octave pedal

[john3696]

:oops:
Hello, I am new to this forum, but I have made a couple of pedals and I am currently looking to build an octave pedal.  I have searched around and found many octaves with distortion, but what I am looking for is a pedal that just does an octave up or down.  I don't want any distortion if I don't have to.  

If someone could point me to a schematic or perferably a PCB drawing of a simple octave up/down pedal I would be greatful.

thanks,
graham

[smoguzbenjamin]

That's gonna be tough! Check out http://hammer.ampage.org I believe mark there has an octave project floating around...

[Mark Hammer]

Octave up and octave down boxes use different technology.  Octave down will be strictly mono (play more than one note at a time and it turns to garbage), and while octave up won't be as sweet with 6 notes as it will with only at a time, at least it will let you hear 6 notes.

Octave up boxes can be made fairly simply and still deliver reasonable tone (the Bobtavia is an excellent example of just how simple it can be) whereas octave down boxes need a LOT of help to provide anything reasonably acceptable.

As far as learning how to use one, octave up boxes are less finicky about what you feed them than are octave down boxes.

There are several octave-related choices on my site as well as at www.runoffgroove.com and www.generalguitargadgets.com

Pick your poison and we"ll take it from there.

[john3696]

i suppose I would rather have an octave down pedal.  I will look at the sites you guys mention to see If I can find anything out

thanks a bunch

[smoguzbenjamin]

If you look around hard enuogh there's plenty of BOSS OC-2 octave down pedals around for quite cheap. I have one and it has the best tracking out of all the octave boxes I tried, and it sounds good too!

[R.G.]

what I am looking for is a pedal that just does an octave up or down. I don't want any distortion if I don't have to.

This question comes up regularly. There are no simple circuits that produce octave up or donw without distortion. The simple circuits all make octave with distortion, in varying degrees.

But the simple circuits are fun - dig in and experiment.

[kroushl]

Here's a post a while back that has a good explanation of the octave technologies by Hammer, Keen, and others I believe.

http://www.diystompboxes.com/sboxforum/viewtopic.php?t=16321&highlight=clean+octave

Later,
Brad

[Mark Hammer]

The esteemed "Commander" Keen is quite correct in his comments about octave-generation and distortion being twins joined at the hip.

On the other hand, by reasoned use of lowpass filtering, the raspiness of what starts out life as a square wave in octave-down boxes can be tamed so that it starts to sound more like a low note with a bit of coloration and less like a fuzz.  Further advice about how and where to implement this can be provided once you have chosen a design to build.

The Anderton/PAiA/Guitar-Player Rocktave project, posted at my site and JDSleep's site is an excellent octave-down box with very good tracking (some say the best out of what's available).  It also has an on-board Rat-style tone control for rolling off highs from the octave signal.  This can be further modified to take out more of the highs and mids and make the octave-down smoother sounding.

If this one grabs your interest, note that the board layout on my site is for using a 4136 quad op-amp, and the one that Dean Hazelwanter produced (posted on JD's site) is for a TL074 quad op-amp.  These two chips have different pinouts, so you can't use one and expect to use the other chip.  Neither is "better" but some folks have difficulty getting the 4136 locally, where there are many more pin-for-pin substitutes for the TL074 than there are for the 4136.  In that sense, Dean's board layout is handier.  Dean also included the alternate-interval mods that are described in the PDF at my site.  I note that none of the other commercially available octave-downs, for whatever pluses they have, will provide this option.

[john3696]

thanks for the great info, I think I will try the Roctave from General Guitar Gadgets.  I assume that everything but the opamp(Tlo74) is the same as the Anderton Rocktave?  as far as the parts list goes at least.  because I downloaded the pdf and am making a list to order my parts!!!!

again thanks a ton!!!!!!!!!!!!!!!!!

[Eb7+9]

There are no simple circuits that produce octave up without distortion.

... wrong

My Harmonic Multiplier is a reconfiguration of the EMS VCS3 ringmod circuit which is based on a LM1496 4-quadrant multiplier ... it produces an octave effect based on linear multiplication of two AC signals and includes fine-tuning to null-out the fundamental carrier and higher order overtones - something you can't do with a starved differential pair or rectifier circuit ... as long as you stay within the linear range of the Gilbert cell you'll get pure octaving unlike other types of circuit can ...

... jc

[Mark Hammer]

Easy to get sucked into the vortex by all those rectifier-based octave units, isn't it?  :wink:

Yep, we ALL forgot about multipliers for a moment there.

[R.G.]

... as long as you stay within the linear range of the Gilbert cell you'll get pure octaving unlike other types of circuit can ...

... wrong

Your "harmonic multiplier" is a straightforward implementation of a multiplier.  If guitar signals were pure sine waves, you'd have a chance. They're not, though. Using a gilbert cell multiplier on a real guitar signal produces intermodulation distortion, as you know. You get multiples of the fundamental, the harmonics and multiples of the sums and differences of the fundamental and harmonics as well.  (By the way, Moto always said that the 1496 is a modulator, and the 1495 the multiplier.)

It is less intense distortion than other means, like full wave rectification, but it is distortion nonetheless.

You'll notice that I did not mention my JFET and MOSFET doublers. That was not an accident. He asked for distortion free octaves.

The MOSFET and JFET doublers do exactly what a multiplier does - produce an almost pure squaring of the input signal - but exactly like the doubler, it produces intermodulation distortion of the signal with itself. In fact, the doublers do exactly the same math that a gilbert cell does on the signal.

The only distortion free real time octaves up or down that I know of are DSP based. It's something that has to be done in the time domain.

[anyuser00]

You'd probably want to try a pitch shifter. http://www.generalguitargadgets.com/v2/index.php?option=displaypage&Itemid=115&op=page&SubMenu
Have no idea how good it sounds but it's probably the closest to what you're looking for.

I was going to try and design something using phase vocoding based on Richard Dobson's pvtransp VST plugins since I get really good results.  Doing this analogly would get pretty complicated though.

[Eb7+9]

If guitar signals were pure sine waves, you'd have a chance. They're not, though. Using a gilbert cell multiplier on a real guitar signal produces intermodulation distortion, as you know.

I don't think you understand how these analog circuits are classified to begin with ... they are defined in mathematical terms based on a pure sinewave input - of course if you stick a multi-tone waveform in there you get sum and difference side-terms ... that has nothing to do with distortion per se ...

The question of distortion in squaring circuits is properly addressed when you stick a pure sine wave in and observe how many overtones are present ASIDE from the octave component ... this is the only framework where one can consistentally define and measure distortion in such circuits ... the need and consideration for this definition comes historically and practically from the fact that PLLs can lock on to these overtones instead of the  "double frequency" tone as used in phase-locked communication circuits ...

... Moto always said that the 1496 is a modulator, and the 1495 the multiplier.

I know it tends to throw off technicians all the time, design engineers interchange terms modulator and multiplier loosely depending on the application ... whatever you think Moto wants you to believe is irrelevant, what matters here is that I'm refering to a 4quadrant Gilbert cell - the daddy "linear" version of the multiplier/modulator family which is necessary and sufficient for obtaining large-signal linear multiplication ... the term linear refering to a low "product" distortion ...

It is less intense distortion than other means, like full wave rectification, but it is distortion nonetheless.

... this is a groundless generalization ...

The MOSFET and JFET doublers do exactly what a multiplier does - produce an almost pure squaring of the input signal - but exactly like the doubler, it produces intermodulation distortion of the signal with itself. In fact, the doublers do exactly the same math that a gilbert cell does on the signal.

... wrong again ...

the starved-pair doublers you are refering to don't simply cancel out all terms other than the second harmonic, again in relation to a pure sinewave input, but rather cancel out "odd" order harmonics (including the fundamental) provided each side of the pair are gain matched ... what you are left with is not a pure 2nd harmonic but all "even" order harmonics summed together - albeit the 2nd harmonic being dominant ... you end up with a dirty octave even if your differential circuit is perfectly matched (side-to-side) and all odd terms have been cancelled out - which it can't be perfectly in practice using discrete devices ... this mechanism is fundamentally different than a multiplier based squaring circuit which can be adjusted for residual on-chip offsets as devised by the designers of the EMS Synth circuits ...

The only distortion free real time octaves up or down that I know of are DSP based. It's something that has to be done in the time domain.

... you must mean the z-domain ... and even then you have quantization noise, alliasing noise and phase-distortion ... granted, DSP octaving is an effect completely different than what any continuous-time squaring circuit will produce ...

... jc

[Paul Perry (Frostwave)]

There are no simple circuits that produce octave up without distortion.

... wrong
My Harmonic Multiplier is a reconfiguration of the EMS VCS3 ringmod circuit which is based on a LM1496 4-quadrant multiplier ... it produces an octave effect based on linear multiplication of two AC signals ... jc

...well 'it all depends what you mean by "octave up" ' as the late Professor Joad would say. If you play two notes at once, you are going to have a shitload of intermodulation distortion, no matter what, using a 1496 or any other multiplier.
All this confusion arises because it wasn't clear whether peopel were talikng about a Whammy-type DSP pitch shifter, or a single note processing analog setup.
My advice to the original poster is build the Anderton octave down unit, and play one note at a time