The AOLI low-order non-linear curvature wavefront sensor

Transcription

1 The AOLI low-order non-linear curvature wavefront sensor A method for high sensitivity wavefront reconstruction Jonathan Crass Institute of Astronomy, University of Cambridge SPIE Astronomical Telescopes and Instrumentation 2 nd July 2012, RAI, Amsterdam

2 Outline Motivation AOLI: combining two correction techniques AOLI adaptive optics system Non-linear curvature wavefront sensor Reconstruction algorithms Future work Supported by St Edmund s College, Cambridge, Cambridge Philosophical Society & SPIE Student Travel Grant

3 Motivation Challenging to produce diffraction limited imaging in the visible on large ground based telescopes After Hubble, there are no plans for future optical space telescopes

4 Adaptive Optics and Lucky Imaging AO currently is capable of diffraction limited imaging up to the near-ir Lucky Imaging can produce diffraction limited images in the optical on 2.5m telescopes Removing the strongest atmospheric effects allows Lucky Imaging to work on larger telescopes Tried and tested!

5 Adaptive Optics and Lucky Imaging HST - ACS Lucky 10% Fourier 20% Fourier 50%

6 Adaptive Optics Lucky Imager (AOLI) Initially for the 4.2m William Herschel Telescope Lucky Imaging based science instrument comprises: square EMCCDs providing px imaging region Pixel scale of 15 milliarcseconds in I-band Resolution of 40 milliarcseconds AO component: ALPAO 97 actuator deformable mirror (DM97-15) Non-linear curvature wavefront sensor Comprises 2 EMCCDs (E2V CCD201)

7 Adaptive Optics Lucky Imager (AOLI) WFS assembly Deformable mirror WHT focus De-collimating lens Collimating lens Fold mirror Collimating lens Pickoff mirror

8 Non-linear Curvature Wavefront Sensor nlcwfs - initially proposed by Guyon Differs from a conventional curvature wavefront sensor Conventional CWFS Images around focal plane Uses two images Works in linear regime Non-linear CWFS Images around pupil plane Uses four images Works in non-linear regime nlcwfs offers: High sensitivity to low and high orders Reconstruction with fewer photons than conventional techniques

9 Non-linear Curvature Wavefront Sensor Pupil p 3 p 1 plane p 2 p 4 Collimated beam -z 2 -z 1 +z 1 +z 2 Distance from pupil plane Direction of light propagation

10 Non-linear Curvature Wavefront Sensor Simulations at 650nm for a 4.2m telescope with 1m central obscuration

11 Non-linear Curvature Wavefront Sensor L3 CCD L3 CCD Imaging lens Cube beamsplitter Imaging lens Pentaprism beamsplitter Cube beamsplitter Lateral displacement beamsplitter L3 CCD L3 CCD Imaging lens Lateral displacement beamsplitter Cube beamsplitter Pentaprism beamsplitter

12 Chromatic effects 10km 50km 100km 500km

13 Reconstruction Algorithm Initial estimate of pupil phase Propagate to p 1 Calculate new candidate and impose input intensity from p 4 Propagate to p 1 Calculate new candidate and impose input intensity from p 1 Propagate to p 4 imposing pupil shape Calculate new candidate and impose input intensity from p 3 Propagate to p 3 Propagate to p 2 imposing pupil shape Calculate new candidate and impose input intensity from p 2 Propagate to pupil plane and impose pupil shape Unwrap phase estimate

14 Reconstruction Algorithm Previously, reconstruction for four planes undertaken with Error Reduction method Work at Cambridge has used the Input-Output method Need to improve Speed of reconstruction Reconstruction at low photon levels

15 Low photon number reconstruction Conventionally use binning on CCD in low photon situations For phase reconstruction algorithms, this increased grid size increases residual error Better to use pre-processing of image data before reconstruction

16 Low photon number reconstruction

17 Future work Further simulations of chromatic effects Laboratory demonstrator To test algorithm on camera data To investigate chromatic effects Reconstruction algorithm speed No prior information used currently Statistical reconstruction (e.g. Bayesian) Further information from previous reconstructions Parallel processing with GPUs

18 Summary Reconstruction algorithm for nlcwfs Input-Output method works well Low photon regime use pre-processing No one size fits all solution Chromatic effects For low order correction use algorithm to handle effects AOLI - [ ] Craig Mackay Thursday, 16:10 By using two correction techniques, AOLI will demonstrate near-diffraction limited imaging in the visible on large groundbased telescopes.

19

20 Future work CCD Camera Deformable Mirror Light Source SHWFS