Alex's Octave-Up Rabbit Hole

finebyfine

Well-known member

ex post facto introduction:

Analog octave up is easily my favorite effect. Throw a good fuzz on top or in it and you have the basis for my favorite guitar sounds. I love how it works with doom metal (eg, the EQD Life Pedal using a Green Ringer, the Boss Super Fuzz being used all over the place in the genre, etc), psych rock (Octavia, etc) and anything in between and all over. I love the textures they make with chords, and love how they make single notes shine.

In this thread I've documented some of the digging I've done into analog octave up and related circuits & topographies for frequency doubling and squaring. Other than this intro, no other posts have been edited that much after the fact.



Can’t find too much from googling but has anyone seen an op-amp precision full wave rectifier used in any octave circuits?

I was digging into phase inversion correction and got introduced to the concept. am gonna be breadboarding a few when I have some free time. Hard to beat an octave fuzz in my opinion and I don’t recall seeing it done this way in a circuit I’ve looked into.

Overview: https://circuitdigest.com/electroni...wave-precision-rectifier-circuit-using-op-amp


TDPRI thread with examples: https://www.tdpri.com/threads/the-groble-green-ringer-octave-blend.326950/#post-4169518

Circuit from above that author recommends with a buffer and low pass filter, with variable R3 allowing for a similar effect as the green ringer nuller mod

768426F4-58B1-42A7-9776-43F05FD72691.gif
 
Last edited:
I wanted to document some research I've done on this topic and also ask for some guidance.

The late Time Escobedo had several op-amp rectification based octave up effects in his Folk Urban circuit collection.

His Rambler (and Marginally Better Rambler) are fuzz heavy octave fuzzes, and the Octup Blender is the closest to a textbook op-amp rectifier as far as I can tell.

Octup_Blender.gif

When I breadboarded it, it's octave effect was less dependent on typical octave up playing conditions (neck pickup, playing above 12th fret, tone knob all the way down) than others, including Anderton's Green Ringer (EQD Tentacle, PedalPCB Squidward) which is the gold standard in my opinion.

My main question about the Octup Blender is about the voltage divider Vr: this is done so that the input signal is not clipped to ground like it would be in a single supply op amp, right? If this is true would there be any benefit to using a voltage inverter to create a larger supply rail difference instead of a divider? I've breadboarded other op-amp rectifier circuits using this method but couldn't tell much of a difference in effect.
 
I'm a little confused by your question. Using Vc/2 as the signal reference is SOP for single supply op-amps. If you were to use a single supply op-amp (V+ = 9V, V- = GND) and your signal was referenced to ground, you'd only have a positive swing. You could use a dual supply for the op-amps and eliminate the voltage divider and reference the signal to GND if you wanted, but I'm not sure if you'd need the added headroom/the design warrants that. I don't believe the signal is approaching the rails here.
 
I'm a little confused by your question. Using Vc/2 as the signal reference is SOP for single supply op-amps. If you were to use a single supply op-amp (V+ = 9V, V- = GND) and your signal was referenced to ground, you'd only have a positive swing. You could use a dual supply for the op-amps and eliminate the voltage divider and reference the signal to GND if you wanted, but I'm not sure if you'd need the added headroom/the design warrants that. I don't believe the signal is approaching the rails here.

Yep headroom is exactly what I meant and the phrasing that was on the tip of my tongue. Thanks!
 
A few updates for this thread:

Another pedal circuit based on precision rectifiers is JC Maillet's Super Full Wave Octaver:

1658865187114.png

Super clean octave up, even on bright bridge pickups. Crazy fun build.

I've also been digging into ring mod circuits to do octave-ish up. This LM13700 one is up next for the breadboard.

1658865737635.png
Trying to see where this gets me because AD633 ring mods and similar (EHX Frequency Analyzer - madbean freekout is a clone) are pretty complex for what I'm going for, plus analog multipliers ICs are not cheap.
 
I have been really wanting to avoid spending $12 on an AD633 but I'm waiting on that, and also an SMA5101 - which is 4 NPN transistors arranged as a Gilbert Cell - which is the basis for as far as I can tell, most analog multipliers, but especially the AD633.

This is what I'll be starting with on the AD633, from the AD633 datasheet for frequency squaring

1661723730601.png
 
Last edited:
Some more updates mostly for my own record. No demos unfortunately.

I've been playing around with two analog multiplier ICs: Analog Device's AD633, which is at the heart of a lot of more recent ring modulator pedals, and the obsolete, but cheap (like $1-$2 a pop, compared to $15-$20 AD633) and plentiful Exar XR2208. The bounceless frequency squaring application note in the AD633 datasheet is pretty good out of the box running at ±9V - which has been a relief considering all of the schematics I've seen using ±15V. For the XR2208 I've used ixtern's circuit from their amazing ring modulator thread.

I was really hoping to be more blown away by the AD633 in terms of not having to do the typical octave-up playing considerations (neck pu, above 12th) but I am mostly chalking that up to just needing to dial it in better and spend more time with it.

Because of the price and availability despite being obsolete, I'm really trying to nail the circuit with the XR2208. The only datasheet with application notes I've found is the datasheet for the XR2228 multiplier which is similar but has a few quirky differences.

Also next up after/alongside tightening up the 2208 circuit is to play around with frequency division. My ideal octave up ringer is still the end goal but it's listed in the AD633 datasheet and looks like a fun one.
 
Last edited:
Another update for posterity!

  • I haven't gotten around to playing with the XR2208 and AD633 as much as I've been wanting to. Still on the short list
  • I've read through Ray Ring's H11F1M optofet designs documented on his blog, Circuit Salad (I've built this compressor he designed) and am especially interested in his H11F1M Ring Modulator that uses an H11F1M as a mixer. I still don't really understand how it works, nad with the XR2208 and AD633 it's low priority, but interesting nonetheless.
I got boards made for JC Maillet's "Green Ringer Redux". I'm having trouble finding a link to a blog post from him about it, but the URL at the bottom of this schematic is at least his source of the circuit.

This is not his previous Green Ringer Mod circuit with a tone toggle and nulling mod, drawn on the GGG schem, that is easy to find on his lynx site. In his schematic of this version, in fact, he has omitted the nulling mod altogether.

The Fuzz control turns it into a wild octave fuzz. But because my goal here is really more towards semi-clean Octave Ringing, I haven't used it much while playing with it and there are better octave fuzzes if that's the goal. I do turn it up in the demo below.

I kept potentiometers for both sides of the common "nulling" mod. Having both of these pots external is definitely overkill. If I redid the boards/schem I'd still keep one external, and have the other either be a set resistor value like in traditional null mods, or keep the other as an internal trimpot.

gr-redux.png

Here is a very sloppy (and long) demo of it in use.


At around 3:50 in I switch to my amp's dirt channel - which is really why I love from cleaner octave up circuits, that is, having the octave up be on the cleaner side so I can choose how to color / fuzz it up.
 
Another update for posterity!

  • I haven't gotten around to playing with the XR2208 and AD633 as much as I've been wanting to. Still on the short list
  • I've read through Ray Ring's H11F1M optofet designs documented on his blog, Circuit Salad (I've built this compressor he designed) and am especially interested in his H11F1M Ring Modulator that uses an H11F1M as a mixer. I still don't really understand how it works, nad with the XR2208 and AD633 it's low priority, but interesting nonetheless.
I got boards made for JC Maillet's "Green Ringer Redux". I'm having trouble finding a link to a blog post from him about it, but the URL at the bottom of this schematic is at least his source of the circuit.

This is not his previous Green Ringer Mod circuit with a tone toggle and nulling mod, drawn on the GGG schem, that is easy to find on his lynx site. In his schematic of this version, in fact, he has omitted the nulling mod altogether.

The Fuzz control turns it into a wild octave fuzz. But because my goal here is really more towards semi-clean Octave Ringing, I haven't used it much while playing with it and there are better octave fuzzes if that's the goal. I do turn it up in the demo below.

I kept potentiometers for both sides of the common "nulling" mod. Having both of these pots external is definitely overkill. If I redid the boards/schem I'd still keep one external, and have the other either be a set resistor value like in traditional null mods, or keep the other as an internal trimpot.

View attachment 38914

Here is a very sloppy (and long) demo of it in use.


At around 3:50 in I switch to my amp's dirt channel - which is really why I love from cleaner octave up circuits, that is, having the octave up be on the cleaner side so I can choose how to color / fuzz it up.
Rad! I must try this out. I’m a big ringer fan
 
Got any clips of JC Maillet's Super Full Wave Octaver? Your description of it sounds very intriguing.
 
I was surprised I didn't have one, so here's a quick one. The noise floor on this build seems a lot higher than I remember it being but my amp also sounds more noisy than it did when I recorded the redux demo earlier today. I switch to dirt around 1:45
That sounds super fun, thanks for the demo! Do you find the "second" control to be a useful enough to put external?
 
That sounds super fun, thanks for the demo! Do you find the "second" control to be a useful enough to put external?

It's much more useful with a log pot than a linear one - with a linear in this build I've mostly just kept it dimed. I just didn't have any A100ks on hand (somehow! I probably have 30 in my parts cabinet now) when I got the PCBs in for it. With a log pot you can really hear it focus in and out more like nulling a green ringer.
 
Since you guys are octave up enthusiasts, I thought you might appreciate this. I recently built a couple double-analog-octave up pedals. Basically it’s just 2 green ringers in series, with a toggle to select one or both and of course the footswitch to bypass the whole thing.

The second circuit brings in a whole new sound. You can just barely hear the 2nd octave on certain notes, but mainly it just makes it more pinchy / nasty / aggressive (awesome with fuzz after it). It behaves similarly to the regular green ringer in that it works best with neck pickup playing up the neck.

Once I get around to making a little recording I’ll post a build report. You could of course use the concept with the other octave up circuits.

I attached a graph of what I think is approximately happening with the signal when the 2nd octave circuit is engaged. I did change the input capacitor of the 2nd circuit from 47n to 10n to help that high pass filter (DC coupling cap) work a little faster in theory. The difference is subtle but I think it’s there.

I can’t say if I like it better than the single octave, they’re both cool in their own way, which is why it's great to have the toggle. I have couple more boards I could send if anyone is interested!
 

Attachments

  • face.jpg
    face.jpg
    326.9 KB · Views: 28
  • gut 1.jpg
    gut 1.jpg
    307.3 KB · Views: 21
  • gut 2.jpg
    gut 2.jpg
    335.1 KB · Views: 27
  • Double analog octave graphs output.JPG
    Double analog octave graphs output.JPG
    36.9 KB · Views: 30
Since you guys are octave up enthusiasts, I thought you might appreciate this. I recently built a couple double-analog-octave up pedals. Basically it’s just 2 green ringers in series, with a toggle to select one or both and of course the footswitch to bypass the whole thing.

The second circuit brings in a whole new sound. You can just barely hear the 2nd octave on certain notes, but mainly it just makes it more pinchy / nasty / aggressive (awesome with fuzz after it). It behaves similarly to the regular green ringer in that it works best with neck pickup playing up the neck.

Once I get around to making a little recording I’ll post a build report. You could of course use the concept with the other octave up circuits.

I attached a graph of what I think is approximately happening with the signal when the 2nd octave circuit is engaged. I did change the input capacitor of the 2nd circuit from 47n to 10n to help that high pass filter (DC coupling cap) work a little faster in theory. The difference is subtle but I think it’s there.

I can’t say if I like it better than the single octave, they’re both cool in their own way, which is why it's great to have the toggle. I have couple more boards I could send if anyone is interested!

Oh wow this sounds awesome - I'd definitely appreciate a demo if you can. Your graph seems right on to me in terms of what's happening. How does it play with multiple notes and chords?

I'm for sure gonna try something like this. I'm interested in how a blend on the second octave would sound, or hell, maybe feeding the output of a ringer back into itself a bit. The Carlin Ring Modulator can kind of do that if I'm remembering correctly.
 
Back
Top