thru zero FM on 3340

[Haible_Juergen#Tel2743]

Hi!

Built thro-zero-FM into my 3340-based dual VCO module this weekend.
I used the method suggested in the synthesource (sp?) reprint of
Barry's IC book, with a rectifier and some XNOR gates. It was
intended to be with the 3345, but there was a workaround shown for
the 3340 as well.
It works as it was described: Fine, but with some unpredictable
beatings on several notes with different modulation depth.
I really like this kind of beatings - they give some life to the sound,
and it's very pleasant. Sounds a little like some type of sync
(well, sync *is* part of this method!), but without the boring
phase-lock of sync.
There's another effect, that wasn't described in the paper: as long
as you don't go thru zero, everything is fine and smooth. When
you go heavily thru zero, everything is fine as well - not the slightest
shift in frequency, just a change of timbre! - but at the very edge
to zero, when the rectifier path just starts to come in with tiny tips,
things are getting a little disturbed and noisy. So there's some
limitation for a useful range of dynamic sweep of FM depth.
OTAH, it sounds quite pleasant with fast envelope times - gives
a certain "attack" to the sound.
All in all, I am very happy that I  have retrofittet this in my VCO module.
I can produce some Tangerin-Dream-like solo lines now that
I couldn't do before.
But anyway, I'd like to learn about the reasons for these side effects I
have described. Any idea anybody?

Thanks,

JH.

PS: If you use VCO1 to make thru-tero-FM on VCO2, and then you
        sync VCO1 to VCO2 (i.e. closing some weird loop), you get
        very strange and chaotic sounds!