Gene's DIY OB-8/CS-80 Ribbon Controller

[WeAreAs1 at aol.com]

Regarding his DIY OB-8 ribbon controller, Gene Stopp wrote:

<< And the ribbon - I'll try it like this for a while (linear seemed intuitive
to me). Does a CS-80 ribbon have a non-linear response, like as you go
towards the higher frequencies you need more distance travelled for the same
musical interval? >>

It's been a while since I last played a CS-80, so I don't remember for sure 
about the linearity of the ribbon's pitch response.  The ribbon is two 
octaves in physical length, and I'm pretty sure that its pitch pretty closely 
matched the pitch of the keyboard notes that were right below it.  That is to 
say, you could closely approximate chromatic melodies (Theremin-style), 
simply by sliding your finger up and down to the location above the keyboard 
notes you wanted to play.  Again, it's been a while, so I could be wrong 
about this.  However, I do have schematics, so here's some tech info on the 
ribbon circuit:

Each CS-80 voice has a crude exponential amplifier (in order to achieve the 
keyboard-position-matching response), since the Yamaha VCO's had linear CV 
response.  This exponential amp is used to provide the VCO's with the ribbon 
CV and octave transpose CV (there's a separate tuning trimmer for each of the 
four octave ranges and another trimmer for the ribbon scaling, though, 
eliminating the need for a real accurate converter).  This CV is fed into the 
"FT" input on the VCO IC's, not their Keyboard CV input (which has linear 
response, and gets its CV directly from the 8-channel D/A output via some 
S/H's).

Strangely enough, though, the CS-80 ribbon doesn't feed this expo amp 
directly.  It connects to one of the Yamaha custom IC's, a custom 8-channel 
D/A chip called a YM27600, and sums with the octave transpose info at that 
point (I think... the documentation is confusing as heck), then it goes on 
through some S/H's, then to the expo amp/octave select gates, then to the 
VCO's.

However, the most interesting thing about the CS-80 ribbon is its ability to 
make wherever you first place your finger the "zero" pitch point.  For 
instance, you can put your finger down at the middle of the ribbon, and then 
you can bend up/down one octave in either direction sarting from there, or 
you can put your finger down at the top of the ribbon, and that becomes the 
zero point, from which you can bend DOWN the entire ribbon's two-octave range 
(or conversely, UP two octaves, from the bottom of the ribbon).  This also 
means that you don't ever have to be real careful about exactly where you 
place your finger when you first touch the ribbon (the pitch won't "jump" to 
your finger's absolute position, it just thinks of that position as the new 
relative zero point).  It does this with a clever little pair of simple FET 
S/H circuits and a pair of opamp comparators.  I'd be happy to send you a 
drawing of this circuit if you're interested in trying it on your homemade 
ribbon.  It's not very complicated at all, and could probably even have a 
switch to disable/enable its "finger position memory" function, if you wanted 
to read absolute position.  And since your OB-8 VCO's are exponential, you 
could also have a much larger than two octave range, if you wanted.

Corrections from CS-80 experts encouraged...

Michael Bacich