MultiConjugate Adaptive Optics

1
Multi-Conjugate Adaptive Optics

• Presented to the TMT
• Instrument Working Group
• Donald Gavel
• Santa Cruz, March 19, 2004

2
Simulations of Image Correction on a Star
Cluster Using Various modes of Adaptive Optics
Correction
No correction blurred long exposure image due to
atmospheric turbulence
Corrected image using adaptive optics with one
laser guide star
Corrected image using adaptive optics with five
laser guide stars
Image courtesy of Brent Ellerbroek, Gemini
Observatory
3
Multi-conjugate Adaptive Optics Concept
Sodium layer
Cone shape return beam
Atmospheric turbulence
Laser projectors
Telescope primary/secondary
Focal plane, constellation of guidestars
4
MCAO Concept (continued)
AO Bench
Telescope focal plane (LGS focus behind infinity
focus)
AO Relay (Collimating parabola)
DM conjugate to upper altitude
DM conjugate to lower altitude
DM conjugate to aperture
5
MCAO Concept (continued)
Deformable Mirrors
6km
3km
0km
wfs1
Visible guide star light
wfs2
Dichroic Split
Wavefront sensors
IR Science light
6
Laser Guide Star Focal Plane is Behind the
Natural Star Focal Plane

• So how to split the light
• Probes?
• Dichroic (large)?

7
MCAO Optics Design Example

• CELT AO concept (Greenbook Ch9)
• 4 DMs
• 2 optical spaces

4-5m?
8
MCAO Design Example

• MCAO system parameters
• 4 DMs 0, 3, 5.8, 12 km
• actuators 8,7,4,2,000
• 9 guidestars, 10W each
• Tip/tilt stars, mH23
• IR Tip/tilt stars to get sky coverage

• Requirements
• 1-2 arcminute field
• S0.5 at l1m (133 nm rms)
• Full sky coverage

• Error budget

1 arcmin field, Cerro Pachon model Cn2 profile
nm
Fitting
70.00
Gen. Aniso
84.00
Tomography
20.54
Meas
40.00
Bandwidth
42.00
125.47
9
MCAO Field Uniformity - Simulation Results
3 DMs, 5 laser guide stars, 65 arcsec radius
constellation, Cerro Pachon profile
Field, qy
0.5
0.25
0
Converted to nm
0.25
0.5
0
Field, qx, arc-min
Brent Ellerbroek, Wavefront reconstruction
algorithms and simulation results for
multi-conjugate adaptive optics on giant
telescopes, Backaskog, 2003 (on AOWG web page
tmt.ucolick.org/participants/AOWG)
10
Scaling laws for error budget terms number of
guidestars
CP profile Cn2 model
MK profile Cn2 model
Cn2 profile models
11
Scaling laws for error budget terms number of DMs
12
State of the Technology

• DMs need 8000 actuators per
• Largest actuators on conventional DM to date
  2000
• Large (gt30cm) DMs are needed to satisfy Lagrange
  invarient (MEMS too small)
• Wavefront sensors need 512x512, 1 khz, low
  noise
• Present technology 128x128, 2 khz
• Need to be linear
• IR tip/tilt sensor
• 30e- read noise is state of the art
• IR APDs an option
• Computers present technology most likely
  sufficient, if we use fast algorithms
• Fast algorithms need to be proven stable in
  closed loop
• Algorithms must measure and adapt to changing Cn2
  profile
• Large optics

13
MCAO Summary

• Error scaling laws are basically known and agree
  with limited simulations
• We will continue to work with the SAC to define
  MCAO system requirements
• Simulations on super computers and lab tests
  needed to prove closed loop stability,
  robustness, and test algorithms that measure and
  adapt to seeing changes
• Considerable technology development is needed for
  all key components. AODP is a start