Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0135-1-0.ps.gz Äàòà èçìåíåíèÿ: Mon Jan 20 12:57:29 2003 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 22:18:08 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: carl sagan

XMM­Newton CCF Release Note
XMM­CCF­REL­135
EPIC MOS Spectral Response Distribution
R. D. Saxton, S. Sembay
10 Dec 2002
1 CCF components
Name of CCF VALDATE EVALDATE List of Blocks changed CAL VERSION XSCS flag
EMOS1 REDIST 0020.CCF 1999­12­10 2000­07­15 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0021.CCF 2000­07­15 2000­11­09 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0022.CCF 2000­11­09 2001­04­18 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0023.CCF 2001­04­18 2001­08­18 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0024.CCF 2001­08­18 2001­09­26 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0025.CCF 2001­09­26 2001­11­25 CCD REDISTRIBUTION­n 3.152 NO
EMOS1 REDIST 0026.CCF 2001­11­25 ­ CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0020.CCF 1999­12­10 2000­07­15 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0021.CCF 2000­07­15 2000­11­09 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0022.CCF 2000­11­09 2001­04­18 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0023.CCF 2001­04­18 2001­08­18 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0024.CCF 2001­08­18 2001­09­26 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0025.CCF 2001­09­26 2001­11­25 CCD REDISTRIBUTION­n 3.152 NO
EMOS2 REDIST 0026.CCF 2001­11­25 ­ CCD REDISTRIBUTION­n 3.152 NO
2 Changes
Two major changes are incorporated in this release. For the first time an observing mode
dependency has been introduced in the redistribution parameters to describe the difference
between the Timing and Burst mode response and that of the imaging modes. This has
been achieved by adding a column called MODE ID to the CCD REDISTRIBUTION­n
extensions.
The second change is the introduction of an epoch­dependency to describe how the line
resolution has deteriorated with time. In this release, the response is given for seven
distinct epochs which parallel the CCF for gain and charge transfer inefficiency (CTI),
which are the fundamental drivers for the response change.
These changes are only useful in conjunction with the cal package version 3.152 or later.
1

Figure 1: The evolving line resolution of the MOS cameras
3 Scientific Impact of this Update
The separation of the response into different epoch files allows the MOS line widths to
be correctly modelled for a given observation date. In previous releases the response has
been fixed at the launch value which has given rise to an underestimate of the detector
line widths and potentially a false detection of broad lines. Figure 1 shows the changing
resolution with time.
Calibrated values for the MOS Timing mode response allows spectra from these observa­
tions to be analysed. MOS Burst mode is currently uncalibrated.
4 Estimated Scientific Quality
Two closed­cal observations taken from revolutions 136 and 455 (MOS1, CCD1) have
been compared to check the width of the Aluminium K ff line (Fig. 2). Using a response
matrix appropriate to revolution 136, a spectral fit correctly returns a narrow line for the
first observation but gives an apparent line width of 10 eV for the rev 455 data (Table 1.)
If the CCF element, EMOS1 REDIST 0026.CCF, is used to generate the correct response
for the revolution 455 observation then the fit shows a narrow, unresolved line as expected.
2

Figure 2: Line spectra from two cal­closed, MOS­1, observations
The line centre energy is seen to be accurate to 3--4 eV at 1.5 keV and the line widths are
estimated to be correct to ¸ 3% with this CCF release.
5 Expected Updates
The MOS CCDS have been cooled to improve their performance [1]. This has improved
the charge­transfer efficiency and led to a narrowing of the line resolution. A new CCF
will be needed to describe the post­cooling epoch for both cameras.
6 Test procedures
The changes introduced here are only used within rmfgen. They are designed to allow
the SAS to produce RMFs which are equivalent to the LUX canned matrices. Tests will
check that they are as similar as possible.
1. Mos­1, epoch 1 (rev 62) imaging mode response, all patterns.
2. Mos­2, epoch 7 (rev 408) imaging mode response, pattern 0 only.
3

Table 1: Spectral fits to the Al­K ff calibration line
Cal. Obs. a Response[0021] b Response[0026] b
(revolution) Energy Width Energy Width
(keV) (keV) Energy Width
136 1:490 \Sigma 0:001 0:0008 \Sigma 0:0012 ­ ­
455 1:491 \Sigma 0:001 0:0104 \Sigma 0:0045 1:491 \Sigma 0:001 0:0009 \Sigma 0:0062
a The revolution number of the calibration observation.
b Spectral fits to the Al calibration line using the response appropriate to revolution
136 (from EMOS1 REDIST 0021.CCF) and the correct response for revolution 455 (from
EMOS1 REDIST 0026.CCF).
3. Mos­1, epoch 7 (rev 408) timing mode reponse.
7 Test results
The difference between the MOS standard RMF and the SAS responses generated using
these new CCF elements are shown in Figures 3--5. The MOS­1 and MOS­2 responses are
duplicated perfectly.
References
[1] Abbey, A.F., Bennie, P.J., Turner, M.J.L., Altieri, B., Rives, S., (in prep.)
4

Figure 3: Comparison of MOS­1 canned response function v SAS equivalent for a 0.5 keV line from an
imaging mode observation of revolution 60, using all patterns.
5

Figure 4: Comparison of MOS­2 canned response function v SAS equivalent for a 0.7 keV line from an
imaging mode observation of revolution 408 using pattern 0.
6

Figure 5: Comparison of MOS­1 canned response function v SAS equivalent for a 1.0 keV line from a
timing mode observation of revolution 408.
7