Next: Images Up: XRTSUB Prev: Input and output

Method

The details of the method vary slightly, depending on whether the source file is an image or not, but in all cases the procedure is as follows:
  1. Open the source and background files and check that they are compatible. They are compatible if they have been taken over the same pulse height range and over the same times.
  2. Sum the background counts. Background pixels with bad quality are ignored.
  3. Calculate the area of the source and background boxes. If the background file contains image axes, its area is calculated by adding up the number of GOOD image pixels and multiplying by the area of a pixel. Otherwise the area is calculated using the SPATIAL information in the SORT box of the file.
  4. Calculate the expected particle count (as a function of PH channel) in the background box. This calculation uses the formula of Steve Snowdon (Ap.J. submitted). It turns out to be a function of Master Veto Rate which is a function of time and is contained in the eventrate (_EVR) file.
  5. Calculate the photons in the background box by subtracting the particles.
  6. 'Correct' the background photon counts to the source box position, using the vignetting functions contained in the effective area file and do the same for the particles but using the spatial form in Steve Snowdon's paper. Both of these corrections are energy dependent and are applied separately for each spectral bin in the background file.

  7. Normalise the background counts for the difference in area between the two boxes and subtract from the source counts.
Photon vignetting is a function of energy. If the background file hasn't got an energy axis then by default the program assumes a mean photon energy of 0.2 keV when it produces the background image. This may be overridden on the command line. However, if the background file contains a spectral axis, then each pulse height bin is vignetted individually to produce the output image. This means that it is good practise to use either a spectrum or a spectral image as the background file when subtracting an image (see the `advice' section)
Subtopics:

Next: Images Up: XRTSUB Prev: Input and output