Transcript Slide 1

Development of surface metrology for the
Giant Magellan Telescope primary mirror
J. H. Burgea,b, W. Davisona, H. M. Martina, C. Zhaob
aSteward
Observatory, University of Arizona
bCollege of Optical Sciences, University of Arizona
GMT primary mirror segments
• 25 meter telescope, requires 7 mirror segments
• Each mirror segment is 8.4 meters in diameter
• The off axis segments have 14.5 mm aspheric departure
Principal measurement using interferometer and
reflective null corrector with CGH
3.75-m M1 fold sphere
tilted 14.2°
Interferometer
interferometer
computer-generated
hologram
CGH
of pa
rent e
llipsi
od
25
meters
0.7575-cm
m spheremirror
Axis
Sam
GMT
segment
GMT
segment
M1
center of curvature
Interferometer
at M1 center
of curvature
Test tower at Steward Observatory
Mirror Lab
Original tower
New tower
New tower
28 meters tall, 80 tons of steel
floated on 400 ton concrete pad
accommodates other UA projects (LBT, LSST)
lowest resonance of 4.8 Hz with 9 ton 3.75-m fold sphere + cell
New test tower at Mirror Lab
Test optics for
GMT segment
Test of 3.75 m
fold sphere
28 m vibration-isolated tower was installed
2006-07. Supports all GMT tests, plus LSST,
future 6.5 m and 8.4 m mirrors.
GMT off-axis
segment
Measurement of center segment
The center segment can be measured by tilting the fold sphere to point
straight down, then a small computer generated hologram will
compensate the residual errors.
Cone defined by
light from outer
edge of mirror
Cone defined by
light from edge of
central hole
50 mm CGH
compensates only
20µm aspheric
departure
Vibration
insensitive
interferometer
Optics of Sam
Insert a CGH to test Sam
Point source
microscope
aligned to M2
Interferometer for
GMT measurements
CGH
M2
Invar cradle provides stable reference for M2 and CGH
M2 is aligned to CGH
Point Source
Microscope
M2 CoC
reference ball
M2 CoC
reference ball
To
M2
Computer
generated
hologram
CGH and M2 CoC reference ball are aligned using
CMM to 10 µm
M2 aligned to CoC reference ball using PSM
M2 mount
Use of a Point Source Microscope to align M2
• Use cradle to locate ball at location
where M2 center of curvature
should be
(cradle geometry defined by CMM)
• PSM is adjusted to the ball
• The ball is then removed. The
PSM is looking at mirror directly.
• Adjust the mirror until reflection
from it is focused on the same spot
as the ball on the camera
Interferometer alignment to CGH
Use return into interferometer from reference patterns on CGH for
• tilt (using fold flat)
• shifting interferometer for focus
CGH test of Sam
• CGH inserted into light coming from Sam
• Reflection back through system is used to verify
wavefront
• CGH mounted on invar plate with other references for
M1 alignment
Alignment of M1, GMT
• M1 is aligned to Sam with
~100 µm tolerances
• Reference hologram is aligned to
Sam. Then it is used to represent
Sam.
• A laser tracker measures the 3space position of the reference
hologram and M1.
• M1 is aligned to the reference
hologram according to the
measurements.
• The laser tracker also provides
the reference for the GMT
location in the test
References co-aligned with CGH
CGH, coaligned with:
Corner cube tracker reference
Flat mirror, angular reference for
tracker
Alignment error budget
Effect on primary mirror segment in telescope
correction
force
(N rms)
residual
rms surface
(nm)
0
2.8
2.4
0.0
0
5.8
5.4
Reference Hologram
0.6
2
6.0
6.9
M1
1.0
5
7.2
9.1
GMT
0.2
0
2.7
3.2
Sam not measured by
reference hologram
0.2
3
7.1
7.4
System total
1.2
6
13.4
15.4
radial shift
(mm)
clocking
(arcsec)
Interferometer
0.0
M2
3.75 m fold sphere
•
•
•
Figure of fold sphere will be measured in situ and subtracted.
Accuracy of correction depends on slope errors, and magnitude of small-scale structure that cannot be subtracted.
Finished fold sphere meets requirements:
•
•
•
< 2 nm/cm rms slope error
small-scale errors < 15% of GMT segment specification
Overall accuracy < 20 nm rms over clear aperture.
Cast in the Mirror Lab
spinning oven
Polished at the Mirror Lab
Coated at Kitt Peak
Support of fold sphere
3750 mm
mm
Hangs from “Active” support, allowing quasi-static
force adjustment based on in situ measurement
455
mm
Scanning pentaprism test
Pentaprism rail lies in plane
perpendicular to parent axis.
Image at CCD
Hub rotates rail to scan
different diameters.
CCD camera at
focus of paraboloid
Axis
of pa
lo
rabo
id
d
mate
Colli
Scanning
pentaprism
pa
rent
Fixed reference
pentaprism
with beamsplitter
laser
parent paraboloid
Off-axis mirror
Scanning pentaprism measures slope errors
by producing collimated beams parallel to
parent axis. Displacement of focused spot is
measured with camera in focal plane.
Scanning pentaprism test as implemented
for GMT off-axis segments. Pentaprism
rail is suspended from tower.
Pentaprism test of 1.7 m off-axis NST mirror
•
•
•
•
1/5 scale GMT pentaprism test
This was done in late 2007 before
the mirror was finished.
The pentaprism test only samples
lowest order aberrations
The PP results agree with results
from interferometry
Poster paper by P.Su et al
interferometer
pentaprism
nm rms surface
nm rms surface
astigmatism 0°
8
9 ± 23
astigmatism 45°
0
-2 ± 23
coma 0°
-87
-98 ± 12
coma 90°
-4
16 ± 12
trefoil 0°
-50
-32 ± 35
trefoil 30°
9
23 ± 30
spherical
-32
-35 ± 8
nm surface
aberration
interferometric test
pentaprism measurement
Laser Tracker Plus
laser tracker & distance-measuring
interferometers (DMI)
sphere-mounted retroreflector for laser tracker
laser tracker
DMI laser and remote
receivers
PSD
10% BS
DMIs
DMI retroreflector
Retroreflector for interferometer
and position sensing detector (PSD)
assemblies in 4 places at edge of
mirror
Poster paper by T. Zobrist et al
Laser Tracker Plus measurement of
3.75 m fold sphere M1
• R = 25.5 m, tracker distance = 22 m
• 93 sample points, measured 4 DMIs with each sample
• Subtracted best-fit sphere (R = 25.497 m)
before DMI correction: 1.4 μm rms
after DMI correction: 0.75 μm rms
Shear test
Each segment has axisymmetry
about parent axis
Rotate segment about this axis
under the optical test and
separate effects that move with
the mirror from those that remain
with the test.
Summary
• We are building the hardware to measure the GMT segments.
• We expect to meet a tight error budget
• Low order modes controlled by active optics, using < 5% Force
• Uncorrectable features fit well within the allotted GMT PM structure
function
• We take this problem seriously and have implemented a
comprehensive set of crosschecks
• Scanning pentaprism system
• Laser tracker Plus
• Shear test
We have invested in metrology for making the first of the
segments. Others will be made at low risk, low cost.