Negative frequency

[Dan Gendreau]

> well, I don't know if I get it or not. why is it called "negative
> frequency"
> when all that happens is you turn the frequency down to zero and them back
> up again?

Think of a VCO thats being controled by a CV. Imagine moving the CV from 1V
to -1V. Youre not turning the CV down to 0 and back up. Your running it from
1V to -1V.

The VCO will harmonically the same at 1V and -1V, but the trick is that a
thru-zero VCO responds with the correct mathematical waveform even as the CV
crosses through 0V. Many VCO designs cant handle this situation and their
behavior is undefined.


> the most you'll see is a phase change, surely?

You are right. A negative CV just causes the VCO to run in reverse. And in
the case of a stationary CV this just results in a phase reversal and the
exact same harmonic content.

But when you modulate rapidly back and forth across the 0 point, say at
1kHz, you can get some unusual waveshapes and harmonic content which are not
possible when modulating back and forth across, say 2V. The harmonic content
is also drastically affected by the waveshape of the CV waveform as well.


> and don't "real" rotary algorithms use a combination of vibrato
> (for doppler effect) and pan?

The original Yamaha DX-7 was a mono FM synth. Several of the better DX-7
organ patches simulate the Hammond B3 organ sound, but not in stereo. They
did it by modulating 6 sine wave VCOs in a particular arrangement. What Don
was explaining was that the CV inputs of some of those oscillators actually
go between negative and positive CVs. This causes some VCOs to go quickly in
and out of phase with eachother causing the shimmering Hammond sound.


> sorry for being stupid, I've never come across this phenomenon before.

Youre asking very valid questions. FM synthesis is a very weird realm
compared to the standard VCO->VCF approach. The DX-7 for example has no
filters whatsoever. All the harmonic content is created purely through FMing
those 6 sine-wave VCOs.

-Dan G.