If two source photons reach the CCD within the same readout frame then there is a chance that they will be interpreted as a single event with artificially high energy or that they will form an illegal pattern and both be ignored. The probability that this will occur is directly proportional to the count rate of the source. A table giving the count rate limit where this becomes important, for each EPIC instrument and observing mode, is given in the user handbook.
rmfgen includes an option to correct for the flux loss and spectral distortion caused by the pile-up of photons, within a single frame, for observations made with the EPIC-pn camera in imaging mode, i.e. FullFrame, ExtendedFullFrame, SmallWindow and LargeWindow modes. The correction is made with the following call:
rmfgen spectrumset=spectrum.ds rmfset=myresp.rmf
correctforpileup=yes raweventfile=events04.dat
The task needs to read a raw event file to make the correction. This should contain all the charges which landed on the CCD where the source is centred. This is not created by default by the SAS processing but may be produced by either:
which produces a raw event file for each CCD with the name "events
ccd
.dat".
which produces a raw event file for each CCD with the name
"*_ccd_PileupEvts.ds", e.g.
0537_0136540701_EPN_S008_04_PileupEvts.ds .
The correction technique used is very compute-intensive. The execution time is of the order of hours for typical observations.
