lgs spot truncation mitigation in elts: optimizing...
TRANSCRIPT
LGS spot truncation mitigation in ELTs: optimizing the pixel usage
L.Blanco1, C. Correia1, T. Fusco2, B. Neichel1, Y. Ono1
1 Laboratoire d’Astrophysique de Marseille (UMR7326 - CNRS-INSU, Université d’Aix-Marseille) 2 ONERA - The French Aerospace Lab
LGS Spot elongation
LLT
Sodium layer
D e t e c t o r plane
T ~ 20km
H ~ 90km
Distance from launch site
P u p i l plane
Predicted spot elongation pattern
LLT
2AO4ELT 2017
3
LLT
D e t e c t o r plane
P u p i l plane
25”
1”
Ideally, we need subapertures with 25x25 pixels of ~1”
For 80x80 subapertures, we need 2000 x 2000 pixels
Most likely, we will have no more than 10x10 pixels
On a ~40m telescope
LGS Spot elongation
Strong truncation
AO4ELT 2017
4
• À la Diolaiti/Schreiber, we simulate elongated spots (normalized by the Na profile intensity)
• Compute the centroid of each sub-aperture, and project the resulting wave-front on Zernikes
• No elongation: no aberration
• No truncation: Tip/Tilt/Focus only
Impact of truncation on a single WFSAl
titu
de
Time
AO4ELT 2017
5
(blue = peak; error bar = rms ; histogram = avg ; red = cumulative of avg)
Impact of truncation on a single WFSAl
titu
de
Time
• Select 11 Na profiles showing a large variety of configurations
• For each profile, and each time, compute the residual aberrations (for a single WFS)
AO4ELT 2017
6
Impact of truncation on LTAOAl
titu
de
Time
• 6 LGS • No turbulence • MMSE reconstructor • 42’’ radius asterism
(histogram = avg ; red = cumulative of avg ; blue = 30nm and 50nm cutoff)
AO4ELT 2017
7
Need for a Truth Sensor but aberrations are mostly low order => 10x10 subapertures on a truth sensor should be enough on
average but on very perturbed profiles…
Impact of truncation on LTAOAl
titu
de
Time
(histogram = avg ; red = cumulative of avg ; blue = 30nm and 50nm cutoff)
AO4ELT 2017
• 6 LGS • No turbulence • MMSE reconstructor • 42’’ radius asterism
(more on tomographic errors in T. Fusco’s talk this afternoon)
Elongated spots cropping
What if we completely discarded the biased measurements from the subapertures with the most truncated spots ?
8AO4ELT 2017
What if we completely discarded the biased measurements from the subapertures with the most truncated spots ?
Spot elongation∝distance from laser launch telescope DLLT ->we ‘crop’ the subapertures at DLLT > D x (1-pupcrop)
Highly truncatedPartially truncated
Not truncated
pupcrop=0
9
Elongated spots cropping
AO4ELT 2017
What if we completely discarded the biased measurements from the subapertures with the most truncated spots ?
Spot elongation∝distance from laser launch telescope DLLT ->we ‘crop’ the subapertures at DLLT > D x (1-pupcrop)
Not truncated
Highly truncatedPartially truncated
pupcrop=25%
10
Elongated spots cropping
AO4ELT 2017
What if we completely discarded the biased measurements from the subapertures with the most truncated spots ?
Spot elongation∝distance from laser launch telescope DLLT ->we ‘crop’ the subapertures at DLLT > D x (1-pupcrop)
Not truncated
Partially truncatedHighly truncated
pupcrop=50%
Killer for 1 WFS…
11
Elongated spots cropping
AO4ELT 2017
12
What if we completely discarded the biased measurements from the subapertures with the most truncated spots ?
Spot elongation∝distance from laser launch telescope DLLT ->we ‘crop’ the subapertures at DLLT > D x (1-pupcrop)
pupcrop=10%
pupcrop=30%
WFS1 WFS2 WFS3 WFS4
But we have redundancy
12
Elongated spots cropping
AO4ELT 2017
13
LLT
D e t e c t o r plane
P u p i l plane
25”
1”
On a ~40m telescope
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM (Fried
geometry) Subaperture pitch : 50cm
Subaperture FOV : 10’’ (1’’ pixels) 6 LGS-Asterism diameter : 42’’
30% central obscuration (Y. Ono poster P3059, Thursday)
E2E simulations (OOMAO)
AO4ELT 2017
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 14
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 15
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 16
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 17
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 18
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 19
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 20
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Elongated spots cropping
AO4ELT 2017 21
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Performance increases as we discard subapertures with too large an elongation/truncation
WFE due to elongation reduced from 152nm to 106nm RMS
Elongated spots cropping
223nm
270nm247nm
AO4ELT 2017 22
152nm106nm
Differential subaperture weighting
- à la Tallon 2008/Robert 2010 - Takes into account the elongation in the noise covariance matrix - Takes into account cross-coupling between X and Y slopes
AO4ELT 2017 23
Differential subaperture weighting
- Tallon 2008/Robert 2010 - Takes into account the elongation in the noise covariance matrix - Takes into account cross-coupling between X and Y slopes
AO4ELT 2017 24
Differential subaperture weighting
- Tallon 2008/Robert 2010 - Takes into account the elongation in the noise covariance matrix - Takes into account cross-coupling between X and Y slopes
AO4ELT 2017 25
Differential subaperture weighting
- Tallon 2008/Robert 2010 - Takes into account the elongation in the noise covariance matrix - Takes into account cross-coupling between X and Y slopes
AO4ELT 2017 26
223nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
270nm247nm
WFE due to elongation reduced from 152nm to 83nm RMS when using subaperture differential weighting
but Na profile not always bimodal !
238nm
Stre
hl in
K-b
and
(%)
AO4ELT 2017 27
SODIUM DENSITY PROFILE VARIATIONS
21Pfrommer and Hickson 2011
13.7% 18.6% 16.7%
15.8% 15.3% 11.2%
8.7%
Subaperture weighting+cropping
AO4ELT 2017 28
226nm
234nm
223nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Narrow spot mostly within the subaperture even opposite from LLT
Weighting over-regularizes already good information
Stre
hl in
K-b
and
(%)
AO4ELT 2017 29
223nm
270nm
238nm247nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
WFE due to elongation reduced from 152nm to 83nm RMS when using subaperture differential weighting
Stre
hl in
K-b
and
(%)
AO4ELT 2017 30
223nm
275nm
226nm
257nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Differential weighting gives almost ideal performance
Stre
hl in
K-b
and
(%)
AO4ELT 2017 31
223nm
265nm
237nm247nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
WFE due to elongation reduced from 143nm to 80nm RMS when using subaperture differential weighting
Best ‘cropping’ (20%) WFE due to elongation =106nm
Stre
hl in
K-b
and
(%)
AO4ELT 2017 32
223nm
335nm
238nm 250nm233nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
WFE due to elongation reduced from 250nm to 83nm RMS when using subaperture differential weighting
WFE reduced to 67nm RMS when combining weighting and cropping
Stre
hl in
K-b
and
(%)
AO4ELT 2017 33
223nm
349nm
318nm330nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Very perturbed profile with sharp peak at high altitude Unstable loop when taking into account all subapertures Weighting mandatory for full pupil High cropping (40%) improves performance over weighting but far from ideal
Stre
hl in
K-b
and
(%)
AO4ELT 2017 34
223nm
400nm
300nm
337nm
Subaperture weighting+cropping
Full E-ELT simulation: Pupil diameter : 37m
74x74 SH LGS WFS 75x75 actuator DM
Subaperture pitch : 50cm Subaperture FOV : 10’’ (1’’ pixels)
6 LGS-Asterism diameter : 42’’ 30% central obscuration
High flux
Stre
hl in
K-b
and
(%)
Very perturbed profile with sharp peak at high altitude Unstable loop when taking into account all subapertures Weighting mandatory for full pupil High cropping (40%) improves performance over weighting but far from ideal
AO4ELT 2017 35
Truth sensor
How many modes do we need to measure to remain below a prescribed WFE due to spot elongation ?
AO4ELT 2017 36
Truth sensor
How many modes do we need to measure to remain below a prescribed WFE due to spot elongation ?
AO4ELT 2017 37
8x8 truth sensor sufficient to remain below 10nm at all times
Truth sensor
How many modes do we need to measure to remain below a prescribed WFE due to spot elongation ?
AO4ELT 2017 38
Single Top hat Double Very
narrowMulti peak
Very wide
Top hat +peak
10nm z1 z10 z14 z14 z28 z42 z56
20nm z1 z1 z1 z1 z1 z42 z42
30nm z1 z1 z1 z1 z1 z42 z42
40nm z1 z1 z1 z1 z1 z28 z28
50nm z1 z1 z1 z1 z1 z28 z28
223nm
270nm247nm
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
Corner Shack-Hartmann
AO4ELT 2017 39
223nm
270nm247nm
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
Corner Shack-Hartmann
Not needed
AO4ELT 2017 40
223nm
270nm247nm
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
Corner Shack-Hartmann
AO4ELT 2017 41
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 42
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 43
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 44
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 45
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 46
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 47
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 48
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 49
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner Shack-Hartmann
WFS
sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional S
H
Corner S
H
Pupil cropping increases
AO4ELT 2017 50
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without
truncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS sensorWithout truncation
160 pixels10pix/subap
160 pixels10 pix/subap
160 pixels12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
Without tr
uncation
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventional S
H
Corn
er
SH
Pupil
cro
ppin
g incre
ases
WFS
sen
sor
With
out t
runc
atio
n
160
pixe
ls
10pi
x/su
bap
160
pixe
ls
10 p
ix/s
ubap
160
pixe
ls
12 p
ix/s
ubap
Con
vent
iona
l SH
Cor
ner S
H
Pup
il cr
oppi
ng in
crea
ses
WFS sensor
Without truncation
160 pixels
10pix/subap
160 pixels
10 pix/subap
160 pixels
12 pix/subap
Conventional SH
Corner SH
Pupil cropping increases
WF
S s
ensor
With
out tru
ncatio
n
160 p
ixels
10pix
/subap
160 p
ixels
10 p
ix/s
ubap
160 p
ixels
12 p
ix/s
ubap
Conventio
nal S
H
Corn
er S
H
Pupil c
roppin
g in
cre
ases
6 LGS configuration
4 LGS configuration
Corner SH always better than conventional SH for a given # of subapertures
51
->Completely discarding 10-40% of the truncated subapertures actually increases Strehl Ratio (depending on Sodium profiles)
->Combining subaperture differential weighting and cropping allow us to reach quasi-ideal performance on most sodium profiles
-> Help release constraints on Truth Sensor design: 8(x8) is enough !
-> Corner Shack-Hartmann to optimize pixel usage
AO4ELT 2017
Conclusion
AO4ELT 2017 52
- Thorough analysis of more parameters (truncation, d/r0, central obscuration, asterism diameter, #LGS…)
- Advanced centroiding algorithms/strategies
- Actual ELT M4 geometry (see N. Schwartz talk tomorrow 12:30)
- Temporal aspects, sodium profil evolution during the night
On-going work
The authors thanks the European Research Council for its support through the grant ERC-STG-2015 ICARUS 678777
AO4ELT 2017 53
Gracias