Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/docs/documents/CAL-TN-0043-1-1.ps.gz Дата изменения: Thu Jun 24 17:17:22 2004 Дата индексирования: Mon Oct 1 21:19:15 2012 Кодировка: Поисковые слова: zodiacal light

XMM-Newton Calibration Technical Note
MOS optical loading
XMM-SOC-CAL-TN-0043 Issue 1.1
B.Altieri
November 13, 2003
1 Abstract
The MOS optical PSF has been analyzed with diagnostic frames acquired on the RGS calibration
star AB Doradus in the THIN lter, with the aim to assess the MOS optical loading.
This analysis provides better constrains than the previous technical note (XMM-SOC-CAL-TN-
0040, June 27, 2003), based on OPEN lter measurements and supersedes it.
2 Introduction
MOS diagnostics frames (integration time = 2.6s) were performed on the V=7 AB Dor star
(spectal type: K1IIIp) in revolution 709 (RGS calibration observation). These data were used
to characterize the optical PSF, by plotting the EEF (Encircled Energy Function) and estimating
the fraction of the optical ux in the central pixels and to compare D. Lumb's model implemented
in PHS tools (PHS Tools - EPIC Optical Loading, XMM-PS-TN-40, D. Lumb, Nov. 16, 2000).
3 Analysis
As AB Dor is also an X-ray source (!), a special dedicated processing ("deglitching") had to be
applied to remove the X-ray photons from the optical diagnostics frames (integration time =
2.6s). The MOS1 and MOS2 optical PSFs though the THIN lter are shown in gures 1 & 2
respectively (averaging of all diagnostic frames after deglitching). Note that most X-ray events
from the source have been removed, but not the cosmic rays splashes. Only the frames where
no cosmic ray was impinging closer that 30 arcsec from the centre of the source were used for
the analysis.
1

european space agency
agence spatiale europИenne
XMM Science Operations Team
Document No.: XMM-SOC-CAL-TN-0043
Issue/Rev.: Issue 1.1
Date: November 14, 2003
Page: 2
Figure 1: MOS1 optical PSF : AB Dor through THIN lter
Figure 2: MOS2 optical PSF : AB Dor through THIN lter

european space agency
agence spatiale europИenne
XMM Science Operations Team
Document No.: XMM-SOC-CAL-TN-0043
Issue/Rev.: Issue 1.1
Date: November 14, 2003
Page: 3
4 Results
The derived optical EEF is plotted for MOS1 and MOS2 in gures 3 and 4 respectively. Although
the optical PSFs display the same characteristics as in the X-ray regime (MOS2 PSF triangular
and somewhat more peaked), the EEF are remarkably similar.
The core optical PSF is sligthly larger than the X-ray PSF, with a HEW (Half Equivalent
Width) of about 11"(radius), vs 9". On the other hand the 80% or 90% EEF is more peaked
than in X-ray (less scattered optical light in the PSF wings).
The optical loading in the central pixels is about 130 ADUs (123 ADUs for MOS1 and 137 ADUs
for MOS2) and amounts to about 0.5% of the total optical ux for both MOSs.
It is also interesting to note that the MOS2 optical ux is 7% higher than MOS1. This is likely
due to small variations from the lter transmission, (e.g. if 0.99 for MOS2, 0.9907 for MOS2) .
5 Impact on optical loading
The optical loading assessment performed by PHS tools, derived from D. Lumb's model (PHS
Tools - EPIC Optical Loading, XMM-PS-TN-40, Nov 16. 2000) assumes that the fraction of
light that is detected in the central pixel of the PSF distribution is 2% for MOSs (and 15% for
EPIC-pn). It predicts for AB Dor a pixel loading of  1500ADUs, i.e. more than 10 times larger
(11.5) than what is measured here on AB Dor.
This is due partly to the overestimation of the fraction of the optical ux in central pixel by
a factor 4 (2% vs 0.5% found here), The other factor (about 3) is probably coming from an
overestimation of the lter transmission, as the OPEN measurements in revolution 90 showed a
good agreement with the model.
The optical loading assessment in PHS tools, is hence overestimated by a factor 10. The V-
magnitude limits in Lumb (2000) can be decreased by more than 2.5 magnitudes in the THIN
lter (table page vi).
For instance, the optical loading of a V=12 K-star (like AB Dor) will be of 1 ADU in the central
pixels, while Lumb (2000) gives a limit of V=15.4 (but for 1 electron per CCD frame time)
A relative safe magnitude limit to observe with the THIN lter is V11. This yields to about
3 ADUs of optical loading in the central pixels. This level of optical loading is corrected in the
SAS, by a local background correction, using E3 and E4 energies. So only a minimal spectral
distorsion is expected, due an increase of double events and some shot noise due to the increased
"bumped' background.
For sources brighter than V=9, a "dead area" is the core of the PSF appears in THIN lter,
when several adjacent pixel are all above the detection threshold (25 ADUs).

european space agency
agence spatiale europИenne
XMM Science Operations Team
Document No.: XMM-SOC-CAL-TN-0043
Issue/Rev.: Issue 1.1
Date: November 14, 2003
Page: 4
Figure 3: MOS1 optical EEF, compared to the X-ray EEF (dashed line)
Figure 4: MOS2 optical EEF, compared to the X-ray EEF (dashed line)

european space agency
agence spatiale europИenne
XMM Science Operations Team
Document No.: XMM-SOC-CAL-TN-0043
Issue/Rev.: Issue 1.1
Date: November 14, 2003
Page: 5
6 Conclusion
Based on this measurement of the optical loading though the THIN lter and extrapolating to
other lters by assuming an additional optical blocking of 10 2 and 10 5 for the MEDIUM and
THICK lters respectively, the magnitude limits in the XMM-Newton User's Handbook shall
be revised safely to :
 THIN lter: V=12
 MEDIUM: V=7
 THICK V=0
Acknowledgment
Thanks to Stephane Rives for acquiring these MOS diagnostic frames via manual OCRs.
References
PHS Tools - EPIC Optical Loading, XMM-PS-TN-40, Issue draft, D. Lumb, November 16, 2000
EPIC optical PSF, XMM-SOC-CAL-TN-0040, B. Altieri, June 27, 2003