Re-appearance of the
MOS1 CCD1 meteorite column
during revs. 1792-1801
Madrid, 23. March 2010
● Presumably micro-meteorite impact into MOS1 in rev. 0961.
Anomaly frame [energy] Subsequent frame [energy]
Review: Impact on hardware
● CCD6 inactive.
● Damaged pixels in CCD1 causing hot column close to boresight.
First diagnostic exposure taken
in rev. 0961: CCD1 detail
11 dead pixels:
RAWX 323 RAWY
New hot column RAWX 323:
3 (9) pixels next to pn (RGS)
prime boresight position
Review: Corrective actions
● Investigated meteorite column offset in FF, LW and SW mode
• Same increased column offset (by ~19 ADU) measured for all
● Investigated impact of meteorite column offset for science
• Calibration line energies in calclosed observations were shifted
by ~20 ADU.
• Similar offset as in diagnostic exposures.
• Thus just an additive offset.
● Correction can be done by changing of onboard-offset table to make
the meteorite column available for science again.
Review: Monitoring of column offset
● Regular FF, LW and SW diagnostic exposures in routine calibration.
• Column offset showed exponential long term decay.
● On-board offset table value decreased by 10 ADU (123 =>113 ADU)
after ~600 revolutions (active since rev. 1690).
Topic of this talk
Re-appearances since on-board offset table change
● Detections usually by Blanca Juárez (last remaining InsCon).
● During first eclipse phase after correction in single observations.
● Few single observations, e.g. slew 9152300002.
● Short phase in revs. 1537+1538.
Offset in on-board table decreased (~rev.1690).
● After change, column was mostly hot in revs. 1698-1701.
● Longer phase in revs. 1792-1801.
Trend during re-appearance in revs. 1792-1802
● Diagnostic exposures to measure the column offset during a “hot
phase” are only available for latest occurrence due to coincidence
with routine calibration observation in rev. 1796 (Capella).
500 sec bins
First diagnostics during re-appearance
● Diagnostics in routine cal. observation 1796-0510780401 (Capella):
column offset about 7 ADU higher (119 ADU) than expected by
the decay model (112 ADU).
Detection due to telemetry anomaly
● Independent notification of hot meteorite column by Pedro Calderon
via strange behaviour of MOS1 telemetry, e.g. in rev.1800.
● MOS1 telemetry
● Meteorite column is
cause of telemetry
Frequency search in SW obs.1800_0412600801
● Except readout time and its harmonics, no significant frequency was
found inside the data.
0.3 s = readout
Harmonics of readout frequency
Frequency search in FF obs.1798_0605580601
● Column modulation at long timescales or feature of harmonics?
2.6s = readout
XMM-Newton 1000 x
● If the meteorite column appears hot, it is due to increase of the column
● No time correlation was found for frequencies higher f>0.01 Hz.
● Possible correlations at very low frequencies, might be just features in
harmonics of light curve binning/readout frequency harmonics(?).
Still open questions:
● Why does the meteorite column occasionally re-appears hot for some
observations / time periods?
● What causes the (short term) variation of the column offset?