iraf.net

Hello,
I've been trying to fix this problem a while now and have run out of ideas. The problem is, that I have data that is a mosaic of 8 CCD's. I can use mscred/ccdred/ccdproc to bias and trim the images. Before I trim the images the DATASEC is [6:2049,5:4100].
Next I unpack the mosiac in order to use noao/ccdproc to flat and zero correct. (mscred wont do flat and zero)The problem I am having is when I try to correct the images with a Zero and Flat frame. I get the following error in CCDPROC. "ERROR: Size of DATASEC and CCDSEC do not agree."I've checked the trimmed image frames and the flat/zero frames and neither has a defined DATASEC.
CCDSEC is [1:2040,14:4086] for the image frames and
[1:2039,14:4086] for the Zero and Flat frames. If I change CCDSEC to the same value for all three, namely [1:2040,14:4086], I get a new error,
"ERROR: Pixel subscript out of bounds (Zero.fits)" where Zero.fits is my zero frame. I've also tried deleting CCDSEC in all the headers and that resulted in DATASEC and CCDSEC do not agree.
Here is an example of one of the trimmed image headers.
FILENAME= 'run537 ' / Base filename at acquisition
MSTTIME = 'Mon Mar 15 01:49:02 MST 2004' / Local time in Arizona
IMAGESWV= 'CFHT DetCom v3.214 (Dec 10 2002)' / Image creation software version
CCD = 'SITe 2Kx4K CCD'
MAXLIN = 65535
SATURATE= 65535
TIMESYS = 'UTC ' / Time System for DATExxxx
DATE-OBS= '2004-03-15' / UTC Start of observation
TIME-OBS= '08:29:02.00' / UTC Time at start of observation
JD = 2453079.85350 / JD at start of observation
LST-OBS = '12:35:40.77' / Sidereal time at start of observation
OBSTYPE = 'OBJECT ' / Observation/ Exposure type
EXPTYPE = 'OBJECT ' / OBSTYPE is the new name for this
EXPNUM = 537 / Exposure Sequence Number
OBSID = 537 / Observation Sequence Number
EXPTIME = 1200.0 / Integration time (seconds)
DARKTIME= 1200.0 / Dark current time (seconds)
TCSSTAT = 'Enabled' / TCS link status
TELESCOP= 'MDM 2.4m' / MDM 2.4m Hiltner Telescope
LATITUDE= 31.95000 / Latitude (degrees N)
LONGITUD= -111.6167 / Longitude (degrees E)
EQUINOX = 2000.00 / Equinox of RA and DEC
RA = '16:34:38.15' / Right Ascenion
DEC = '+46:19:45.66' / Declination
HA = '-03:59:04.97' / Hour Angle
SECZ = 1.475 / Airmass at start of exposure
TELFOCUS= 4645 / Telescope focus encoder readout
ROTANGLE= -0.040000 / Teescope rotator angle
COMMENT
COMMENT Detector array and CCD Information
COMMENT ----------------------------------
DETECTOR= 'MDM8K ' / MDM 8K x 8K Mosaic CCD camera
DETSIZE = '[1:12450,1:8230]' / Total data pixels in full mosaic
NCCDS = 8 / Number of devices
NAMPS = 8 / Number of amplifiers
CCDSUM = '1 1 ' / Binning factors
CCDBIN1 = 1 / Binning factor along first axis
CCDBIN2 = 1 / Binning factor along second axis
RASTER = 'FULL '
PIXSIZE1= 15.0 / Pixel size for axis 1 (microns)
PIXSIZE2= 15.0 / Pixel size for axis 1 (microns)
PIXSCAL1= 0.2060 / Pixel scale for axis 1 (arcsec/pixel)
PIXSCAL2= 0.2060 / Pixel scale for axis 1 (arcsec/pixel)
INSTRUME= 'MDM8K Mosaic' / Instrument used for observation
CONSWV = 'MDM=48,DCU=65,MDS=50,DCU=47' / Controller DSPID and SERNO versions
DETSTAT = 'detector temp is low' / Detector temp range (-91..-85)
DETTEM = -423. / Detector temp deg C = 773 + -0.292 * 4095
BIASSEC = '[2050:2080,2:4128]' / Overscan area of frame
SHUTOPEN= 0.0000 / Shutter blade opening time (msec)
SHUTCLOS= 0.0000 / Shutter blade closing time (msec)
FILTER = 'I ' / Name of filter in position
FILTERID= 4 / Filter index (see also FILTER)
CCDMEAN = 4242.767
XTENSION= 'IMAGE ' / Image extension
BITPIX = -32 / Bits per pixel
NAXIS = 2 / Number of axes
NAXIS1 = 2040 / Axis length
NAXIS2 = 4073 / Axis length
PCOUNT = 0 / No 'random' parameters
GCOUNT = 1 / Only one group
ORIGIN = 'NOAO-IRAF FITS Image Kernel July 2003' / FITS file originator
EXTNAME = 'chip00 ' / Extension name
INHERIT = T / Inherits global header
DATE = '2009-03-17T21:12:48' / Date FITS file was generated
IRAF-TLM= '17:12:51 (17/03/2009)' / Time of last modification
OBJECT = 'SDSS1004' / Name of the object observed
COMMENT file run53700, raster FULL, etype OBJECT, etime 1200, filter I
COMMENT 04 05 06 07
COMMENT Image data for chip 00
COMMENT ---------------------- [00]01 02 03
IMAGEID = 0 / CCD chip number
CHIPID = 0 / Use IMAGEID instead
GAIN = 1.50 / Amplifier gain (electrons/ADU)
CTYPE1 = 'RA---TAN' / WCS Coordinate type
CTYPE2 = 'DEC--TAN' / WCS Coordinate type
CRPIX1 = 6340.3 / WCS Coordinate reference pixel
CRPIX2 = 4059.6 / WCS Coordinate reference pixel
CD1_1 = 5.6075E-05 / WCS Coordinate scale matrix
CD2_2 = -5.6668E-05 / WCS Coordinate scale matrix
LIM1_1 = 1.0 / WCS
LIM2_2 = 1.0 / WCS
WAT0_001= 'system=image' / WCS
WAT1_001= 'wtype=tan axtype=ra' / WCS
WAT2_001= 'wtype=tan axtype=dec' / WCS
WCSDIM = 2
CDELT1 = 5.60750000000001E-5
CDELT2 = -5.6668000000000E-5
LTV2 = -13.
LTM1_1 = 1.
LTM2_2 = 1.
CCDSEC = '[1:2040,14:4086]'
TRIM = 'Mar 17 17:12 Trim is [6:2045,18:4090]'
OVERSCAN= 'Mar 17 17:12 Overscan is [2057:2070,100:4000], function=minmax'
CCDPROC = 'Mar 17 17:12 CCD processing done'
PROCID = '537V1 ' Thank you,
Shelly

It helps to think about the defaults for DATASEC and CCDSEC (i.e. when they are missing from the header). These default to the size of the image such that if your image is 2040x4072 then the sections would be [1:2040,1:4072]. A consequence is that if the images (object vs calibration) are different sized you will get an error even if these keywords are missing.The purpose for these keywords are to possibly allow different trimming for the object and calibration (where the calibration is bigger). CCDPROC would then figure out what piece to use for matching the data in "CCD" coordinates.Anyway, I think the sizes must be different?Does this help?Yours,
Frank Valdes

I appreciate the help. Working on this further the problem presented itself with the two, as you said, not matching. What I had to do to fix this was to display the images and get the pixel scale, ie x1, x2 and y1, y2, which for my case was [1:2039,1:4073]. Then I went into hedit and changed both the CCDSEC and DATASEC (this had to be added to the header to prevent it from setting it to the default) to that value. This fixed the problem.Thank you!
Shelly