AW: Formant (and jbv!)

[Tony Clark]

> What you need is thermal coupling, plus regulated heating,  or 
> 3300ppm temperature compensation.
> 
> Sorry for the bad news. But buying expensive dual transistors
> is completely in vain without temperature compensation. (You need
> it even with perfectly ideal transistors).

   I agree with this!  This is why I use an MAT-04 in my VCO design.  You 
can achieve nice regulated heating by using a transistor as a heater and 
then use the transistor opposite the heater on the chip die as the 
regulator.  That way heat is distributed evenly to the two transistors 
making up the current mirror.
   It works very nicely as far as I can tell.  I've monitored the output 
frequency and the output of the regulator and watched it do its work 
while externally heating and cooling the chip.  I observed no 
fluctuations in frequency at all.  Also, the regulator does not cycle but 
is a constant DC source, so there are no frequencies being induced into 
the circuitry.
   You can use the National AN-299 Figure 1 to get an idea about how to 
do it.

   Tony

-----------------------------------
I can't drive (my Moog) 55!
-----------------------------------
Tony Clark -- clark at andrews.edu 
http://aupe.phys.andrews.edu/~clark
-----------------------------------