vlti tutorial bko for web - eso · vlti tutorial: vlti concept & technical aspects, b. koehler,...

VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching

VLTI Tutorial

The VLTI conceptual design


VLTI Concept...The birth

From a 16m telescope(1980) …nice dream, but no interferometry …

…to a linear array of four 8m telescopes...(1988)

…the baby was born! …

…the troubles could start! …☺

…to a mature VLT Interferometer layout(1990)


The Final VLTI Array Layout4 UT’s as ‘trapezoidal’ as possible (wind shadow, ground quality)

♦ Baselines 47 — 130m ⇒ 1.5 milli arcsec

30 Stations for AT’s

♦ Baselines 8 — 200m ⇒ 1 milli arcsec

Delay Line Tunnel (160 x 8m) positioned to minimize path differences. Can house 8 DL’s

Central combining Lab (20 x 7m) with specific VLTI building / control room

All elements are located on an 8m grid

Light travelling underground for high thermal stability (vacuum alternative not selected for cost reasons)


The VLTI Optical Layout


The ‘Appetizers’(2 Siderostats)♦ Installed early 2001♦ Relocatable in day(s)♦ Primary: 0.4 m♦ Autoguiding only♦ Airy disk (in K): 1.1”

The ‘Workhorses’(3 Auxiliary Tel.)♦ Relocatable in 3h♦ Primary: 1.8m♦ Fast Tip-Tilt♦ Limited Chopping♦ Airy (in K): 0.25”

The telescope familyThe ‘Kings’(4 Unit Tel.)♦ Fixed position!♦ Primary: 8 m♦ Adaptive Optics♦ Airy disk (in K):

0.06”


Telescope Optical DesignIntermediate Focus below M4:

♦ Various artificial light sources for calibration & alignment

Intermediate pupil image on M8:

♦ Deformable mirror for MACAO

Coude Focus:♦ Fast Tip-tilt sensor (STRAP) and

later Wave Front Sensor (MACAO)♦ TCCD for Field acquisition♦ Field of View at Coude: 2 arcmin.

(FoV 2 arcsec in laboratory)


Relay Optics


Relay Optics‘Grasping’ the UT-photons♦ M9: large dichroic.

» Reflects IR to VLTI » transmits Visible to STRAP and later MACAO Curvature sensor. » Optimized for polarization.

The Relay Optics (M9, M10, M11) inside UT Coude room)

♦ M10: convex spherical. » Re-image telescope pupil

(M2) in Tunnel center. » Articulated mount to

adjust lateral pupil position in Lab.

♦ M11: off-axis parabola.» Collimates beam and send

it into light duct towards M12 in DL Tunnel Location for STRAP/

MACAO Wave Front Sensor


The Delay Lines


The Delay Lines

The ‘Paranal-Express’♦ Stroke: 60 m (120m in OPL)♦ Resolution: <5nm♦ Max. velocity: 0.5 m/s♦ Stability (jitter): <14nm rms♦ Power dissipation: <15W

Variable Curvature Mirror (VCM) ♦ re-image pupil inside the

Laboratory♦ mounted on piezo translator for

fast OPD correction

The first two Delay Lines (#I & II) inside the tunnel


Transfer Optics


Transfer Optics

The beam bending ‘gadgets’♦ Very high optical quality flats (λ/60

surf.)♦ Coating optimized for transmission

and polarization (45° incidence)♦ Very stiff mounts and tables for

immunity to µ-seismic noise♦ M12 eventually on robotic arm for

automatic array reconfiguration♦ M16 on translation stages for beam

switching inside lab.♦ Support future dual-feed (PRIMA)

The M12 flat folding mirror

The M16 flat folding mirror


Beam Make-up

STRAP (Tip-Tilt)or

MACAO (Adaptive Optics

FINITO (Fringe Tracker)


STRAP (Tip-Tilt)Quad-cell detector with APD at telescope Coude focus

Pure tip-tilt correction is optimal when D/r0<4, i.e. UT/10µm (MIDI) or AT/2 µm (VINCI,AMBER, MIDI)Gain on Strehl up to 5.Will be used on UT till MACAO is availableWill be resident on AT

10 11 12 13 14 15 16 17 180

0.1

0.2

0.3

0.4

0.5

AT lab results, diffr. limit at 2.2 microns = 0.255 arcsecUT lab results, diffr. limit at 10 microns=0.26 arcsec

Tilt loop residual rms angular jitter

NGS V-magnitude, K0 star

Diff

r. lim

it un

its


MACAO (Adaptive Optics)

X-Y Table allows reference source different from target

A must for UT in K-band (AMBER). Gain in coherent flux ≈10060 elements Curvature System at telescope Coude

♦ WaveFrontSensor using APD coupled with optical fibers♦ Bimorph Deformable Mirror on Tip-Tilt mount ♦ vibrating membrane, ♦ radial geometry micro-lenses,


MACAO performance

667k 266k 106k 42.1k 16.8k 6.67k 2.66k 1.06k 421

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.65" seeing, magnitude 20.5 sky, 1" FoV, 250 dark current, including error budget

Specifications

1.0

Count/subaperture/second

Guide Star Magnitude

0.2

0

0.6

0.4

0.8

1816141210

0.65” @ 500 nm, τo~4ms, V=20.5 sky background

Stre

hlR

atio

at 2

.2 µ

m


FINITO (Fringe Tracker)


The VLTI Laboratory

High thermal stability; limited access, low power dissipationTelescope Exit Pupil re-imaged into instrumentsOptical design for up to 8 beams (4 UTs with dual feed)

5 m

VINCI

PRIMA FSU

M16

MIDI

MIDI

VisitorInstrument

PRIMA DDLs

BeamCompressors

Switchyard

AMBER

DL IDL II

DL IIIDL IV

VLTI Laboratory

ZOPD

Ref.FSU

Telescope Exit Pupil

Delay Line Tunnel


Inside the Laboratory

The Beam SwitchyardThe Beam Compressors

♦ 3-mirror design (compress 80 18 mm)♦ High optical quality off-axis parabolas (surface

error 7 rms)

LEONARDO Reference sources♦ Laser and white-light sources♦ Light can be sent to any instrument or

backwards to telescope♦ Provide reference axis for each beam and

reference OPD between beams♦ Enable instrument to obtain fringes in ‘autotest’

VINCI Test Instrument♦ Used for VLTI commissioning & later as

reference for performance tracking.

Fringe Tracker FINITO♦ Co-phases (3 beams) to allow long exposures♦ mH<12 (UT)

The Beam Compressors (3 mirrors each)

The Switchyard & Leonardo reference source


VINCI


VLTI Tutorial

The VLTI selected critical technical aspects


VLTI Error Budget

LEGEND:

OPD VARIATION ACROSS PUPIL

OVERALL

OPD VARIATION WITH TIME

POLARIZATION EFFECTS

UNEQUAL BEAM INTENSITIES

VLTIINSTRUMENT

FRINGE TRACKER

ADAPTIVE OPTICS OPTICAL DESIGN OPTICS IMPERFETION

FINE IMAGE TRACKER

Atmosphere

Internal seeing

Coating

Internal seeing

Figuring errors

Alignment errors

Fringe sensor

Actuator (Delay Line)

Control elect.

Wave Front Sensor

Deformable mirror

Control elect.

Fine Image Sensor

Actuator (M16)

Control elect.

LINEAR RETARDATION

PARTIAL POLARIZATION

ROTAT. OF FRAME OF REFERENCE

ON AXISCO-PHASING MODEUT with ADAPTIVE OPTICSλ = 2.2 µmWith spatial filter φ=17 airy radii

LINEAR RETARDATION

Systematic visibility loss (%) Random visibility loss (%)

Error source

Fringe sensor = Allocation to be extracted from a closed loop analysis

39.3 5.4

2.0 0.4 31.0 5

6.0 0.0 0.06 0.0 0.1 0.0 0.0 0.0

0.0 0.0

0.14 0.0

6.16 0.0 0.14 0.0 10.0 ? 0.0

2.0 0.4 ?

4.5?22.7 0.02 0.0 6.3 0.1 2.0 0.4 ?

0.1 0.0

4.2 0.0

2.0 0.1?

VLTI INSTRUMENTS = Error sources not part of this error budget

OPD variation with time(vibrations)

OPD variation across pupil(optical quality)

Polarization

Uneq. beam intensities

Detector resolution, etc.

Seeing: 0.66 arcsec (@ 0.5 µm)t0: 10 msec (@ 0.6 µm)Mv: 11Cn2: 10e-15 m-2/3

VISIBILITY LOSS: ∆V/V


OPD variation with time (Piston errors)Error sources:♦ Atmosphere♦ Internal seeing♦ Vibration (wind, natural & man-made

seismics, acoustics, pumps, etc.)

Compensated (Partially) by fringe tracking.

00,10,20,30,40,50,60,70,80,9

1

0,2

0,8

1,4 2

2,6

3,2

3,8

4,4 5

5,6

6,2

Fringe Phase (rad)

Nor

mal

ised

Frin

ge In

tens

ity

Fringe pattern at time t Fringe pattern at time t + dt

Resulting Fringe patern

Piston variation

Top Level requirement: Instrumental errors (internal seeing & vibrations) < atmosphere

Wavelength 0.6 µm 2.2 µm Exposure time 10 msec 48 msec OPL requirement (each arm) in [nm] 21 75 Telescope (UT or AT, each) 14 50 Delay Line (each) 14 50 Beam Combiner 6 21 Transfer Optics 3 10 Internal Seeing 3 10


Unequal beam intensity

Visibility loss due to intensity mismatch:

Error sources:♦ Atmospheric scintillation (negligible)♦ Coatings (small)♦ For fiber-fed instrument (large):

» Tip-tilt errors » Instantaneous Strehl fluctuation

21

21 ..21

IIII

VV

+−=∆

VINCI measure I1 & I2 (photometry channels) to correct for this effect. ⇒


Polarization EffectsLinear Retardation♦ Error sources: Coatings

» Different coatings in the two arms» Differential incidence angles (between arms)» Differential coating characteristics (e.g. thickness of the protecting layer)

♦ Can be very critical in Near IR/Visible and for multi-dielectric coatings (M9).


Polarization EffectsRotation of polarization frame♦ Error sources:

» Differential frame-of-reference/pupil rotation due to optical design and/or misalignment

♦ Can be avoided by proper optical design and alignment.


VLTI Tutorial

Some results from on-going VLTI commissioning


Measured micro-seismic vibration

OPD as measured with accelerometersat the back of Transfer Optics mirrors (M12, M16, Switchyard, Beam Compressor)

Accelerometer on back of M12-G0 mirror

Error budget:Wavelength VIS NIR TIRTransferOptics

3 10 45


OPD stability inside UTMeasured with high sensitive accelerometers on UT#1&3Few improvements on sub-system identified and on-goingOverall stability excellent

M1

M2

M3M4

M5

M6

Overview - Vibrations of subsystems

0.0

0.1

1.0

10.0

100.0

All O

FF

Air C

o. in

Pow

er c

abin

et O

N

(Alt+

Az+N

asm

.) LC

U's

ON

M2

Uni

t LC

U O

N

Cas

segr

ain

Adap

tor L

CU

ON

M1

Cel

l LC

U O

N

HBS

ON

Coo

ling

Pum

ps O

N (T

ube

valv

e C

LOSE

D)

Coo

ling

Pum

ps O

N +

Tube

val

ve O

PEN

Azim

uth

Mot

or O

N

Altit

ude

Mot

or O

N (C

able

Wra

p O

FF)

Altit

ude

Mot

or O

N +

Cab

le W

rap

B O

N

Altit

ude

Mot

or O

N +

Cab

le W

rap

A&B

ON

Subsystems

Cont

rast

Los

s [%

]

VISNIRTIR

Spec.


Measured OPD spectrum

0 10 20 30 40 50 60 70-30

-25

-20

-15

-10

-5

0

5

10

15

20File: opd_vinci_on_sky

Time [sec]

OP

D [u

m]

OPD as measured on a starby VINCI with Siderostats

10-2 10-1 100 101 10210-8

10-7

10-6

10-5

10-4File: opd_vinci_on_sky

Frequency [Hz]

PS

D o

f OP

D [m

/sqr

t(Hz)

]

Measured by VINCI Kolmogoroff (seeing=0.65" to=6 msec


Measured Internal seeing (tip-tilt)

0 200 400 600 800 1000 1200-0.06

-0.04

-0.02

0

0.02

0.04

0.06

0.08Centroid Motion X & Y from ALIU CCD. File: vinciTCCD_UT3_NasBea_2.dat

Time [sec]

Cen

troid

[arc

sec/

sky]

X = 0.0115 [arcsec/sky] rmsY = 0.015 [arcsec/sky] rms

X CentroidY Centroid

10-2 10-1 100 101 10210-6

10-5

10-4

10-3

10-2

10-1

100Centroid Motion X & Y from ALIU CCD. File: vinciTCCD_UT3_NasBea_2.dat

Frequency [Hz]

PSD

of X

& Y

Cen

troid

[px]

X CentroidY Centroid

Nasmyth laser beacon (UT3) imaged on VINCI CCD(after 25 mirrors and 200m distance)


Measured Image Quality

Image Quality from UT Nasmyth laser beacon to VINCI CCD:

FWHM < 0.040” (limited by pixel size)


ConclusionSpecific and stringent error contributions need to be taken into account for an Interferometer:♦ OPD stability (vibrations from equipment, wind buffeting,

micro-seismic noise, residual atmospheric piston,...)♦ Polarisation (optical layout, coatings)♦ Image Quality (mirror figuring, alignment, internal

seeing,...)

The two major technical risks have been:♦ OPD stability within the UT♦ The internal seeing

Both proved to be according (or very close) to the required performance

Ready for the next step: PRIMA + ...

vlti tutorial bko for web - eso · vlti tutorial: vlti concept & technical aspects, b. koehler,...

Documents