iraf.net

From: Marcos Huerta <marcosh@rice.edu>
Date: August 19, 2005 3:17:42 PM MST
To: fvaldes@noao.edu
Subject: ccdproc and CCDSEC
Hello,So, I was using ccdproc to process some images... and eventually come
across some objects that were taken with only a partial row CCD
readout - to save readout time. The section of the full CCD that was
read out is stored in a CCDSEC keyword in the fits header.ccdproc seemed to process this all fine, even though the flat and
bias were different sizes than the objects. So, is it doing what it
should and using the appropriate section of the bias and flat based
on CCDSEC? Or is it doing something alltogether different. PHil
Massey seems to think it's doing the right thing, but he said you'd
know for sure. It doesn't seem to be in any of the documentation.thanks,
Marcos----
Marcos Huerta
Physics & Astronomy Department, Rice University
Herman Brown Hall 344, MS-108; Phone: x3514
From: Frank Valdes <valdes@noao.edu>
Date: August 26, 2005 12:04:46 PM MST
To: Marcos Huerta <marcosh@rice.edu>
Subject: Re: ccdproc and CCDSEC
Hello Marcos,CCDPROC was intended to handle this case and is the main reason why
the CCDSEC keyword is used. However, it is easy to get things mixed
up. So, if you get no error messages then I think it is quite likely
it did the right thing of applying the part of the calibrations that
match the pixels in the smaller readout. But if there is an error in
header keywords then it might be wrong. So it makes sense that you
check. One way to check might be to make copies of the data and
insert a dummy value such as a very high or low flat field pixel and
then see where its effect is seen in the data after you apply the
flat field. If it is in the right place taking into account the
relative areas of the CCD in the two images then you will be reassured.Yours,
Frank Valdes