Bad Pixels

CCD#40 has relatively few bad pixels, mostly around the edges and in 3 bad columns. These can be removed from data, if necessary, by means of a bad pixel mask. Bad pixel masks show those pixels which do not respond linearly. You can create your own bad pixel mask by taking appropriate flat field observations, or download the most recent mask image below.

Generating a Bad Pixel Mask

They are generated by taking flat field exposures at high and low intensity levels, and identifying those pixels where the ratio of the normalised flux is more than 15%. In detail (using IRAF commands):

  • Take: 5 Bias images, 5 Domeflats with low counts (~5000 counts), 5 Domeflats with high counts (~40,000 counts)
  • Zerocombine the Bias images
  • Bias subtract the Domeflat images
  • Flatcombine the domeflats with high counts - FLATHIGH
  • Flatcombine the domeflats with low counts - FLATLOW
  • imarith FLATLOW.fits / FLATHIGH.fits FLATLHRATIO.fits
  • ccdmask FLATLHRATIO.fits BADPIX_MASK_FILE with the following parameters set:


                    Image Reduction and Analysis Facility
PACKAGE = ccdred
   TASK = ccdmask

image   =                       Input image
mask    =                       Output pixel mask
(ncmed  =                    7) Column box size for median level calculation
(nlmed  =                    7) Line box size for median level calculation
(ncsig  =                   15) Column box size for sigma calculation
(nlsig  =                   15) Line box size for sigma calculation
(lsigma =                  20.) Low clipping sigma
(hsigma =                  20.) High clipping sigma
(ngood  =                    5) Minimum column length of good pixel seqments
(linterp=                    2) Mask value for line interpolation 
(cinterp=                    3) Mask value for column interpolation 
(eqinter=                    2) Mask value for equal interpolation
(mode   =                   ql)

How to use the BADPIX masks

If you use one of the BADPIX masks provided below the file you download is a single badpix mask with no trimming. You must not trim your images in order to use it.

Simply enter the name of the badpix file in the FIXFILE keyword of CCDPROC and run CCDPROC. That's all there is to it!

filter/median raw_image median 7,1
replace/image raw_image fixed_imag badpix_1x1.fits -1,-1 median.bdf

i.e. this will replace each bad pixel with the median of the 7 pixel surrounding the bad pixel in the same row. An example of an U flat-field fixed this way is shown in this figure. The right image is an U flat-field before processing, while the left image is the same flat after applying the mask

BADPIX masks for EFOSC2 at the 3.6m

Note that the change of telescope to the NTT does not affect the CCD, so the 2007 bad pixel masks are still OK.

January 2002

January 2006

March 2007

*The 2002 gif image was boxcar-averaged with a 5x5 window to improve visibility, so it exagerates the extent of each bad pixel cluster. This was needed to make them visible on a compressed plot.