NIO GSMT Overview

S. Strom, L. Stepp 23 August, 2001

AURA NIO: Mission

• • – In response to AASC call for a GSMT, AURA formed a New Initiatives Office (NIO) collaborative effort between NOAO and Gemini to explore design concepts for a GSMT NIO mission “

to ensure broad astronomy community access to a 30m telescope contemporary in time with ALMA and NGST, by playing a key role in scientific and technical studies leading to the creation of a GSMT.”

Goals of the NIO

• • • • • Foster community interaction

re

GSMT Develop point design Conduct studies of key technical issues and relationship to science drivers Optimize community resources: – – – emphasize studies that benefit multiple programs, collaborate to ensure complementary efforts, give preference to technical approaches that are extensible to even more ambitious projects.

Develop a partnership to build GSMT

AURA New Initiatives Office

Management Board Matt Mountain Jeremy Mould William Smith Engineering Oversight Jim Oschmann Project Scientist Steve Strom Mechanical Designer Rick Robles Program Manager Larry Stepp Systems Scientist Brooke Gregory Part time support NOAO & Gemini TBD Administrative assistant Jennifer Purcell Instruments Sam Barden (NOAO) Structures Paul Gillett Site Testing Alistair Walker (NOAO)

Objectives: Next 2 years

• • • • • • • Develop point design for GSMT & instruments Develop key technical solutions – – – Adaptive optics Active compensation of wind buffeting Mirror segment fabrication Investigate design-to-cost considerations Involve the community in defining GSMT science and engineering requirements Involve the community in defining instrumentation options; technology paths Carry out conceptual design activities that support and complement other efforts Develop a partnership to build GSMT

Resources: 2001-2002

• NIO activities: $3.6M – Support core NIO staff (‘skunk works team’) • • Analyze point design Develop instrument and subsystem concepts – Support Gemini and NOAO staff to • • Explore science and instrument requirements Develop systems engineering framework – Support community studies: • • • Enable community efforts: science; instruments Enable key external engineering studies Support alternative concept studies

Objectives: Next Decade

• • • Complete GSMT preliminary design (2Q 2005) Complete final design (Q4 2007) Serve as locus for – community interaction with GSMT partnership – ongoing operations – defining; providing O/IR support capabilities – defining interactions with NGST

Resources: Next Decade

• $15M in CY 2003-2006 from NOAO base – Enables start of Preliminary Design with partner • $25M in CY 2007-2011 from NOAO base • Create a ‘wedge’ of ~$10M/yr by 2010 – Enables NOAO funding of • • • Major subsystem Instruments Operations

Key Milestones

• • • • •  2Q01: Establish initial science requirements 3Q01: Complete initial instrument concepts 3Q01: Complete initial point design analysis 1Q02: Identify and fund alternate concept studies 2Q02: Identify and fund key technology studies 2Q02: Identify potential US and/or international partners • • • • • • 1Q03: Complete concept trade studies 2Q03: Develop MOUs with partner(s) 2Q03: Establish final science requirements 4Q03: Initiate Preliminary Design 2Q05: Complete Preliminary Design 4Q05: Complete next stage proposal

FIRST STEP: UNDERSTAND SCIENCE REQUIREMENTS

Developing Science Cases

• • • • • Two community workshops (1998-1999) – Broad participation; wide-ranging input Tucson task group meetings (SEP 2000) – – – Large-scale structure; galaxy assembly Stellar populations Star and planet formation NIO working groups (MAR 01 – SEP 01) – Develop quantitative cases; simulations NIO-funded community task groups (CY 2002) NIO-funded community workshop (CY 2002) – Refine performance goals and requirements

• • • • •

Tomography of the Universe

Goals: Map out large scale structure for z > 3 Link emerging distribution of gas; galaxies to CMB Measurements: Spectra for 10 6 galaxies (R ~ 2000) Spectra of 10 5 QSOs (R ~15000) Key requirements: 20’ FOV; >1000 fibers Time to complete study with GSMT: 3 years Issues – Refine understanding of sample size requirements – Spectrograph design

Mass Tomography of the Universe

Existing Surveys + Sloan Hints of Structure at z=3 (small area) z~0.5

100Mpc (5 O x5 O ), 27AB mag (L* z=9), dense sampling GSMT Gemini NGST 1.5 yr 50 yr 140 yr

Tomography of Galaxies and Pre-Galactic Fragments

• • • • • Goals: Determine gas and stellar kinematics Quantify SFR and chemical composition Measurements: Spectroscopy of H II complexes and underlying stars Key requirements: Deployable IFUs feeding R ~ 10000 spectrograph Wide FOV to efficiently sample multiple systems Time to complete study with GSMT: ~1 year Issues – Modeling surface brightness distribution – Understanding optimal IFU ‘pixel’ size

Tomography of Individual Galaxies

• Determine the gas and stellar dynamics within individual galaxies • Quantify variations in – star formation rate

Tool

: IFU spectra [R ~ 5,000 – 10,000]

out to z ~3

GSMT 3 hour, 3 s limit at R=5,000 0.1”x0.1” IFU pixel (sub-kpc scale structures) J H K 26.5 25.5 24.0

Origins of Planetary Systems

• • • • • Goals: – Understand where and when planets form – Infer planetary architectures via observation of ‘gaps’ Measurements: Spectra of 10 3 accreting PMS stars (R~10 5 ; l ~ 5m ) Key requirements: On axis, high Strehl AO; low emissivity Time to complete study with GSMT: 2 years Issues Understand efficacy of molecular ‘tracers’ Trades among emissivity; sites; telescope & AO design

Probing Planet Formation with High Resolution Infrared Spectroscopy

Planet formation studies in the infrared (5-30 µm):

   

Probe forming planets in inner disk regions Residual gas in cleared region low

t

emission Rotation separates disk radii in velocity High spectral resolution high spatial resolution S/N=100, R=100,000,

l

>4

m

m Gemini out to 0.2kpc sample ~ 10s GSMT 1.5kpc ~100s NGST X

 8-10m telescopes with high resolution (R~100,000) spectrographs can detect the formation of Jupiter-mass planets in disks around nearby stars (d~100pc).

• • • • •

Stellar Populations

Goals: Quantify IMF in different environments Quantify ages; [Fe/H]; for stars in nearby galaxies Develop understanding of galaxy assembly process Measurements: Spectra of ~ 10 5 stars in rich, forming clusters (R ~ 1000) CMDs for selected areas in local group galaxies Key requirements: MCAO delivering 2’ FOV; MCAO-fed NIR spectrograph Time to complete study with GSMT: 3 years Issues – MCAO performance in crowded fields

Derived Top Level Requirements

Field of View Principle wavelegths PSF

Strehl Photom. accuracy(derived)

Astrometric accuracy Narrow field AO 1 2 3 MCAO

10 arcsec + + 1.0 - 2.5 microns Resolution Diffraction limited Stability 1% 80% 1% 10^-4 arcsec + + + + + 2 arcmin + + 1.0 - 2.5 microns Diffraction limited + 5% 50% 5% 10^-3 arcsec

1 2 3 Low order AO

2 arcmin + 1.0 - 20 microns 0.1-0.2 arcsec + 2% <10% 2% 10^-2 arcsec

1 2 3 Seeing limited (PF) 1 2 3

20 arcmin 0.4 - 2.5 microns 0.4-0.7 arcsec + 2% 0% 1% 0.05 arcsec

Stability timescale Emissivity Maintenance/Ops Reliability Science Efficiency

3,600 s <20% <15% 90% 90% 3,600s <20% <15% 90% 90%

How to r ead this table:

Four telescope “Operating regimes” ar e defined and the specs f or the telesc ope (not the instr ument) in each regime are cited. There are columns at the right of each regime labeled 1,2,3 for the th ree science pr ograms discuss ed in the NO AO panel W orkshop. In these columns the spe cs are asses sed in term s of the adequacy for each sc ience program .

3,600s 10% <15% 90% 10,000 s 15% <15% 90% 90%

Matt: I think this mode is important, its our only "thermal IR" mode, and we may find some so photon starved they

90%

spectroscopy modes are 1 Galaxy Evolution and LS Structure 2 Stellar Populations 3 Star and Planet Formation Key:

+ meets needs does not meet needs irrelevant or not critical

NIO Approach

Parallel efforts – – Address challenges common to all ELTs • Wind-loading • Adaptive optics • Site Explore “point design” • Start from a strawman motivated by science • Understand key technical issues through analysis – NB: Point design is simply a starting point!

• NIO will explore alternate approaches

Enemies Common to all ELTs

• Wind…..

• The Atmosphere……

Wind Loading

• • Primary challenge may be wind buffeting – More critical than for existing telescopes • • Structural resonances closer to peak wind power Wind may limit performance more than local seeing Solutions include: – Site selection for low wind speed – – Optimizing enclosure design Dynamic compensation • • Adaptive Optics Active structural damping

Gemini South Wind Test

Performed during integration of Gemini Telescope on Cerro Pachon – Enclosure has large vent gates – permits testing a range of enclosure conditions – Dummy mirror has properties equivalent to the real primary mirror, but can be instrumented – M1 is above the EL axis - open to the wind flow (similar to concepts for future larger telescopes)

3-axis anemometers

Instrumentation

Pressure sensors, 32 places

Ultrasonic anemometer

Sensor Locations

Ultrasonic anemometer Pressure sensors

AVERAGE Pressure (C00030oo) -0.5

-1 -1.5

-2 0 1.5

1 0.5

0 50 100 150 200 Time History: time (second) 250 300 AVERAGE Pressure (C00030oo) 10 7 10 6 10 5 10 4 10 3 10 2 10 1 10 0 10 -3 SUM = -226 10 -2 10 -1 10 0 Frequency Response Function: frequency (Hz) 10 1

Animation

Wind pressure: C00030oo test_2, day_2, Azimuth angle=00, Zenith angle=30, wind_gate:open, open; wind speed=11 m/s

Wind Pressure Structure Function

C00030oo Average Structural Function for C00030oo 4 3.5

3 2.5

2 1.5

1 0.5

0 0 Prms = 0.076124 d ** 0.4389

1000 2000 3000 4000 5000 sensor spacing, d (mm) 6000 7000 8000

Extrapolation to 30 Meters

0.25

Pressure variation on 30-m mirror about twice 8-m 0.2

0.15

0.1

0.05

0 0 Prms = 0.04d^0.5

5 10 15 20

Sensor separation (meters)

25 30

Summary and Conclusions

• Wind loading on M1 is strongly dependent on vent gate openings • Wind loading on M2 is not strongly dependent on vent gate openings • Control algorithm will maintain wind speed at M1 < 3 m/sec • With vent gates closed, M1 deformations remain within error budget even in high winds • Pressure variations on M1 are larger than average pressure • M1 wind deformations are dominated by astigmatism • M1 deformations are proportional to RMS pressure variations on surface • • M1 deformations ~ proportional to (wind velocity at mirror) ² Pressure structure functions fit 0.5 power law • Structure functions allow extrapolation to larger telescopes – pressure range for 30m twice that of 8m

AO Technology Constraints:

DMs and Computing power (50m telescope; on axis) r 0 (550 nm) = 10cm No. of Computer CCD pixel Actuator pitch S(550nm) S(1.65

m m) actuators power rate/sensor (Gflops) (M pixel/s) 10cm 74% 97% 200,000 9 x 10 5 800 25cm 50cm 25% 86% 30,000 2 x 10 4 125 2% 61% 8,000 1,500 31 SOR (achieved) 789 ~ 2 4 x 4.5

Early 21 st Century technology will keep AO confined to l for telescopes with D ~ 30m – 50m > 1.0

m m

AO Technology Constraints:

Guide stars and optical quality MCAO system analysis by Rigaut & Ellerbroek (2000): • • 30 m telescope 2 arcmin field • • 9 Sodium laser constellation 10 watts each • 4 tip/tilt stars (1 x 17, 3 x 20 Rmag) • Telescope residual errors ~ 100 nm rms Instrument residual errors ~ 70 nm rms System performance: l ( m m) Delivered Strehl 1.25

0.2 ~ 0.4

1.65

0.4 ~ 0.6

2.20 0.6 ~ 0.8

PSF variations < 1% across FOV

Site Evaluation

• • ELT site evaluation more demanding Evaluation criteria include – surface and high altitude winds – – – – – – – – turbulence profiles transmission available clear nights (long-term averages) cirrus cover (artificial guide star performance) light pollution accessibility available local infrastructure land ownership issues

Site Evaluation

• NIO gathering uniform data for sites in: – Northern Chile – Mexico and Southwest US – Hawaii • • Effort led by NIO staff Parallel effort in Mexico, aligned with NIO developed criteria

POINT DESIGN APPROACH

GSMT System Considerations

Science Mission Adaptive Optics Active Optics (aO) Instruments Full System Analysis Site Characteristics Support & Fabrication Issues Enclosure protection GSMT Concept (Phase A)

Point Design: Philosophy

• Select plausible design – must address key science requirements • • • Identify key technical challenges Focus analysis on key challenges Evaluate strengths and weaknesses – – guide initial cost-performance evaluation inform concept design trades • Point design is only a strawman!

Point Design: Motivations

• Enable high-Strehl performance over several arc-minute fields – Stellar populations; galactic kinematics; chemistry • Provide a practical basis for wide-field, native seeing-limited instruments – Origin of large-scale structure • Enable high sensitivity mid-IR spectroscopy – Detection of forming planetary systems

Choices for NIO Point Design

• • • Explore prime focus option – – attractive enabler for wide-field science cost-saving in instrument design Assume adaptive M2 – compensate for wind-buffeting – – reduce thermal background deliver enhanced-seeing images Explore a radio telescope approach – possible structural advantages – possible advantages in accommodating large instruments

Point Design: End-to End Approach

• • • • • • • • Science Requirements (including instruments) Error Budget Control systems Enclosure concept – Interaction with site, telescope and budget Telescope structure – Interaction with wind, optics and instruments Optics – Interaction with telescope, aO/AO systems and instruments AO/MCAO – Interaction with telescope, optics, and instruments Instruments – Interaction with AO and Observing Model

30m Giant Segmented Mirror Telescope Point Design Concept Typical 'raft', 7 mirrors per raft

GEMINI

1.152 m mirror across flats Special raft - 6 places, 4 mirrors per raft

Key Point-Design Features

• Paraboloidal primary – Advantage: Good image quality over 20 arcmin field with only 2 reflections • • Seeing-limited observations in visible Mid-IR – Disadvantage: Higher segment fabrication cost

Key Point-Design Features

• F/1 primary mirror – Advantages: • • • Reduces size of enclosure Reduces flexure of optical support structure Reduces counterweights required – Disadvantages: • • Increased sensitivity to misalignment Increased asphericity of segments

Key Point-Design Features

• Radio telescope structure – Advantages: • • • Cass focus can be located just behind M1 Allows small secondary mirror – can be adaptive Allows MCAO system ahead of Nasmyth focus – Disadvantage: • • Requires counterweight Sweeps out larger volume in enclosure

GSMT Control Concept

Deformable M2 : First stage MCAO, wide field seeing improvement and M1 shape control LGSs provide full sky coverage Active M1 (0.1 ~ 1Hz) 619 segments on 91 rafts



M2: rather slow, large stroke DM to compensate ground layer and telescope figure,



or to use as single DM at

l

>3

m

m. (~8000 actuators)



Dedicated, small field (1 2’) MCAO system (~4-6DMs).

10 20’ field at 0.2 0.3” seeing 1 2’ field fed to the MCAO module Focal plane

Enabling Techniques

Active and Adaptive Optics – – Active Optics already integrated into Keck, VLT, Gemini, Subaru, Magellan, … Adaptive Optics “added” to CFHT, Keck, Gemini, VLT, … Active and Adaptive Optics will have to be integrated into GSMT Telescope

and

Instrument concepts from the start

30m GSMT

Initial Point-Design Structural Model Z X Y Output Set: Mode 1, 2.156537 Hz, Deformed(0.0673): Total Translation Horizon Pointing - Mode 1 = 2.16 Hz

Response to Wind

Need to quantify wind effects, and develop control system responses to correct pointing, focus, and optical aberrations

Modeling of Active Corrections

We are considering two approaches: – – Partition mode shapes + line-of sight equations Ray-tracing mode shapes

Structural Mode Shapes

M1 Surface

Modal Partitioning

•  For each vibration mode shape – – Separate group of nodes representing M1 Determine Zernike polynomials for M1 mode shape by least squares fit – Determine rigid-body motions of M2, calculate effect on image motion and aberrations Combine participation of all modes in random vibration analysis based on a defined wind input, to calculate the PSD, and RMS amplitude of: – – – – image motion piston and focus astigmatism coma

Ray Tracing of Mode Shapes

•  For each vibration mode shape – – Determine rigid-body motions of M1 and M2 Develop an “interferogram” phase map from the residual aberrations on M1 – – Input information to an optical analysis program (i.e., Code V) Determine image motion and Zernike polynomials that describe the wavefront at the focal surface Combine participation of all modes in random vibration analysis based on a defined wind input, to calculate the PSD, and RMS amplitude of: – – – – image motion piston and focus astigmatism coma

~100 ~50 ~20 ~10

Controls Approach:

Offloading Decentralized Controllers

LGS MCAO

spatial & temporal avg

AO (M2)

spatial & temporal avg

aO (M1)

temporal avg spatial avg spatial avg

Secondary rigid body

2

spatial & temporal avg

0.001

Main Axes

0.01

0.1

1 10 100

Enclosure studies

Initial enclosure studies will focus on: – – Ability to protect telescope from wind Understanding cost issues

Some Possible Enclosure Designs

AO Systems

• • • • Adaptive secondary mirror Adaptive mirror in prime focus corrector Multi-conjugate wide-field AO High-order narrow-field conventional AO

Adaptive M2 Intrinsic to Point Design

Options: low- to high-order • Compensate for wind driven M1 distortions • Deliver high Strehl, mid-IR images with low emissivity • Deliver high-Strehl, near-IR images Goal: 8000 actuators 30cm spacing on M1

Key Point-Design Features

• 2m diameter adaptive secondary mirror – Advantages: • • • • Correction of low-order M1 modes Enhanced native seeing Good performance in mid-IR First stage in high-order AO system – Disadvantages: • Increased difficulty (i.e., cost)

Optical “seeing improvement”

using low order AO correction

Image profiles are Lorenzian 16 consecutive nights of adaptive optics at CFHT

Boundary Layer Compensation

Median natural seeing (upper solid line) and median compensated seeing at V band for a single deformable mirror with actuator pitches of 25 cm (solid), 50 cm (dotted) and 1-m (dashed).

(From Rigaut, 2001)

MCAO/AO foci and instruments

Oschmann et al (2001) MCAO optics moves with telescope

elevation axis

Narrow field AO or narrow field seeing limited port MCAO Imager at vertical Nasmyth 4m

MCAO System

Instruments

• • • • • Telescope, AO and instruments must be developed as an integrated system NIO team developing design concepts – – – – Multi-Object, Multi-Fiber, Optical Spectrograph Near IR Deployable Integral Field Spectrograph MCAO-fed near-IR imager Mid-IR, High Dispersion, AO Spectrograph Build on extant concepts where possible Define major design challenges Identify needed technologies

Multi-Object Multi-Fiber Optical Spectrograph (MOMFOS)

•20 arc-minute field • 60-meter fiber cable • 700 0.7” fibers • 3 spectrographs, ~230 fibers each • VPH gratings • Articulated collimator for different resolution regimes Resolution Example ranges with single grating • R= 1,000 350nm – 650nm • R= 5,000 470nm – 530nm • R= 20,000 491nm – 508nm • Detects 13% - 23% of photons hitting the 30m primary

Prime Focus MOMFOS

Barden et al (2001)

Key Point-Design Features (MOMFOS)

• MOMFOS located at prime focus – Advantages • • Fast focal ratio leads to reasonably-sized instrument Adaptive prime focus corrector allows enhanced seeing performance – Disadvantages • Issues of interchange with M2

MOMFOS with Prime Focus Corrector

Conceptual design fits in a 3m dia by 5m long cylinder

MOMFOS Spectrograph Micro-Lens Relay

500 mm diameter fiber fed by two micro-lenses with spherical surfaces 100% of light couples into fiber.

Field stop limits light to 0.72” aperture

MOMFOS Spectrograph

R=20000 mode R=5000 mode 500mm pupil; all spherical optics R=1000 mode

350 nm 440 nm 500 nm 560 nm 650 nm On-axis Spot Diagrams for MOMFOS Spectrograph R=1000 case with 540 l/mm grating.

Circle is 85 microns equal to size of imaged fiber.

On-axis R=5000 case with 2250 l/mm grating.

470 nm 485 nm 500 nm 515 nm 530 nm Barden et al (2001)

MOMFOS Spectrograph Predicted Efficiency

Near Infra-Red Deployable Integral Field Spectrograph (NIRDIF)

• MCAO fed • 1.5 to 2.0 arc-minute FOV • 1 – 2.5 m m wavelength coverage • Deployable IFU units • 1.5 arc-second FOV per IFU probe • 31 slices per IFU probe (0.048” per slice) • ~26 deployable units

Near Infra-Red Deployable Integral Field Spectrograph (NIRDIF)

Relay optics contained in deployable arm.

1.5 by 1.5 arc-second field of view.

f/38 to f/128 converter from MCAO field to image slicer.

Telecentric input and output.

Cold stop located within relay.

Near Infra-Red Deployable Integral Field Spectrograph (NIRDIF)

f/128 image slicer with 31 slices converted to f/11.5 for the spectrograph.

Side view showing top, middle, and bottom slices.

Top view showing top, middle, and bottom slices.

Near Infra-Red Deployable Integral Field Spectrograph (NIRDIF)

Spectrographs • 2 IFU’s per spectrograph • ~13 spectrographs • R = 1000 to 10,000 • Z, J, H, and K spectral coverage (not simultaneous) • 2 K detector format assumed

Mid-Infrared High Dispersion AO Spectrograph (MIHDAS)

• Adaptive Secondary AO feed • On-Axis, Narrow Field/Point Source • R=120,000 • 3 spectrographs • 2-5 m m (small beamed, x-dispersed), 0.2 arc-second slit length • 10-14 m m (x-dispersed), 1 arc-second slit • 16-20 m m (x-dispersed), 1 arc-second slit • 10-14 m m spectrograph likely to utilize same collimator as 16-20 m m instrument. Different Gratings and Camera.

• 2-5 m m spectrograph may require additional AO mirrors.

Mid-Infrared High Dispersion AO Spectrograph (MIHDAS)

16-20 mm spectrograph will be large. Diffraction limit at 20 microns is about ¼ arc-second, comparable to native seeing limit.

Echelle grating is 1.5 meters in length!

Overall instrument is expected to take up a volume of about 3 by 3 by 3 meters!

All of which needs to be cryogenically cooled.

Instrument to be located at Cass location and move with the telescope.

Mid-Infrared High Dispersion AO Spectrograph (MIHDAS)

45 l/mm X-disperser 26.5° blaze ~360 mm R10 Echelle (84°) 7 mm/line 150 by 1500 mm\ m = 696 - 804 16 to 20 m m 1” slit length .28” slit width 1024 by 1024 Si:As detector

MCAO Near-IR Imager

(1:1 Unfolded) 40 mm Pupil ~ 2 arc minute field

MCAO Near-IR Imager

• f/38 input with 1:1 reimaging optics • 1.5 to 2 arc-minute field of view Monolithic imager  5.5 mm/arc-second plate scale!

 685 mm sized detector array for 2 arc-min field!

   28K by 28K detector!

7 by 7 mosaic of 4K arrays 0.004 arc-second per pixel sampling Alternative approach is to have deployable capability for imaging over a subset of the total field.

Instrument Locations on Telescope

• Prime Focus • Co-moving Cass • Fixed Gravity Cass • Direct-fed Naysmyth • Fiber-fed Naysmyth View showing Fixed Gravity Cass instrument

Instrument Locations on Telescope

View showing Co-moving Cass instrument

Instrument Locations on Telescope

Fiber-fed

MOMFOS

MCAO-fed

NIRDIF

or MCAO Imager Cass-fed

MIHDAS

Information on AURA NIO activities is available at: www.aura-nio.noao.edu