Xpander VCF

[Martin Czech]

:::But on *this* I disagree !
:::
:::In *time domain* you can simulate an evolving self oscillation.

Ok, of course you can, but you have to be very carefull.  I should have
chosen other words, I should have said that the "results" for instable
circuits should be crtitically observed. The "solution" might be totally
wrong, may have nothing to do with reality. It depends on the actual
mechanics of the integration process.

A stable circuit should simulate similar on different integration
machines. Sometimes even stable circuits won't show any convergence.
Unstable circuits will simulate different on different integration
machines.

Closed loop simulation is not accepted for any statement about analog
circuits in our analog dept. Certainly not for prooving stability.
In fact, spice simulation is not accepted to proove anything here ;->

Too often too wrong.


m.c.