Museum of Lost Effects:
Interfax Harmonic Percolator

Few guitar pedals can rival the cult cachet of the Harmonic Percolator, a singularly ugly distortion stompbox produced in minuscule numbers in the early ’70s by Interfax, a small company based in Milwaukee, Wisconsin. And guess what? They sound ugly too, though they’re ugly in a cool and useful way.

They don't get much rarer — or uglier.

They don’t get much rarer — or uglier. (This is a cosmetically faithful reproduction from Theremaniacs.)

I’ve been wanting to write about these for years, but was hindered by the fact that I don’t have access to one. No one does! Well, except the pedal’s best-known user, producer/guitarist Steve Albini. (Steve has posted several popular YouTube videos in which he sings the praises of the original and evaluates it against modern clones.)

But I revisited the idea recently when Christian Magee, who runs Tube Depot, sent me a couple of old 2N404A transistors from a stash he recently acquired. This rare PNP germanium transistor appeared in the original, along with an NPN 2N3565 (also rare, but not as ridiculously rare as the 2N404A). Yes—this pedal uses both a positive-ground germanium transistor and a negative-ground silicon transistor in the same circuit. (Another Fuzz Face/Tone Bender clone, this ain’t!)

I whipped up several variants:

      • a clone using the original parts


      • a near-clone using more readily available alternatives


      • a Harmonic Jerkulator, an all-silicon/no-diodes variation created by DIY stompbox titan Tim Escobedo


    • an experimental version with extra controls

Survey the wreckage:

Post-mortem after the jump.

I’ve always found this circuit compelling, and not just because of its weird architecture. It truly has a unique distortion character. Chuck Collins, who makes the cosmetically faithful clone pictured above, talks about the pedal in Tom Hughes’ Analog Man’s Guide to Vintage Effects:

But it was only a matter of time for someone to notice that this pedal does the same thing that tube-type amplifiers do; that is to create even-order harmonics when pushed into saturation or overdrive.

Man, I do not hear what Collins hears! To my ear, the distortion is much more akin to solid-state distortion, like overdriving a tape recorder input. Fortunately, most of us have matured beyond old-school distortion sensibilities (“Tube distortion good, all other distortion bad!”), and many players can put this razor-edged timbre to good use. I totally get why Albini digs it—it has the same uncompromisingly hard edge you hear in his productions. It’s a sound that won’t get lost in the mix! At the same time, it’s as responsive to input dynamics as any number of more conventional transistor distortion pedals.

Regarding parts: Using the exact original transistors matters little, IMHO, and I’d be leery of any product that uses their inclusion as the main selling point. Substituting other germanium/silicon pairs tended to sound at least as good as the original. Same with the diodes: Every diode type sounds a bit different, but there’s nothing particularly magical about the original germanium ones. I’d like to experiment more with silicon diodes and LEDs to see whether the add even more nasty edge.

This circuit is ripe for experimentation. I’m going to play with it more and see whether I can come up with an especially useful and exciting alternative — unless you beat me to it! :)

28 comments to Museum of Lost Effects:
Interfax Harmonic Percolator

  • the almost-clone sounds AWESOME!

  • Oinkus

    Pretty neat and right up your alley ! Fun stuff to play with and then terrorize wheeee! I think that LP looks and sounds great , good to see you playing it too. Where are the vacation pics ?

  • smgear

    and…. another one for the list. I’ve never heard of this one so thanks for the post and info!

  • Arnoud Philip

    Joe, you might like the marriage between the Harmonic Energizer and the Harmonic Jerculator I did over at freestomps:

  • Lefteris Sfak

    (First post here! )
    A great video as always!

    Percolator is my favorite fuzz. I have the Theremaniacs clone. I play single pickups only, and with both sliders halfway it is very dynamic, stacks greatly and cuts through the mix very well. At higher gain settings it can sound very synth-like as it is obvious in the video.

    Its only drawback is noise, an audible hiss, which is a problem when you play softly.

    Judging from the sound of the video, what comes out my speaker is closer to the almost-clone. I haven’t done any pedal building before but i’ve been tempted to try a clone as well, with various mods/and hmm… experiments!

  • Jermaine Eyum

    Awesome! I was checking out some of these Jerkulator discussions over at diystompboxes a few months back. Also saw the Albini videos.

    Few questions/comments:
    1) How do you expect us to take you seriously if you’re not walking around on stilts?

    2) Remind anyone else of Neil Young? Black Les Paul? Sound like running through barbed wire fence?

    3) Anyone have the slightest idea of “how it works”? This circuit could be a stumper for any aspiring electronics student. As far as I can tell it uses no known topologies for transistor circuits! Truly innovative!

    4) How’s it sound on barre chords? You played mostly riffs and lead type things.

    5) As always kudos for playing in a way that shows how the effect behaves, rather than demonstrating your awesome shredding prowess. I wish everyone else demoing effects on YouTube could pick up a hint.

    • joe

      LOL — work at a guitar magazine for a week or two, Jermaine, and you’ll never shred again! :)

      • Jermaine Eyum

        Ooops, forgot one question – does it work OK with an active guitar input (or otherwise buffered, e.g. running a compressor first)?

        • joe

          Hmm — I haven’t tried, ’cause I don’t have any active guitars. It doesn’t have the fatal feedback resistor that makes Fuzz Faces so awful downstream from buffers, but I’d expect it to sound a little brighter and harsher when not “hearing” a passive pickup. But then, so does everything else. :)

        • “does it work OK with an active guitar input (or otherwise buffered, e.g. running a compressor first)?”

          The guitar is heavily loaded at the input — input impedance is 51K (feedback resistor from the base to emitter of the PNP)/100K (input pot)/some loss from the Ge transistor being biased quite low on the emitter (. It gets very bright and harsh after a buffer. You can build it with a switch that puts a 10K or 22K in series with the input and a 1nF cap to ground to correct the issue and imitate the input loading when using a buffer or active guitar, but it might ALSO be unhappy with the increased input.

          • Jermaine Eyum

            Thanks for the info. I don’t really want to use it with an active guitar, it was just my sneaky way of getting an answer without giving away my swell idea (which will most likely take several years to realize).

  • Cool to see a side-by-side demo of a couple different versions of these to get a sense of what effect the changes have.

    I think you’ll like experimenting with the diodes, especially because you said that you liked it best backed off a bit. The transistor distortion in this one is, I think far more magical than anything the diodes are doing. I don’t even find that the use of the resistor to get asymmetrical clipping to be that important — most of the magic is in the interaction of the two transistors, and raising the Fv of the diodes lets you hear more of it without going as far as removing them like in the Jerkulator.

    You can hear a diode comparison in my demo here:

    p.s. Jermaine, I play plenty of chords there. This circuit does chords super well.

    Never mind that the higher Fv diodes are OA126 — silicon works just as well.

    My only complaint with this circuit has always been the noise. :(

    • joe

      Thanks, Jon. I agree that the diode portion is the circuit isn’t nearly as significant as the transistor portion. (As you point out, the Jerkulator doesn’t even have diodes.) To my ear, the diodes add a touch of compression and focus the lows, and the circuit sounds a bit better with them. But I didn’t even bother adding a diode on/off switch to the experimental version because the contrast just isn’t that interesting. I agree the circuit works pretty well with chords, though Jermaine is right — I didn’t play many in the demo. But there is a brief string of barre-chord triads at around 2:37 that’s quite representative.

      • joe

        And oh — this probably isn’t new info to some of you, but once again, this experiment reaffirms my belief that the part number of germanium transistors means next to nothing. Any two germanium transistors or equal hFE yield very similar results. Anyone who sells stompboxes based on the inclusion a specific model Ge transistor is probably full of it. There’s vastly more difference in sound between two Ge transistors of the same model and varying hFE (gain) value than there is between Ge transistors of different models but equal hFE.

        Furthermore, much of the conventional wisdom about optimal hFE levels in vintage fuzz designs is just … wrong. That’s not to say that the commonly recommended hFE values don’t sound great — just that values that lie outside the ranges deemed necessary/optimal often sound fantastic. I wish more people relied on their ears rather than their multimeters.

  • amen,joe…

    that mojo shit is just that… bullshit. differences in tone are pretty minimal, gain range is way more important.

    fwiw, in my experiments with this circuit (which led to the 4 knob juergulator) all ge sounds even better. on mine, i have a switch that chooses 2n404 or a simple 3906. gain is similar, but the ge is smoother.

    better to socket and use your ears. people ask me what hfe i use in things, i say socket and experiment.. best way to go.

    hi hiss in a perc is almost always from leakage… if anyone is having an issue, i’d try a different transistor.

    to my ears, i get more tone/distortion with LOW gain, leaky as hell ge’s than the higher gains usually specified. higher gains tend to distort more, probably cuz it’s overdriving the second transistor more, giving more of a thick saturated compressed distortion… but hear me out… this often will NOT reach unity. but for some reason, lower gain ones will. my theory is that by not overdriving the second stage into clipping as hard, the sound isn’t as compressed and allows the second stage to amplify more rather than being strictly for more distortion.

    ymmv… just my observations on this circuit. the juerg is interesting if ya gotta perc on the breadboard… the extra knobs can make a huge difference in the sound, depending on the other variables. with some things, they’re barely noticeable, with others they are ridiculously interactive.. depends on whether you’re fuzzing a clean amp or playing into some dirt..

    anyways.. onwards and upwards… peace out

  • On a very tangental note, the name of this pedal makes me think about how coffeemaking techniques have changed since the 70s. If it had been created in more recent times, would it have been dubbed the Harmonic French Press or the Harmonic Clover?

    • joe

      Hehe — I had EXACTLY the same thought. (Does a 20-year-old even know what a percolator is?)

      I wonder how this circuit compares to the Harmonic Melita of the the ’80s? Or those new-fangled Harmonic Nespresso Pods?

      • mwseniff

        How about a vacuum coffee maker? My grandparents had one that sounded like a rocket ship taking off that coffee pot was around till the late 70′s? On the other hand my other granddad used a percolator to make caboose coffee. The caboose coffee took two or three times long percing and also had some salt in it (nasty stuff even old burnt 7-11 coffee is more drinkable). Caboose coffee would kill most starbuck customers in a single sip. I liked your take on the jerkulator with no diodes and the extra controls and all the controls sounded pretty useful. That being said I have been randomly tweaking controls on my pedal board lately (within reasonable limits) and then adjusting my playing till I find a good sound. It’s amazing to me how much “touch” can pull a good useable tone out.

  • Shizmab Abaye

    Whether it generates more even order harmonics doesn’t affect the possibility of a “razor” sound. I interpret this as “increased higher harmonics” (whether even or odd). That depends on the sharpness of the corners of the waveform where it clips. So it could do both and still sound a lot different from tubes for just that reason. I don’t know whether it does or not. But I love the raw sound!

    • joe

      I just have to call out that “even order harmonics” thing because it gets misused too often. Yeah, you can measure it when you’re working with something pure, like a synthesized waveform. But a guitar’s sound is so complex that spectral analyses rarely reveal anything of the sort. Plus, the notion that even-order harmonics sound pleasant while odd-order ones sound nasty may be conventional wisdom, but it’s not especially wise. ;)

      • Shizmab Abaye

        It is just one of many factors. I have a very basic 5W class-A amp and I am not overly fond of its overdriven tone (like when using a booster at the input). To me it sounds “cluttered”.

        • I have a recent issue 5 watt/EL84 amp (Gibson GA-5), and I’m actually in the process of tweaking some of the component values to tame the nasty overdrive & make it a bit more pedal friendly. I suspect that these amps were designed with the expectation that people wanted raunchier tones from them.

          • joe

            Totally! And I’ve noticed the same thing about practice amps and such. Let us know how the mod goes — I’ve heard interesting things about that little amp. :)

      • Shizmab Abaye

        Not to cast aspersions on Mr. Albini, after all he’s well known for his musical and production exploits, while I’m not (sigh), but to hear him explain it, it seems like he’s not quite sure what it does and is just repeating what someone else told him.

        You can take any signal into an audio editor and clip one side at zero to get the effect he is talking about. And it does not sound that distorted, even harmonics and all.

  • Jay Quackenbush

    Definitely try the LEDs. I used green ones in my build of the madbean clone PCB and it’s a nasty and very usable sound.

  • Joe P.


    Which Percolator schematic did you go off of? The standard or the “Sardonic Albinator” one?

    • joe

      I haven’t seen the “Sardonic Albinator.” There are two versions of the original schematic floating around, with very slight differences. For my “origin” segment, I worked from those two, and when the disagreed, I just went with whichever option sounded best to me. For the mutated version, I changed a lot of stuff, and it doesn’t conform to any extant schematics. Hope that helps! :)

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

:1up: :alert: :ban: :beer: :borg: :coffee: :cuckoo: :cuss: :finger: :goombah: :stupid: :megaman: :mad: :pity: :noshake: :oogle: :pacman: :pill: :poison: :poop: :rant: :satansmoking: :shake: :shiftyeyes: :shroom: :sick: :smirk: :spammer: :stfu: :thumbdown: :thumbup: :turtle: :what: :whatever:

Click to upload a JPG