How to process EPIC slew data


This thread explains how to process slew data taken with the EPIC-pn camera.

Expected Outcome

At the end of this thread a set of filtered event files and images will have been created. These represent contiguous sections of the slew.

SAS Tasks to be Used


It is assumed that a slew data file (SDF) has been unpacked into the current directory. The SDF may be obtained from the archive.

Useful Links


Last Reviewed: 25 MAY 2023, for SAS v21.0

Last Updated: 16 January 2017




  • Set up your SAS environment (following the SAS Startup Thread)


  • Create an EPIC-pn event list from the slew data file. The procedure for this is identical to the initial processing of a pointed observation (see How to reprocess ODFs to generate calibrated and concatenated EPIC event lists).


  • Tell the SAS to use the Raw Attitude File (RAF) for the attitude reconstruction of the slew by:
    1. csh: setenv SAS_ATTITUDE RAF
    2. sh: export SAS_ATTITUDE=RAF
  • Run eslewchain

    This task does not need any parameters. It reads the EPIC-pn event list from the directory pointed to by the environment variable SAS_ODF and processes the whole slew (which can take several hours for a long slew) producing a set of output files.

The output files are given below where the band number b means:

  1. Attitude corrected images of name PobsidPNS003IMAGE_biii.ds Where, obsid is the 10-digit observation number, b is the energy band identifier and iii is the image sequence number in hexadecimal which typically increments from 000 to 04A or so.
  2. Exposure maps of name PobsidPNS003EXPMAPbiii.ds
  3. Filtered event files for each image section of name: PobsidPNS003PIEVLI0iii.ds
  4. Unfiltered images of name PobsidPNS003UNFDAT8iii.ds

Note that the images produced have a small overlap to avoid problems with sources located at the very edge of an image. Due to this, some sources will be found in two consecutive images.

  • Optionally run eslewsearch to perform source searching on the produced slew images

        This task has no input parameters.

        It produces four output files:

        - PobsidPNS003OMSRLIb000.FIT  a source list for bands 6,7,8

        - PobsidPNS003OMSSLI0000.FIT    a combined source list



  • Instrument selection during a slew:

    Initially the EPIC-MOS collected scientific data during slews. These data were of poor quality because the frame time in the MOS detectors is long compared with the motion of the satellite during a slew (90 degrees per hour). MOS sources are hence stretched out into long streaks with a consequent increase in the background level and reduction in the signal-to-noise ratio. The EPIC-pn frame times are small and give a relatively small perturbation to the effective point spread function (PSF) of sources.
  • Practical issues

    Slew data is particularly susceptible to problems caused by background flares, low-energy noise and halos around bright sources. For a practical guide of the issues to be aware of when interpreting slew data please see Saxton et al. 2008, A&A 480, 611.
  • Use SAS 13 or later:

    In SAS v13 a much improved treatment of time jumps in EPIC-pn data was introduced. Previously, it was possible for sources to be assigned to the wrong sky position due to the attitude reconstruction software, which relies on an accurate knowledge of the arrival times of the photons.

    In SAS v14 the algorithm was changed to divide the slew by time rather than spatial coordinates. This solved a problem seen in about 1% of slews where very large subimages were created.

  • The source searching task eslewsearch is only available from SAS v16 or later.

  • It is strongly recommended to use SAS v13 or later to process slew data.