Source-linked AI summary
Large Binocular Telescope Adaptive Optics System: New achievements and perspectives in adaptive optics
Simone Esposito, Armando Riccardi, Enrico Pinna, Alfio Puglisi, Fernando Quirós-Pacheco, Carmelo Arcidiacono, Marco Xompero, Runa Briguglio, Guido Agapito, Lorenzo Busoni, Luca Fini, Javier Argomedo, Alessandro Gherardi, Guido Brusa, Douglas Miller, Juan Carlos Guerra, Paolo Stefanini, Piero Salinari
TL;DR
Historically, limited efficiency and sky coverage for early NGS AO systems helped motivate LGS systems. This paper reports LBT commissioning results and compares upgraded NGS performance with LGS, finding results that support reconsidering their roles.
Problem
Early NGS systems had limited guide-star flux efficiency and sky coverage, motivating LGS development for AO science applications.
Source: §5 · p. 7Method
Results
The LBT-like NGS system has 54% sky coverage versus 84% for LGS and better coverage at Strehl ratios above 25%; commissioning achieved H-band Strehl above 80%.
Source: §6 · p. 9Takeaways & Limitations
The results suggest revisiting the choice of LGS over NGS for 8 m and ELT-class AO systems.
Source: §5.2 · p. 8, §6 · p. 9For guide stars with MR >13, the online seeing estimate can be non-negligibly underestimated because of a calibration-to-operation PSF mismatch.
Source: §3.2 · p. 4
Abstract
from arXiv · showhide
The Large Binocular Telescope (LBT) is a unique telescope featuring two co-mounted optical trains with 8.4m primary mirrors. The telescope Adaptive Optics (AO) system uses two innovative key components, namely an adaptive secondary mirror with 672 actuators and a high-order pyramid wave-front sensor. During the on-sky commissioning such a system reached performances never achieved before on large ground-based optical telescopes. Images with 40mas resolution and Strehl Ratios higher than 80% have been acquired in H band (1.6 micron). Such images showed a contrast as high as 10e-4. Based on these results, we compare the performances offered by a Natural Guide Star (NGS) system upgraded with the state-of-the-art technology and those delivered by existing Laser Guide Star (LGS) systems. The comparison, in terms of sky coverage and performances, suggests rethinking the current role ascribed to NGS and LGS in the next generation of AO systems for the 8-10 meter class telescopes and Extremely Large Telescopes (ELTs).
1. INTRODUCTION
The LBT pairs two 8.4 m optical trains with adaptive-secondary-mirror correction and high-order pyramid wave-front sensing. The paper reports FLAO commissioning performance and uses the results to consider future AO systems.
1. INTRODUCTION
- Each LBT optical train uses a 672-actuator adaptive secondary mirror and a modulated pyramid wave-front sensor with up to 30×30 subapertures.The system is designed to balance correction performance with the limited guide-star photons available to astronomical AO.Source: §1 · p. 1
- The authors report H-band correction above 80% Strehl on bright stars and efficient loop closure on a roughly 17th-magnitude star.For the faint star, the system operated with about 15 photons per subaperture per frame.Source: §1 · p. 1
- The paper analyzes FLAO#1 commissioning data and considers future AO systems based on the achieved LBT results.FLAO#1 was installed on the telescope’s right side; commissioning began in May 2010 and was scheduled to finish in fall 2011.Source: §1 · p. 1
2. FLAO SYSTEM OVERVIEW
The FLAO overview describes the adaptive secondary mirror, pyramid wave-front sensor, adjustable sampling, modal controller, and calibration approach used on sky.
2. FLAO SYSTEM OVERVIEW
- The adaptive secondary mirror uses 672 contactless voice-coil actuators; commissioning results were obtained with 641 after 31 defective actuators were deactivated.Its thin shell is 1.6 mm thick and 911 mm in diameter, with settling times of 0.7–1 ms.Source: §2 · p. 2
- The pyramid sensor samples the mirror with up to 30×30 subapertures, and CCD binning adjusts pupil sampling from 30×30 to about 7×7.The four listed modes are 30×30, 15×15, 10×10, and approximately 7×7 subapertures; pyramid modulation also adjusts sensitivity.Source: §2 · p. 2
- On-sky configurations vary with equivalent R magnitude and include binning, frame rate, controlled modes, pyramid modulation, and median read-out noise.The table summarizes the typical settings used for each guide-star magnitude range.Source: Table 1 · p. 3, §2 · p. 3
- The controller uses an ASM-fitted Karhunen–Loève modal basis and a modal integrator with gains estimated automatically after guide-star acquisition.Gain values are not changed during normal AO operation.Source: §2 · p. 3
- Daytime modal interaction matrices were calibrated at the telescope using retro-reflector optics below the ASM to return the calibration beam to the wave-front sensor.A separate matrix was calibrated for each binning-mode and pyramid-modulation combination.Source: §2 · p. 3
3. OBSERVING CONDITIONS
The section defines how guide-star brightness and turbulence conditions are estimated for interpreting on-axis FLAO performance. It also notes that the online seeing estimate can be biased low for faint guide stars.
3. OBSERVING CONDITIONS
- The performance analysis focuses on on-axis H-band Strehl as a function of guide-star magnitude and atmospheric seeing, among several factors that limit AO performance.The stated limiting factors include received guide-star flux, turbulence, object–guide-star separation, detector read-out noise, and telescope vibrations.Source: §3 · p. 3
3.1 Equivalent star magnitude in R band
- An equivalent R-band guide-star magnitude is used to compare reference stars, based on WFS photon flux measured from CCD frames and instrument constants.The pyramid sensor receives light over approximately 600–950 nm, and the guide star’s luminosity in that band drives correction capability.Source: §3.1 · p. 3
3.2 Estimation of the seeing value
- Seeing is estimated by reconstructing open-loop modes from ASM commands and residual WFS signals, then fitting their variance distribution; estimates for MR >13 are non-negligibly low.The fit omits the first three radial orders and last two radial orders; a mismatch between calibration and closed-loop PSF sizes underestimates modal residuals and hence seeing.Source: §3.2 · p. 4, Figure 2 · p. 4
4. ON-SKY COMMISSIONING RESULTS
FLAO#1 commissioning measurements broadly matched simulated Strehl performance across guide-star magnitudes. The observations also demonstrated diffraction-limited resolution at brighter magnitudes, high-contrast correction with a bright star, and reduced image width at the faint end.
4. ON-SKY COMMISSIONING RESULTS
- 597 Strehl Ratio estimates across star magnitudes ~7.5–18 were in accordance with simulated performance, showing that FLAO#1 met expectations.Figure 3 plots estimates by guide-star magnitude and seeing, with different binning modes and simulated performance curves.Source: §4 · p. 4, Figure 3 · p. 5
- 40 mas diffraction-limited resolution was achieved down to guide-star magnitude ~12, with examples spanning magnitudes 8–17 at about 0.8 arcsec seeing.The examples include AO-corrected PSFs and an open-loop PSF for comparison.Source: §4 · p. 5, Figure 4 · p. 5
4.1 Performance at the bright end: high-contrast imaging
- SR >80% was achieved in H band with an MR=6.5 guide star under 0.6–0.8 arcsec seeing, using 1 kHz control of 400 KL modes.The reported contrast was better than 104 at 0.4 arcsec radius, before the turbulence residual halo; the authors identify this as the first on-sky confirmation of the deep annular region where pyramid-sensor non-aliased correction reaches maximum contrast.Source: §4.1 · p. 6, Figure 5 · p. 6
4.2 Performance at the faint end
- At MR=17.2, the AO-corrected image had ~3.5% H-band SR, 1.7× lower FWHM, and a peak ratio of 2.6 versus the seeing-limited image.The acquisition used binning mode #4 and controlled 10 modes; seeing was estimated as 0.72 arcsec at 500 nm assuming an outer scale of 40 m because no DIMM estimate was available.Source: Figure 6 · p. 6, §4.2 · p. 6
5. NGS AND LGS ADAPTIVE OPTICS SYSTEMS AFTER LBT RESULTS
The LBT results motivate renewed comparison of NGS and LGS adaptive optics, including an LLLCCD-upgraded NGS system. The comparison finds a sky-coverage trade-off and highlights potential NGS advantages at high Strehl ratios and shorter wavelengths, while LGS-based ELT designs face added complexity.
5. NGS AND LGS ADAPTIVE OPTICS SYSTEMS AFTER LBT RESULTS
- Early NGS-based AO systems were relatively inefficient for science applications, typically achieving only one to a few percent sky coverage; fast-readout CCDs were noisier, and wave-front-sensor efficiency at low flux needed improvement.Source: §5 · p. 7
5.1 Improving NGS AO performance and Sky Coverage with an LLLCCD
- LLLCCD simulations used 0.1 e- RMS CIC noise and 0.03 e- RMS read-out noise at 1 kfps, with 0.6-arcsec seeing, 40 m outer scale, and 15 m/s wind.Figure 7 compares simulated LLLCCD performance with the LBT FLAO system and shows Keck LGS AO for comparison.Source: §5.1 · p. 7, Figure 7 · p. 7
- At 0.6-arcsec seeing, NGS+LLLCCD exceeds LGS performance through magnitude 16.8 and reaches over 70% K-band SR through magnitude 15.2, while LGS reaches 19.0 at 10% SR versus 18.4 for NGS+LLLCCD.The figure compares simulated K-band SR against guide-star magnitude for FLAO, its LLLCCD upgrade, and Keck LGS AO.Source: §5.1 · p. 7, Figure 7 · p. 7
- For required Strehl ratios above 25%, NGS sky coverage averages about 15% higher than LGS, while at 10% the studied sample gives 54% for NGS+LLLCCD versus 84% for LGS.The comparison uses about 850 high-galactic-latitude galaxies from the Steidel sample under 0.6-arcsec seeing.Source: §5.1 · p. 8, Figure 8 · p. 8
5.2 Future developments for AO systems
- NGS correction can extend to J and visible wavelengths if sufficient photons and actuators are available; LBT achieved about 50% SR in J and 20% at 850 nm.Figure 9 depicts SR scaling with wavelength for H-band correction levels of 85%, 70%, 55%, and 40%.Source: §5.2 · p. 8, Figure 9 · p. 9
- For visible AO, a single LGS becomes less effective at shorter wavelengths because of focus anisoplanatism; LGS correction there requires multiple LGS and multiple NGS for low-order modes.The passage contrasts this with NGS systems, which require no major changes for shorter-wavelength operation.Source: §5.2 · p. 8
- For ELTs, LGS designs must contend with multiple guide stars, spot elongation, and sodium-density variability, whereas NGS designs are described as scaling from 8 m systems by adding actuators and subapertures.The authors suggest reconsidering LGS for 8 m and ELT systems whose corrected field is limited by the atmospheric isoplanatic patch.Source: §5.2 · p. 8
6. CONCLUSIONS
The LBT FLAO#1 commissioning results set a new performance standard and support the use of adaptive secondary mirrors and pyramid sensors in future 8 m and ELT AO systems. An LBT-like NGS system with an LLLCCD shows sky coverage comparable to LGS in the studied extragalactic case and better coverage above 25% SR, prompting reconsideration of the NGS–LGS choice.
6. CONCLUSIONS
- The LBT FLAO#1 commissioning achieved H-band SR above 80%, image contrast better than 104, and loop closure down to a 17.5-magnitude star.The authors present these results as setting a new standard for ground-based astronomical AO and as evidence that adaptive secondary mirrors and pyramid sensors are mature for future 8 m and ELT systems.Source: §6 · p. 9
- In the studied extragalactic case, NGS+LLLCCD achieved 54% sky coverage versus 84% for LGS, while NGS coverage was better above 25% SR, motivating a renewed look at choosing LGS over NGS.The conclusion notes LGS limitations including cone effect and the need for NGS sensing of low-order modes.Source: §6 · p. 9