Nice Analysis of Minimoog Filter and Oscillators

[Joachim Verghese]

Martin Czech wrote:

> I can't understand how a darlington output stage may alter the feedback
> at *low* frequencys, but this is *linear* thinking, and this darlington
> differential might introduce considerable distortion when the amplitude
> rises (feeback rises).

I'm not sure if this answers your question, but the darlington
buffers in the Minimoog filter do load the top stage of the
ladder, effectively forming a high-pass with the top ladder
capacitor. This high-pass attenuates feedback at low frequencies
and lowers the amount of resonance as the cut-off frequency is
lowered.

In some Moog designs (Taurus I, etc) the ladder signals are coupled
directly into a CA3080 which again loads the ladder. All ARP versions
of the ladder filter, as well as later Moog designs, used FET buffers
(discrete or op-amp) which do not load the ladder, and thus make
resonance virtually independent of frequency.

Thanks to Sean for forwarding the original message -- nice reading
indeed.

BTW, I was surprised to notice that there's no info on the SSM22x0
dual transistors at ADI's web site. They're not in the obsolete
parts list either. Farnell and Maplin still have them in stock,
I think.

-joachim