Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0047-1-0.ps.gz Äàòà èçìåíåíèÿ: Wed Jan 3 15:10:24 2001 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 21:47:25 2012 Êîäèðîâêà:

XMM­Newton CCF Release Note
XMM­CCF­REL­47
EPIC Filter Transmission
D Lumb
November 16, 2000
1 CCF components
Name of CCF VALDATE List of Blocks
changed
CAL VERSION XSCS flag
EMOS1 FILTRANSX 0007 2000 01 01T00:00:00 FILTER­THICK,
THIN ,MEDIUM,
EBINS
NO
EMOS2 FILTRANSX 0007 2000 01 01T00:00:00 FILTER­THICK
,THIN ,MEDIUM,
EBINS
NO
EPN FILTRANSX 0010 2000 01 01T00:00:00 FILTER­THICK
,THIN ,MEDIUM,
EBINS
NO
2 Changes
2.1 THICK Filter
The previously submitted data set was determined to have come from the flight spare PN thick
filter. The Flight Model unit data sets measured at HASY laboratory have been identified, and
these used to populate the CCF. The tranmission is somewhat lower than for the previous set.
However the HASY data have an incorrect Al K absorption edge energy, and the apparent
EXAFS structure in transmission does not match well that seen at BESY and with other filters.
Therefore I splice in the BESY data for the Flight Spare (suitably scaled) close the the 1550eV
absorption features.
Furthermore there were no HASY data for higher energies, whilst there ARE data for the
1

XMM­Newton CCF Release XMM­CCF­REL­47 Page: 2
Sn absorption feature around 4keV from the BESY measurements. Hence, taking a model for
the polypropylene/Sn/Al layers that best matches the HASY data, we extrapolate to energies
–2keV,and scale the BESY data to that
Figure 1:
2.2 MEDIUM and THICK Filters
EPIC measurements for the MEDIUM and THICK filters were made at Orsay synchrotron for a
limited set of energies. In addition the manufacturer (MOXTEK) has estimates of the polyimide
and Aluminium thicknesses based on point measurements, which indicate very close matched data
for all filters. A model based on the Henke coefficients has been taken and reconciled with the few
Orsay data points. In order to provide a better match at the critical absorption edges, however, the
inferred attenuation coefficients were extracted from the measurements made on the similar ACIS
blocking filters, and an edge structure spliced into the EPIC data sets.

XMM­Newton CCF Release XMM­CCF­REL­47 Page: 3
Figure 2:
2.3 Energy points
The format of the filter files requires a common energy bin structure, so the above data sets were
then interpolated with a Hermite polynomial technique.
3 Scientific Impact of this Update
Better low energy spectral fitting should result. However we note that some of the mirror and CCD
efficiency data has already been inferred with in­flight measurements, ASSUMING previous filter
transmission data. Therefore some re­iteration on all 3 data sets will be necssary.

XMM­Newton CCF Release XMM­CCF­REL­47 Page: 4
Figure 3:
4 Estimated Scientific Quality
We are still lacking the knowledge of which THICK filter was flown on the MOS cameras and given
these filters varied significantly there could be 10% (energy dependent ) discrepancies for the MOS
THICK filters.
There are known spatial variations around the C edge of THICK filter, and that not all filters
are identical. These features will have to be updated, and a spatial dependence in the TELCO­
ORD co­ordinate system established, with suitable defocusing accounted for. The figures compare
measurements with the previous CCF contents, and the new data sets.
5 Expected Updates
ffl Revised data when improved in­orbit filter measurements are made (RXJ0720 for exasmple)

XMM­Newton CCF Release XMM­CCF­REL­47 Page: 5
ffl Spatial variation of THICK filter transmission
ffl Revised MOS THICK filter data when the piece numbers are identified.