On Wed, 08 Dec 2004 21:45:19 +0000, Pooh Bear
wrote:

The value as of 1991 (for the charge on the electron) is 1.60217733 (49)
x 10¯19 coulombs.

http://dbhs.wvusd.k12.ca.us/webdocs/...on-charge.html

So a gain stage operating at 1mA would have 6.25 x 10^15 electrons
passing through it per second.

That's 1940 times more 'equivalent resolution' than perfect 24 bit
192kHz sampling. Not much granularity there !

The potential issue, as I (poorly, no doubt) understand it, is not
in granularity of collector current, but rather in granularity of
base signal current. And DC values aren't really relevant anyway.
Also, I think the results should be independent of frequency.

What's interesting should be the number of injected electrons or
holes for a typical modern device at typical modern currents at
signal levels near the noise floor. You're probably a better
authority than I about what some good talking values might be
for those three numbers.

For convenience, maybe we could start with defining signal level
as -134dBu minus 6 dB for balanced input, for a working level
of -140dBu. Nice round number.

Next, how much base signal current does this cause in a typical
modern device at typical modern operating points?

Thanks for your thoughts,

Chris Hornbeck
"Shi mian mai fu"