Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0138-1-0.ps.gz Дата изменения: Mon Jan 20 12:57:29 2003 Дата индексирования: Mon Oct 1 22:19:11 2012 Кодировка: Поисковые слова: http astrokuban.info astrokuban

XMM­Newton CCF Release Note
XMM­CCF­REL­138
EPIC MOS quantum efficiency
R. D. Saxton, S. Sembay
11 December 2002
1 CCF components
Name of CCF VALDATE Blocks changed XSCS flag
EMOS1 QUANTUMEF 0013.CCF 2000­01­01 FRACTION ENERGY, CHBINS FRACTION NO
EMOS2 QUANTUMEF 0013.CCF 2000­01­01 FRACTION ENERGY, CHBINS FRACTION NO
2 Changes
Support for Timing and Burst mode Energy pattern fractions have been added in this
release. In Timing mode pixels are binned together in the Y direction. This leads to
Timing mode single pixel events being comprised of imaging mode pattern 0, 1 and 3
events. Similarly, double pixel events are actually comprised of imaging mode patterns
2,4 and 5 to 12. This is handled within the CCF structure by setting the FRAC S column,
of the FRACTION ENERGY extension, for Timing mode data equal to the sum of the
fractions of imaging mode pattern 0, 1 and 3 events and FRAC D to the sum of the
fractions for patterns 2, 4 and 5 to 12.
Burst mode (known as Timing Compressed) data contains no pattern information. This
means that the standard data selection will contain events with imaging mode patterns
0­12. This is handled by making the FRAC S column equal to the sum of the fractions
of patterns 0­12 for Burst mode.
An additional extension, CHBINS FRACTION, has been added to quantify the PI chan­
nels over which the FRACTION CHANNEL arrays are defined. This extension has the
column PI CHAN.
1

3 Scientific Impact of this Update
This change allows the effective area file (ARF) to be calculated correctly for timing and
burst mode data, taking into account any user pattern selection.
4 Estimated Scientific Quality
If the Timing mode is well calibrated it should return the same spectrum during spectral
fitting as the imaging modes. To test this, the observation 290/0089960301 of the AGN
NGC 5548 has been analysed. In this observation the source was simultaneously observed
with MOS­1 in timing mode and MOS­2 in small window mode. In small window mode the
source was piled­up and the inner 10 arcseconds of the core have had to be excised when
extracting the spectrum. Despite this complication, a simultaneous fit of the two spectra,
using ARFs produced by arfgen with these CCFs and RMFs produced by rmfgen, shows
that the two spectra agree to within ё 10% from 0.3 to 9 keV (Fig. 1). The spectrum is
complex and has been fitted here with a model of an absorbed power­law, two black­bodies
and nineteen spectral lines; giving ь 2
r
= 1:3 for the combined fit.
5 Expected Updates
The pattern fractions in channel space for timing mode have been set to be the same as
for full frame mode. They will need to be measured from in­flight data and updated in a
future release of these CCF files.
6 Test procedures and results
There are no standard matrices to check this release against. The analysis in Section 4
shows that the calibration is good to 10%, relative to the calibration of the imaging modes
at all but the highest energies.
References
2

Figure 1: A combined fit of a MOS­1 timing mode (black) and MOS­2 small window mode observation
(red) of NGC5548.
3