. : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0112-1-0.ps.gz
: Fri Mar 1 13:35:01 2002
: Mon Oct 1 22:11:05 2012
:
: rho ophiuchi
XMMNewton CCF Release Note
XMMCCFRELnonumberassigned
EPIC Spectral Response Distribution
R. D. Saxton
27 Feb 2002
1 CCF components
Name of CCF VALDATE CAL VERSION XSCS flag
EMOS1 QUANTUMEF 0010.CCF 20000101 NO
EMOS2 QUANTUMEF 0010.CCF 20000101 NO
EPN QUANTUMEF 0010.CCF 20000101 NO
2 Changes
Four new extensions, FRACTION CHANNEL, FRACTION ENERGY, QE TOTAL and
EBINS FRACTION have been added. These have been added to allow the response
generation routines, rmfgen and arfgen, to produce matrices which duplicate those of the
instrument teams. This new method of calculating the RMF was instigated at MPE to
create response matrices which gave less weight to doublepixel events at low energies [1].
The new RMF matrix generation scheme is as follows:
For each energy
--
For each event grade (single, double, triple or quad)
--
create the redistribution function (rmf) as normal
multiply the rmf with the pattern fractions for this
event grade [FRACTIONCHANNEL]

Add together the rmfs for each grade
Normalise the total rmf to 1.0
1


The new quantum efficiency (QE) part of the ARF generation is:
For each energy
--
get the QE of the detector in all patterns [QETOTAL]
multiply the QE by the pattern fraction for the event grades
in the spectrum [FRACTIONENERGY]

2.1 Structure of the new extensions
The new extensions have the following columns:
FRACTION CHANNEL
MODE ID: The observing mode of the instrument
THRESH: The thresholding used within the detector in ADU channels.
REGION: The spatial region over which these fractions apply
FRAC S: The fractions in each channel for singlepixel events
FRAC D: The fractions in each channel for doublepixel events
FRAC T: The fractions in each channel for triplepixel events
FRAC Q: The fractions in each channel for quadruplepixel events
FRAC SD: The fractions in each channel for single and doublepixel events
FRAC SDTQ: The fractions in each channel for single,double,triple and quadruplepixel
events
FRACTION ENERGY
MODE ID: The observing mode of the instrument
THRESH: The thresholding used within the detector in ADU channels.
REGION: The spatial region over which these fractions apply
FRAC S: The fractions in each energy for singlepixel events
FRAC D: The fractions in each energy for doublepixel events
FRAC T: The fractions in each energy for triplepixel events
FRAC Q: The fractions in each energy for quadruplepixel events
FRAC SD: The fractions in each energy for single and doublepixel events FRAC SDTQ:
The fractions in each energy for single,double,triple and quadruple pixel events
2

QE TOTAL
QE TOTAL: The total quantum efficiency, over all patterns, for each energy.
EBINS FRACTION
ENERGY: The energies used to define the QE TOTAL array.
3 Scientific Impact of this Update
This update allows the SAS to produce matrices which give a response which is within
1% of the equivalent files produced by the instrument teams.
4 Estimated Scientific Quality
The pattern fractions in this release have been calculated from inorbit data. The PN
values were calculated from a rev 0082 observation of MS1229.2+6430 for the Full Frame
mode, from merged observations of the Coma cluster for Extended Full Frame mode, from
observations of Mkn421 (rev 0171) for Small Window mode, from a rev 0075 observation
of Mkn 205 for Large Window mode and from an observation of GX13+1, rev 0057, for
Timing mode.
The fractions are not expected to be significantly mode or position dependent for the MOS
detectors and so have been calculated from a Full Frame mode, rev 0082, observation of
MS1229.2+6430. These values are used for all modes and at all detector positions in the
current release.
The total QE values in this release, for the PN, are an extrapolation of the current QE 0,
QE 1 values calculated from the initial ground calibration runs, so there has been no
change here. The values for the MOS have been taken directly from the total QE used to
generate the standard instrument matrices by LUX. These numbers are expected to be
superceded within days so no more details will be given here.
The number of energy bins in the standard PN matrices has been increased from 972 to
1319. These new energies are contained in the EBINS FRACTION extension.
3

5 Expected Updates
The MOS QE is under intense review at the moment. A new QE curve for MOS1 and
MOS2, which will be used by the SAS and LUX, is expected to be released shortly.
6 Test procedures
The changes introduced here are only used within arfgen, rmfgen and calview. The whole
point of it is to produce a response which is the same as the instrument standard responses
and so this is the check which needs to be applied.
The routines, eexpmap and lccorr use extensions in this file which should not have changed
in this release.
7 Test results
The difference between the MOS and PN standard responses and the SAS responses
generated using these new CCF elements are shown in Figures 1 and 2. The agreement
is very good except at the extremes of the spectrum.
References
[1] Haberl, F., Briel, U.G., Dennerl, K. and Zavlin, V.E., ''Spectral response of the EIC
PN detector: Basic dependencies'', Proceedings of the symposium 'New visions of the
Xray Universe in the XMMNewton and Chandra era'.
4

Figure 1: Comparison of MOS1 response to a slope=1.5 powerlaw spectrum using event paterns 0--12
5

Figure 2: Comparison of PN response to a slope=1.5 powerlaw spectrum using a spectrum in FF mode
and pattern 0 only
6