[sdiy] New System Solution Diagram - Analog Music Synthesizer

[Ian Fritz]

At 03:49 PM 7/7/2004, Harry Bissell Jr wrote:
>--- Ian Fritz <ijfritz at earthlink.net> wrote:
>
> > Fighting words!  I totally disagree with you.  What
> > is important in a noise
> > transistor is lack of 1/f noise.  I once selected
> > through a batch of 50
> > 2N3904's to find the flatest spectrum.  The weakest
> > signal one was the
> > best.  I made the mistake of  paying for a 2N2712,
> > and it is terrible.  A
> > strong output, but a horrid spectrum.  I never
> > thought of trying a
> > supertransistor, but I should have.
> >
> >    Ian
>
>Triple Muah-ha-ha then.  I jumped on National Semi
>about it.  Here is their reply...
>
>Hello Harry,
>Thank you for taking the time to send your comments on
>the Analog Music Synthesizer system diagram.  You are
>correct, the low-noise transistor pair
>LM394 is not a good choice for a noise source.  Based
>on your feedback, we are are updating the query that
>produces the parts list, so that it only
>shows the LM3046 general-purpose transistor array.  In
>a noise generator, one of the LM3046 transistors could
>be a noise source, hooked up as a zener or as a diode,
>depending on the circuit used.
>
>Regards,
>Wanda Garrett
>Web Business Group
>National Semiconductor Corp.
>
>Now I have to say... Ian may have some valid points.
>OTOH National Semi WILL agree with my opinion, as
>they want to SELL the product as "low noise" and not
>imply that they might be "high noise"
>
>H^) harry

Wanda is wong.
It's not "low" or "high" that matters, it's the spectrum. Since you have to 
amplify the signal anyway, you don't even have to waste any opamps in the 
circuit.