Oh man — my friends at Premier Guitar just posted a video of me demoing four of my new pedal prototypes at NAMM. I didn’t even think I’d have these ready by showtime, but I powered out at the last minute. (Maybe ’cause I needed something to take my mind of the inauguration.)
It’s been a busy month since we filmed this. We’ve settled on names, graphics, and specs, and everything is in development. (Though they’re still a few months away from shipping.) I’ve included the first-draft enclosure graphics as well.
Purr is a minimal one-knob optical vibrato. Yep — one knob, which means no independent rate and depth controls. Sound crazy? I agree. But it just sort of works! (I explain my questionable reasoning in the video.)
It’s not a deep, wobbly pitch-shift effect — more like cross between a really warm, pretty tremolo and a subtle optical vibrato. It’ll have the same large knob as my Duh pedal, so you can make adjustments with your foot, assuming you’re not as clumsy as I am.
It’s my fave modulation circuit, one I’ve used on a number of my YouTube videos. Here it’s on throughout at a very subtle setting.
In this video it’s set more strongly, though I toggle it on and off as I loop additional layers.
Screech, a mutant spinoff of the Octavia and Green Ringer octave fuzzes, is an outgrowth of the experiments I did did a few years ago, when I built models of every major octave fuzz design, and then attempted a variation of my own. (It’s not too far removed from the final pedal in this video, which appears at the 10-minute mark.)
It’s got the most extreme octave effect I’ve ever heard from an analog octave fuzz, and unlike on an Octavia or Green Ringer, the effect works in all neck positions and at all pickup settings. You can also bypass the octave portion of the circuit for a straight distortion sound. (That’s not a new idea—it’s a popular Octavia mod. But that non-octave Octavia sound is dull as dirt, whereas this, I think, has a bit more character and impact.)
Porkolator also springs from a video demo/experiment of a few years back. It’s my oddball spin on the Interfax Harmonic Percolator, which is already pretty odd to begin with. It uses the same weird combination of negative- and positive-ground transistors for that sort of gravelly, decidedly non-tube-like distortion that Steve Albini loves so much. But all the part values differ, and the gain stages work very differently. There’s also an independent boost stage that can generate tons of extra level if desired. Again, it’s not that far removed from the final example in my octave fuzz video. (It appears starting at 5:55.)
Unlike the other three pedals, Cult Germanium Channel is pretty much finished. (We were originally going to include it among the 2016 releases, but decided that three new products were enough.) Its heart is the same primitive germanium overdrive circuit as in my Cult pedal, but with lots of added doodads: a tone-shaping pre-gain control, great-sounding active 2-band tone control, and an output trim. You can read more about Cult Germanium Channel here.
The NAMM video also features a demo of Kitty Boy, my imaginary vision of a germanium fuzz that should have existed in the 1960s. It’s sort of a cross between a Maestro Fuzz Tone and a Tone Bender Mk. I, which can go from lightly overdriven “Satisfaction” tones to hyper-saturated Ziggy Stardust glory. (It’s inspired by a conversation with Lyle Workman, so thanks, Lyle!)
I hope folks dig these. I’ll keep you posted about release dates and final prices.
Thanks, Jason Shadrick and Perry Bean, for doing such a nice job with the video and squeezing me into a brutal production schedule at the last minute.
Jet and Kaiju say: “When Joe plugs in those horrible fuzz pedals, we hide in the closet. But that new Purr pedal with the black cat on it isn’t so bad.”
Firstly, it’s still strange to hear Joe talking in a video. (But not in a bad way, I must add.)
Secondly, now I want a Purr. Reminds me a little of the ‘harmonic’-style tremolo.
1. That one knob vibrato is awesome.
2. I still have to get a cult.
3. Love the version of Wichita Lineman
Thanks, guys. I know what you mean, Julian — it’s sort of like “tremolo with more harmonic action,” which connotes brownface trem. But actually, the mechanism works pretty differently. “Harmonic” trem has a phasing effect ’cause it modulates two frequency ranges independently. This is simply a 1-band wobble. 🙂
Joe, these sound absurdly amazing. Fucking amazing. Since devouring ever audio clip/review, I have been fantasizing about a full-Gore pedalboard. If there was such a thing, what would it be comprised (& what non-Gore pedals would you add)?
1) The Purr sounds sweet. Your unaccompanied version of “God ONly KNows” gets me a little farklempt. Simply gorgeous.
2) Like yourself, I’ve built clones of nearly all the analog octave-up units, and the Foxx is nearest and dearest to my heart. Putting another one together the other week, I subbed Schottky diodes for the germanium ones used to provide the octave. MUCH better octave sound over more of the finger board, for my money. And note that the clipping diode pair just before the tone control don’t actually need to be there to have fuzz. Quite the roar without them. The Scrambler really needed a better front end…so I stuck a sort of modded DOD250/Dist+ and a Scrambler inside a single box, using the 250/Dist+ clone to adjust the tonal quality and “push” of the Scrambler. The front end was tweaked for more bass. The range of tones achieved out of the Scrambler, by simply adjusting the sustain/compression of the clipped front end, and signal level fed to it, is astonishing, from normal-sounding grunt to distortion tones that implode on themselves. Recommended.
3) Finally, if you have anything that is a derivative of the basic Fuzz Face “engine”, do yourself a favour and play with the feedback path between the emitter of Q2 and base of Q1. Increasing the feedback resistance will yield all sorts of interesting glitchiness.
Hiya, Mark! Pardon the brief reply, ’cause I’m off on a little vacation. I just wanted to say, thanks for your posts. You writing taught me a lot when i was starting out, and I’m flattered to bits that such an important figure in the DIY community is hanging out here. 🙂
Yes — an extra boost stage does wonders in front of the Ringer/Scrambler circuit. I’ve never tried it with a DOD 250, but I wish I’d thought of it! I’ve mostly used simple little one-transistor JFET boosters.
I agree about Foxx Tone Machine. In fact it was the Prescription Electronics Experience pedal — a near Foxx clone — that got me into this shit in the first place 20 years ago. When I went to the UK to record with PJ Harvey, I took several of them as gifts. I tried one out at the first rehearsal. After a long silence, Flood said, “Well, it sounds as if Norman Greenbaum has arrived.” It wasn’t intended as a compliment.
If your curious: A couple of years ago I made video with side-by-side comparisons of a half-dozen fuzz circuits. (My soon-to-be Screech pedal is descended from the experimental final example. https://www.youtube.com/watch?v=pTrLGuARBYY
I have a JOYO copy of the Foxx Tone Machine which I believe Fulltone later ‘copied’ as the Ultimate Octave.
JOYO call it the Voodoo Octave. I checked through the circuitry on mine and found several electrolytic capacitors had been inserted the wrong way round. The PCB was pretty mucky – covered in flux, which I cleaned off. I then entered the circuit into LTSpice, which revealed all sorts of interesting things about how the various stages were biased, leading me to make a few mods. For example the phase splitter in the octave section of the Foxx isn’t perfectly balanced, which degrades the octave performance. I have found a number of interesting quirks in many of the classic effects through LTSpice simulation.
The Armstrong Green Ringer has better performance as an octave / full wave rectifier, but it needs a healthy gain stage in front of it.
A number of people have reviewed the JOYO on YouTube and pretty much dismissed it. At least one reviewer actually said it was because they didn’t understand fuzz or octave pedals. I think it can sound amazing running into a wah or an envelope filter.
Mark – thanks for the tip about Schottky diodes, I must try that out. My JOYO has a pair of authentic looking, large glass envelope, supposed germanium diodes, in it. However I have found a number of supposedly germanium diodes which I received in some Chinese effects kits recently have quite high forward voltages of 0.6 or 0.7 volts. All my NOS germanium diodes are around 0.35V.
Anybody ever used an op amp based full wave rectifier? Maybe with a high-pass and 2x gain on the end to deal with the DC offset and level. Seems way easier than these scrambler circuits if the goal is “accurate” full wave rectification. If that’s not the goal then I’ll just be quiet.
Now that I have a breadboard I may give it a shot.
https://www.indiabix.com/electronics-circuits/full-wave-rectifier-with-op-amps/
1) I’m blushing, Joe.
2) I did see that video with the octave-up units. One of these days, somebody has to make a “round robin” comparison unit so that a player can insert a bunch of pedals into individual send-return loops, and hit a momentary footswitch while they play to do instant comparison between different pedals in round-robin fashion. We are often such prisoners of the tricks our auditory memory plays on us!
3) Terry, there are so many ways that the two daughter signals from the phase splitter can be imbalanced and result in poorer octaving. I would imagine some of it has to do with using 5% tolerance resistors. Bernie Hutchins had a comment in the Electronotes series back in the 70’s regarding one of the two equal-value resistors needing to be just a tad lower than the other, and I’ll be damned if I can remember which one, or by how much. In the case of the Foxx unit, as well, one would want the two diodes to have very similar forward voltage. Again, the failure of the Joyo unit to live up to expectations may simply be a matter of part tolerances. Well, that and backwards caps!
4) Hi Larry! Fancy meeting you here! 🙂 My sense is that the audibility of an octave may require more than mere full-wave rectification. There are a lot of things that “count” as FWR, but the octave doesn’t necessarily jump out at you. That doesn’t mean an op-amp-based unit could not do the trick. I think it just requires appropriate parts-value selection so that the two rectified half-waves are of equal value, something which is not always done. I made an Elektor octave circuit back in the early 80’s that used an op-amp-based FWR and was a big disappointment. Of course I knew much less than I do now and paid no attention whatsoever to matching or selecting parts to optimize audible doubling.
5) Both analog up and down octave units get easily confused by harmonics, and unstable pitches. I think that’s one of the reasons why tracking is usually only well-behaved above the 7th fret. The shorter the string length, the stiffer the string and the less harmonic content. That points to use of heavier-gauge strings, and maybe even flatwounds, as the potential secret to more dependable tracking. Something to consider.
Mark – I would imagine that the original Foxx pedals used 5% or worse resistors but the JOYO pedals have 1% metal oxides.
In the case of the Foxx circuit making the collector load of the phase splitter 4.5K and leaving the emitter load as 4.7K results in balanced outputs at those points. However the output of the full wave rectifier is still far from balanced because the D.C. offsets of the two signals developed across the two 100K pull down resistors from the cathodes of the two diodes are not the same. This results in the two rectified half waves being different amplitudes.
As far as the input from the guitar influencing the purity of the octave is concerned the full wave rectifier trick works best, as you say, with signals that are as close to a pure fundamental as possible. Its true that the higher strings tend to produce fewer harmonics, but the position of the pickup along the vibrating length of the strings also has a big effect. The closer the pickup is to the middle of the vibrating string the stronger the fundamental component of the signal. So the notes with the strongest fundamental are around the 12th fret where the neck pickup is closest to being in the middle of the vibrating string length.The fundamental can also be enhanced by always picking in the centre of the vibrating string length.
Hi Mark,
I looked up a schematic for a clone of the Elector octave and its topology is slightly different than the one I posted above (the op-amps in the Elektor are configured as non-inverting). I’m too old/lazy to determine what a circuit does just by looking at it, if it’s anything remotely unusual, so I can’t tell if that would matter and I didn’t really want to take the time to simulate it.
But yeah, using 1% or better resistors will go a long way to making sure that the thing is balanced, and the high gain and feedback of the op-amp around the diodes greatly suppresses any effect of diode forward voltage. I have no reason to suspect that this circuit would want minor tweaks to resistor values.
As usual, I am going off into theoretical babble instead of hands-on results here, but I have noticed that rolling off a lot of the highs when putting guitar through a ring modulator suppresses a lot of the garbly hash, and it might also be advised in the case of a FWR for octaving. And by that I mean roll off the highs BEFORE the effect. Afterwards too possibly, but for different reasons.
Larry,
This is the thing people too often forget about ring modulators. They were developed using essentially pure tone generators as both input signal and modulation source. People thought it was an interesting effect, and wondered whether it could be applied to guitars, neglecting that the circuit does not overlook all the harmonics which would be absent with an oscillator but generated by a guitar string.
So yeah, careful tone shaping prior to rectification is important in octave generation. Personally, I think part of the magic of the Foxx unit is the feedback path between the second and first transistors in the circuit, which appear to impose a midscoop in the feedback path, that is pretty distinctive for octave fuzzes.
Terry,
1) I think you identified which of the two resistances needs to be lower than the other! 🙂
2) Yep, everything that leads to perfectly balanced amplitudes of the two complementary half-rectified signals, AND everything that makes the fundamental stick out like a sore thumb (including pickup, guitar tone, where you pick, string gauge, fret, etc.) is a contributor to more robust octave doubling.
Yes, the Ultimate Octave is a Foxx Tone Machine clone. (I used to call it the FTM, but nowadays that acronym tends to mean “female-to-male,” as in “He’s an FTM transexual.”) I believe that one of those super-cheap plastic Danelectro pedals of the late ’90s was another. (French Toast, maybe?) But as often mentioned here, I discourage friends from purchasing Dano products because their parent company, the EVETS Corporation, has a record of appropriating ideas from the boutique/DIY community and donating large sums of money to anti-equality causes.
I’d mentioned the Prescription Electronics Experience Pedal above, which melted my mind in 1994. I had no idea that it was a Foxx clone! I had no idea what a Foxx was! I believed it was something fresh and original. See, I’d grown up at exactly the wrong time to experience the great transistor fuzzes of the ’60s. My first distortion pedals were IC-based MXRs and DODs. They convinced me not to use distortion pedals for 10 years. (Sorry! Just personal taste!) Actually, I’ve come to admire the Distortion+ and 250 circuits, which can be so much more dynamic than Screamers. It’s just … come on, man! Those IC things sounded so boxy and compressed. Prescription Experience was the Piledriver of the Gods!
And that story made me remember this one, forgotten for decades: My first electric guitar was a late-’60s CBS Jazzmaster, a bar mitzvah present, accompanied by a beaten, black-painted ’52 Deluxe amp. (I was too civilized a kid to show my disappointment at receiving that crappy old box instead of some newfangled solid-state turd.) The seller was the son of one of my mom’s fellow elementary school teachers. We visited his mobile home, where he was selling three lightly used guitars: a Paisley Tele, my black Jazzmaser, and (I think) a Fender XII. (Hey, if anyone reading this might know the guy, speak up! I can’t remember, and Mom’s gone. But I’d probably recognize his name if I heard it. This was 1971.)
ANYWAY .. the deal also included a Fender Blender! Which didn’t work at first. Or later. It got lost.
Which was probably a good thing. I built a couple of those from the schematic and, man, do they sound shitty. The distortion is harsh and un-dynamic, with peaky resonances in all the wrong places. You could maybe use it as a deliberately ugly effect, but I predict a meager ROI.
The longwinded point being, that’s yet another diode/octave/fuzz thingy.
My bar mitzvah gifts included an unexpected delivery of a box containing Beatles VI, The Rolling Stones Now, and Herman’s Hermits On Tour. In retrospect, you made out better. Years later I did, however, inherit the considerably better bar mitzvah gift received by my now-deceased cousin, who got a ’64 “batwing” Epi Coronet, that has been modded a hundred ways from Sunday since he gave it to me 40 years ago (I gave the P90 it came with to Marshall Crenshaw). Absolutely terrible neck, since it appeared to have been made for 12 year-old hands. Gotta go up past the 7th fret to reach a width most necks would have at the nut. But probably lighter than a Parker Fly…or most ukeleles.
But anyway, in principle, the Fender Blender *ought* to be a better version of the Superfuzz, given how much is in common with it. The ability to blend clean and octave-fuzzed signal is a plus. My sense is that the basic design simply needs some fine-tuning, and matching of components. I’ve only seen it used live once, back in 1974 or so, by a local jazz guitarist. It was interesting, but not memorable. To be honest, what I found most memorable was the cool chassis.
It was my understanding that the current owner or co-owner of Danelectro is a former owner of Foxx, such that the French Toast pedal is a direct clone of the Tone Machine…minus the furry chassis, and plus the e-switching.
Playing around with the Tone Machine I recently made, using Schottkys, I found that I could get decent octaving as low as the 5th fret, even for the wound strings. But here’s the thing: it would only “blossom” into octaving if the string sustained for a little while. Given that strings have their most harmonic content at the outset of the pluck, and descend to mostly fundamental after a little while, that gives further support to the notion that satisfying octaving results from whatever one can do to make the fundamental stick out head and shoulders above everything else.
Great story Mark — thanks! 🙂
I’ve got my Screech producing a clear octave on all frets and pickup settings. But I can’t really tell you why — I just fucked around with different components till it did that. Which is pretty typical of my ignorant empiricism. 😉
It seems the the Foxx Tone Machine was designed by Steve Ridinger when he was 19. Mr Ridinger is the president of Evets Corp. which owns and operates the Danelectro brand.
Anyway I opened up my JOYO Voodoo Octave and changed the two Chinese ‘germanium’ octave diodes (0.6V forward voltage) for IN5819 Schottky diodes and the two clipping diodes (also 0.6V ‘germanium’) for a matched pair of NOS germanium diodes (0.35V forward drop. Now my Voodoo Octave really sings and sounds nothing like it did out of the box and probably nothing like an original Foxx pedal either.
I fooled around in LTSpice tweaking resistors in the full wave rectifier section and I could get pretty close to perfectly balanced operation for both half waves, but that circuit lacks symmetry, so as soon as I change the input signal level the balance changes.
It does seem to me that any of the full wave rectifier circuits probably need carefully matched components and even a trimmer, to get close to perfect, balanced rectification and therefore the purest octave sound.
Empiricism – now what happens if I do this …. oh crap! I’m never going to do that again. 😉
I thought that auditioning zillions of diodes while developing Gross Distortion would give me a better ability to predict the sonic effects of varying diodes. Maybe it would have done so for someone smarter, but I haven’t moved far beyond what I already knew: Ge sounds smooth and tubby, Si sounds crisp and defined, and LEDs sound crisper, more defined, and louder. But basically, it’s still all roulette to me. (Oddly, as much as I adore Ge transformers, I almost always go with Si or LED for clipping diodes. That is, if I use them at all. Gross is an anomaly for me.)
Terry,
Pleased to hear your experience replicates my own. One of the unrecognized perks of using Schottkys lies in the distortion sound. The Tone Machine defeats the octaving by simply lifting one of the rectifying diodes, obliging the entire audio to pass through a single diode in series with the signal, producing crossover distortion. I had taken to using a 3-way toggle to produce octave-doubling, octave defeat (middle toggle position), and use the 3rd “side” position of the toggle to bridge the remaining diode. Using Ge diodes, there is an audible difference between bridged and unbridged mode in the quality of the distortion. Using Schottkys, though, I can barely hear a difference. The low forward voltage of the Schottky is insufficient to pose a major barrier to the signal.
So, a followup.
In discussion with my friend Dino, who resurrects the obscure (https://www.youtube.com/watch?v=lYV2yWIbLnw), the topic of the role of the diode pair to ground came up. I suggested that, since the FTM remains distorted even without that pair, conceivably the diode pair was really intended to be a crude compressor, and “clamp” the level. Since the note fundamental tends to be obscured by all that harmonic content at pick attack, and since string level quickly declines as the fundamental becomes the dominant part of the signal, holding the level relatively constant *after* the initial pick attack, allows the octave to appear to “bloom”.
Dino was in accord, but noted that, in his own experiments, adding a small resistance (a 2k trimmer) between the diode pair and ground improved audibility of the octave. Makes sense. The diode pair WILL hold level constant, but adds harmonic content that impairs audibility of the octave as the note sustains and becomes primarily doubled fundamental. The added resistance keeps the compression effect, but softens the clipping to keep the obscuring harmonics to a blessed minimum, thereby making the octave stand out more. I also found that a small cap in parallel with the diode pair and resistance further shaves off unwanted harmonics to make the octave stand out a little more.
Thanks for the info, Mark. (And sorry for taking so long to reply. I was swamped by rehearsals for a few weeks there.) So … can you explain how the “swell” function on the Experience pedal (Prescription Electronics’ FTM derivative)?
Super interesting, Mark! Yes, I’ve noticed that on pre-IC designs, the diode-clipping pair makes remarkably little difference. (Harmonic Percolators are another great example.) The simpler the circuit, it seems, the more apparent the diode effect. Which is why I opted for a bonehead-basic Electra derivative for Gross, my pedal with the zillions of diode combinations. I’m totally going to try that trimmer trick — thanks for the tip. (And just how small a parallel cap are we talking about?)
The diodes can certainly operate as crude compressors, but (as discussed in another chat with the stupefyingly knowledgeable Terry Relph-Knight) I’m skeptical about the degree to which this was planned. Just speculation, though.
I have been looking for an octave for quite awhile now since I started doing a 3pc thing sometimes to do the dual guitar bits and etc… I tried a TRex Quint machine and it did not work out of the box , so much for buying anything else from that company. I have been looking at the Foxrox Octron but just can’t bring myself to buy another pedal. I might just have to get one of yours now , when I can scrape up some cash.
I do wonder with a lot of the older effects just how much of their creation is down to carefully reasoned step by step design and how much is down to ‘here are a few circuit elements I half understand, I’ll knock up a breadboard and then circuit bend it until I think I have something’. In the 60’s they did not have fancy digital scopes with built in analysis functions, or computer circuit simulation, or over 50 years of innovation in effect design. A fair number of the ‘classic effects’ were thrown together by teenagers. On top of that the ‘engineers’ had management breathing down their necks saying ‘You have too many parts in there, those resistors cost 2 cents each you know, do it with less’
If you analyze something like the Fuzz Face circuit today there is no way of knowing what the intentions of the original designer were. Today, simply because the Fuzz Face circuit is so prone to variability, it is usually assumed that the circuit is intended to produce asymmetric clipping. Most of the time if you put one together with the accepted values, asymmetric clipping is what you get. However what if the original designer really intended the Fuzz Face to produce symmetric 1:1 square waves? Or perhaps just didn’t care as long as every Fuzz Face built produced some form of clipping. Many of the rivals to the Fuzz Face certainly look like someone said three transistors must be better than two, lets just stick a few bits together and see what happens.
I suspect the FTM falls in this category – the designer had a bunch of half understood ideas, threw together a prototype, tweaked it a bit and called it done.
How could it NOT be choice #2? As far as I can tell, a bunch of young hackers (or whatever they called a hacker in the 1960s — maybe “ham radio buff?”) swiped some old amp designs from WWII-era electronics books. Most of these people weren’t good guitarists — they were just stabbing in the dark, and relying on feedback from the players for whom they built these gizmos. If they could find a cheaper part, or could substitute some other part in lieu of spending more on new stock, they always went that route.
It goes beyond a tech question, and becomes a philosophical/cultural one — I guess you’d call it “technological determinism.” I tend to believe that our guitar aesthetics are more defined by our tools than the other way around. This stuff came out. Great guitarists made some great music with it. The world retraced their steps. And the circuits went from “whatever” to canonic.
Spot on conclusion, which makes innovation in this area pretty frustrating.
Anybody ever used an op amp based full wave rectifier? Maybe with a high-pass and 2x gain on the end to deal with the DC offset and level. Seems way easier than these scrambler circuits if the goal is “accurate” full wave rectification. If that’s not the goal then I’ll just be quiet.