[sdiy] Buchla 295 10-band comb filter topology

[Gordonjcp]

On Fri, Nov 25, 2022 at 03:42:02PM -0800, brianw wrote:
> 
> A quick visit to Wikipedia says that PM does not alter the frequency, so I'm wondering how the DX-7 implemented that.

It does alter the frequency, it just always ends up back in the same place.

Think about it this way - if you measured the speed of a rotating disc by timing how long one mark on the rim takes to go past, you will measure its average speed per revolution but you will not be able to detect if its speed changes during one revolution. If the disc was driven from a shaft with an angled universal join it, then every 180° it will be going faster, and offset by 90° from that it'll be going slower because with the shaft at an angle its end is rotating through an ellipse.

If you put marks all around the wheel, every 10° say, 36 of them except one is missing to tell you where the "start" of one turn is, you'd be able to see that a graph of the speed would look like a sine wave. You'd also have invented a 36-1 trigger wheel as used in car electronic ignition systems!

In any case the trick works because the modulator is symmetric above and below the zero line, and if there was a DC offset it actually would change the frequency.

-- 
Gordonjcp