X-Message-Number: 19391
Date: Tue, 2 Jul 2002 15:37:02 EDT
Subject: Re: #19388

Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

> Unfortunately, 100 Ghz sampling rates are NOT a possibility. Remember, 
> light
> is a particle as well as a wave, and if you're sampling at higher than the
> rate photons are arriving, most of your samples will be empty. Naturally,
> the dimmer the source, the lower the rate at which photons arrive...
> Brett Bellmore
> Allows me a small computation: From memory, the Sun has a magnitude near 
-26 (between minus 26 and minus 27). Adding five magnitudes divide energy by 
100 in round numbers. Now the solar constant is 1.3 kJ per second per sq. 
meter at the Earth's orbit distance. Typical optical photons have an energy 
of 3 electron-volts and 1 eV = 1.6 x 10^-19 Joule, that is, the Sun light is 
a shower of 1300/(3 x 1.6 x 1O^-19) = 2.7 x 10^21 photons per sq. meter per 

Take a 4th magnitude star as an example, this is 30 mag. under the Sun or 6 x 
5, as 5 mag divide the number of photons by 100,  30 mags. divide it by 
10^12. So, a 4th magnitude star send us a shower of 2.7 x 10^9 photons by sq. 
meter by second. The Narrabri interferometer light collectors was 7 m in 
diameter or 38 sq. m. So they collected something as 100 billions photons per 
second from a 4th magnitude star.

With a sampling rate of 100 Ghz, there would be one photon per bin, this is 
the most efficient way to detect bunching. Beyond that frequency indeed we 
would gain nothing as all bunch would be resolved. So I maintain what I have 
said: for a typical ii observation a 100 Ghz sampling rate is not only 
possible, it is too the best possible.

Sorry, I like computing and giving definite values :-)

Yvan Bozzonetti.


 Content-Type: text/html; charset="US-ASCII"


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=19391