Craig Anderton Octave-Doubling Fuzz

[Joe Viau]

Hi all:

Has anyone built Craig Anderton's Octave Doubling Fuzz?
I heard that the sound was close to the Octavia.
Any opinions welcome (pertaining to ODF!)

TIA

[Mark Hammer]

I've re/built it a few times.  Like most of Craig's effects, it begs for mods.  I couldn't say how close it is to an Octavia, but the octave effect isn't bad. I still like my Foxx Tone Machine clone best of all although that may well be because of some constellation of quirks.

A few things to note for the CA-ODF:

1) The stock design uses a master output volume and a switch to enable/defeat the normal fuzz sound.  It may be more flexible to leave the output volume fixed at max with a simple 10k resistor to ground instead of a pot, and have a normal distortion level pot *before* the final mixing stage.  Switch the 15k input resistor to 10k, like the octave channel, and replace the toggle switch with a 10k log pot.

2) The stock design has no lowpass filtering on the output stage to keep hiss/noise from the output, even though it introduces lots of gain.  Craig should know better, even 20 years ago.  Stick a 68 to 100pf cap in parallel with the 220k resistor in the output stage (100pf = 7.2khz rolloff, 68pf = 10.6khz rolloff).

3) The stock design uses 1N4001 diodes all round.  These have higher clipping thresholds, so they require a much hotter signal to deliver pleasing clipping and octaving.  If you are using a hot signal, fine, but mere mortals may wish to consider using 1N914/4148's and even Ge diodes for the octaving section.  Other options for the clipping section include use of warp controls or "softening" resistors in series with the diodes; basically anything you might use in a Tube Screamer.  You can even play with the value of the feedback resistor or stick in a trimpot for fine-tuning.  Just as long as the octaving section receives a signal that is big and flat enough to let it do its thing.

4) There are some signal-phase issues to be mindful of.  The input stage is non-inverting.  The clipping stage and full-wave rectifier section used for the octaving are inverting.  The output/mixer stage is inverting.  This means that : a) the fuzz section is in phase between input jack and output jack since it is inverted twice. (during clipping and in mixing stage), b) the octaving section is out of phase at the output since it is inverted 3 times.  This also means that some cancellations occur when blending fuzz and octave due to those two signals being inverted, relative to each other.  That's not a weakness or anything that necessarily requires "curing", but just be aware that turning the knob up a bit does not always produce "more".

5) The stock design uses a unity-gain buffer stage at the input.  No reason why you couldn't add some gain there and make it a little more sensitive.  Stick a 47k resistor and 330pf cap in the feedback loop of that first op-amp, and a 22k resistor and .22 cap from the inverting pin to ground, and you'll have a gain of just over 3 and some extra oomph and tone-shaping.

6) The cap switching arrangement in the clipping stage is OK for tone modding, I suppose, but the problem is that the octave section gets its signal from the output of that stage, and whatever cuts treble for a mellower sound also makes it harder to produce octaves on higher notes.  For example, a 220pf cap in that feedback loop rolls off around 720hz, meaning that by 1400hz what the octaver gets is a full 6db lower than stuff from the A and E strings.  That's one reason why I would prefer to do the "mellowing" at the output stage, rather than at the clipping stage.  personally, I wouldn't go any larger than 100pf in that clipping/fuzz stage.

7) The design can probably be adapted to work off a single ended supply and Vref.  I'm not sure how to calculate the needed components, but it would essentialy be the standard equal-value resistors plus cap from Vref to gnd for the floating ground.  The 220k resistor to ground on the input stage goes to Vref as do the non-inverting pins of all the other stages (instead of ground as they currently show).

[Aharon]

I have a PCB express file for the Octave Douling Fuzz if anyone is interested.I would send it to someone for posting and you guys can debug if needed.I have never built it but made the PCB as an exercise.
Aharon

[Joe Viau]

Mark Hammer:
Thanks for the great ideas.  I'm going to try them a few at a time, if I did them all at once I wouldn't know what mods sounded better than others.  They're fairly simple changes and should be fun.

Aharon:
I'd like to look at the PC layout, I'm using the Express Schematic and PCB layout software, also as an exercise.  How can I see what you've done?

Thanks,
Joe

[Aharon]

Send me a private message with your email,I'll send it to you.
Aharon

[Mark Hammer]

Joe,

Just so we're clear on this:

a) It doesn't NEED any particular op-amp like a 4739 or 4136.  Singles, dual, quad, whatever you have in your bin.

b) The fuzz itself is unspectacular and the octave alone will not elicit a "Yeah, baby!" from anyone.  Where this thing works well is in variable blending, as opposed to the more basic octave/no-octave found in many of the classics.

I think your one-mod-at-a-time approach is sound.  Like most of Craig's projects, it is an excellent vehicle for learning about how things work and foiguring oput how to tailor effects to your own specific needs.

[Elektrojänis]

b) the octaving section is out of phase at the output since it is inverted 3 times.

Hmmm... I always thought that if you full wave rectify a signal, then every other half wave is in phase with the original and every other is out of phase.

Edit. Ooops... It was an old thread... Sorry... Got to it through a link in a newer thread and mixed them up.

[jmusser]

Mark, I was wondering if you've ever built the Tim's Octup Blender? When I think of about the ultimate in blend and boosted stock guitar signal, I have not built a better one to date. It is unique sounding, and can be adjusted so that you can have the boosted guitar signal with an up octave break up or anything in between. The thing is, it's distinct, and you know they are both there at the same time. The other difference is, that when you have octave up only, you have it EVERYWHERE. It don't just start on the 10th fret. It's hard to pin down, but I would guess it is square wave down the neck transistioning to squared up octave up the neck, but it is such a smooth transistion, that you'll be hard pressed to hear when it happens. It really fools the ear into hearing that it's at every fret, and I have not heard another pedal, that does what it does.

[Mark Hammer]

I actually started that one, but like a lot of perfed almost-done projects, you get interrupted, come back to it a month later, and it's juuuuuust hard enough to remember what connections were and weren't made that you say to yourself "I'll set this aside for a less hectic night/weekend".  Of course, those never come. :cry:  You've made me want to dig through my bin of unfinished boards for it though.

[jmusser]

I believe it would be worth your time, because it's not like other up octaves I've heard. It's way more buzzy/synth sounding than say the FTM, OUSB, Tycho, Scrambler, etc. If I remember right (and it's been awhile) it doesn't seem to care if the tone knobs are rolled off or not, or where the pic up selector is to perform it's octave feat. that in itself is different