Re: Noise floor in Disk 16 data

[Tom Droege <tdroege@veriomail.com>]

I think part of the I camera problem is that it has not been as cold as the 
V camera because of a deteriorating coolant flow problem.  Now that I 
understand this, I can keep the tubes clear of bug parts with the 
filter.  I also have a better chiller setup since TN-65 so possibly it is 
time to do that experiment again.

In any case, I urge Arne and Michael to put a water filter from the 
hardware store on their units before the tubes clog up.  Because of the 
automatic closing disconnects, they are a pain to clean if they clog up.

Tom Droege

At 12:43 PM 3/16/01 -0500, you wrote:

>   I've spent a little more time working on the images from
>Data Disk 16.  Arne provided some suggestions for the
>source of the "noise floor" of about 0.005 magnitudes.
>I'm checking them out.
>
>   One possibility was noise in the master dark frames.
>Each one was created from only 5 dark images.  The histograms
>of pixel values show a "messy" peak, with several sub-peaks.
>Tom has explained this effect as due to the chip being run
>at higher-than-optimal temperatures -- see
>
>     http://a188-l009.rit.edu/tass/technotes/tn0065.html
>
>   Anyway, I calculated the contribution of the noise in the
>dark frames to the total uncertainty in aperture photometry.
>It turns out that the noise from the dark frames is a bit
>smaller than the noise from the sky, and contributes only
>about 0.0005 mag to the overall uncertainty for bright stars
>in both V and I.
>
>   Arne also mentioned that I cut off all magnitude measurements
>at thousandths of a mag.  That introduces another 0.0005 mag
>of noise.
>
>   So, these first two sources don't add up to the 0.005 mag
>"floor" I found in the data.  I'll keep working ....
>
>                                        Michael Richmond