[sdiy] Cheapest Matched Pair

[harrybissell]

Hmmm... i wonder.

I have never tried the 3046 for a dual VCO...

but I have used the MAT-04 and had no soft syncing
problems.  I used the pairs of 1,2,3 and 12,13,14 (iirc)
for the two expo converters...

and I paralleled the pair at 5,6,7 and 8,9,10 ... diode connected
I hoped that this would give the best read of the chip temperature...

It worked well.  Maybe the 3046 has some better and worse pairs
to use in a dual...

but hell, they're cheap so why not use two and not worry

H^) harry

Grant Richter wrote:

> Merry Christmas to All,
>
> One other caution on the CA3046. If you use two sets of transistors
> internally to make the expo converter for two VCOs. The VCOs will tend to
> "soft synch" probably due to capacitive coupling in the CA3046.
>
> Using Rossum compensation (diode and resistor) you can make an expo VCO that
> tracks +/- 5 cents or better over 8 octaves.
>
> >
> > Joe, Oren etal --
> >
> > The CA3046 is indeed speced to 5 mV matching, but in practice it is usually
> > much better due to improvements in Si technology in the years since the device
> > originally came out.
> >
> > Also the beta is usually much better than speced.
> >
> > The main drawback of this array is the relatively high emitter series
> > resistance.  This needs to be compensated for in critical converter
> > applications.
> >
> > If you can find it, the CA3083 is quite a bit better than the CA3046 in this
> > regard.
> >
> > Seasons greetings to all!
> >