XMM-Newton Users Handbook

3.3.11 EPIC event grade selection

Event selection is performed on-board in order to allow transmission of useful data only (i.e., removal of empty pixels from the data stream). Also a significant number of X-ray events are expected to be split between pixels, and these must be recognised as such. The Science Analysis Subsystem (SAS) software allows these data to be reconstructed to a single energy value in an event list. Recommendations on pattern (similar to the Chandra or ASCA “event grade”) selections for the spatial and spectral analysis of XMM-Newton event lists are given in the document XMM-SOC-CAL-TN-0018. Users must choose carefully the data sub-set for the science of interest, and further beware that not all pattern combinations may be calibrated as well as some default sets (see Table 7. Further details about EPIC patterns are available e.g. from the descriptions of the SAS tasks emevents and epevents).

Table 7: List of EPIC event patterns

Camera	Mode	X-ray generated pattern					no X-ray$^1$
		singles	doubles	triples	quadruples	higher
MOS	imaging	0	1-4	5-8	9-12	13-25	26-31$^2$
	timing	0	1				2,3
pn	imaging	0	1-4	5-8	9-12		$>$ 12
	timing	0	1-4			5-12	$>$ 12

Notes to Table 7:
Calibrated patterns are flagged by bold text font.
1) no genuine X-ray events.
2) 26-29: diagonal pattern due to Si-fluorescence or pileup of two monopixel events; 30-31: cosmic rays extended tracks.

Users should also be aware of the different methods of processing events that are present in the EPIC MOS and pn cameras. The MOS camera uses an event pattern recognition scheme and all detected events are transmitted to ground with a pattern identification already attached to them. The pn camera conversely telemeters all pixels above a certain threshold, and the X-ray event selection and processing is performed wholly on-ground, i.e., through the use of the SAS.