Analog Octave Effects

[jcuempire]

I have a noob question.  How and where does an octave circuit (Green Ringer, Foxx Tone Machine, Octane 3, etc) work?  In other words, what part of the circuit is doing the octave?  Thanks

[earthtonesaudio]

Full wave rectification.  The FTM and GR use a phase splitter driving two diodes, not sure about the Octane.  It might use a differential pair but it's still full wave rectification.

[Mark Hammer]

All of them use what is called "phase-splitting", or production of two identical inverted (relative to each other) copies of the signal.  This is always easily identified by the observation of a transistor with identical-value resistors between the emitter and ground, and the collector and V+.  Doing that results in one version of the signal available at the emitter, and the opposite version at the collector.  If that transistor were seeing a pure steady tone, and you were viewing them both on a scope, there would be a peak in one where there was a trough in the other.

If you "chop off" one half of each signal (making sure its the appropriate half), and combine them, you effectively "fill" the gaps between successive peaks in one with the peaks of the other.  bada-bing, bada-boom, you now have twice as many peaks stuffed into the same time period: frequency doubling.  Does it look, smell, taste, feel like "true" frequency doubling?  Not at all, but it sorta sounds like it.

The "chopping off" is done in various ways, but they all use either diodes in the path of each copy of the signal, or the diode properties of other transistors at the point/node where the two halves are mixed.  Remember that the diode will permit one half cycle to pass, but block the other.

One of the reasons why it will never really be exactly like pure frequency-doubling is precisely because of the action of those diodes.  They all require that the signal not only be of a certain swing-direction to pass through (the negative or positive half-cycle), but that it be of a certain amplitude.  So if we go back to the example of a pure steady sine wave fed into such a circuit, the desired signal will start to go in the intended direction but will not pass until it reaches the critical threshold for the diode to conduct it.  What might be a gradual increase in the waveform gets turned into something with flat sides, even though the peak of that half wave is unaffected.  This is often referred to as "cross-over distortion".

One of the mods I like to do to Foxx Tone Machines addresses this.  The FTM uses the phase-splitter, with a diode in series with each output.  When the two diode-rectified outputs are mixed, you get nice octaving.  But when you turn the octaving off so that the output only receives one of the phase-splitter outputs, are you getting a "normal" signal?  No.  the output gets one copy of the signal, yes, but only one half cycle of that signal, and also a half cycle that has this cross-over aspect applied to it via the series diode.  If the octave on/off switch shunts this diode (places a direct connection across it), then you get the entire signal, with both half-cycles, and no cross-over distortion.  Much louder, more body, and potentially much cleaner.  Since there is some utility in the original partially-clipped version too, I use a 3-position topggle so that I get octave, "traditional" non-octave, and pure non-octave.  provides a bigger pallete.

[jcuempire]

Wow.  Thanks to both of you for answering my question.  Now that my brain is all soupy, I *think* I'm getting the gist of what you're saying.  Is the octave a natural occurrence of the process or was it designed with that in mind?  Mark, are there projects with cloned FTM's that incorporate your mods?  It sounds like a cool pedal.  Thanks again

Keith

[amptramp]

This question has come up before and the following gives a graphical account of the harmonics created by fullwave rectification.  It is not just the second harmonic:

If you want to check up on the harmonics and their relative strengths derived from full-wave rectification, check this site:

http://www.falstad.com/fourier/e-fullrect.html

It shows the DC value is 0.63662 times the input
2nd (desired) harmonic = 0.42442 = -7.44db
4th harmonic = 0.08489 = -21.42db
6th harmonic = 0.03638 = -28.78db
8th harmonic = 0.02021 = -33.89db

There are no odd harmonics, but the harmonics above 2nd are usually undesirable.  The waveform coming into the device will behave this way if it is a pure sine wave - other waveforms may just become a fuzzy mess.  The site quoted above will show what happens in fullwave rectification of a triangular sawtooth function under the "next" link on that page.  In order to avoid this, it may be best to split the signal into frequency bands and rectify the signals in each band separately before recombining them.

A better way to generate an octave up would be to square it (multiply the signal linearly by itself) using the trig identity that:

cos (F) * cos (F) = 0.5 + 0.5 * cos(2F)

This has the advantage that only the second harmonic is generated.  Once again, it may be best to split the signal into frequency bands and square the signals in each band separately before recombining them.  Linear multiplication can be done in an MC1496 (readily available) or MC1495 (rare - I last used one in the early 1970's).

[Mark Hammer]

Now my brain is "all soupy".

Is the octave a natural occurrence of the process or was it designed with that in mind? 

Yes and no.  The phase-splitting, followed by chopping one half-cycle does provide two copies of the signal that can be combined to produce twice as many peaks per chunk of time.  But to be heard as an octave - that is a "note" one octave higher, as opposed to mere a "brightening via a harmonic", the two signals to be combined need to be balanced, such that the extra peaks are just as tall as the ones they are wedged between.  One example of this can be found in JC Maillet's mod for the Green Ringer - http://www.lynx.bc.ca/~jc/pedalsGreenRinger.html - where a trimpot is used to adjust the level of the collector output, relative to the emitter output.

Mark, are there projects with cloned FTM's that incorporate your mods?  It sounds like a cool pedal.  Thanks again

Not as such.  It's not really a "project", just a tweak of an existing switch in the Foxx Tone Machine.  Doesn't require any additional parts, just a different type of toggle.  Build one and I'll tell you how to wire it.  And....you're quite welcome.

[jcuempire]

And here I thought this was going to be a pleasant little hobby to pass the time. 

amptramp - I can't begin to understand all the math involved (I was a history major), but on some level, I think I almost get it.  Thanks.  I really went into this hobby pretty blind, so I think I need to get myself some education on electronics in general and what all the components do relative to each other.  So maybe someday I won't see it as "magic"

Mark - I will contact you when I build it.  I have to finish troubleshooting my smallbear tremelo and I just started the madbean Rump Roast preamp, so I am looking to do the octave device or a phase shifter next. 

I am amazed at the ready help the community offers.  It makes it a heck of alot of fun, if not additively expensive.

Keith

[Mark Hammer]

yeah, right.  Pleasant little hobbyyyyyyyyyyyyy
y
y
y
y
y
y
y
y
y
y
y
y
y
y

Alice said as she fell down the rabbit hole. 

[garcho]

"Linear multiplication can be done in an MC1496"

I have a few MC1496P's from Small Bear laying around but I think the only circuits I've ever seen them in (I'm a newbie) are some rather complicated ring mod synth modules. Any octave/harmonic circuits you know of use the 1496? I thought I would just use those chips for arts & crafts or something. 

[runmikeyrun]

yeah, right.  Pleasant little hobbyyyyyyyyyyyyy
y
y
y
y
y
y
y
y
y
y
y
y
y
y

Alice said as she fell down the rabbit hole. 

dude, you have NO idea!!  lol.

[runmikeyrun]

I guess I have an intelligent question for this thread.  How well do these types of octave up circuits work on bass?  Especially an active 5 string bass tuned B-E-A-D-G.  Am I only going to get octaves on higher notes?

[egasimus]

"Linear multiplication can be done in an MC1496"

I have a few MC1496P's from Small Bear laying around but I think the only circuits I've ever seen them in (I'm a newbie) are some rather complicated ring mod synth modules. Any octave/harmonic circuits you know of use the 1496? I thought I would just use those chips for arts & crafts or something. 

This. I remember this device being pointed out to me by amptramp, too, but I couldn't really figure it out, either.

[garcho]

I've read in the datasheets and application notes that one of the uses for the MC1496 is as a 'frequency doubler', which sounds great for audio, except the frequencies the chip appears to made for are well beyond human hearing. Under 'typical applications' on the datasheet there's a circuit for a 150MHz to 300MHz doubler. Is there a way to change that to the audible range of frequencies? And what happens if you don't feed the chip 150MHz (in this example)?

[amptramp]

I've read in the datasheets and application notes that one of the uses for the MC1496 is as a 'frequency doubler', which sounds great for audio, except the frequencies the chip appears to made for are well beyond human hearing. Under 'typical applications' on the datasheet there's a circuit for a 150MHz to 300MHz doubler. Is there a way to change that to the audible range of frequencies? And what happens if you don't feed the chip 150MHz (in this example)?

Check figure 30 from here:

http://www.onsemi.com/pub_link/Collateral/MC1496-D.PDF

Audio frequency 1496 doubler.

[blackcorvo]

if you wanna go crazy, there's this project for a Tube Octave Up effect called Otávio:
http://www.youtube.com/watch?v=9GTvd1zSinE
http://www.youtube.com/watch?v=aJ1AzYSBmE8&feature=related
you can see the schematic for it in the pic on this post on the brazillian forum where it was being discussed:
http://www.handmades.com.br/forum/index.php?topic=2564.msg50778#msg50778