Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200502/db_cticolumns.pdf Дата изменения: Fri May 16 13:27:15 2014 Дата индексирования: Wed Apr 13 13:27:00 2016 Кодировка:

MOS monitoring
Fading Cal sources

MOS CTI Column Correction

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS CTI Column Correction
Problem spotted in 1E0102...

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Problem spotted in 1E0102...

-Variation in column-to-column counts - cause by variations in column-tocolumn CTI

left: Jean's plot col320 Vs neighbouring 4 columns <500eV, effects embadpixfind

Two main problem columns ­ MOS1 c320, MOS2 c291

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Time dependency? Energy dependency?

time

Time dependency... Predictably, it evolves with time as Jean spotted... But we can track the time variability in the CalClosed data

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Method: - Combine many CalClosed observations (interflare periods, per >~100 rev) - Produce a data cube of RawX Vs RawY Vs Energy (after SAS CTI correction) - Fit Gaussian to each bin (1 column wide, 60 p ix e ls lo n g )

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Al Gaussian centroid position offset from the average ­ individual columns over 6 time periods

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 Col319 Al Deviation from CCD average (adu)

Orbital revolution
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Col 290 Col 291

Energy Offset (adu)

Col 289

MOS2 Al

Time dependency

Orbital revolution

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Energy dependency?

Do the size of the offsets depend on energy MOS1 column 320

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Stacked rev 718-920 CalClosed's, MOS1 CCD1, sas6.1.0 Each point represents a single column

Al offset

ence depend Energy 0.25x -0.069+ y=

y rg e e en enc No nd e ep d

Mn K offset

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Which equation should we use? CTIC PHA=PHA+rawX.(ctiX+colcorr)+rawY.ctiY
but this an approximation to...

CTIC PHA=PHA*(1+ctiX/PHA)^rawX...

& rawY

for energy of ~1keV : evts.pha~300 ADU with low CTI of 0.01 ADU/transfer ... Approx method = 300+300*0.01=303 Better method = 300*(1+0.01/300)^300 = 303.19 <1ev difference For a bad CTI=0.2 ADU/transfer (i.e. a 60 ADU/200 eV shift at 1keV) Approx method = 300+300*0.2= 360 Better method = 300*(1+0.2/300)^300 = 366.391 ~20eV difference
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 col. 320 no Energy dependence in the column offset correct

Energy (10 eV bins)
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 col. 320 Energy dependence in the column offset correct

Energy (10 eV bins)
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS2 col. 291 no Energy dependence in the column offset correct

Energy (10 eV bins)
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS2 col. 291 Energy dependence in the column offset correct

Energy (10 eV bins)
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


EPIC-MOS Monitoring
eV=cons+linr*adu eV

Gain
adu

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 - Al line energy

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 - MnK line energy

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 - Linear Term
MOS2 linear term also fine...
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS1 - Constant Offset Term

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


MOS2 - MnK line energy

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


EPIC-MOS Fading Cal sources
MOS1 Mnk

MOS1 Al

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


EPIC-MOS Fading Cal sources
· IDL script to merge several CalClosed observations into one · Should work with all SAS tasks · If anyone wishes to test it...!

Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Conclusions...
· Monitoring ­ update CCFs with new gain constant offset for a couple of CCDs?

· MOS column-by-column doable & accurate ­ need SAS programming · Should column-by-column correction be just an offset or an energy dependent offset?

...and finally....
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/


Please send me your calibration talks! dbl@star.le.ac.uk
Mallorca EPIC Cal Meeting - February 2005 Darren Baskill: http://www.star.le.ac.uk/~dbl XMM project: http://www.src.le.ac.uk/projects/xmm/