ODP: thru zero VCO questions

[Rene Schmitz]

At 13:15 10.09.99 +0200, Roman Sowa wrote:
>1. whatever will be starting position of FF, VCO's integrator will
>travel from 0V (hopefully) to one of thresholds. When it reaches
>any of them, FF toggles, and integrator changes direction.
>I see no problem here. Noise and metastability - those are
>the problems we're trying to solve. Maybe using 2 sets of
>comparators, and pulse "spray" can fix it completely.

The second set of comparators is perhaps the way I'll go, so a second set of 
higher and lower voltages is monitored as well. 
Maybe even still check if the oscillator has stopped via a watchdog like
Jorgen Bergfors suggested, a simple monostable might be enough. Brute force
discharging the cap to 0V if it doesn't get reset.
I wonder what the probability of such a condition is. It could occur once a
few seconds or minutes or days... The faster the logic and comparators, the
less likely. 

>2. AFAIK FM input will be used only for modulation, so stability
>or distortion isn't such a problem. I don't have objections using
>OTA input for FM. Done that in my quadVCO, where it is set
>to a certain voltage, and modulated from FM input. Works fine.
>I can even get through-zero FM, but very limited.

I see it that way, too. The if distortion of the OTA is 1-2%, its still
good enough. But I'm tempted to try the SSM2164, it suits this circuit.
(Has no linear FM input for the exponentiator, but to the gain cell.) It
would be cool to avoid a PNP exponentiator and all that buzz.  Once the
latchup problem is solved, it'll be of course a matter of taste.

Bye
 René


error:asciiartisnotfully  \ uzs159 at uni-bonn.de
supportedwithyourcurrent   \ http://www.uni-bonn.de/~uzs159
displaydriversettings...    \ http://members.xoom.com/Rene_Schmitz