Moog 904A lowpass filter Question

[tomg@efm]

I've seen the pin 13 rule broken quite a few times. My experience is that
if you use a transistor for the chip heater pin 13 needs to be more negative,
otherwise not necessarily. When using just the 12345 pair I have found
making just pin 13 more negative causes severe drift..... sometimes..... It is 
"fixed" by putting all unused pins to gnd..... sometimes.....I don't like 
leaving pins floating on the 3046 at all anymore and when making pin 13 
more negative (about -1V) tie 12,13 and 14 together and then gnd the rest. 
An example of this is the VCO4 version on the SBM. You can use a 1583 
or a 3046 12345 pair. All unused pins on the 3046 are at gnd because it is 
more stable that way.
 
Gene told me once just to use it whenever I wanted. I have found that pretty
much any circuit that I use all 5 transistors in seems to work regardless of
the voltage on pin 13.....I'm sure I've just had a very long streak of good luck 
and one of these days it'll bite me. The only real advice I can give is the app-
notes say it should be more negative. Try this and if it works fine if not then
ignore the app notes and do whatever it takes to get it to work....That helps
a lot huh?....

Tom
  

> > BTW, does anyone know for sure it is OK to leave the substrate connected
> > transistor (pins 13,12, and 14) floating rather than hard wiring pin 13 to
> > v- ?  In that case it would seem to me that diode action with the other
> > transistors would bring the substrate at least as negative as any other
> > transistor.
> 
> I don't remember in which circuit, but I'm quite sure that ARP were so
> bold (and probably so experienced) that they even broke the rule
> of pin 13 being the most negative of the array. In that circuit pin 13
> was slightly more positive than the emitter of another transistor of the
> array, but not so much more positive that a significant current would
> be drawn. Joachim, maybe you remember that circuit ? I only remember
> that we talked about that ...
> 
> As for ARP's "Moog" filter, maybe the extra diodes were just in there
> to confuse patent lawyers ? Might be worth a thought, given the situation
> at the time ...
> 
> JH.
> 
> (just back from Poland, switching 20 MegaWatts at 110kVolts, so
> expect a confused mind being directed to microAmperes again (;->) )