Debugging harmonic percolator

[tjdracz]

Hi there. Tried building Harmonic Percolator yesterday, using this schematics http://topopiccione.atspace.com/pjimages/HarmonicPercolator.sch.png adapted to veroboard: http://www.fredric.co.uk/misc/interfax%20harmonic%20percolator%20stripboard%20veroboard%20layout.gif I've replaced a number of values to conform to the MadBean pedals "Albini" ones: http://www.madbeanpedals.com/projects/PepperSpray/docs/PepperSpray.pdf (that includes additional 1n5 cap to the ground). And now, surprise, surprise, it doesn't work. In the bypass mode it's all fine, but when the effect is engaged, no dice. Tranny voltages seem to be all over the place, but since I'm starting out and got no idea what should be like, I need help of you pros.
Oh, one more thing, didn't have the original transistors so replaced Q1 with BC307 and flipped it around in relation to veroboard diagram to conform to its pin layout and Q2 is 2N5088.

Now for the voltages:
Q1:
C: 0.00 B: 0.17 E: 0.81
Q2:
C: 9.01 B: 1.25 E: 0.81

So, where it did all go wrong?

[pinkjimiphoton]

i may be wrong, bit it looks to me like your q1 may be in backwards. if your q1 was opposite, and a low gain leaky kinda ge, your voltages look great but backwards. at least in my humble opinion.

[garcho]

C: 0.00   C: 9.01

Those are way, way off.

Try this schematic:

[tjdracz]

.

i may be wrong, bit it looks to me like your q1 may be in backwards. if your q1 was opposite, and a low gain leaky kinda ge, your voltages look great but backwards. at least in my humble opinion.

It should not be. BC307 is PNP with CBE pin layout (AFAIK) and it's situated on the board accordingly.

C: 0.00   C: 9.01

Those are way, way off.

Try this schematic:

This is the same schematic that I've linked in my first post, only difference that I can see is a different value resistor to the ground. Veroboard was adapted from it and nothing was changed in position, only the values. Shouldn't Q2 be getting around 9 volts on the collector since it's connected to the 9v line?

[garcho]

Don't trust your instincts on this one, I guess. Electricity is complicated.

LucifersTrip:
Quote from: Mark Hammer on October 18, 2012, 08:12:02 PM


I used a 2N3565 (hfe around 150) and a 2SB33 (hfe 52-54)

The voltages, however, are as follows (battery = 9.7V):

Q1: E-1.58  B-1.49  C-1.39

Q2: E-1.58  B-2.12  C-3.24    



Finally breadboarded and my voltages are very close to Mark's

Q1 (GE 100 hfe / 50uA leak)
Q2 (SI 100 hfe)

9.69V supply

E        B       C
1.71,  1.60,  1.43
1.71,  2.39,  3.24

Those guys know stuff, your voltages should be like theirs if you want this to work properly.

[tjdracz]

Any idea of a good place to start looking for a fault then? Already replaced trannys just to be sure it's not them

[Gus]

If Q1 collector is at 0VDC I would look for a short from the collector to ground to start

The HP is not the best beginner build.  The way the two transistors bias make the circuit harder to set up
You need to select the transistors to get the HP sound.

You need to read all the threads at this site to understand what is going on with this circuit

EDIT check that both emitters are connected together  Q2 collector at 9VD and Q1 at 0VDC

[pinkjimiphoton]

it shouldn't be all THAT hard... despite the "mojo" parts, this circuit is pretty forgiving, really.
looking at the voltages, again, i can see the collector of q1 being at or near zero volts, depending on the transistor's gain there.

but that said, you should be getting the same voltage at each emitter pretty much.

this is also a totally different schematic from any of the others i've seen.

put your meter on the lowest dc voltage range, and check the collector voltage again on q1. wondering if it's actually reading just a fraction of a volt..

did you socket your transistors? gain can make remarkable differences.
in my builds, every single one, it's sounded better with gains of around 40-60 hfe in q1 and 120-350 or so in q2.

[duck_arse]

show us some photos of your work, it always helps.

check your connections for R2 and C3. there are some connection-dots missing on the posted diagram.

the bc307 pinout is the opposite to the 2n5088, (or should I say "the 2n5088 pinout is backwards to the bc307"?)

[Gus]

IMO the HP is not a good beginner project unless you understand how the resistors bias each stage and how it is a two gain stage in series, series powered powered circuit so there are interactions.  
So you measure and select your transistors and/or adjust resistor values OR change transistors until it sounds good

[Bret608]

Hey, I'm hoping to jump onto this thread as I've got a non-working "stock" specs Percolator built on Madbean's Pepper Spray board. Basically, I'm getting this grating hum/oscillation, and I can literally get the pitch to change by manipulating the balance and harmonics knobs, like some kind of synth. I can hear some slight change if I strum the strings hard. Board is getting a consistent 8.96 volts DC from my test rig's power connection. Continuity looks good across the board. And I've checked my resistor and cap values over several times, even going at it with a capacitance meter yesterday (to the extent that works while they're already on the board, anyway!).

I've successfully built an Albini-specs previously, so this is weird. The only part subs I have are a 150pf silver mica instead of 100pf and I've used a 2n1305 with 49 hfe and about 150 mA leakage for the PNP germanium. My 2n3565 is about 260 hfe.

Here are my voltages:

Q1 (2n1305)
E: 1.12v
B: 1.08v
C: 0.81 v

Q2 (2n3565)
E: 1.16v
B: 1.52v
C: 4.62v

I've posted my questions at Madbean's forum too, but I notice there seem to be lots of folks here with Percolator knowledge! Love your various versions, pinkjimiphoton!

Thanks if you can shed any light on this!

Cheers,

Bret

[pinkjimiphoton]

my best reccomendation is to build a juergulator, but i'm kinda biased.

i am wondering, did you by chance reverse any of the connections to the pots? i did and got a similar phenomenon just earlier today!!

[Bret608]

No, I didn't do that luckily! Brian's layout makes that a difficult mistake to make.

I listened to your Juergulator and checked out the layout at Tagboard. Believe me, it's got my interest peaked. Just running a little dry on parts, that's all!

Say, I forgot to mention, I also tried a 2n1376 in here with exactly the same result. I even double-checked my test rig is still working!    No worries, it is...

[pinkjimiphoton]

yah, bean rocks.

i couldn't believe it this morning... i wire it all up, fire it up... no sound. i reversed the sides of the vero. brain death!!!

i'll post vid of the new version of the juerg asap. i think you'll diggit. good luck with the debugging!!! worst part of the hobby sometimes, but often one of the most rewarding and educational!!

[duck_arse]

as we always say, bret, post some photos of your work. we love looking at stuff, and different eyes are often helpful.

[Bret608]

I appreciate it man, let me take one more crack at this thing this weekend and then I might have to go that route.

I work at a technical college, and I just had a really interesting debugging section at the side of one our electronics faculty who is also a guitarist. He was pretty fascinated by the circuit and wanted to hear my working version (Albini specs). He hit my non-working one with an oscilloscope and all other sorts of techniques. He's pretty sure the transistor I was using for the PNP germanium (2n1305, 49 or so hfe) is not liking the 20k-220k-750k combination of resistors for the biasing. I will try a high-gain 2n404 I have and see what that does. Then I will think about trying out some different resistors for R1, R2 and maybe R3.

[pinkjimiphoton]

check the juergulator bro!! seriously... you can put damn near anything in it and get it to work!

[duck_arse]

bret, I used a 2sb156 hfe ~43, leak 101uA, 22k/220k/680k (parallels with a 220k) and 22k + 50k trimm to V+. works fine. do you have any idea of the leakage of yr transistor? and my 2n3565 is hfe ~290.

[pinkjimiphoton]

i must concur; i found that if the gain of q1 is too high, for some reason, you LOSE volume.. and balls.

if the hfe were say 500 or so i suspect the perc won't even fire.

hfe of 30-60 is probably good. i went down to hfe of 42 on the juerg  and got above unity gain with it, trying higher gain transistors got me LESS output!

[duck_arse]

the weirdness never ends with this circuit