[sdiy] Matched transistors - critical parameters?
Oren Leavitt
obl64 at ix.netcom.com
Tue Feb 7 04:09:31 CET 2012
On 2/6/2012 8:07 PM, Matthew Smith wrote:
>
> What are the critical parameters of old matched transistor arrays, such
> as CA3096?
>
> I've been looking at various modern ones, and the only thing that seems
> to stand out is that the hFE of the CA3096 is somewhat higher. (I'm
> comparing with the DMMT5551.)
>
> Is this a critical factor in, say, a VCF, or is it the actual matching
> that is of greatest importance?
>
> Just trying to figure out what the "magic" is that makes the old parts
> desirable over a nice little SOT-26 part costing somewhat less than half
> a dollar.
>
>
In the CA3xxx arrays, the transistors are on the same chip which is
important for best thermal tracking. Vbe match is typically spec'd to
5mV for the first two transistors in some of these. In reality the match
is often much better than that, and often all 5 transistors match
closely, making them great for mirroring control current to sections of
a four pole filter. The CA3083 arrays worked really well and are cheap.
Some of these small devices are multiple single 2N3904, 2N5551, etc,
transistor die encapsulated in one package.
It can still work well your design. The Vbe's are likely to be pretty
close. Put something together and try it - your ears are the ultimate
judge (assuming it's synth related).
- Oren
More information about the Synth-diy
mailing list