The 0.3V refers to the emitters. I did my own layout as I didn't have the correct resistor value for a 750K so I had to series a 470K with a 300K

UPDATE: Did a little more nosying and found tha R4 was not the value I'd put in. So I scribed the life out of it and my new voltages are:


Q1

E 1.28
B 1.27
C 1.28

Q2

E 1.28v
B 2.71
C 3.28

There are no shorts on Q1, I double checked.

[pinkjimiphoton]

if it's on veroboard, first thing i suspect is continuity. check every trace, every connection for continuity.

dumb question, but i've done this too many times to count... do you have the board ground hooked up to the starground?
no chassis ground can make all kinds of weird shit happen.

gonna go check mine right now...take a quick walk with me, will ya?


ok...

battery voltage 9.40

q1 (mp16b, russian germ hfe 115 leakage 6ma)
c   5.62
b   5.60
e   3.96

q2  (2n5089) (hfe 626)
c   5.62
b   5.89
e   6.21

but wait, i gotta problem here...

the voltages change depending on the settings of the damn knobs!!
what's above was what i found when i first fired it up.

but looking at it, the 5089 is in backwards. like, flipped 180 degrees from how it SHOULD be, but it SOUNDS BETTER that way!!

wtf.. dude, i gotta play with this some...lol

with the trannys in the PROPER way, it has way less balls...but here's the readings:

q1
c  2.63
b  2.56
e  .69

q2
c  2.63
b  3.15
e  5.76

looks like q2 is backwards, doesn't it? wtf? but my meter says the pinout's right...

here's the readings with q2 wrong:

q1
c   3.74
b   3.62
e   1.01

q2

c   3.74
b   4.16
e   4.36

man..this thing has me thrown for a loop now!! wtf!!

gonna put q2 in backwards, it has way more fuzz and volume.

something weird!!!

[Jazznoise]

Wow, Jimi! This is very interesting. If what you say is true than the input and output loading makes a big difference. I'll try and test with amp and guitar in future - though I have both pots installed. No shorts, just went line by line again with the DMM. The only think I'm thinking at this stage is some of the caps having a really low ESR - but we're talking values of 200K+. Still, if I turn my knob to 0 I get an audible "bump" type noise. It's not quite DC crackle, but it's weird.

Jacks and PSU are hooked to the same grounding, no chassey ground as of yet. Bit of a catch 22 to go drilling out a box that may never work!

[garcho]

Post your layout? Might be something a fresh set of eyeballs could notice.

[Jazznoise]

First thing tomorow night. It must be something weird I didn't spot. Anyway I've college, band practice and Pancake tuesday to tackle in about 7 hours so I'd better get my rest!

Peace!