[sdiy] So just what *is* the Buchla 291?

[Donald Tillman]

On May 8, 2010, at 10:19 AM, Aaron Lanterman wrote:

> I first heard this in person, thanks to Charles Cohen. It sounds pretty wicked.
> 
> I had a student build it, but I don't understand it.

But does the student understand it?   :-)


> I've heard it called a state variable filter, but when I look at the schematic I can't quite map it to any particular state variable arrangement I am familiar with. I don't see obvious variable gain integrator structures, and the "business capacitors" (i.e. the ones that look like the appropriate values for doing audio filtering) actually all seem to be in the feedback loops.
> 
> Here's the schematic:
> 
> http://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_2910_150.jpg


Oh man, that's strange... 

I don't know definitively, but I think this is a Friend Biquad topology (a couple of feedback paths around a single inverting amp stage).  If true, then:

The main filter is composed of R6/R7, C5, C6, R14/VT3 and opamp 3.

In a normal Friend Biquad there's a resistor across the inverting opamp (opamp3).  That's not exactly the case here.  That resistor has been replaced by R14/VT3, but instead of being across the opamp, it's connected to opamp2.

C2/R87 through opamp2 function to adjust the tuning by virtually adjusting the values of C5, C6, and the effective opamp resistor depending on the resistance of their vactrols (VT2 and VT3).

Bandwidth is adjusted by vactrol VT4 changing the gain of opamp3, and vactrol VT1 adjusting the input gain to compensate.

That's my theory.

  -- Don

--
Don Tillman
Palo Alto, California
don at till.com
http://www.till.com