Multiple pitches from Rocktave?

[gez]

Thinking about this further, it's interesting to note that the fifth harmonic of the overtone series, and subsequently the fifth drawbar of the Hammond organ, is 2 octaves and a major third above the fundamental. As you note, it would be expected that this would create some dissonance when playing, for example, a B in the key of A minor. However, reality tells us that organs using this stop don't automatically sound dissonant, nor do instruments with prominent fifth harmonics, like triangle wave synths or woodwinds, sound nasty in the same situation. It seems that having notes separated by octaves makes them progressively less dissonant - for example the minor ninth is dissonant, but less so in my opinion than the minor second.

There's a big difference between the harmonics within a note forming intervals and fundamental pitches doing the same.  Harmonics tend to have lower amplitude and tend to be widely spaced in relation to their fundamental, so are less obtrusive and don't dominate the harmonic 'mix'.  An interval comprising two fundamentals will not only have the harmonic relationship between both fundamentals (which will be of similar amplitude within an octave span), but also any clashes resulting from the two sets of harmonics generated by said fundamentals.  These clashes will be more noticeable if the fundamentals are close together.

Range also has a big part to play.  You can play 'closed' triads over a large range on a keyboard, but as you get progressively lower it sounds muddier as the harmonics enter the range of human hearing and all those clashes become more noticeable.  That tends to be when arrangers/musicans shift one of the notes up an octave, to reduce said harmonic clashes.

These considerations can be accounted for when you're using a pitch shifter.

They can?  Even if you reduce the amplitude of the harmonised note, it still has it's own set of harmonics that have a harmonic relationship with the other/original note and its harmonics.  Remember that you're dividing down so notes are being generated below the fundamental.  Now remember what I said about range. 

[Taylor]

Thinking about this further, it's interesting to note that the fifth harmonic of the overtone series, and subsequently the fifth drawbar of the Hammond organ, is 2 octaves and a major third above the fundamental. As you note, it would be expected that this would create some dissonance when playing, for example, a B in the key of A minor. However, reality tells us that organs using this stop don't automatically sound dissonant, nor do instruments with prominent fifth harmonics, like triangle wave synths or woodwinds, sound nasty in the same situation. It seems that having notes separated by octaves makes them progressively less dissonant - for example the minor ninth is dissonant, but less so in my opinion than the minor second.

There's a big difference between the harmonics within a note forming intervals and fundamental pitches doing the same.  Harmonics tend to have lower amplitude and tend to be widely spaced in relation to their fundamental, so are less obtrusive and don't dominate the harmonic 'mix'.  An interval comprising two fundamentals will not only have the harmonic relationship between both fundamentals (which will be of similar amplitude within an octave span), but also any clashes resulting from the two sets of harmonics generated by said fundamentals.  These clashes will be more noticeable if the fundamentals are close together.

Range also has a big part to play.  You can play 'closed' triads over a large range on a keyboard, but as you get progressively lower it sounds muddier as the harmonics enter the range of human hearing and all those clashes become more noticeable.  That tends to be when arrangers/musicans shift one of the notes up an octave, to reduce said harmonic clashes.

These considerations can be accounted for when you're using a pitch shifter.

They can?  Even if you reduce the amplitude of the harmonised note, it still has it's own set of harmonics that have a harmonic relationship with the other/original note and its harmonics.  Remember that you're dividing down so notes are being generated below the fundamental.  Now remember what I said about range. 

No, this is not true, because in this case we are synthesizing a note, which at least in my case would be a sine wave, which of course has no harmonics to clash.

[gez]

No, this is not true, because in this case we are synthesizing a note, which at least in my case would be a sine wave, which of course has no harmonics to clash.

There have been a number of approaches mooted in this thread, so it's difficult to keep up with who's doing/suggesting what.  Even if you limit things to sinewaves, there are notes within the octave span that will form dissonant intervals with one another.  Unless you develop a way of shifting notes up in pitch as well as down, you're going to end up with some pretty dissonant harmonies.  Which is why I asked whether you need all 12 notes within an octave span: perhaps it's better to consentrate on consonant intervals?  And to reiterate what I typed in a previous post, the fixed intervals/chords generated won't necessarily be diatonic. 

All this assumes that you're not creating a keyboard synth here, but are trying to create a harmoniser to be used with something like a monophonic guitar signal.  I thought that was what this thread is all about.  Apologies if this isn't the case.

[Taylor]

Did you read my post above about the fifth harmonic?

[gez]

Did you read my post above about the fifth harmonic?

OK, I've re-read your posts in this thread.  My apologies, all this talk of top octave generators left me with the impression that you wanted access to all 12 tones within an octave at once.  You clearly state that this isn't important to you. 

OK, now the engineer in me is taking over.  In the analogue domain there's no easy way to do this.  In order to obtain pitches higher than the original note - other than an octave - phase-locked-loops would have to be the obvious choice.  As mentioned, you're going to have to parallel those in order to get the full range of tones...or at least parallel a few to get some of the more useful ones.  The latter would keep the parts count down (which is why I was talking about being selective in the intervals you choose).  No matter what, it's going to be a big circuit.

As for digital, I'll bow out now as others have more insight into that area. 

[Taylor]

Is it possible to make the divider chips in the harmony generator (4526BE) do other numbers besides major thirds and fifths? I realize the end result might be out of tune, but I'm curious if the division ratios in the HG are set that way to protect the sensibilities of less experimental users or if that is the inherent nature of the chip.

[gez]

Could you link to the schematic you mention? (I can't find it).

[Lurco]

http://hammer.ampage.org/files/EMMHarmoGen.PDF

[StephenGiles]

This is all very well and very scholarly (not easy to say after 3 gins!)......but as a veteran of the EH Guitar Synth, I can only (once again ) repeat from other similar threads - possibly even this one, that without extracting a good fundamental signal, and I'd wager a hatful of guineas (21 shillings in old English money) against all of the songs you can sing, that the fundamental extracted in the Roctave is pretty crappy, you may as well not bother. Why reinvent the wheel when EH effectively did the work for you nearly 30 years ago, and my tracing of the circuit still sits on Marks site. Tracking VCOs are the answer surely

[gez]

Thanks for the link Lurco. 

Is it possible to make the divider chips in the harmony generator (4526BE) do other numbers besides major thirds and fifths? I realize the end result might be out of tune, but I'm curious if the division ratios in the HG are set that way to protect the sensibilities of less experimental users or if that is the inherent nature of the chip.

Stick your 4017 in the loop of the 4046 PLL (see the article you linked to in your 1st post).

Stephen, yes, most of these circuits are pretty useless without fundamental extraction.  Having said that, yodelling in an upwards direction tends to be less noticeable than downwards, so PLL designs might not be too bad without the extraction.

[Taylor]

This is all very well and very scholarly (not easy to say after 3 gins!)......but as a veteran of the EH Guitar Synth, I can only (once again ) repeat from other similar threads - possibly even this one, that without extracting a good fundamental signal, and I'd wager a hatful of guineas (21 shillings in old English money) against all of the songs you can sing, that the fundamental extracted in the Roctave is pretty crappy, you may as well not bother. Why reinvent the wheel when EH effectively did the work for you nearly 30 years ago, and my tracing of the circuit still sits on Marks site. Tracking VCOs are the answer surely

Easy there. I am planning on putting the FE from the Oc-2 in front of it, as well as filtering and compression. I have looked at the EH guitar synth as well - I may yet try it.

All that said, if I wanted perfect tracking, I would just get an EHX HOG. I actually like the glitchiness. Chalk it up to being young enough to not remember a time when stuff like the harmony generator was "the best we can do right now" - to me, these artifacts aren't just unwanted side effects, they are an important feature.

[Mark Hammer]

All that said, if I wanted perfect tracking, I would just get an EHX HOG....to me, these artifacts aren't just an unwanted side effects, they are an important feature.

Now THAT'S the spirit!!

[Taylor]

Can anyone tell me the pinout of the 4040b IC? Obviously I would normally look at the datasheet, but I'm on a really bad internet connection today and the PDF keeps timing out.

In the harmony generator, the 4040b is doing the octave dividing once the 4526's have done the 3rd/5th division. The 4040b puts out division by powers of 2. What would be cool is to sweep through all integer divisors, which I realize, of course, could give some "bad" sounds, but I like that. Is it possible with this chip to do divisors besides powers of 2?

If this question is dumb/naive, please forgive, I'm a noob with electronic components; my effects designs have previously all been in the computer world, where anything is possible, so I'm coming to grips with the limitations of real life electronics.

[Taylor]

After finally being able to read the datasheet, nah, doesn't look possible. I would edit the above post but I guess my window to do so has past.

So, since you clicked on this thread, but it has no new information, enjoy this: