Examining Some Diagnostics

The first step after emchain/epchain is seeing what is in the data. For this purpose we suggest forming some images with which to diagnose potential problems:

evselect table=mos1S001.fits withimageset=yes
imageset=mos1S001-diag-det-unfilt.fits
filtertype=expression
expression="(PI in [300:1000])&&
(PATTERN$<=$12)&&((FLAG & 0x766aa000)==0)"
ignorelegallimits=yes imagebinning=imageSize
xcolumn=DETX ximagesize=780
ximagemax=19500 ximagemin=-19499
ycolumn=DETY yimagesize=780
yimagemax=19500 yimagemin=-19499
evselect table=mos2S002.fits withimageset=yes
imageset=mos2S002-diag-det-unfilt.fits
filtertype=expression
expression="(PI in [300:1000])&&
(PATTERN$<=$12)&&((FLAG & 0x766aa000)==0)"
ignorelegallimits=yes imagebinning=imageSize
xcolumn=DETX ximagesize=780
ximagemax=19500 ximagemin=-19499
ycolumn=DETY yimagesize=780
yimagemax=19500 yimagemin=-19499
evselect table=pnS003.fits withimageset=yes
imageset=pnS003-diag-det-unfilt.fits
filtertype=expression
expression="(PI in [300:1000])&&
(PATTERN $<=$ 4)&&(#XMMEA_EP)"
ignorelegallimits=yes imagebinning=imageSize
xcolumn=DETX ximagesize=780
ximagemax=19500 ximagemin=-19499
ycolumn=DETY yimagesize=780
yimagemax=19500 yimagemin=-19499
evselect table=pnS003-oot.fits withimageset=yes
imageset=pnS003-oot-diag-det-unfilt.fits
filtertype=expression
expression="(PI in [300:1000])&&
(PATTERN $<=$ 4)&&(#XMMEA_EP)"
ignorelegallimits=yes imagebinning=imageSize
xcolumn=DETX ximagesize=780
ximagemax=19500 ximagemin=-19499
ycolumn=DETY yimagesize=780
yimagemax=19500 yimagemin=-19499

This is a large amount of typing, which is why it is a good idea to script much of the reduction.

For the MOS we have used ((FLAG & 0x766aa000)==0) which is not equivalent to #XMMEA_EM but is the equivalent to #XMMEA_EM with the addition of the out-of-FOV events. For the pn we use ((FLAG & 0xcfa0000)==0) which is the equivalent of #XMMEA_EP, which does include the out-of-FOV events. (The flag for the pn which excludes the corner data would be ((FLAG & 0xfb0000)==0).)We have extracted the standard patterns. We have restricted the images to the soft band, as that is where most of the problems can be detected.

If the circular FOV is easily seen in all the instruments, then there is a significant soft proton flare problem. Lower amounts of soft proton flares may not be readily visible to the eye in these images. If, for the MOS images, some individual CCDs appear brighter than others (typically MOS1-4, MOS1-5, MOS2-2, or MOS2-5) then those CCDs are in their “anomalous states” and may have to be removed from the analysis, but only if the analysis includes data from $E<1.5$ keV. Comparison of the pn image with the pn-oot image will show whether there are any bright objects that are likely to cause OOT problems in the later analyses.

If the analysis is to include data at higher energies, it would be wise to repeat these diagnostic images at those higher energies. Scattering arcs due to bright sources just outside the field of view tend to be stronger at higher energies, so a quick check at those energies is encouraged.

Given the large number of files created by ESAS, we suggest limiting the number of files in your analysis directory; it is a good idea to either delete these diagnostic files after they have been created, or to move them to their own subdirectory.