PWM a la PS3x00 (was: Re: [sdiy] why 10V

[jhaible at debitel.net]

> > >(PWM), but in practice there we keep the peak-to-peak voltage
> > >stable and let the RMS voltage drop
> > 
> > Notable exception: Korg PS series. Here, the pulse amplitude is changed
> with
> > pulse width. Unfortunately, it doesn't preserve the RMS value, either,
> > because it makes shorter pulses lower instead of higher.
> 
> OUPS!
> 
> Somebody have had to do something like it anyway.
> 
> Now, did they spend extra hardware to get it or is it a consequence of the
> hardware design?

It's a consequence of the hardware design. An increadible piece
of circuit design, actually: A single (!) transistor, plus
two resistors and one diode are enough to create
triangle, saw or (fixed and variable) pulse waves, depending
on two control voltages. 

Se my redrawn (changed, also, but the waveshaper is almost
like the original) circuit at  
http://www.oldcrows.net/~jhaible/polykorg/sg_1sch.pdf

Q12 (upper right corner) is the waveshaper for one note 
and oscillator (96 such circuits in the PS-3200). Feeding
the right CVs over "WFD" and "WFR" will either
* fold the saw input into a triangle (like the Moog 
  triangle converter), or
* pass the saw unchanged, or
* use the transistor as a voltage clipper.

The latter mode (clipper) will not only change the pulse
width, but also the pulse height, depending on the
"WFD" voltage thar sets the clipping level.

Clever, these Korg designers, aren't they?

JH.


-------------------------------------------------
debitel.net Webmail