Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/external/xmm_sw_cal/icwg/presentations/XMM_KIRSCH_POLLOCK.pdf Äàòà èçìåíåíèÿ: Mon Jun 19 14:03:05 2006 Äàòà èíäåêñèðîâàíèÿ: Sat Dec 22 15:11:28 2007 Êîäèðîâêà: ISO8859-5

ESAC

XMM-Newton -
Calibration strategy, status and factors limiting success

M.G.F. Kirscha),and the EPICb,c,d CAL Team A.M.T. Pollocka) and RGSe CAL Team
aEuropean

bMP

E,

cLeiceste

Space Agency, r University, dCEA/Saclay & APC eSRON,

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 1


The Mission

ESAC

Xray-Multi-Mirror-Mission
Cornerstone Mission of ESA's Horizon 2000 Program
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 2


XMM-Newton hardware features
Ç
E PIC-M OS -Camer a 2 E PIC-M OS -Camer a 1

ESAC
EPIC:

RG S

Ç
RG S-C amera 1 RG S-C amera 2

2 Reflecting Grating Spectrometers

- 3 independent CCDcameras (2 MOS & 1 PN), observing simultaneously the same field - 3 different light filters for both camera types - different modes to accommodate brightness and timing - high-resolution spectroscopy of bright sources in the energy range from 0.3 to 2.1 keV - Extends the spectral coverage of XMMNewton into the UV and optical - Six broadband filters - Two grisms, one in the UV and one in the optical
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 3

Optical Mo nitor

EP IC-p n -Camer a

Ç

Optical Monitor

Mi rro rs


general approach of calibration
In orbit calibration (O/C) using - SNR: N132D, 1E0102-7219, Cas A - Continuum sources: PKS2155-304 - Isolated neutron stars: RXJ1856 - Stars: zeta Puppis
Ç Ç

ESAC
refinement of ground calibration system now exposed to orbital environment (radiation, temperatures, ...) necessity of tracking time dependent effects

Ç

Ç Ç Ç

Ground calibration (G/C) at - PANTER/Germany - Lure/ France - Bessy/Germany

main calibration of all system components system still virginal no time dependence measurements possible
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 4


G/C facilities
LURE/ORSAY Synchrotron

ESAC

pn: 6 weeks

2100 man-hours=262 man-days

+ 5 people for integration and 6 test facility MOS: 12 weeks 6050 man-hours=756 man-days Anecdote: note: G/C coincided with world-cup 1998

PANTER/MUNICH X-ray tube test facility

pn: 5 weeks 2200 man-hours =250 man-days Kirsch & Po + 5 people for integration and 4 test facilityllock, XMM-Newton European Space Astronomy Centre MOS: 4 weeks 2200 man hours =250man-Page 5 days


calibration topics
Mirrors:

ESAC

Ç imaging Ç effective area Ç energy redistribution Ç Gain/CTI Ç timing Ç background

Eff. Area PSF Astrometry Vignetting

Filter: eff area

CCD:
QE, CTI, Gain, redistribution, Astrometry

electronics:
Gain, Timing, Modes

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 6


Point Spread Function:
Def.: spatial distribution of light in the focal plane in response to an observed (monochromatic) point source. The PSF integrates to 1 over the infinite focal plane.

ESAC
mirrors

MOS1
110 arcsec

MOS2

pn
filter CCD electronics

G/C: - measurements at PANTER for various monochromatic lines - simulations with SCISIM - analytical model: King function PSF=A[1+(r/r0)2]- problems with ground cal concerning spectral shape and normalization of spectra for different extraction regions

O/C: - refinement of parameters using the BHC MCG-06-30-15

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 7


effective area
Def.: the effective area is the collecting area of the optical elements and detector system of the EPIC cameras as a function of energy. 1. Collecting area of mirror 2. Gratings 3. Filter transmission 4. Quantum Efficiency of CCDs 5. Vignetting

ESAC
mirrors

filter CCD electronics

1. mirrors

2. Filter transmission

4. Vignetting

3. QE of CCDs

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 8


effective area
Def.: the effective area is the collecting area of the optical elements and detector system of the EPIC cameras as a function of energy. 1. Collecting area of mirror 2. Gratings 3. Filter transmission 4. Quantum Efficiency of CCDs 5. Vignetting

ESAC
mirrors

filter CCD

G/C: - mirror effective area measurements at PANTER for all mirror modules and simulations - thick filter troughput measurements at Bessy - thin and medium filter measurements at the Osservatorio Astronomico di Palermo - long measurements for QE at PANTER for various monochromatic lines - edge scans at LURE for Si and Au

O/C: - pn-QE refinement due to other thickness of wafer and SiO2 layer - refinement of mirror parameters around edges using very bright sources (much higher statistical accuracy possible than on ground) - vignetting refinement needed due to uncertainty in optical axis position: various pointings of 3C58

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 9


energy redistribution
Def.: The energy profile recorded by the detector system in response to a monochromatic input. Ç mode/time dependent different rmfs required Ç difficult at low energy (large interplay between redistribution and efficiency)

ESAC
mirrors

Cu-L
filter CCD electronics

pn-response to monochromatic ground calibration source ibration

G/C: - long measurements at PANTER for various monochromatic lines - edge scans at LURE for Si and Au

O/C: - refinement using various Blasars and the isolated neutron star RXJ1856 isolated neutron - MOS spatial and time dependent RMFs needed due to evolving patch

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 10


CTI & Gain
Ç Ç CTI (Charge Transfer Inefficiency) is the imperfect transfer of charge as it is transported through the CCD to the output amplifiers during read-out. Gain is the conversion (amplification) of the charge signal deposited by a detected photon, from ADU (Analogue to digital unit) charge into energy (electron-volts).

ESAC
mirrors

filter CCD electronics

G/C: - long measurements at PANTER for various monochromatic lines - measurements with a floureszence tube at LURE

O/C: - refinement using line rich SNRs N132D and Cas-A - CTI degradation in orbit due to radiation: Time dependence is monitored with internal calibration source (Al-K and Mn-K lines)
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 11


EPIC-Charge Transfer Efficiency
Ç

ESAC

EPIC-MOS Mn
Ç

EPIC-pn CTE degradation is slight and in agreement with pre-launch predictions no clear correlation between the EPIC-pn CTE degradation and proton flares solar flares created a series of jumps in the EPIC-MOS cameras CTE EPIC-pn CTE is degrading independently of the solar flares with a nearly constant rate

EPIC-MOS Al
Ç

EPIC-pn Mn
Ç

EPIC-pn Al

see M. Kirsch, SPIE 5898-29

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 12


timing
Ç Ç relative time accuracy of EPIC-pn: P/P=(Pradio-Px-ray)/PRadio absolute time accuracy: TRadio-TX-ray

ESAC
mirrors

filter CCD electronics

G/C: - chopper measurements down to 1 ms Period at University of Tuebingen and Panter

O/C: - monitoring of Timing accuracy with the Crab (Period: 33 ms)

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 13


background
Ç Ç Ç Ç low energy electronic noise soft proton "flares" quiet time high energy proton induced astrophysical background

ESAC
mirrors

MOS Al K

PN Cu K
filter CCD electronics

G/C: - long CLOSED measurements at PANTER

O/C: - CLOSED measurements - various BG modells
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 14


major calibration steps and improvements
Ç Ç Ground calibration (1997-2000)
- PANTER, Orsay, Bessy

ESAC

CAL-PV-phase (2 Feb - 24 June 2000)
- - - 22 calibration 29 performance verification targets CTI orbit calibration, RMF tuning pn offset calculation method

Ç

Routine phase
- 2001: pn-Timing modes energy calibration - 2002: - relative timing problem solved - MOS CTI degradation requires epoch dependent energy calibration - cooling of the MOS cameras to slow down degradation process - 2003: - vignetting recalibration (optical axis) - major cross calibration campaign started - pn-QE refinement - 2004: - recalibration of PSF - astrometry recalibration - 2005: - discovery of spatial and time-dependent redistribution change in both MOSs epoch and spatially dependent RMFs and recalibration of pn effective area and rmf

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 15


EPIC cross cal differences
Ç

ESAC
EPIC shows at low and high energies difference that can lead to
- differences in slope of up to 0.05 (M1
Example plot

0.50-0.85 0.85-1.50 1.50-4.00 -5.4 2.4 6.8 (M1-pn)/pn [%] -1.6 4.1 7.3 (M2-pn)/pn [%]
Ener gy Band [keV]

4.00-10.0 11.4 7. 4
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 16

Averaged flux differences from 17 blazar observations


ESAC

XMM-Newton

Ç Ç Ç

Reflection Grating Spectrometer(s)

line-rich stars

smooth-continuum AGN

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 17


An RGS instrument

ESAC

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 18


Ground Calibration++

ESAC

mirror

grating



CCD

=

effective area

Ç physical instrument model coded in CCF and SAS to encompass all parameter space Ç consolidation during early-flight calibration and verification phase Ç |m| > 1 Ç scale Ç cross-dispersion scattering Ç RGS cooling campaign Ç noise reduction & fewer and cooler hot pixels Ç RGS instrument monitor dynamic CCD offsets Ç on-board calibration sources gain Ç effects of radiation damage Ç synthetic background spectra
Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 19


RGS SAS & CCF components
BORESIGHT LINCOORDS MISCDATA

ESAC

rgsproc

HKPARMINT ADUCONV BADPIX CROSSPSF CTI EXAFS LINESPREADFUNC QUANTUMEFF REDIST EFFAREACORR TEMPLATEBCKGND

Ç Ç Ç Ç Ç Ç Ç Ç Ç Ç Ç Ç Ç Ç

atthkgen rgsoffsetcalc rgssources rgsframes rgsbadpix rgsevents evlistcomb gtimerge rgsangles rgsfilter rgsregions rgsspectrum rgsrmfgen rgsfluxer

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 20


What didn't go as expected ?
Ç inconsistency between RGS1 and RGS2
Ç instrumental oxygen Ç other Henke improvements Ç < 7 scalar EM theory inadequate Ç > 25 ground calibration difficult Ç effective area changes - probably due to hydrocarbon contamination = Ç failure of readout electronics of RGS1 CCD7 & RGS2 CCD4 Ç small systematic scale errors

ESAC

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 21


How were these problems tackled ?
Ç RGS1 || RGS2 Ç repeated observations of "calibration standards" Ç power-law blazars : PKS2155-304 & (1Ms of) Mkn421 Ç SNR : 1ES0102-7219 & N132D Ç stellar coronae : Capella, AB Dor & HR1099 Ç RXJ1856-3754 = Ç C ra b Ç blank fields Ç ATOMDB & CHIANTI Ç RGS1 || RGS2 || EPIC

ESAC

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 22


RGS "rectified" fluxed spectra of two blazars

ESAC

g()=2-f()exp(+NH())

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 23


RGS response to SNR 1ES0102-7219

ESAC

CVI Ly 33.734

OVIII Ly 18.967

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 24


The Mkn421 Ms with RGS2

ESAC

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 25


Mkn421 Ms consequences

ESAC

Of the 5000 pixels of the RGS Ç Ç Ç Ç

...

85% obey Poisson statistics 14% exceed e-ÅÅn/n! by <5% 1% hot pixels or columns 1% cool columns

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 26


finally... SAS v7.0 RGS vs EPIC statistics

ESAC

Kirsch & Pollock, XMM-Newton European Space Astronomy Centre Page 27