merging amps in SOAR/SOI images
Ron Marzke wrote on Jun 18, 2007
Hi,
I am working on a set of images from SOAR/SOI taken in April 2007, and I am having trouble merging amplifiers in CCDPROC. The images are binned 2x2 and include both prescan and overscan for each of four amps (two amps per CCD). The SOI headers appear to include the correct DETSEC, CCDSEC, DATASEC, and BIASSEC header keywords. They also appear to include the relevant DTM/DTV elements. I compared the header parameter values to those recommended by Frank Valdez in his response to a similar posting by Eduardo Cypriano in March 2004, and the headers look ok.
In CCDPROC, I have set overscan=yes, trim=yes and merge=yes. All other actions are set to no. The resulting MEF file should have two image extensions, each 1024x2048. Instead, I get two image extensions, each 512x2048.
I've placed an example raw image at
http://www.physics.sfsu.edu/~marzke/soar/s07a017_0417.285.fits
I've tried various combinations of LTM/LTV, DATASEC, DETSEC, DTM/DTV etc but have not had any luck. Any suggestions would be much appreciated.
Thanks,
Ron
I am working on a set of images from SOAR/SOI taken in April 2007, and I am having trouble merging amplifiers in CCDPROC. The images are binned 2x2 and include both prescan and overscan for each of four amps (two amps per CCD). The SOI headers appear to include the correct DETSEC, CCDSEC, DATASEC, and BIASSEC header keywords. They also appear to include the relevant DTM/DTV elements. I compared the header parameter values to those recommended by Frank Valdez in his response to a similar posting by Eduardo Cypriano in March 2004, and the headers look ok.
In CCDPROC, I have set overscan=yes, trim=yes and merge=yes. All other actions are set to no. The resulting MEF file should have two image extensions, each 1024x2048. Instead, I get two image extensions, each 512x2048.
I've placed an example raw image at
http://www.physics.sfsu.edu/~marzke/soar/s07a017_0417.285.fits
I've tried various combinations of LTM/LTV, DATASEC, DETSEC, DTM/DTV etc but have not had any luck. Any suggestions would be much appreciated.
Thanks,
Ron
Francisco Valdes wrote on Jun 18, 2007
Hi Ron,
Thanks for providing an example. The problem turned out to be surprising to me. It is due to formatting the keywords LTV1 and LTV2 as string valued. For mscred.ccdproc it doesn't matter what these are initially because they get set based on the CCDSEC keywords (which are correct in your data). Once they are set they are used for the merging. For a reason I don't understand the updating of these keywords does not work when they are formated as strings.
The workaround is to delete these keywords from the headers in the raw data before running CCDPROC:
Yours,
Frank Valdes
To avoid this in the future you might forward your message and my answer to the SOAR folks and maybe they can change the header formating code. Note that the DTV/DTM, ATV/ATM keywords have the same problem though these are not currently used for data reductions.
Thanks for providing an example. The problem turned out to be surprising to me. It is due to formatting the keywords LTV1 and LTV2 as string valued. For mscred.ccdproc it doesn't matter what these are initially because they get set based on the CCDSEC keywords (which are correct in your data). Once they are set they are used for the merging. For a reason I don't understand the updating of these keywords does not work when they are formated as strings.
The workaround is to delete these keywords from the headers in the raw data before running CCDPROC:
msc> msccmd "hedit $input ltv* del+ verify-" <list of files>
Yours,
Frank Valdes
To avoid this in the future you might forward your message and my answer to the SOAR folks and maybe they can change the header formating code. Note that the DTV/DTM, ATV/ATM keywords have the same problem though these are not currently used for data reductions.
Ron Marzke wrote on Jun 18, 2007
Hi Frank,
Many thanks for the quick reply! I'll pass it along to the SOAR folks.
Best,
Ron
Many thanks for the quick reply! I'll pass it along to the SOAR folks.
Best,
Ron
Last post on Jun 18, 2007