Minimoog VCF

[Rene Schmitz]

Hi Jörgen,

At 09:07 12.03.99 +0100, you wrote:
>Hi Rene
>
>I have a couple of questions regarding the filter.
>I get the impression that you have actually built it and that it worked as
> it should.

Well, I can't test that of course, 'cos I don't have a mini! 
(Send all unwanted minimoogs....)
But given, the sounds I've heard from the mini, and what has been said by
Sean, I now think that it must be very close.

>I'm a bit surprised that you could change the supply voltage from 10 to 15
volts without changing a lot of resistors.

Well the moog-VCF of Rick Jansen works with 15V and the same resistor
values for the ladder-transistors. 

>The only changed resistors that I can find are R70, R74, R6 and R7. This
> surely has to do with the CV range and the resonance.

I've also decreased the cap values. This raised the maximum frequency. 
I've found that with R70 being 1k8 and the CV-summing resistors of 100k I
could not get a 1V/oct setting, so I reduced R70. I increased R7 to get
more loop gain, and I increased R6 to get a more symetrical current flow
thrugh the diffamp. With the collector of Q6/Q5 connected to +15V I found
that Q5 got a little warm, so I decided to reduce the current. 
I suspect that this can cause large Vbe differences in the diffamp,
shifting the operation point. Initially I had trouble with the
self-oscillation when Q5 had warmed up. (And I have killed a 548 when
probing arround, shorting Q5s emitter to GND!)

>The voltage at the range pot is still 10 volts. Is this intentional?

Nope, I've overlooked it, I've corrected it and uploaded the new image.

>Have you tested the temperature stability? Should Q26 and Q28 be in
thermal contact? Would it be a good idea to use a tempco resistor for R70?

I've not made many thoughts about it, since the amplitude of the
selfoscillation depends on the frequency, one would hardly use it as a
"sinewave-VCO". When used as a tracking filter with added envelope
modulation, it would hardly matter if its stable or not. I'd say that Q26
and Q28 should be coupled thermally, but I think full compensation is not
necessary. 

Bye
 Rene

             ,   : (uzs159 at uni-bonn.de)
|") [" |\ | ["   : (http://www.uni-bonn.de/~uzs159)
|"\ [_ | \| [_   : (http://members.xoom.com/Rene_Schmitz)