VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
VLTI Tutorial
The VLTI conceptual design
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
The VLTI Optical Layout
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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”
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
Relay Optics
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
The Delay Lines
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
Transfer Optics
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
Beam Make-up
STRAP (Tip-Tilt)or
MACAO (Adaptive Optics
FINITO (Fringe Tracker)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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,
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
FINITO (Fringe Tracker)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
VINCI
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
VLTI Tutorial
The VLTI selected critical technical aspects
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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. ⇒
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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).
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
VLTI Tutorial
Some results from on-going VLTI commissioning
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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.
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
Measured Image Quality
Image Quality from UT Nasmyth laser beacon to VINCI CCD:
FWHM < 0.040” (limited by pixel size)
VLTI Tutorial: VLTI Concept & technical aspects, B. Koehler, 20 November 2001, Garching
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 + ...