TN-67

[Tom Droege <droege@wwa.com>]

I would like to thank Jure very much for his nice TN-67.  The diary format 
was very clever, and conveys a lot of information about what is involved in 
doing this work.

It is all I can do to resist asking Jure to send me a copy of his 
system.  I have a 600 MHz Athelon, and it should just run on my 
system.  Well, it will not be easy, it never is.

But I first have to do hardware.  So I will resist for the time being until 
I can get the first few systems out the door.

For the May 21 entry, the exposures alternated short and long.  For 
example, .818, .822, .858, and .862 are all short exposures.  There is more 
noise than change in the mean level.  The noise is higher than I would 
like, but it does decrease with lower temperature (TN-65).  A better number 
for the scale factor is 2.4 e- per ADU.  I need to get some good cameras 
and run them really cold to make a good measurement.

I can take bias frames and use them regularly for test.  Well, a bias frame 
is about a 1+46 second mean exposure.  i.e. it takes 46 seconds to read a 
frame and roughly 1 second to take a minimum dark exposure.  So each pixel 
has been leaking dark current for 47 seconds.  But note the effect of two 
successive read outs is to smear out the dark current from the individual 
pixels.  So the (Richmond) effect is not seen in a bias frame but is seen 
in an only slightly longer exposure - say 20+46 seconds.

Tom Droege