Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/sas/6.1.0/doc/emenergy/node9.html
Дата изменения: Mon Nov 22 23:37:55 2004
Дата индексирования: Sat Dec 22 09:05:48 2007
Кодировка:
Поисковые слова: annular solar eclipse

XMM-Newton Science Analysis System


emenergy (emenergy-8.1.2) [xmmsas_20041122_1834-6.1.0]

Meta Index / Home Page

Comments

• The flag map built by EVFLAG could be of interest to eexpmap and arfgen. To that end it would need to be supplemented with small 5x5 maps (one per pattern) describing the effective extension of a bad pixel. Computing the area lost to bright pixels declared on-board is another matter. It depends on energy, through the bright pixel level.
• If the telescope's attitude were to vary significantly (more than the PSF) with time, then it should be considered using X,Y (attitude corrected) as coordinates for CCDBKG, since the optical background from stars would follow the attitude variations. Note that this would need accumulating the map over a larger area than the current window to accommodate pointing excursions. The large scale optical background (e.g. stray light from the earth) does not come through the telescope's optics and therefore has no reason to follow the attitude variations. On the other hand it would be expected to vary along the orbit. The way to deal with that would be to model the large scales (e.g. by a quadratic shape) separately in several time bins and then model the parameters' variations with time (quadratic).
• The subroutines CTI_CORR and ENERGY only call the CAL routines CAL_mosCTIcorrection and CAL_gainCorrection.
• The scheme does not allow for the possibility of measuring the gain in one observation (with the calibration source) and passing this on to emenergy (to supplement or replace the CAL).
• The baseline is to perform the ENERGY subroutine at the end. It could also be considered performing ENERGY (converting all E$_i$ into PI) before CTI_CORR. This would be the best option if the ADU to PI conversion were significantly non linear. On the other hand it is somewhat more difficult to calibrate and may induce some additional error.