[sdiy] Buchla 257 is an extremely strange circuit

[Tom Wiltshire]

These triangles are quite pretty, but you finish using three  
parameters to control the mix. Which isn't any advantage over using a  
simple level control for each one.

If you can cope with only having a mix of two of the signals  
simultaneously then you can do it with JH's interpolation circuit, as  
has been mentioned. This can be thought of as arranging the signals  
equidistantly on the circumference of a circle, and then using an  
"angle" parameter to define the mix.

It's a similar problem to the Prophet VS vector mixer, which provides  
a good subset of all-possible-mixtures, but doesn't allow every  
combination. What limitations are you willing to accept?

T.

On 16 Dec 2009, at 14:03, Jerry Gray-Eskue wrote:

> A second thought if you do not like the resistive triangle..
>
> Use three slide pots arranged as the triangle with signals on the  
> points and
> ground in the center.
> Add a plate with 3 slots at 90 degrees to each slide travel with the
> matching wiper through each slot.
> Add a central control knob to the plate and allow free movement of  
> the plate
> but prevent rotation.
>
> - Jerry
>
>
> -----Original Message-----
> From: Jerry Gray-Eskue [mailto:jerryge at cableone.net]
> Sent: Wednesday, December 16, 2009 7:49 AM
> To: Synth-diy at dropmix.xs4all.nl
> Subject: RE: [sdiy] Buchla 257 is an extremely strange circuit
>
>
>
> This approach captures the way I visualize a three way mix. See the  
> diagram
> labeled
>
> [Barycentric coordinates (?1,?2,?3) on an equilateral triangle.]
>
> What I would be tempted to try is forget about all the math  
> involved and use
> the triangle in physical form.
> Imagine:
> (1) the diagram as a resistive material like used for ribbon  
> controllers.
> (2) Each Point (0,0,1) (0,1,0) and (1,0,0) is the signal injection  
> point for
> the three signals.
> (3) Add a conductive wiper allowed free movement over the entire  
> area of the
> triangle.
> (4) Feed the signal from the wiper into low impedance voltage  
> follower.
>
> Bingo! a pure analog mechanical solution of very simple order that  
> should
> not be too difficult to fabricate.
>
> - Jerry
>
>
> -----Original Message-----
> From: synth-diy-bounces at dropmix.xs4all.nl
> [mailto:synth-diy-bounces at dropmix.xs4all.nl]On Behalf Of Oren Leavitt
> Sent: Wednesday, December 16, 2009 2:10 AM
> To: sdiy
> Subject: Re: [sdiy] Buchla 257 is an extremely strange circuit
>
>
> Hmmmnnn....
>
> It be interesting to try an analog implementation of Barycentric
> interpolation. Just an idea.
>
> http://en.wikipedia.org/wiki/Barycentric_coordinates_%28mathematics%29
>
> - Oren
>
> Derek Holzer wrote:
>> And to make it a bit more complicated, what if I wanted to cross-fade
>> between *three* signals, lets say x, x^2 and x^3? Any ideas out there
>> for that one?
>>
>> D.
>>
>> Derek Holzer wrote:
>>
>>> Hi Aaron,
>>>
>>> my apologies of this thread is dead and buried, but I'm  
>>> interested in
>>> analog computer functions and ways of "interpolating" or cross- 
>>> fading
>>> between them, or between the input signal and the processed signal.
>>> This PWM AM sounds rather complicated to implement, would there be a
>>> simpler VCA-based solution? Something like the Serge sound  
>>> processing
>>> module with it's voltage controlled cross fader? or am I missing
>>> something important in the concept of a CV crossfader/interpolator?
>>>
>>> Best!
>>> Derek
>>>
>>> Aaron Lanterman wrote:
>>>
>>>> I know using "Buchla" and "extremely strange" in the same sentence
>>>> isn't exactly news, but this is even stranger than usual.
>>>>
>>>> The 257 is the Dual Control Voltage Processor. It has a means of
>>>> crossfading between two CVs by means of a third signal.
>>>>
>>>> The way it does this is totally odd. There's a fixed frequency
>>>> triangle wave oscillator running at 21 kc, and then the controlling
>>>> CV is added to that. This is then run through a comparitor, so you
>>>> get a pulse wave whose duty cycle is determined by the CV.
>>>>
>>>> This pulse wave then controls some CMOS switches, set such that you
>>>> get a signal that is CV1 part of the time, and CV2 part of the  
>>>> time,
>>>> depending on that pulse wave. So depending on the duty cycle, it
>>>> spends a particular amount of time at CV1 vs. CV2.
>>>>
>>>> Then, this goes through a boatload of lowpass filtering to  
>>>> smooth out
>>>> things and average the signal.
>>>
>>>
>>>
>>>
>>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy