Upload
others
View
8
Download
3
Embed Size (px)
Citation preview
Christian Kirsch & Maximilian Lorenz, Remeis Observatory& ECAP
Simulating Athena with Sixte:The WFI and X-IFU instrument
Sixte Workshop IFCA, Santander — February 2019
The Athena Mission
• to be launched to L2 in ∼2030 as the second ESA L-class mission• large consortium, members from Europe, US, and Japan
(> 200 members)• two instruments on board:• Wide Field Imager (WFI)• X-ray Integral Field Unit (X-IFU)
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 2
The Athena Instruments
WFI (Imager)
• high count-rate, moderate spectralresolution• large field of view
MPE
X-IFU (Calorimeter)
• for high-spectral resolutionimaging• calorimeter operating at 50 mKelvin
CNESChristian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 3
The Wide Field Imager (WFI)
Large Detector Array and Fast Detector (35 mm defocused)
• DePFET activepixel technology(similar to CCDwith line-by-linereadout)• spectral resolution:≤ 170 @ 7keV• high count-rate
capabilities(10 Crab)• large FOV:
40′ × 40′
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 4
Example: The Galactic Center with Athena
5′
25
20
15
10
5
kcts
5′
10
8
6
4
2
kcts
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 5
Example: The Chandra Deep Field South with Athena
5′
30
25
20
15
10
5
kcts
5′
20
15
10
5
kcts
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 6
Example: The Chandra Deep Field South with Athena
A. Rau
dithering efficientlyremoves gaps be-tween the chips
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 7
WFI Effective Area
ARF (100µm Be filter)ARF (with filter)ARF (no filter)mirror area
520.50.2 101
10000
1000
100
Photon Energy [keV]
Are
a[cm
2]
thick Be filter⇒ removes photons below 2 keV
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 8
Different modii of the WFI (in SIXTE)
Name Size (rows× columns) time resolution defocusinglarge 512× 512 5018µs —w64 64× 512 627µs —w32 32× 512 314µs —w16 16× 512 157µs —
w64df 64× 512 627µs 35 mmfast 64× 64 80µs 35 mm
fastBe 64× 64 80µs 35 mm
• Large Detector Array and Fast Detector• Fast Detector defocused by default• Option for a thick Be filter
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 9
Defocusing of the Fast Chip
40mm35mm25mm10mm
543210
1
0.1
0.01
10-3
10-4
10-5
Radius [mm]
Surf
ace
brig
htne
ss[a
.u.]
10 mm10 mm10 mm10 mm10 mm 25 mm25 mm25 mm25 mm25 mm
35 mm35 mm35 mm35 mm35 mm 40 mm40 mm40 mm40 mm40 mm
⇒ defocusing distributes photons over larger area (35 mm is optimal)
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 10
Event Detection in SIXTE for the WFI
1
3 4
2
y
x
ccrd
d
=⇒
single quadrupletripledouble
patterns are recombined for each frame in the pattern analysis⇒ invalid patterns rejected
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 11
Pile-up in the WFI
energy pile-up pattern pile-up
pile-up events distort spectral shape
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 12
WFI DePFET read-out implementation in SIXTE
0
Inte
gra
tion f
act
or
t0 t1
Cle
ar
t0 t1 t3 t4t2
t3 t4t2
0
Exp. Readout Exposure
Misfits
If photon hits during the read out: measured charge is affected⇒ Energy E wrong (“Misfit”)
this is most relevant for window modes or the fast detector
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 13
Bright Sources with the WFI
fastBefastw64dfw16w32w64large
1010.10.0110-310-4
1
0.8
0.6
0.4
0.2
01010.10.0110-310-4
0.1
0.01
10-3
10-4
Flux [Crab]
Thr
ough
put
Flux [Crab]
Pile-upFraction
among
ValidE
vents
throughput: ratio of valid events to number of incident photons
• slightly better bright source performance with smaller mirror• fast chip allows to observe sources above 1 Crab• up to 10 Crab with Be filter possible
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 14
Summary: The WFI with Sixte
• DePFET technology: active pixels, no line shifts→ misfits if pixel is hitduring readout• observations possible up to a few Crab, plus a thick filter for even brigther
sources• large 40’ FoV made of 4 chips→ requires dithering• simulations possible for the full 4 chip LDA (athenawfisim), or only a
single chip (LD, or the 35 mm defocused FD) with runsixt
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 15
The X-ray Integral Field Unit (X-IFU)
• very high spectral resolution imaging (2.5 eV FWHM and a 5’ FoV)• 3168 TES (Transition Edge Sensor) pixels
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 16
The X-ray Integral Field Unit (X-IFU)
Pixels are single Transition Edge Sensors, operated at 50 mK⇒ measure temperature increase of photon hitting the pixel
0.5350.530.5250.520.5150.51
1
0.8
0.6
0.4
0.2
0
time [sec]
signal/tempera
ture
[a.u.]
• numerical solution of differential equations for T (t), I(t) (Irwin & Hilton,2005),
CdTdt
= −Pb + PJ + P + Noise and LdIdt
= V − IRL − IR(T , I) + Noise
• linear resistance, R(T , I;α, β); noise: Johnson of circuit, bath, excess noise• input parameters: C, Gb, n, α, β, m, R0, T0, Tb, Lcrit
pulses with smaller separation yield lower energy resolution⇒ Event Grading depending on the source flux
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 17
The X-ray Integral Field Unit (X-IFU)
Pixels are single Transition Edge Sensors, operated at 50 mK⇒ measure temperature increase of photon hitting the pixel
0.5350.530.5250.520.5150.51
1
0.8
0.6
0.4
0.2
0
time [sec]
signal/tempera
ture
[a.u.]
• numerical solution of differential equations for T (t), I(t) (Irwin & Hilton,2005),
CdTdt
= −Pb + PJ + P + Noise and LdIdt
= V − IRL − IR(T , I) + Noise
• linear resistance, R(T , I;α, β); noise: Johnson of circuit, bath, excess noise• input parameters: C, Gb, n, α, β, m, R0, T0, Tb, Lcrit
pulses with smaller separation yield lower energy resolution⇒ Event Grading depending on the source flux
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 17
The X-ray Integral Field Unit (X-IFU)
Pixels are single Transition Edge Sensors, operated at 50 mK⇒ measure temperature increase of photon hitting the pixel
0.5350.530.5250.520.5150.51
1
0.8
0.6
0.4
0.2
0
time [sec]
signal/tempera
ture
[a.u.]
pulses with smaller separation yield lower energy resolution⇒ Event Grading depending on the source flux
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 17
X-IFU Implementation in the end-to-end simulator SIXTE
xifupipeline:• full detector array• full imaging implemented• fast detection simulation using
response matrices (workssimilar to CCD-typesimulations)
⇒ science simulations
tessim/sirena• Simulation of TES physics
and pulse reconstruction• Slower than xifupipeline,
but much better physics• pixel interaction (crosstalk)⇒ Input for xifupipeline
⇒ physics-based tessim results converted to be used in the fast andgeneral xifupipeline simulation (event grading, crosstalk, . . . )
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 18
X-IFU Implementation in the end-to-end simulator SIXTE
xifupipeline:• full detector array• full imaging implemented• fast detection simulation using
response matrices (workssimilar to CCD-typesimulations)
⇒ science simulations
tessim/sirena• Simulation of TES physics
and pulse reconstruction• Slower than xifupipeline,
but much better physics• pixel interaction (crosstalk)⇒ Input for xifupipeline
⇒ physics-based tessim results converted to be used in the fast andgeneral xifupipeline simulation (event grading, crosstalk, . . . )
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 18
Example: SIXTE X-IFU simulation of a Galaxy Cluster
520.5 101
10-12
10-13
10-14
0.950.900.85
10-13
10-14
1′
520.5 101
10-12
10-13
10-14
0.950.900.85
10-13
10-14
Energy [keV]
νF
ν[erg
/cm
2/s]
Energy [keV]
νF
ν[erg
/cm
2/s]
SIXTE describing, simulating, and analyzing complicated sourcesX-IFU spatially resolved high-resolution spectroscopy
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 19
Performance at High Countrates (grading effect only)
defocusing of the Athena optics allows observations up to 1 Crab
High resolutionTotal throughput25 mm35 mm
High res goal (10 mCrab)
Broa
dban
d th
roug
hput
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Source flux [mCrab]1 10 100 1000
No filter100 µm Be filter25 mm defocus35 mm defocus
7 eV
thro
ughp
ut in
the
5-8
keV
band
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Source flux [mCrab]1 10 100 1000
Grade ∆t since previous pulse ∆t until next pulse Energy res.(1) High res. > 7.9 ms > 45.3 ms 2.5 eV
(2) Medium res. > 7.9 ms > 2.3 ms 3 eV(3) Limited res. > 7.9 ms > 1.0 ms 7 eV
(4) Low res. > 7.9 ms – ∼ 30 eV
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 20
Crosstalk in SIXTE
unintended transmission of information between signal channels
Different types of crosstalk:• thermal coupling of two pixels (physical neighbors)• electrical coupling due to multiplexed readout (frequency neighbors)• non-linear amplification of the read-out SQUID
→ implemented in SIXTE
crosstalk effect on events is predictable
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 21
How does Crosstalk affect X-IFU Events?
simulation of a narrow emission line (1 Crab)
Ext < 4 eVExt < 0.2 eVall events
6.426.416.46.396.38
7×105
6×105
5×105
4×105
3×105
2×105
1×105
0
Energy [keV]
Flux[X
-IFU
Counts
/keV]
1000100101
5.5
5
4.5
4
3.5
3
2.5
Flux [mCrab]
Energ
yReso
lution
(Ext<
4eV)[eV]
⇒ remove events which are strongly effected by crosstalk
trade-off between energy resolution and throughput⇒ 10 eVresolution with 50% throughput @ 1 Crab
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 22
Summary: The X-IFU with Sixte
• 3168 TES pixels in a hexagonal shape• 5’ FoV• higher flux (>10mCrab) reduces energy resolution and throughput• science simulations with xifupipeline, taking the most important TES
physics effects into account• physics input to the simulation pipeline by tessim
Christian Kirsch & Maximilian Lorenz, Remeis Observatory & ECAP 23