Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Adaptive Optics for the Thirty Meter Telescope Brent Ellerbroek Thirty Meter Telescope Observatory Corporation AO4ELT3 Florence, Italy May 27, 2013 TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 1 Contributors Brent Ellerbroeka, Sean Adkinsb, David Andersenc, Jenny Atwoodc, Arnaud Bastardd, Yong Boe, Marc-André Boucherc, Corinne Boyera, Dmitri Budkerf, Peter W. G. Byrnesc, Kris Caputac, Jeffrey Cavacog, Shanqiu Chenh, Carlos Correiac, Raphael Coustyd, Joeleff Fitzsimmonsc ,Luc Gillesa, James Gregoryi, Glen Herriotc, Paul Hicksonj, Alexis Hillc, Zuo Junweie, Zoran Ljusicc, N. Marchetd, Angel Otarolaa, Dan O’Marai, Leo Myerk, Benoit Neichell, John Pazderc, Hubert Pagesd, Spano Paoloc, Robert Priorc, Vladimir Reshetovc, Simon Rochesterf, Jean-Christophe Sinquing, Matthias Schoeckc, Malcolm Smithc, Kei Szetoc, Jinlong Tangh, Jean-Pierre Veranc, Lianqi Wanga, Kai Weih, Ivan Weversc, and Sylvana Yeldak aTMT Observatory Corporation, bW. M. Keck Observatory, cNRC Herzberg, dCILAS, eTechnical Institute of Physics and Chemistry, fRochester Scientific, gAOA Xinetics, hInstitute of Optics and Electronics, iMIT Lincoln Laboratory, jUniversity of British Columbia, kUniversity of California at Los Angeles, lGemini Observatory TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 2 Presentation Outline TMT Project highlights First light AO requirements review Derived architecture and technology choices Subsystem status report – NFIRAOS; LGSF Component development – Deformable mirrors; wavefront sensing detectors; guidestar lasers; real time controller Modeling and algorithm development Related papers at AO4ELT3 Acknowledgements TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 3 TMT Project Highlights Site permit approved by the Hawaiian BLNR – Geotechnical studies and ground preparation to begin this year – Construction start date slated for April 2014 Construction Phase funding proposals under review in Canada, India, and Japan, and China Yale University has committed to joining the Project NSF Cooperative Agreement signed Instrument design work progressing well – IRIS Preliminary Design Phase initiated May 2013 – IRMS delta Design Review (from MOSFIRE) held April 2013 – MOBIE Conceptual Design Review in September 2013 First light scheduled for 2022 TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 4 First Light AO Requirements Review 3 science ports at f/15 with 2 arc min unvignetted field High throughput (85% in J, H, K, and I bands) Low thermal emission (15% of sky + telescope) Diffraction-limited IR image quality on a moderate FoV – [187, 191, 208] nm wavefront error over a [0,17,30] arc sec field High sky coverage (50% at galactic pole) High photometric accuracy – 2% over 30 arc sec at l=1 mm for a 10 minute observation High astrometric accuracy – 50 mas over 30 arc sec in H band for a 100 second observation High observing efficiency TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 5 How First Light Performance Requirements Drive AO Architecture Decisions High throughput Minimize surface count Low thermal emission -30C operating temperature Diffraction limited performance in J, H, K bands Order 60x60 wavefront sensing and correction 30”corrected science field Atmospheric tomography + MCAO High Sky coverage Laser guide star (LGS) wavefront sensing NGS tip/tilt/focus sensing in the near IR MCAO to “sharpen” NGS images High precision astrometry and photometry on 30” fields Distortion-free optical design form MCAO for uniform, stable PSF AO telemetry for PSF reconstruction High-order LGS MCAO with Available at TMT first light with Utilize existing and near term components low risk and acceptable cost and system concepts possible NGS tip/tilt/focus sensing in thewhenever Near IR TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 6 First Light AO System Architecture and Technology Choices Laser Launch Telescope Beam Transfer Optics Laser Systems TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 Laser Guide Star Facility (LGSF) – Nd:YAG or Raman fiber laser technology – Lasers mounted on telescope elevation journal – Conventional beam transport (mirrors) – Center-launch laser projection 7 First Light AO System Architecture and Technology Choices Narrow Field IR AO System (NFIRAOS) NFIRAOS facility AO TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 – Piezostack deformable mirrors and tip/tilt stage – “Polar coordinate” CCD array for the LGS WFS – HgCdTe CMOS arrays for low order, infra-red NGS WFSs (in client instruments) 8 UBC, Vancouver (Sodium LIDAR) Project Participants TOPTICA, Munich (Laser Systems) TIPC, Beijing (Laser Systems) HIA, Victoria (NFIRAOS) DRAO, Penticton (RTC) MIT/LL, Lexington (WFS CCDs) AOA/Xinetics, Devens (Wavefront Correctors) CILAS, Orleans (Wavefront Correctors) IOE, Chengdu (Laser Guide Star Facility) Keck Observatory, Waimea (WFS readout electronics) Also Rochester Scientific (Berkeley, Sodium Atomic Physics) TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 TMT, Pasadena (Management and SE) 9 NFIRAOS Status Update “Pre-Final” Design Phase is now underway Opto-mechanical layout Prototyping/test activities include: – DM drive electronics development – DM breadboard testing – WFS lenslet testing – Instrument rotator interface NFIRAOS on Nasmyth – Component testing at -30C – MCAO lab bench development Design work will include: – NGS WFS optics bench – LGS WFS zoom optics – NFIRAOS instrument support tower – FEA for vibration/thermal/seismic effects TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 10 NFIRAOS Prototypes and Testbeds 96-channel DM drive electronics board MCAO test bench Phase Screen Location DM’s LGS WFS Path Source simulators TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 “Science” Path 11 Recent LGSF Design Activities Design updated for new MELCO telescope structure: – (Yet another) optical path trade study – 2-axis launch telescope flexure compensation system – Laser system and laser bench layout on telescope elevation journal – Details of beam tube diameter, relay lens design, and top end layout Beam tube turbulence modeling underway LGSF Preliminary Design phase scheduled to begin November 2013 TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 12 LGSF Optical Path Trade Study •Baseline OP •New OP 13 TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 Work in Progress on Beam Duct Turbulence Modeling CAD model of Fold mirror array and relay lenses Difference from mean interior air temperature TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 14 AO Component Requirement Summary Deformable mirrors 63x63 and 76x76 actuators at 5 mm spacing Tip/tilt stage 500 mrad stroke with 0.05 mrad noise 10 mm stroke and 5-10 % hysteresis at -30C 80 Hz bandwidth NGS WFS detector 240x240 pixels, 4x4 pixels per subaperture LGS WFS detectors 60x60 subapertures with 6x6 to 6x15 pixels each Low-order IR NGS WFS detectors 1024x1024 pixels (subarray readout on ~8x8 windows) Sodium guidestar lasers 25W (20W with backpumping), M2 < 1.17 Real time controller Solve 35k x 7k reconstruction problem at 800 Hz ~0.8 quantum efficiency,~1 electron at 10-800 Hz ~0.9 quantum efficiency, 3 electrons at 800 Hz ~0.6 quantum efficiency, 3 electrons at 10-200 Hz Coupling efficiency of 130 photons-m2/s/W/atom TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 15 DM Development Initial testing of the CILAS 6x60 actuator breadboard has been completed in early 2013 – Good actuator-level performance at ambient and low temperature (-30C) – Not all requirements met at the full breadboard level – New round of development underway New study initiated at AOA Xinetics – Evaluate actuator performance at -30C – Develop DM conceptual designs meeting TMT performance and interface requirements TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 16 One-Quadrant Prototype LGS WFS CCDs Front- and back-illuminated devices now tested – Meet specs for QE and CTE – Read noise of 2.7 to 3.7 eat 3.5 MHz (3 erequirement) – Dark current sufficiently low for 800 Hz frame rate BI devices have good yield – 3 of 8 probed CCDs show very good performance – 99% of outputs functional, 99% low noise Black: QE spec and allowed nonuniformity Red: Measuremented QE (Courtesy Keck Observatory) TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 17 NGS WFS CCDs for NFIRAOS MIT/LL CCID-74 Engineering Grade Device in Test Station – 256x256 pixels – 64 planar JFET amplifiers Testing of back-illuminated engineering devices now in progress at Keck – High QE as above (for Polar CCD) Measured read noise at -15C with 32 Amplifiers – Dark current of 26 electrons/pixel/second – Read noise approaches 1 electron at 100 FPS – Candidate science grade CCDs will be packaged for testing and potential use as the deliverable NGS WFS CCD for NFIRAOS TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 18 Guidestar Laser Systems TMT continues to follow two laser development efforts: Toptica/MPB frequency-doubled Raman fiber laser meets all TMT requirements for output power, line width, beam quality, volume, and power dissipation – some TMT interfaces will require development Work on TIPC prototype SFG Nd:YAG laser is continuing: – Currently 18W@800Hz power for 100μs pulse – M2 ~ 1.5 – Line width of 0.6 GHz, with 0.2 GHz wavelength stability On-sky tests in early 2013 confirm high sodium coupling efficiency for a large spot size without repumping Further on-sky tests (with repumping) planned later this year TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 19 TIPC Nd:YAG SFG Guidestar Laser System Laser System and Optical Schematic Lijiang Observatory, February 2013 TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 20 Laser Coupling Efficiency (Sce) vs. Irradiance Results from Lijiang, China, February-March 2013 Results were obtained for: – – – TMT.AOS.PRE.12.073.DRF01 Various Power Levels 800 Hz or 600 Hz Pulse Repetition Rate Pulse lengths 95 to 100 micro-secs Results, as expected, show the effect of optical saturation TMT Specification: 120 Photos/W/s/(atoms/m2) Sce at higher Irradiance levels can be improved with D2b repumping and may reach the TMT Specification More tests with repumping and smaller spot sizes planned 21 Real Time Controller (RTC) Architectures RTC architecture study now underway at NRC Herzberg and TMT to update RTC conceptual designs from 2008-09 Benchmarking and design of a GPU-based architecture is currently most advanced – 2 GPUs per WFS implement gradient computation and matrix-vector-multiply (MVM) wavefront reconstruction – Matrix updated at 0.1 Hz Benchmark results: 0.95ms mean latency, 1.05 ms peak Timing includes gradient computation, MVM computation, data transfer over 10 Gig ethernet TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 22 Recent Work in AO Modeling and Control Algorithm Development AO Error budget maintenance and sky coverage analysis – Incorporates improved modes for optical surface errors and sodium range tracking Performance trade studies for OIWFS passband and detector choices High precision astrometry High contrast imaging PSF reconstruction lab tests with GeMS Analysis of LGS Fratricide amplitude for GeMS On-instrument WFS guidestar acquisition Distributed (computationally efficient) Kalman Filter tomography LGS SLODAR for Cn2 _and wind velocity_ profile estimation MCAO Lab Bench development at NRC-Herzberg also continuing TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 23 Term WFE, nm RMS Delivered Performance 191 Higher-order error (LGS modes) First-order turbulence compensation Generalized fitting error AO Performance Estimate (Zenith, median seeing, 50% sky coverage) 163 132 106 Noise-free estimation error 59 Servo lag 17 WFS noise and nonlinearity 52 Opto-mechanical implementation errors 71 TMT pupil misregistration 12 Telescope/enclosure wavefront 37 NFIRAOS wavefront 53 Science instrument wavefront 30 AO components and higher-order effects 66 Deformable mirrors 49 LGS WFS and sodium layer 39 Control algorithm implementation 21 Tip/tilt and plate scale (NGS controlled modes) 58 Telescope windshake 17 Telescope vibration and tracking error 20 Turbulence and measurement noise TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 Contingency 52 80 24 AO Sky Coverage vs. OIWFS Detector Option and Spectral Passband Median seeing Galactic pole guidestar densities Zenith angles from 0 to 50 degrees Widening the spectral passband from JH to JHKs improves sky coverage by ~10% Switching from the H2RG detector to SELEX APDs would yield another ~5% improvement TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 191 25 Median AO Performance vs. Galactic Latitude and Longitude (at Transit) • Median seeing TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 26 Median AO Performance vs. Galactic Latitude and Longitude (at Transit) • Good (25%) seeing TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 27 High Precision Astrometry for Observations of the Galactic Center Many sources of error have been investigated: – Photon, detector, thermal noise – Differential tip/tilt jitter – Distortions: Probe arm positioning error Geometric (static) – PSF estimation – Confusion Single-epoch error budget: – Bright stars (K < 15): distortion dominates (~8 μas) – Faint stars (K > 15): confusion dominates (> 8 μas) TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 28 High Contrast Imaging IRIS imager optics now included in modeling NFIRAOS+IRIS equivalent to ~1h of Gemini/Keck in 30 s Approaches GPI performance in 1-2 h, assuming 50x of SSDI speckle suppression (hard) – Better astrometry and higher SNR for high res. spectroscopy – Fainter and more distant targets? TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 29 Progress in PSF Reconstruction • GEMs (Canopus) references sources and ground-conjugate DM used to generate long-exposure PSFs and simultaneous WFS telemetry • PSFs with Strehls of ~40% estimated to within 1-2% relative error • Next step: Wide-field PSF reconstruction for MCAO using both DMs TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 30 Related Papers at AO4ELT3 Schoeck 1010 Monday High precision astrometry Hickson 0830 Tuesday Mesospheric sodium layer Marois 1120 Tuesday High contrast Imaging Ellerbroek 1140 Tuesday High precision astrometry Gilles 1200 Tuesday LGS Fourier Tomography Veran 1810 Tuesday RTC architecture study Veran 1130 Wednesday Improved tilt sensing Yelda 1700 Wednesday Galactic center astrometry Thompson 1720 Thursday Vibration requirements Simard 0830 Friday TMT Science Lu 0950 Friday Astrometry with MCAO Wang 1120 Friday GPU-based RTC Wang Poster NFIRAOS sky coverage Herriot Poster T/T/F NGS Acquisition Herriot Poster RTC timing jitter specifications Otarola Poster GeMS fratricide analysis Gilles Poster PSF reconstruction 31 Acknowledgements The TMT Project gratefully acknowledges the support of the TMT partner institutions. They are – – – – – – the Association of Canadian Universities for Research in Astronomy (ACURA), the California Institute of Technology the University of California the National Astronomical Observatory of Japan the National Astronomical Observatories and their consortium partners And the Department of Science and Technology of India and their supported institutes. This work was supported as well by – – – – – – – – the Gordon and Betty Moore Foundation the Canada Foundation for Innovation the Ontario Ministry of Research and Innovation the National Research Council of Canada the Natural Sciences and Engineering Research Council of Canada the British Columbia Knowledge Development Fund the Association of Universities for Research in Astronomy (AURA) and the U.S. National Science Foundation. TMT.AOS.PRE.13.081.DRF01 AO4ELT3, Florence, Italy, 05/27/2013 32