Volt/Octave ADSR

[Martin Czech]

> Hello list --
> 
> I am building some V/Oct modules, and it occurs to me that with V/Oct you want
> an ADSR that produces linear ramps rather than exponential curves that you get
> from charging a capacitor. Does anyone have any info on this concept?

Depends on what you want (like allways).
If you want the same time intervall for every octave you sweep, then yes,
linear.

You simply need an integrator circuit for that.

If you add negative/positive feedback to the integrator circuit, then
the envelope looks like the usual expo stuff (with asymptote) or some
undamped stuff (going off to Vsup), so you need some control anyway to
stop charging at some point.)

The envelope of the very basic harmonic oscillator (state var with
Q=infinity or damping=0) looks just like this: On the start the circuit is
undamped, the envelope is something like c1*(exp{t*k1}+c2) (k1 positive
!), if you stop the oscillation by damping it is like c3*(exp{t*k2}+c4),
now k2 negative and t=0 for the point of time when you start damping.

So both negative and positive damping values make sense somehow.

This is a system where  energy is continously fed in and eaten up by some
damping at the same time. (Of course the damping must have some nonlinear
behaviour to get a finite amplitude).

Other systems like a plucked string need only negative damping,
you have an initial amount of energy which is eaten up
if the string is released.




You should check the archive, someone (J. Vergese) posted a clever
concept and circuitry about that.

All from my head, sorry if any error.

m.c.

------------------------------------------------------------------------
m.c. (alias mad scientist): 3 CDs à 72 min., Elektronic Music info, 
synth do it yourself info, Synth-Diy Mailing List Archive (Grey Pages) 
all @ http://www.geocities.com/SoHo/Museum/4459/index.html