thru zero VCO questions

[Haible Juergen]

	>Hm, why switching at the zero crossing ?! To me it makes only sense
when I
	>rectify the FM input to put it into the expos ref. input. But one
can also
	>directly integrate the FM input. Much like in two opamp LFOs. As I
see it I
	>have to toggle the direction when either the upper or the lower
threshold
	>is exceeded. When the FM input changes sign, so does the
integration
	>direction, I don't need to switch anything for that. I only need
the FF to
	>remember the last direction. When either of the thresholds is met,
then the
	>polarity of the FM input signal is inverted. 
	>
	>So my idea looks like this:     
	>							
	>FM in -> polarity switch -> OTA integrator (Iabc = expo I) -> 2
comparators
	>->ORd together into clk of toggle D-FF, output controls the
polarity switch.

Rene, I'm sorry, I didn't get the idea the first time. It's much clearer
now.

	>Since I don't want to do that, I'll answer with an other question. 
	>
	>Why fullwave rectify the modulation signal and then restore it via
a
	>switched optional inverter. In other words, its not neccessary to
put the
	>FM input into the expo convertors linear FM input. A normal
integrator is
	>already a "thru zero" integrator, but the hysteretic switch doesn't
respond
	>right.

That's a very nice idea indeed !
You still have some switching, don't you, but you keep it outside the main
VCO loop, and only swich the polarity of the modulation signal instead. (?)

I wonder if you really need a flipflop and the pulse / toggle stuff. The
input
of the OTA already carries your up/down information. - why not switch the
polarity of the modulation signal directly from that point ? (should work
with a schmitt trigger - no flipflop needed at all ...)

JH.