Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0256-1-0.ps.gz Äàòà èçìåíåíèÿ: Wed Feb 17 18:51:03 2010 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 22:57:16 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ï ï ï ï ï ï ï

XMM­Newton CCF Release Note
XMM­CCF­REL­256
Rate­dependent CTI correction for EPIC­pn Timing Modes:
SASv9.0 update
M. Guainazzi
August 27, 2009
1 CCF components
Name of CCF VALDATE List of Blocks
changed
Change in CAL HB
EPN CTI 0022 2000­01­01T00:00:00 RATE DEPENDENT CTI NO
2 Changes
2.1 Rate­dependent CTI correction
This update of the EPIC­pn CTI CCF constituent includes a recalibration of the rate­dependent CTI
correction for Timing Modes with SASv9.0. This re­calibration was driven by the fix of an energy­
dependent bug in the calculation of the PSF encircled energy fraction in Timing Modes by arfgen.
This bug may have a#ected in principle the values of the energy­independent gain factor, G corr on
which the calibration of this correction is based (see the CCF Release Note XMM­CCF­REL­248
for further details).
2.2 Calibration results
The SASv9.0­based recalibration of the G corr as a function function of the number of shifted electron
per pixel per second, N e
G corr = a 0 # N a1
e + a 2
is shown in Fig. 1
1

XMM­Newton CCF Release XMM­CCF­REL­256 Page: 2
Rate dependent CTI ­ pn timing mode
10 100 1000
Shifted electrons/pixel/sec
0.85
0.90
0.95
1.00
1.05
Gain
0060740901
0061140101
0061140201 0084020401
0084020601 0085290301
0087350101 0087350601
0087350801 0090340201
0099280101 0109090101
0111040101 0111061601
0111370101
0112900101
0116690801
0117510201
0122340101 0124930301
0127720501
0134120101
0137551001 0144090101
0148220201 0148220301
0149810101 0150498701
0153750401 0153950301
0157960101 0160960201
fit: y=a0(x a1 )+a2
a0= 0.9922
a1= 0.0033
a2=­0.0079
Rate­dependent CTI ­ pn burst mode
100 1000 10000
Shifted electrons/pixel/sec
0.85
0.90
0.95
1.00
1.05
Gain
0112920701
0112920901
0112921401
0112921501
0112921601
0112990101
0117510201
0135730701
0155762501
0160960401
0160960601
0160961301
0202401101
0202401201
0311590501
0312790201
0312790401
0412590301
0412590701
fit: y=a0(x a1 )+a2
a0=0.003
a1=0.305
a2=0.943
Figure 1: G corr versus N e relation for the sample EPIC­pn exposures in Timing (left panel) and
Burst (right panel) Modes, respectively. The solid line indicates the best fit with the functional
form: a 0 # N a1
e + a 2 . The dot­dashed line in the left panel indicates the envelope corresponding to
the 1­# errors on the best­fit parameters.
Table 1: a i parameters in the EPN­CTI CCF version #22
Mode a 0 a 1 a 2
Timing 9.920 â 10 -1 3.300 â 10 -3 -7.900 â 10 -3
Burst 3.200 â 10 -3 3.050 â 10 -1 9.426 â 10 -1
In Tab. 1 the new a i coe#cient are listed.
3 Estimated scientific quality
It was expected that this change might have an impact of at most a few percent in the spectral
residuals.
4 Test procedure and results
The whole testbed of 142 exposures in Timing Mode and 48 observations in Burst Mode used to
test previous versions of this CCF constituent has been re­analysed. An updated evaluation of the
spectral quality with the new CCF is summarised in the following document: ``Evaluation of the
spectral calibration accuracy in EPIC­pn fast modes'' by M. Guainazzi, M. Kirsch, F. Haberl, and
M. de Ovelar (XMM­SOC­CAL­TN­0083).

XMM­Newton CCF Release XMM­CCF­REL­256 Page: 3
5 Expected Updates
Further updates of this extension will include:
. recalibration of the pattern fraction in EPIC­pn Timing Mode, explicitly introducing its RAWY
dependency
. inclusion of an energy­dependent term in the determination of the rate­dependent G corr values
for Timing Mode