[sdiy] Buchla 295 10-band comb filter topology

[mark verbos]

> On Nov 23, 2022, at 12:42 AM, Neil Johnson via Synth-diy <synth-diy at synth-diy.org> wrote:
> 
> Hi Aaron,
> 
>> Then there’s the ultimate question, I often ask when looking at Buchla circuits, which is: HOW DID DON COME UP WITH THIS?
> 
> Without asking him I don't think you'll get a definite answer (so
> that's a "no" then).  My guess is through a lot of actually designing
> and building circuits.  And reading.

Even with asking him, I assure you that you wouldn’t have gotten a definite answer!

His circuits came from books in the analog computing world of the time. I have found cases of him lifting ideas from patent papers that had come out the same year as his circuit. I guess he didn’t worry too much given his tiny quantities! 



> 
>> In some cases, it’s clear he had a basic vocabulary of circuits and generally plunked them down until he got the effect he wanted (there seems to be no effort to go back and simplify things, it seems he was more interested in moving on to something else), but there’s a lot of Buchlaland that just seems like it’s from another dimension.
> 
> Yes - imagination.
> 
>> There’s two main grand mysteries two me:
>> 
>> 1) How did he come up with the “diodeless deadband circuit” in the 259 and Music Easel (I scoured three different books on nonlinear circuits/analog computing trying to find anything remotely like it), and
> 
> It's an optimised deadspace circuit.  Instead of using two diodes
> connected to defined +Vband and -Vband he instead used opamps with
> almost rail-rail outputs and powered them from +/-6V instead of the
> system rails.  That is optimised engineering.  It's the same
> underlying idea of using a single rail low-voltage opamp buffer before
> an ADC running off the same rails as the ADC: you get signal
> conditioning AND range limiting for free.
> 
>> 2) How did he come up with the Buchla 291 topology (I read on a forum that someone asked him and he said something along the lines that it was based on filters Max Matthews used in his vocoder work, but I was never able to find any more information about that to dig in further.
> 
> It's called a bandpass twin-T.  Buchla wrapped an opamp around it.
> 
> The earliest source I have to hand is "Filter Theory and Design:
> Active and Passive", Sedra and Brackett, 1978.  See Figure 8.34 on
> page 467.
> 
> Amusingly, Terry Watson's Masters thesis from 1965 found the bandpass
> twin-T purely by mistake, but didn't find it interesting (see page
> 35):
> 
> https://scholarsmine.mst.edu/cgi/viewcontent.cgi?referer=&httpsredir=1&article=7686&context=masters_theses


On page 44 there is a passive version of the 295 circuit.

Mark