Computing Away

[Tom Droege <tdroege2@earthlink.net>]

For the last several weeks I have been running the computer full time 
trying to discover something about the errors one sees in the data.  It is 
pretty discouraging work.  One computes all day and the result is exactly 
the same as before.

Something constructive is coming out of this.  I am learning the limits of 
the parameter space.  I am also finding things that are OK.  For example, 
the linearity is OK.  It is hard to imagine that it would not be OK because 
of the way CCDs work and the type of ADC used, but now it has been checked.

I am also experimenting with the reference catalog and the aperture 
photometry.  The results are not always intuitive.

It is all worth while since I am learning about the pipeline and what the 
various sections do.

Along the way I discovered that the flat field light box has a light leak 
in the IR.  I had great hopes that this was the cause of the Ic data 
problem.  I made new flats with the light leak sealed up.  These new screen 
flats made no difference.  I also made sky flats. These result in dark 
subtracted and flat fielded .fits files that look much more uniform.  I 
also tried to do a flat field computation with 185 frames which just hung 
for 7 hours.  Unfortunately when I used the sky flats on the data, there 
was no change.  Well, little change.  I am still looking.  As Michael would 
say, rats!

This being said, the data is not all that bad for survey purposes.  I will 
be able to take data over the northern sky that is more accurate and a lot 
more precise than say the AAVSO data.  We will be able to put the data on 
the "standard system" with some error limit.  This might be 0.05 mag at the 
bright end and 0.1 mag at the faint end, but it will be a limit.

 From what I read on the various discussion groups, this is better than we 
have now for the whole northern sky, so I see the work as useful.

Tom Droege