Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200204/Analysis_GRS1758_ferrando.pdf
Дата изменения: Wed May 14 13:30:08 2014
Дата индексирования: Wed Apr 13 13:43:40 2016
Кодировка:
Поисковые слова: m 13

Let's start

Data :

GRS1758-258 MOS 1 & 2 in Small Window PN in Large Window

Thin filters ~ 22 cts/s ~ 6.6 cts/frame ~ 52 cts/s ~ 2.5 cts/frame

Episode 1 : June 2001 meeting (Leicester) « Spectral and spatial deformations on piled-up sources, "a one case study, of MOS data" » Big problems in spectral results as a function of extraction radius Episode 2 : November 2001 meeting (Milano) « Checking encircled energy by spectral fitting » Big problems solved thanks to new PSF : spectral results independent of extraction radius. but strong MOS excess at low and high energy

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

The problem left in Episode 2

MOS excess at low and high energy
6 arcsec cut 24 arcsec cut

Is this an effect of bad background correction ?

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Was really the background to blame ?

Episode 3 Re-analysis of the GRS1758 piled-up source : all problems remain...

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Strategy

Analysis plan : · take the latest emchain processing : developper SAS end of January · stick to one exclusion radius : 6 arcsec for MOS source region : [645] arcsec MOS1, [641] arcsec MOS2 · use the latest matrices (q20 here) · use the CCF corrections for the Encircled Energy · background : from annulus [8.412] arc min (MOS2) · since background far away from source region : separate the particle background correction (non vignetted, but with spatial dependence) and the diffuse background correction (vignetted).

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Equations

Problem equations for « double background substraction » Source spectrum = Raw spectrum source region diffuse bkgd source region particle bkgd source region diffuse bkgd = annulus diffuse bkgd corrected for vignetting · annulus diffuse bkgd = RAW annulus spectrum annulus particle bkgd · annulus particle bkdg = "Lumb" background in annulus region в NORM · NORM = Raw annulus spectrum [1012 keV] / "Lumb" bkgd [1012 keV] source region particle bkgd = "Lumb background" in source region в NORM

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Annulus spectrum raw , vignetting, particle

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Source spectrum and its diffuse and particle parts

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

MOS1 fit

Problem is not solved...
EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002 Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Fitting results

Some impact on fits however Fitting above 0.9 keV, MOS 1 - PL + BB Power law index : 4.18 ± 0.11 with simple annulus bkgd substraction 4.25 ± 0.11 with correct treatment Rise of index as expected (vignetting increase of background substracted)

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp

GRS1758 Episode 3

Conclusion

Conclusion of Episode 3
· Despite very careful background substraction, MOS excess at high energy remains. Causes can be : i) problems with encircled energy ? ii) problems with mono-pixel fraction (selected here) ? iii) residual pile-up effect ? · MOS excess in the ~ 0.50.9 keV range still present. Causes can be ??? · Side result of this analysis : The MOS spectrum is steepened when proper vignetting effect is taken into account. Be careful if sources are in a strong diffuse emission environment.

EPIC Cal-Ops Meeting / Schlo Ringberg / Apr. 25, 2002

Philippe Ferrando, CEA/ SAp