Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0200-1-0.ps.gz
Дата изменения: Thu Nov 10 13:23:14 2005
Дата индексирования: Mon Oct 1 22:39:31 2012
Кодировка:
XMM-Newton CCF Release Note
XMM-CCF-REL-200
Update of EPIC MOS gain
Martin Stuhlinger, Darren Baskill, Stephane Rives, Jenny Carter
October 21, 2005
1 CCF components
Name of CCF VALDATE EVALDATE List of Blocks CAL XSCS
(start of val. period) (end of validity period) changed VERS. ag
EMOS1 ADUCONV 0028 2005-07-24T00:54:36 2005-10-19T19:03:55 OFFSET GAIN NO
EMOS2 ADUCONV 0028 2005-07-24T00:54:36 2005-10-19T19:03:55 OFFSET GAIN NO
2 Changes
A new ADUCONV CCF has been generated to correct for a large step in the line energy positions
of MOS1 CCD1 and CCD5 starting at the last eclipse season.
The voltages of MOS1 CCD's in use on-board during rev. 1030-1073 were not the correct ones.
They correspond to settings prior to March 2000. CCD1, CCD4 and CCD5 were a ected. The
inconsistency was on the following voltages:
MOS1 CCD1: VRD1=18.0 V instead of 17.0 V
CCD RESET 1= 9.0 instead of 10.0 V
CCD RESET 2= 9.0 instead of 10.0 V
MOS1 CCD4: CCD RESET 1= 9.0 instead of 10.0 V
CCD RESET 2= 9.0 instead of 10.0 V
MOS1 CCD5: CCD VOG1=1.5 V instead of 1.0 V
These changes caused a decrease/increase of the gain in CCD1/CCD5 according to gure 1 and
gure 2. CCD4 did not show signi cant changes.
The reason for this undesired change of voltages was the following: Due to the loss of CCD6,
it was decided to set all the voltages of CCD6 to 0.0V. This is done within a sequence that sets
1

XMM-Newton CCF Release XMM-CCF-REL-200 Page: 2
the voltages of all the CCD's. To update an existing sequence, we use a tool that performs the
updates based on the existing sequence. For some reason to be investigated, the tool did not have
the last version of the sequence in its database. The one present was actually prior to March 2000.
Therefore the actualization for the CCD6 voltages has been performed on a sequence with prior to
March 2000 values for some CCD's. This sequence is sent only after switch-o of the instrument,
i.e during the eclipse season. It explains that the problem started to be visible at the beginning of
the eclipse season.
These new CCFs establish a new time period between 2005-07-24 and 2005-10-19 (rev. 1030-
1073). For MOS1, new gain and o set parameters for single and double events were implemented in
the CCF. This opportunity was also used to update the MOS2 gain/o set parameters for the single
events, too. As the parameters for all other pattern types are still to be derived, their parameters of
the latest epoch are used. Because the single and double events dominate the pattern distribution,
the new CCF is able to suppress the residuals present in the energy scale due to the incorrect voltage
settings onboard. Because the central MOS1 CCD1 is a ected, this temporary CCFs are published
to privide the user community the best MOS1 calibration of at least single and double events for
the a ected eclipse period, until all parameters for the other pattern types are available as well.
3 Scienti c Impact of this Update
For MOS1 CCD1 and CCD5 in the new time period the energy scale is now reconstructed again to
about 10 eV or better on the whole energy range.
See gure 3 and gure 4 for the MOS1 CCD1 and CCD5 energy line position of the Mn K and
Al K from the calibration source.
4 Estimated Scienti c Quality
The energy scale accuracy is below or equal to 10 eV on the whole energy range for i) not too bright
sources and ii) ouside of eclipse seasons (at the start of revolutions). In this two cases, as explained
in XMM-CCF-REL-124, the energy scale can be signi cantly over-corrected.
5 Test procedures & results
The new ADUCONV CCFs have been tested with SAS v6.5.0. The results are shown in gure 3
and gure 4.
6 Expected Updates
Current investigations re-de ne the complete energy treatment (CTI, gain) of all MOS CCDs for
both cameras. This re nement will cover all time epochs, especially periods where the CTI/gain
combination is worse than 5 eV.

XMM-Newton CCF Release XMM-CCF-REL-200 Page: 3
Figure 1: Mn K and Al K line energy scale for MOS1 CCD1 with the old ADUCONV CCFs (left:
single events, right: PATTERN 0-12).
Figure 2: Mn K and Al K line energy scale for MOS1 CCD5 with the old ADUCONV CCFs (left:
single events, right: PATTERN 0-12).

XMM-Newton CCF Release XMM-CCF-REL-200 Page: 4
Figure 3: Mn K and Al K line energy scale for MOS1 CCD1 with the new ADUCONV CCFs (left:
single events, right: PATTERN 0-12).
Figure 4: Mn K and Al K line energy scale for MOS1 CCD5 with the new ADUCONV CCFs (left:
single events, right: PATTERN 0-12).