Ping: Gurus. MAT-04 EXPO for (2) VCO

[jhaible]

I've built a SEM VCO clone (with a shared CA3086 for both
VCOs' expo converters) with no unwanted soft sync at all.
OTOH just putting the components of the oscillator core in
close proximity to each other on the PCB caused some trouble.
So I reckon the isolation thru a transistor array in that configuration
is rather good (switching remnants from one VCO to the other
would have to enter the array at one collector, seeing a Ccb
capacitance to GND as a shunt, before it cand find a way into
the other part of the array over a Cbe capacitance to start with)
If you're using a dielectrically isolated array, connect the
substrate to GND. If you want to
be safe, add a JFET on each collector to form 2 cascodes
(as shown in the MS-20 schemos).

JH.

> I've done it using all sorts of stuff. I'm pretty sure mat-04s will work
> too. I discovered once that a 4007 could be used instead of J112s
> and built a dual vco that used a single 4007 for both chopper fets.
>
> This was a bad idea! It seemed to work fine but in fact did not. My
> point being that I was trying to be just too damm cheap! You can
> probably avoid lurking problems and future fixes by using two,
> regardless the fact that one is more than likely fine.
>
> > I'm building the Chris MacDonald "Mini-Modular"
> >
> > I'm considering using the MAT-04 for the expo... in fact I'm even
> > thinking
> > on ONE MAT-04 shared by two oscillators.
> >
> > The MAT-04 has a separate substrate so that should not be a problem...
> > but
> > what about soft-sync, or even thermal effects between transistor pairs