> >I was planning on using a topology that had the LM1973 in the *feedback*
> loop
> >of the microphone preamp stage. Therefore, it would get its drive from
> the output
> >of the mic preamp, after 40 dB or so of gain and attenuate it back to
> microphone
> >level before feeding it to the inverting input.
There's a problem with this approach. The LM1973 has a high output
impedance (of the order 20k), whereas a typical microphone has a low output
impedance (200Ohms or so). Thus, for best noise performance, you would
ideally use a FET amplifier input for the attenuator, and a BJT input for
the microphone. This presents a problem since integrated OP-amps are
either one or the other. In fact, the datasheet stresses that you should
use a FET at all costs, but as I said, this is a lousy ( = noisy!) match
for a low-impedance microphone. You could buffer the output of the LM1973
with a FET opamp, and use a BJT opamp for the gain stage, but you'll have
to watch for stability issues connecting two opamps in a loop like this.
> >They spec these chips at over 100 dB S/N,
Look closely - that's referenced to 4Vrms, which is a HUGE signal! If the
self-noise is 100dB below 4V ( = +12dBV) that puts it at -88dBV. Given the
standard consumer line signal level of -10dBV you've got a best-case SNR of
only 78dB. Decent microphones will do a lot better than this. It is not
clear from the datasheet how this SNR spec has been arrived at, so the
situation could actually be a lot worse.
Christopher Hicks
-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]
