Analog comps, mixturtrautonium

[Dr. Jörg Schmitz]

I have found the URL with the description (scientific paper) of
modelling 
nonlinear processes with a (very very) simple analog computer.
As you will see, the design of the multiplier is much more simple
as some of your comments, it is only a ONE quadrant multiplier.
I think, it is not necessary to use a perfect analog computer for
most of the patches.
The advantage of this design is : it is also possible to DIVIDE,
cause it works via : logarithm -> addition/substraction -> expo ->
result.
(Which is not possible with a ring modulator, if i am not wrong)
I have started some experiments with the multiplier yesterday, and
i found, that the schems in this paper are very simplificated.

http://phyvax.ir.miami.edu:8001/ndl/martin/main.htm


Another question : is here anyone who has done some experiments
with a so called bandmanual, which is used by Oscar Sala in his
"Mixturtrautonium" ?
I have build up a claviature with two manuals, the hardware design
is not so complicated. I have some problems with the design
of the CV output. 
If you use a S/H circuit to hold the last CV played with the string,
the CV-value always jumps a bit down. This is not due to the
S/H-circuit,
I tested it out. I think, the problem has to do with the current
instability,
which occurs in that moment as the contact between ground plate and
string is opened. Exactly at this time, the S/H-gate goes from 5V to 0V.
I know, that the Mixturtrautonium is not constructed for this purpose,
as a note only should occur, if the string is pressed down. But it
would be interesting and musically sinful to include a release phase,
even if the string is released.

Jörg (Sorry for my ugly english)



-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 2943 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/19980905/fee6305e/attachment.bin>