Re: ATM here's another pulsar idea...

[Bob Bond <rgb@ncube.com>]

No, I don't have any active cooling on the diode.  I have noticed that the 
offset errors in particular are reduced as the system cools down through 
the course of the evening.  Even with cooling, however, I think we'd be a 
few orders of magnitude off. Here are some numbers to give everyone a feel 
for what's going on:

A second magnitude star yields a signal of 37MV.  (measured)
The circuit uses 400M ohms in a current to voltage conversion
That  means the diode is generating about 92 pico amps for the 2nd mag star.
Calculating magnitudes, a 15th magnitude star should generate about 0.6 
femto amps
0.6 FA should result in an output of about 200 nano volts...

Aside from the obvious problem of registering 200 NV on a volt meter, there 
are other issues working with signals in the femto amp range.  Circuit 
board layout becomes critical, most op amps won't work, and the photodiode 
is operating in a range where the signal-to-noise ratio is approaching 
one.  I have a 100x gain op-amp on the back of the primary current to 
voltage converter so I should be able to see the 200 NV signal but it is 
swamped by noise in my system.

One obvious solution is more aperture.  Other things to consider would be 
sensors with built-in gain such as photomultiplier tubes and avalanche 
photodiodes.

                         Bob

There are several problems with
At 11:43 AM 8/30/99 -0700, Donald Clement wrote:
>Hello Bob,
>
>Did you try cooling the diode to lower the dark current?
>
>Don Clement
>San Diego/Running Springs