AW: exponential VCO problems

[Eric Barbour]

Haible Juergen wrote:

> How do you build a tube opamp?

See my articles in past issues of GLASS AUDIO:
"Vacuum Tube Op Amps Applied", 3/92, p. 12, and
"Vacuum-Tube AC Feedback Amplifiers", 2/92, p. 14
You need the 1992 back-issue reprint--it is US$23 plus
shipping (I think it is $4.50 in the US, $9 airmail
to Europe.)
It is available from GLASS AUDIO, PO Box 176,
Peterborough NH 03458 USA. You should also subscribe
to GA, it will teach you a lot. (Not about synthesizers,
though....their only interest is audiophile amps and preamps.)
 
> Now, couldn't we build
> the integrators of a state variable with a simple cascode
> stage, with capacitive feedback ?

Yes, but you still need to direct-couple, to avoid the low-frequency
pole. 
See the articles above. The op amp scheme is easier, take my word for
it...
although tube opamps do introduce their own problems; limited frequency
response, which intereferes with the pole-zero response; and the need to
zero out the DC offset occasionally.
 
>         >more if tubes are used as voltage-variable transconductances.
> I have no idea how to do this. Just out of curiosity: How would you
> do it, and how many decades of cutoff frequency would you expect
> to control that way ? 

Supposedly you use them like voltage-variable resistors, with a
keepalive
current always flowing from plate to cathode. Another way is to
use a "remote-cutoff" tube, hooking it up like a regular amplification
stage
(plate-loaded), and using the CV to control the bias point of the tube.
Yet another way would be to use a pentode and vary the screen-grid
voltage
with the CV.

I have yet to try most of this, I did try the variable-resistor scheme
and it did not work very well--quite nonlinear. It is NOT an optimum
method.
The screen-grid method is used in my VCA and to vary Q in the VCF, it
seems
to work OK. Distortion is high, but then, that's part of the unique
sound.

-- 
Eric Barbour
Svetlana Electron Devices
Portola Valley CA USA