Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/calibration/documentation/epic_cal_meetings/200502/hm_effarea.pdf Äàòà èçìåíåíèÿ: Fri May 16 13:29:13 2014 Äàòà èíäåêñèðîâàíèÿ: Wed Apr 13 13:21:42 2016 Êîäèðîâêà:

Effective Area Analysis using the Chandra HETGS
Herman L. Marshall Chandra X-ray Center (MIT)


Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

2


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio update ready -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
3

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


· · · ·

ACIS BI/FI QE
New BI QEs improve matters considerably FI loss due to CR blooms is included At most 3-4% differences in 2-10 keV range No problems within HETGS data

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

4


· · · ·

ACIS BI/FI QE
New BI QEs improve matters considerably FI loss due to CR blooms is included At most 3-4% differences in 2-10 keV range No problems within HETGS data

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

5


· · · ·

ACIS BI/FI QE
New BI QEs improve matters considerably FI loss due to CR blooms is included At most 3-4% differences in 2-10 keV range No problems within HETGS data

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

6


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
7

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


HETGS and LETGS Pileup appears at EA jump · Edge (Rf) ~ 0.01 2.1 keV due to5% loss ct/frame/col., · Rate · Mk 421 (4148): Rf = 0.05, giving jump of 19%

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

8


LETGS -- Pileup

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

9


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- see later talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
10

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


· · ·

Appears when using the new BI QE models Effect is an inverse, reversed edge Jump is about 6%

Si-K Edge in BI QE

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

11


Fitting Si-K Edge in HETGS Residuals
· · ·
Use 13 blazar observations Si-K is near Ir-M

· ·

Fixing Ir-M edge requires good fit to Si-K Use 2 steps: approx. Si-K to get Ir-M, then use good Ir-M fix to finish Si-K fix adjustment

Residual has near edge structure -- use Si-K opacity from ACIS team
Herman Marshall (MIT CXC) Mallorca Feb. 05
12

Chandra Effective Area


Ir-M

Si-K

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

13


HETGS Residuals after fixing Ir-M edge

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

14


Fit to Si-K edge

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

15


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
16

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


· · · ·

N-K Edge in FI QE
Found 2 yr ago in PKS 2155-304 data Objective was to find N-K in contaminant Target was offset to put N-K on BI chip N-K is OK in filter, not in FI chip

ISM ACIS OBF ACIS OBF ACIS FI+OBF

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

17


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
18

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


· Filter above · Composition from edges: C:O:F:N ~ 55:5:4:<2 · ECS gives slower deepening than elemental model

Contamination dominates below .2867 keV, contaminant

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

19


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
20

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


HRMA Ir-M fix & HETGS

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

21


HEG/MEG Comparison
Fit spans full HEG wavelength range

Damp polynomial here

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

22


· · · ·

HETGS Effective Area Calibration ACIS-S
· · · · · · · · · ·
BI QE vs. FI QE -- now down from <15% to <5% Pileup -- impact depends on source brightness Si-K edge -- 6% edge residual N-K edge in FI chips -- more important for LETG/ACIS Contamination -- not this talk C-K edge energy shift -- only important for LETG/ACIS O-K edge -- incomplete, depends on contamination MEG/HEG ratio to be updated -- <7% for E > 0.8 keV Compare LETG to HETG -- incomplete Ir-M edge -- 10% jump at 2.075 keV Herman Marshall (MIT CXC) Mallorca Feb. 05
23

ACIS OBF HETG Efficiency HRMA Effective Area

Chandra Effective Area


HRMA Ir-M edge
· · ·
Jump is about 10% MEG and HEG agree See Diab Jerius' talk

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

24


HRMA Overlayer Study

10 å overlayer

15 å overlayer

25 å overlayer
Chandra Effective Area Herman Marshall (MIT CXC)

20 å overlayer
Mallorca Feb. 05
25


HRMA Overlayer : 20 ± 5 å
(Updated Optical Constants)

· · ·

20 å overlayer is best now Si-K edge was updated, no residual observed Offset of 5% due to poor fit to PL models

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

26


Fit Results

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

27


Fit Results

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

28


Fit Results

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

29


· ·

MEG-HEG Consistency
Fit PL models to MEG and HEG independently Fit parameters show slight calibration biases

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

30


HETGS of 1H1426+42

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

31


HETGS of 1H1426+42

Chandra Effective Area

Herman Marshall (MIT CXC)

Mallorca Feb. 05

32


· · ·

Summary
Most spectral edges and narrow features are fixed (or fixable) HRMA and ACIS EA require 2 more adjustments

· · · ·

Si-K edge Ir-M edge

MEG fix relative to HEG is still not right
Smoother spectra are possible Cross-calibration with XMM will help Herman Marshall (MIT CXC) Mallorca Feb. 05
33

Chandra Effective Area