AW(Andre): [sdiy] analogue phase modulation

[Ian Fritz]

At 05:44 PM 1/11/2007, JH. wrote:

>Actually, I plan to do this:
>
>A = 10V * sin (CV * Pi / 5V),  B = 10V * cos (CV * Pi / 5V),
>
>for +/- 5V control voltage input range, and 10V output swing to drive a pair
>of AD633's.
>
>The sin/cos shaper circuit needs a handful of opamps to create polygon
>approximations, and a 13600 with slight feedback for rounding.

OK, I've got it now. Thanks.

I wonder how accurate the shaper circuits need to be.  I showed before a 
circuit using 2 Zeners and an opamp to give a rough approximation to sin(x) 
or x - x^3.  I used this in my quadrature oscillator and in my 5-phase 
chaos generator.  A similar cos(x) or 1 - x^2 should be easy also. I would 
speculate that these could be trimmed to give proper operation for small 
modulation with increasing inaccuracies at higher modulation levels.

You might also want to think about possibly using an OTA 4Q multiplier 
rather than an AD633. A single CA3280 chip?

Anyway, sounds like an interesting project, especially getting thru-zero 
modulation built in.

   Ian