Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0054-1-0.ps.gz Дата изменения: Wed Jan 3 15:10:24 2001 Дата индексирования: Mon Oct 1 21:49:57 2012 Кодировка: Поисковые слова: comet tail

XMM­Newton CCF Release Note
XMM­CCF­REL­54
EPIC Energy Scale
D Lumb
December 7, 2000
1 CCF components
Name of CCF VALDATE List of Blocks
changed
CAL VERSION XSCS flag
EMOS1 ADUCONV 0010 2000­01­01T00:00:00 OFFSET GAIN NO
EMOS2 ADUCONV 0010 2000­01­01T00:00:00 OFFSET GAIN NO
2 Changes
The User Help Desk reported a bug in the previous conversion. Figure 1 Shows the measured
internal calibration source data for the timing mode. The 3 major emision lines have been fit with a
quadratic which should go through zero. To a good approximation we have a simple linear gain term,
which is significantly different that the nominal gain. Leicester University report a 10% difference
in sampling period between TIMING mode and normal image mode.
Due to the current limitation of the mode dependency of gain, we can only apply a ``fudge factor''
to cover the linear component in TIMING mode. According to the fit of Figure 2, this is actually
8.8% different than the nominal gain.
3 Scientific Impact of this Update
Modifies the gain for the FAST mode of EPIC MOS
1

XMM­Newton CCF Release XMM­CCF­REL­54 Page: 2
4 Estimated Scientific Quality
Without a correction sources will appear red­shifed by 10With this simple correction applied, and
knowing the discrepancy in ACTUAL clock sequence usage, we expect the correction of the error
to be within 1 ­ 2% of the ACTUAL gain. That is 60eV at Fe K lines, and 10­20 eV where photon
spectra peak.
It should further be noted that TIMING mode performs an on­chip binning of pixels and the
resulting change in distribution of split charge will inevitably affect the response distribution in an
as yet undetermined way.
5 Expected Updates

XMM­Newton CCF Release XMM­CCF­REL­54 Page: 3