"Stacked" Power Supplies - any thoughts?

Magnus Danielson magnus at analogue.org
Wed May 20 00:25:11 CEST 1998

>>>>> "S" == Stopp,Gene  <gene.stopp at telematics.com> writes:

 S> Regarding the commutating filter, I intend to find out what it sounds
 S> like! I did test it out when I built it in the lab back at my old job
 S> (lots of spare time right before we closed the doors on the place) but
 S> it was listening to a sawtooth wave from an HP function generator into
 S> the gizmo over headphones. I remember the impression that I got was very
 S> flanger-like. But I reeeeally want to see what it sounds like with a few
 S> VCO's into it, and maybe some nice crisp white noise. You may remember
 S> that it basically consists of a pair of switched-cap integrators in a
 S> state-variable configuration, with a dual pot where the two integrator
 S> R's  would normally go and labelled "Bandwidth", plus a Q control....

What switching frequency?

Can you modulate the switching frequency?

When the switching frequency is well above your signals bandwidth (and
ears ;) then this would act very similar to the digital IIR
filters. They will need the Z-transform to analys correctly (I have
several references to this if someone is die-hard on theory) and they
will certainly follow a diffrent filter-function than normal
state-variables do. You will even find that it's high-end properties
is quite obstructed as compared to the normal state variabler filter,
this is of course due to the wrapping of the jw axis into a circle!

If you can modulate the switching frequency, possibly PWM it you can
get some possibly interesting effects. If you run the frequency down
in the audio range you get a nice wrapping effect which effectively
act as a waveshaper (and a VERY unlinear one).

I can also imagine that some of the above effects would be about the
lamest thing that happend - "Look, we just got ourself an additional
DC term!" or similar... experimentation will show...

Did you use CMOS switches for this one?

 S> I hope it's weird.

Jaaa, I bet'ya! (In that funny "Fargo" way).

 S> Regarding the power supply thing, I'll probably try decoupling at the
 S> loads, or at the far end of the rails, or some combination of that. It
 S> would be interesting to see if the compensated feedback loops work with
 S> each other and be stable, or against each other and ring and wobble.

Try induce instability by suddenly add/remove load and watch your


More information about the Synth-diy mailing list