Harmonic Percolator - make one!

[pinkjimiphoton]

"it's cwazy... the whole thing is cwazy... but it's not as good as the ORIGINAL Libral Komrad...that thing was cwazy..."

dude... that should go down in the anals of diy history as the best parody , like, evah...


sadly, the original liberal komrade gave it's life in the studio fire to the fuzz gods, never to be seen again. snif...

[Morocotopo]

Guys, one question:

My percolator sounds niiiice, octave in the 5th string, all is well.

But....

It´s by far the noisiest OD/dist/fuzz I have. Makes hum, buzz, hiss, all sorts of noises. Not unusable, but would be nice to have it behave more like it´s pedalboard mates. When I compare it to a Guvnor or a Zendrive, it makes them look like they are studio quiet!

What´s your experience on that subject?

[pinkjimiphoton]

they are noisy as @#$%, power supply filter addition helps, but not much. keep your leads wicked short, use shielded cable to/from the switch, and make a sacrifixe to the mexican radio gods and hope for the best.

[inductor]

hey, welcome, inductor!

cool, yah, i could see that working. i didn't try going bigger than 1m on it cuz originally i just wanted to do what tim escobedo did... use common values. 1 meg is close enough for rocknroll with 750k, but the pot thing is hip, you can dial in the output and response of the pedal, make it a little louder if need be, or give it more sustain. too high and it doesn't sing, but get it in the right range and it dilates and swells... it's hip.

~1.4Meg was the sweet spot when I was tweaking my build. It only adds an extra resistor, so imo it's no big deal. I think I posted my schematic in this thread awhile back...

Also try out the Focus "damn stinkin' percolator" mod with a 1n5 cap to ground from either the collector or the wiper. The 1n5 cap was in the Pussy Fuzz that Mr. Bill made for Steve.

[inductor]

It´s by far the noisiest OD/dist/fuzz I have. Makes hum, buzz, hiss, all sorts of noises. Not unusable, but would be nice to have it behave more like it´s pedalboard mates. When I compare it to a Guvnor or a Zendrive, it makes them look like they are studio quiet!

What´s your experience on that subject?

Every time I've built a HP1 and it's unusably noisy its because there are grounding issues or interference from long unshielded leads.

[Digital Larry]

This circuit is fascinating.  While some people have described it as a cascode configuration I do not agree.  I tried to figure out the DC bias conditions of the transistors and as far as I can tell they are biased "OFF" - barring leakage. 

Now the interesting thing is what happens when signal comes in.  But I haven't had the time to figure it out.  Steve Albini's video explanation (cough) of it putting out 2nd harmonic, or, err.. even harmonics, or... uhhh.... whatever leads me to believe it does some type of full wave rectification.  Then some other posters indicated it occasionally puts out an octave down and I am at a total loss for how that would be possible.

Anyone?

[pinkjimiphoton]

i think it's soft clipping the signal with the transistors, and using some kinda SCARY PHASEING MAGIC i totally DON'T UNDERSTAND to make it do all this!

seriously, i think something about the circuit seems to  cancel dissonant harmonics.

i can get octaves out of it now...easily. up. down. both. it's insane.  no idea how/why it works. also discovered adding power supply filtering seems to affect it adversely!!

the latest is the best sounding one... if you make that 91k resistor a 47k from the collector of the pnp to the top lug of a 50k pot, and tie the wiper of that to a 47r resistor, you will get ALL the shit in albini's video. seriously.

it's bizzarre. you can make feedback come out in harmonic chords some times.

it's crazy man, like really crazy how ccrazy this crazy thing really is..

once my girl gets up, i will try and shoot some video. that 91k is the main key to percolation. the other is the 1m feedback resistor on the npn.

[Digital Larry]

OK I wait with (fish) baited breath for the video 

Other questions:

How does it handle a sustained note?
Volume knob manipulations?

I kinda want to know if it "cleans up" at lower levels or sputters, or... recites the Prime Directive?   

[garcho]

I thought it was basically some type of push-pull amp. Does that point you in the right direction, or did you figure that out already?  

Everyone's gotta hate on Albini. Did he ever say, "Hey DIY community, I'm a professional electrical engineer and I'm going to give you a dissertation on my definitive explanation of a goddamn distortion pedal and do all your homework for you because I'm a genius"? Compared to every awful pedal demo online, including EHX, Gearputz, Harmony Central et al., how could this one draw such ire? Because he has a sense of humor as opposed to the classic rock obsessed virtuoso worshiping bedroom shredders who take themselves and their misinformed explanations too seriously? I know the fanboys get sickening, but now y'all know how I feel about Clapton, Wooten, Colaiuta, etc. Rant over.


EDIT: totally stupid of me to rant, sorry y'all. guess i had a computer chip on my shoulder.

[pinkjimiphoton]

go, garcho... sorry gary, petey's video has still got me chuckling, i see a lot of guys i know in that..

yes, they sustain ok. it's not like a big muff tho.

volume changes depend on the transistors and biasing. lower gain pnp's seem to clean up a bit more, but never get super clean.

it's hard to describe. gineen's up, so once she gets up and about, i'll make a little noise, but jesus, ya may as well build one, pretty much any pnp and npn combo will work, and the values aren't all that critical in the couple passives in the circuit. wide open door for experimentation.

[garcho]

I was referring to the over all vibe, not Petey's video or anything specifically, just don't understand the 'mentality' around the gear-web. No need to be sorry!

Larry, take Mark Hammer's advice: build one! Even if you get the mojo parts from Small Bear (I did, and I'm happy I did) this puppy won't set you back much, so no point wondering. I have PCB etch files if you want.

[tca]

I thought it was basically some type of push-pull amp. Does that point you in the right direction, or did you figure that out already?  

Some answers here: http://www.diystompboxes.com/smfforum/index.php?topic=99785.0

Cheers.

[Digital Larry]

I thought it was basically some type of push-pull amp. Does that point you in the right direction, or did you figure that out already?  

Well, it kinda LOOKS like a push pull amp, but those attach the load to the output where the two emitters join together and don't have resistors in the collector circuits.  This has a big capacitor where the emitters join and has resistors in both collector circuits.  Also a push pull amp would drive the bases of both transistors together.  But the top transistor is being driven from the lower one's collector signal!

I've been doing some LTSPICE simulations on it.  I'm not sure I've gotten to the point I want to embarrass myself trying to explain it in detail.  It looks like it creates two pulses per cycle of the input signal.  The pulses are both in the second part of the cycle.  The relative position of the pulses is determined by the level of the signal.  WELL THAT'S PRETTY INTERESTING!  It's almost like a synth oscillator doing amplitude controlled PWM.  I tested it with a square wave input which helps show some of the time-delay pulse shaping behavior.

OK now I tested it with a sine wave and what it looks like it does more than anything else is to create a big pulse when the input signal does a negative going zero crossing.  But it's not doing this by overdriving a typical gain stage.

I'll grant you there are limitations to this simulation.  So I think I will build one and see what it's really doing.  Thanks for the idea.

This circuit is currently defying my ability to describe it as a collection of other well known building blocks.

[garcho]

This circuit is currently defying my ability to describe it as a collection of other well known building blocks.

I think that's partly why it's so popular here.

Take tca's advice and read that thread.
This is from veteran Gus.

[Digital Larry]

Hmmm.... LOL.  Not trying to be a snob, and I don't have a better explanation myself, but none of the given explanations are really jingling my chain at this moment.

One thing I noticed in tca's simulations is that you can see the nulls in the spectrum corresponding to the duty cycle of the pulse.  For example, we all know a square wave (duty cycle = 50%) is missing all of the harmonics multiples of 2 (which is 1/50%).  Here it looks like the duty cycle is 1/6th and so the harmonics which are multiples of 6 are at a minimum.

I don't agree that making the negative pulse bigger adds more 2nd harmonic.  It just makes the whole waveform bigger.  And the little peaks on the top edges of the negative going pulses are higher harmonics than 2nd.  The effect of clipping diodes would be to smooth off any sharp peaks more than anything else.  So compared to clipping a sine wave, which adds harmonics, I'll be so bold as to claim that in this case, the clipping diodes create both level compression and a possible reduction of higher harmonics due to smoothing.

[tca]

I've been doing some LTSPICE simulations on it.

The problem with simulating this circuit is that what defines it does not show by simply putting a 1kHz sine wave as input. This only shows the clipping behavior. There is more to it. The "novel" thing about this little circuit is the intermodulation distortion.

One simple method that can work using an oscilloscope and sims software is La Bel's method. For more detail see: Chapter 14. FIDELITY AND DISTORTION pg. 614

To see this in a quick way feed the circuit with two signals with frequencies f1 and f2 (f1<f2) sufficiently apart and with significant different amplitudes (4:1) and check the frequency response (amplitude spectrum). The output will have frequencies: f1+f2, f1-f2, and other linear combinations. That is why it is easy to get feedback with you guitar and so many "things" comes out of it (you don't need the output diodes for this).


I think Gus explanation is right and concise.

P.S. (edit)
Just added some refs and correct some statments.

[Digital Larry]

Oh god I am SO embarrassed.  I had a connection in the wrong place.  It's clearly an infrared remote control!  

The assumption on which sine wave analysis is deemed "accurate for other signals that can be described as summations of exponentially decaying sinusoids" is that the circuit is linear (this one clearly isn't) and "lumped and constant" (but let's forget about those for now - though it's probably lumpy at least).  But I can't help myself, I just like to see what happens in the simplest case.

Now you and Gus may be entirely correct about the sonic result being due to intermodulation.  But that covers a lot of possibilities and still doesn't explain how the specific circuit works.  I was going back to "when the input signal goes up, then the current goes this way for awhile, until the voltage is sufficient for Q2 to start conducting again, and then the resulting voltage spike though the capacitor turns on Q1 until such time that blah blah blah blah".  That's just trying to understand what voltages and currents are doing in the system without trying to reduce it to a simpler problem.  And I think Steve Albini's description of how it works is (sorry) hogwash, so to have him dancing around in stilts while delivering this doesn't help.  Although I am suitably impressed by his other activities.

Still mystified and impressed.  Thinking about creating a "random circuit generator" software because I'm guessing that there's maybe just ONE, but a good one, made out of 2 or 3 transistors and a handful of R's and C's that will... you know... catapult me to stardom and... z...zzzzzz..zzz.z...  what?  I'm posting on diystompboxes forum?  Oh the shame!  

P.S. muito obrigado para os links...
P.P.S. My "Radiotron Designer's Handbook" is a treasured heirloom given to me by my grandfather 

[tca]

Oh god I am SO embarrassed.  I had a connection in the wrong place.

That happens to all of us... that's part of the fun.

I've been playing with this La Bel's method and really works on a sim! (see link above) A thumbnail calculation gives an almost 100% percentage IM distortion for a full power chord. I don't have much time to put the results here but I encourage someone to try it out (it really deserves an entry on the sims board) and see it.

Cheers.

[Digital Larry]

If you know a lot about music theory (I don't) and you hear a cool piece of music that you like, some people might be tempted to deconstruct it in terms of harmony, arrangement, voice leading, rhythm, etc. so that if they wanted to reproduce some aspect of that specific effect, they'd be able to, knowing what went into the original.  Whether the composer of the original piece had those things in mind at the time cannot be assumed though.  For example, a lot of Todd Rundgren and Frank Zappa tunes have a certain harmonic "sound" that is partially the result of their typical chord voicings. 

So it is when I look at these circuits.  I have textbook and practical experience with designing electronic circuits.  When I see something like this I try to deconstruct it, partially to try to figure out what the designer had in mind when they developed the circuit.  And this circuit is so unique that I conclude "wow that guy was a genius" or "wow happy accident"!

I still think that the designer of the Tube Screamer was TRYING to make a diode clipper by putting back to back diodes in the feedback loop of an op amp circuit.  But because it was a non inverting topology, the diodes don't "clip" the signal.  Rather they create a piecewise linear amplifier where the gain goes from (really high) to one when the diode starts conducting.  Maybe it's a subtle difference sonically but electronically, it's a huge difference.  To me anyway!

[Digital Larry]

I've listened to a few of the demos and my first thought is that it sounds a lot like Neil Young "Rust Never Sleeps" or other Crazy Horse stuff where it sounds like his rig is constantly about to explode.  Anyone know if Neil uses one of these, or is his rig really about to explode?