Walsh circuit now up

[Rene Schmitz]

At 22:33 22.09.99 +0100, Tony Allgood wrote:

>To look at Paul's hex-VCA method, we still have a problem of allowing
>negative value coefficients. That is, a negative CV must produce inverse
>walsh functions. This could be done using the same circuit as mine but
>with decent CMOS switches, eg DG271. However, because 74HC4066s are
>about a tenth of the price and are faster, I have chosen to use them
>instead. To get around the problem of negative coefficients without
>using negative values of CVs, my CVs will produce walsh outputs with
>negative coefficients, when they are lower than a reference voltage,
>Vref. The reference can be varied to allow positive only, or biased
>positive coefficient addition. 

What about this: Using '4053 switches. They have inbuilt level translation,
so they would allow easy interfacing to the 5V/0V logic signals. 
I'd use a mixer with inverting and noninverting inputs.
The two outputs of the switches would switch between one inverting and one
noninverting input. The coefficient voltage would be coupled to the center
of the switch. One would have to use more ICs for the switches because the
'4053 only has three switches per chip, but you save some opamps and
resistors. And the '4053 is also very cheap.

>Actually, since the ref is a voltage, it
>gives some crazy results when this is swept over time. Example, you can
>get a sawtooth changing to a pulse wave. I don't think that the hex-VCA
>can attenuate a signal to some arbitary reference... but to be honest, I
>don't know how that thing works anyway.

I'm not sure if a similar trick could be made with the circuit described.

Bye,
 René 



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