behavior when CV out of range

[Haible Juergen]

	>Basically, the term that limits the CV range is the mixing op amp
	>voltage swing. Say this is typically +/-13.5V with +/-15V supplies.
If
	>you are using 10V CV ranges, then this range is just fine, with a
little
	>headroom leftover. Past 10V, there is typically no further
significant
	>response in most things I design and no "danger" to the rest of the
	<circuit. One reason some manufacturers may use +/-12V supplies is
that
	>this places the limit closer to 10V.

I must admit I'm no friend of this no-headroom approach. I don't remember
who does stuff like +/-10V signal range (so, no flames intended), but even
with 15V supplies this is practically zero headroom.
How will you design a resonant filter to match such specs ? If the peaks
are within the +/-10V limit, the overall volume will be way below, say, a
raw VCO output signal of +/-10V. If the peak voltage exceeds 10V at any
time, you'll get nasty clipping (very unlike the beloved filter *input*
overdrive).
Think of it, even something as "harmless" as a butterworth filter has some
overshot
for fast input transients that will cause problems in such an environment !
The widely used +/-5V nominal level (at +/-15V supplies) are a much better
compromise, as it gives you 6dB of guaranteed headroom, or 8dB headroom
for "typical" opamp swing. Still not splendid, but workable. 

JH.