Instrument characterization from metadata & telemetry

Transcription

1 Instrument characterization from metadata & telemetry Vanessa Bailey Jet Propulsion Laboratory, California Institute of Technology 2018 California Institute of Technology. Government sponsorship acknowledged. The research was carried out in part at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. The decision to implement the WFIRST mission will not be finalized until NASA's completion of the National Environmental Policy Act (NEPA) process. This document is being made available for information purposes only.

2 Gemini Planet Imager 2.8 GPI Credit - Paul Langlois 2

3 y Every GPI image has environment & performance data Bailey raw image 0.25, 0.4, 0.8 ~ WFE λ x ~ AO tip/tilt & focus vibration environment: seeing (Gemini MASS* & DIMM) wind, temperature * MASS quit working April

4 4 Environment parameters alone explain 25-40% of GPI raw contrast variation Tau DIMM seeing dt = abs(ao - amb) I mag 0.25 R 2 = R 2 = R 2 =0.38 real contrast predicted contrast

5 5 Tau governs final GPI contrast more often than raw seeing does faster stronger poorer detection limits turbulence turbulence

6 6 Similar effects seen in other instruments Milli Similar analysis for NIRC2 by Jerry Xuan (Pamona) ongoing

7 7 Temperature disequilibrium degrades GPI performance Melisa Tallis in prep

8 8 What telemetry can we save? LOWFS Science Camera accelerometers temperatures pointing positions: [ z1 zn] WFS DM residual slopes: commands: Reconstructor [ x1,y1 xn,yn] [ z1 zn] or [ m1 mn] modal gains

9 Goal: Use IFS and AO to correct for non-common path errors in system. GPI uses a Fourier modal basis set with Figure: (Left) The normal sinusoid is given by the da individually controlled gains represents the summed m ntosh et al., 2008] and data acquisition represent best fits to I1 + I sign denotes the best foun (spatial) mode PSF Results he DM is determined on A, the complex ion and the DM (u,v ) 2 i (u,v )/ )e e of the sinusoidal erates a speckle pes to cancel these Borde and Traub, Figure: (Left) A pure sinusoid applied to the DM producing (Right) speckles in the image (circled). 9

10 5-60sec Manual AO telemetry sets record detailed information Lisa Poyneer, Dimitry Savransky, Bruce Macintosh y mode reconstruct wavefront error power spectra for each mode (closed & open loop) Full data rate >1GB / min f [Hz] x mode 10

11 Gains optimized every 8 sec ETFs g =0.30 g =0.15 lag errors -> gains lower along wind direction y mode number Hz log scale intensity x mode number

12 12 Telescope tracking Earthquake

13 13 Example: cryocooler controller replace to mitigate M1 60Hz coupling Hartung+, SPIE, 2014

14 14 Vibration analysis example: faulty fan Poyneer+, Appl Opt, 2016 Also LBTI, GEMS, SPHERE

15 15 telescope windshake Connor Beierle in prep

16 Tangent: site characterization Regular AO telemetry = regular site monitoring (postprocessing required!!) Compare to observatory MASS, DIMM, etc. planning upgrades &/or new instruments (AO and seeing-limited) Sri Srinath - SPIE 2016 Adam Snyder - SPIE

17 Reconstructed WFE ~ GPI IFS frames pseudo-closed loop other PSF reconstruction: Veran+1997, Jolissaint+2012, WFE 2 [nm 2 ] IFS img 59 y mode 1.6um x mode _ S S0059_spdc 17

18 18 WFIRST HOWFS HCIT: Eric Cady, Brian Kern HOWFS uses science camera images themselves measured incoherent coherent coherent intensity intensity real E imaginary E 2 = + +i

19 19 realtime x/y centering location of star in every science frame contribution from Z2-Z11 = input to PCA? Fang Shi

20 20 What is the minimal AO data we need to save? Analyze system performance? Complement focal plane WFS? Complement data reduction? ground vs. space? What cadence? Save everything? Realtime process? S/N & error tolerance??

21 21 Summary Reach specs on current systems Develop AO telem pipelines & infrastructure Identify factors limiting astrophysics, not WFE Enable post-processing on future systems Include telemetry in design Save as much data as we can

22 22

23 AO WFE : bandwidth & noise y mode PSD for mode [xi, yi] 10 f [Hz] PSD at WFS Total Bandwidth Noise atmosphere errors bandwidth WFE photon/read noise noise WFE 2 nm / Hz x mode Hz