Miscellaneous Examples


In all cameras:

[000489-PN] [000489-PN] EPIC PN image and exposure map (netscape):
Fainter or brighter CCD or CCD quadrant not represented in exposure map. There are various reasons for this, but the number of affected observations is small.

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).

[000326-PN] EPIC PN image (netscape):
Empty image. There are no counts in the image though the good exposure time as given next to the plot is fine. The image is empty since due to strong flaring the flare GTI is empty (flare GTIs of less than 100 sec are ignored).

Since there is no data for this exposure the exposure is rejected. Proceed with the screening of the exposures of the other cameras (see Table 1.b).

[001567-PN] EPIC PN exposure map (netscape):
Time intervals not matching. In the exposure map (and image) the exposure time is the sum of all GTIs. In this case it is given as 7622.88 sec. However, on the EPIC Processing Summary web-page the total duration of the exposure is 6288 sec, which does not agree. The DPSS-log report gives the times as 'duration 17903.9 s; 7622.9 s in image', that is, the web-page time is wrong.

Check the times in the DPSS-log report (which gets the times from the header of the images); if this is OK proceed with the screening. If not, notify SSC-OPS.

[000617-PN] EPIC PN image (netscape):
High background not screened. The image seems rather bright with very few sources. The keyword 'BKGDSCRN' of the FITS-image says F, which means the flare background has not been screened at all. The corresponding source list is not included in the summary source list.

The source list is unreliable and the exposure is rejected (see Table 1.b).



PN specific:

[000870-PN] EPIC PN image (netscape):
Slight enhancement along the camera center line. An inboard MIP rejection was incorrectly commanded from ground. Check the keyword A_SENDM in the event file: If the above problem occurred it will say 'MIP CORREC.1' (instead of the correct 'MIP CORREC.2'). Only two cases with Rev-nrs 215 and 216 are known.

Screen as normal but look out for possible spurious detections in the brighter background. Use the source comment 'Bright background' in this case and choose the appropriate comment for the DPSS-log report.

[000870-PN ] EPIC PN image (netscape):
Outer CCDs 3, 6, 9, 12 brighter than the other 8 CCDs. Strong background flares caused frequent counting mode intervals in all CCDs in all 4 quadrants. However, this alone should not result in such an effect, and it looks like the data had suffered from some corruption on board, that is, in the end the outer CCDs seem to have different exposure times which is not accounted for in the exposure map.

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).

[001338-PN] EPIC PN image (netscape):
Wrong energy levels. The event energies are shifted down considerably, which means, most of the events are lost (therefore the apparently dark image).

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).



MOS specific:

[001349-M2] EPIC M2 image (netscape):
Central CCD only present in image. The other CCDs were in low gain mode and therefore were not processed. On the web-page the observing mode has been called 'Prime Full Window', though in fact this is a special observing mode with special processing and the observation should not be part of the catalogue.

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).

[001160-M2] EPIC M2 image (netscape):
Central CCD fainter than the others. The GTIs for all CCDs are the same and can therefore not account for the dimmer image on the central CCD. It is suspected that this should have been a similar observing mode as above but something did not work right.

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).

[001264-M1] EPIC M1 image (netscape):
Shifted central CCD. The observing mode 'Prime Full Window' is wrong. The exposure is taken in 'wide timing' mode and is an engineering test. It should not be used for the catalogue.

Do not screen this exposure at all and reject the whole observation on the Phase 1 level (see Table 1.a).

[001476-M1] EPIC M1 image (netscape):
Faint line along CCD gaps. Origin unclear, possibly caused by the different heights of the individual CCDs.

Screen as normal and treat the line as a bright segment.



Flare background time series:

[001569-M1] [001569-M1] EPIC M1 and M2 flare background time series (netscape):
Different counts for flares in M1 and M2. Different filters for the M1 and M2 exposures result in different (apparent) strengths of the flare background (M1 was used with the 'medium' filter and M2 with the 'thick' filter)

No effect on screening.

[000492-M1] EPIC M1 flare background time series (netscape):
Periods of zero counts. This is called counting mode and can occur when the background is very high.

No effect on screening.

[001021-PN] EPIC PN flare background time series (netscape):
Flare background not screened at all. Despite the apparently low counts and good time intervals visible in the flare background time series the flare background was not screened. This probably means that most flares were at low energies, while the plots only show the high energy flares, that is, everything is as it should be. There is no explanation for the negative times, but this can be due to the creation of the PGPLOT-image and has nothing to do with the actual data.

No Phase 2 screening due to high flare background not screened (see Table 1.b).

[000681-PN] EPIC PN flare background time series (netscape):
Low Counts. In this case there is a rapid change between normal and counting mode, that is, between very low and very high flaring. The low counts are of no further significance.

No effect on screening.

[001373-PN] EPIC PN flare background time series (netscape):
No x-axis labels. This is due to a problem in the making of the PGPLOT-image and does not affect the actual data.

No effect on screening.




[000824] [000824]
[000825] [000825] EPIC Observation Summary page and composite images (netscape):
Different observation, same object. Two observation (with consecutive sequence number and same revolution number) are different in target and coordinates, but the images are of the same object, and in fact the observing time is the same!

Do not screen the 'false' observation since it is in fact the same and should have the same sources detected (apart from the the differences that occur in the image creation and source detection process if an observation is being processed twice).



Catalogue products:

[000637] [000637] EPIC ACDS and ROSAT overlay (pdf):
Both ACDS and ROSAT overlays are shifted. The ROSAT and ACDS overlays show a small but systematical shift, that is, the shift is noticeable for all matched sources.

Notify SSC-OPS to check the source position cross-correlations for this observation. Continue with the screening.



Image artifacts:

[000828-M2] EPIC M2 image (netscape):
Shadow of the mirror spider, RGA diffraction spike and pileup. These are typical artifacts of bright sources. The reflection of the RGS array goes through the source and is symmetrical in M1, while in M2 usually one arm M2 is barely visible (see details for Phase 2 screening).

Describe the features in the DPSS-log report and use the flag comments as explained in Table 3. If a source lies close to the shadow of the mirror spider it should be flagged since the background subtraction might not be accurate - talk with the main screener if this occurs.

[001342-M2] EPIC M2 image (netscape):
'Shadow' of a bright source. This only appears in early revolutions. The offset variance map is strongly affected by the very bright source (the hole within the bright source is due to optical loading). The affected area is a projection of the bright source along the x- and y-axes of the CCD where the object is observed.

Screen as normal but flag all sources within or close to the shadowy area since the count rates are probably affected here. Discuss this with the main screener if there are sources affected. Give the appropriate comment in the DPSS-log report.

[001344-M1] EPIC M1 image (netscape):
Reflections. Apart from the normal short and numerous reflection arcs from a strong source outside of the field of view (caused by the multiple layers of the curved mirrors) there can be thick arcs as shown here, which are a combination of these short individual arcs from all the energy bands fused together. The diffuse area at the top of the image is also a reflection and appears only in the MOS cameras, probably caused by the RGA. This is very rare and occurrs only with extremely bright sources (this is the Crab pulsar).

Screen as normal with flags for reflection arcs. Flag sources within the diffuse reflection at the top with 'Bright background'.



Unusual sources:

(see also the examples for Phase 2 screening)
[001282-PN] EPIC PN image with sources overlaid (ds9):
Negative sources. The source in the fat cyan circle has -51 counts. While occasionally slightly negative total counts can occur due to the source being detected and fitted simultaneously in the five different energy bands, it is clear from the image that there is a 'hole' (optical loading?) and no source can be discerned.

Flag the source as false with the comment 'Unknown'.

[1046-PN] EPIC PN image with sources overlaid (ds9):
Split source? Next to the central bright source there is another detection. The bright source looks slightly elongated into that direction, but the position of this second detection does not agree with it. In fact, there is a single black column next to it which might trick the eye in seeing an elongation and triggering this detection.

The cyan detection has been flagged as false with the comment 'Poss split by CCD gap', indicating that there is some doubt (since the split detection is on the same side of the gap as the real source).

[001269-PN] [001269-PN] EPIC PN image with sources overlaid (ds9):
Source apparently split by edge of field of view. The source is split in two and lies very close to the edge of the field of view. A comparison with M1 and M2 confirms that this is a single source. However, by closer inspection a faint dark column is discernible (especially further toward the center) which is the cause of the split source.

The cyan circle has been flagged as false with the comment 'Split by CCD gap'.

Screening Notes Example: Version 29.05.2003