XMM-Newton Science Analysis System


evatt (saslib-1.88.1) [22.0.0-9173c7d25-20250127]

Informational attributes

• DIAGONAL This event was part of a pair of single events touching one another by a corner, but otherwise isolated. A significant number of those hint at pile-up.

• CLOSE_TO_CCD_BORDER This event was immediately next to a border of the CCD, so that part of its charge may have been lost.

• CLOSE_TO_CCD_WINDOW This event was immediately next to a border of the current CCD window (when not in full frame mode), so that part of its charge may have been lost. For EPIC/MOS only events next to the left and lower borders are affected and flagged.

• CLOSE_TO_NODE_BOUNDARY This event was immediately next to the limit between the read-out nodes (when in 2-node mode), so that its energy may not be precise.

• CLOSE_TO_ONBOARD_BADPIX This event was immediately next (to the left, right, above or below but not diagonally) to a bad pixel declared on board. Its energy will be imprecise, but since this may happen only when the bad pixel is inactive it should be a true X-ray.

• CLOSE_TO_BRIGHTPIX This event was immediately next (to the left, right, above or below but not diagonally) to a bright pixel not declared on board. Its energy will be imprecise, and it could be a false event if the bright pixel next to it was active.

• CLOSE_TO_DEADPIX This event was immediately next (to the left, right, above or below but not diagonally) to a dead pixel, so that part of its charge may have been lost.

• CLOSE_TO_BADCOL This event was immediately next to a column declared bad through its offset, so that part of its charge may have been lost.

• CLOSE_TO_BADROW This event was immediately next to a row declared bad through its offset, so that part of its charge may have been lost. It could be a false event if the bad row is bright.

• IN_SPOILED_FRAME This event was detected in a frame which was judged imperfect (like frames with extended integration time). As a result the energy or time of the event may be slightly wrong.

• UNKNOWN_RAWX_PATTERN to cover the case of COMPRESSED TIMING mode events, whose abscissa and pattern are unknown

This is an approximate ordering, in the order of decreasing energy reliability:

DIAGONAL $>$ IN_SPOILED_FRAME $>$ CLOSE_TO_NODE_BOUNDARY $>$ CLOSE_TO_DEADPIX $>$ CLOSE_TO_ONBOARD_BADPIX $>$ CLOSE_TO_CCD_WINDOW $>$ CLOSE_TO_CCD_BORDER $>$ CLOSE_TO_BADCOL $>$ CLOSE_TO_BRIGHTPIX $>$ CLOSE_TO_BADROW