The Richard F. Caris Mirror Lab at the University of Arizona continues the production of 8.4 m lightweight honeycomb segments for the primary mirror of the Giant Magellan Telescope. GMT will have a center segment surrounded by six identical off-axis segments, plus an additional off-axis segment to allow continuous operation as segments are removed for coating. Production highlights of the last two years include the spin-casting of Segment 5, preliminary polishing of Segment 2, and completion of the rear surface processing for Segments 3 and 4. We completed a preliminary design of the significant modifications of the test systems required for Segment 4, the center segment. We finished an upgrade of the 8.4 m polishing machine; both the upgrade and experience gained with Segment 1 have contributed to much faster polishing convergence for Segment 2. Prior to polishing Segment 2, we verified the stability and accuracy of the measurement systems by re-measuring Segment 1, achieving good agreement among multiple independent tests as well as good agreement with the original acceptance tests of Segment 1.
A turnkey observatory with 6.5-m telescope has been developed for a broad range of science applications. The observatory includes the telescope, mount and enclosure, installed on site and ready for operation. The telescope’s primary mirror is an f/1.25 honeycomb sandwich of borosilicate glass, similar to that of the MMT and Magellan telescopes. The baseline optical design is for a Gregorian Nasmyth focus at f/11. A Gregorian adaptive optics secondary that provides a wide-field focus corrected for ground layer turbulence (0.25 arcsecond images over a 4 arcminute field) as well as a narrow-field diffraction-limited focus is optional. Another option is a corrected f/5 focus with a 1° field. The observatory, built by partners from academia and industry with extensive experience, can be delivered within five years at a fixed price.
The Richard F. Caris Mirror Lab at the University of Arizona is responsible for production of the eight 8.4 m segments for the primary mirror of the Giant Magellan Telescope, including one spare off-axis segment. We report on the successful casting of Segment 4, the center segment. Prior to generating the optical surface of Segment 2, we carried out a major upgrade of our 8.4 m Large Optical Generator. The upgrade includes new hardware and software to improve accuracy, safety, reliability and ease of use. We are currently carrying out an upgrade of our 8.4 m polishing machine that includes improved orbital polishing capabilities. We added and modified several components of the optical tests during the manufacture of Segment 1, and we have continued to improve the systems in preparation for Segments 2-8. We completed two projects that were prior commitments before GMT Segment 2: casting and polishing the combined primary and tertiary mirrors for the LSST, and casting and generating a 6.5 m mirror for the Tokyo Atacama Observatory.
The LSST M1/M3 combines an 8.4 m primary mirror and a 5.1 m tertiary mirror on one glass substrate. The combined mirror was completed at the Richard F. Caris Mirror Lab at the University of Arizona in October 2014. Interferometric measurements show that both mirrors have surface accuracy better than 20 nm rms over their clear apertures, in nearsimultaneous tests, and that both mirrors meet their stringent structure function specifications. Acceptance tests showed that the radii of curvature, conic constants, and alignment of the 2 optical axes are within the specified tolerances. The mirror figures are obtained by combining the lab measurements with a model of the telescope’s active optics system that uses the 156 support actuators to bend the glass substrate. This correction affects both mirror surfaces simultaneously. We showed that both mirrors have excellent figures and meet their specifications with a single bending of the substrate and correction forces that are well within the allowed magnitude. The interferometers do not resolve some small surface features with high slope errors. We used a new instrument based on deflectometry to measure many of these features with sub-millimeter spatial resolution, and nanometer accuracy for small features, over 12.5 cm apertures. Mirror Lab and LSST staff created synthetic models of both mirrors by combining the interferometric maps and the small highresolution maps, and used these to show the impact of the small features on images is acceptably small.
The Steward Observatory Mirror Lab is nearing completion of the combined primary and tertiary mirrors of the Large
Synoptic Survey Telescope. Fabrication of the combined mirror requires simulation of an active-optics correction that
affects both mirror surfaces in a coordinated way. As is common for large mirrors, the specification allows correction of
large-scale figure errors by a simulated bending of the substrate with the 156 mirror support actuators. Any bending
affects both mirrors, so this active-optics correction is constrained by the requirement of bending the substrate so both
mirrors meet their figure specifications simultaneously. The starting point of the simulated correction must be
measurements of both mirrors with the substrate in the same shape, i. e. the same state of mechanical and thermal stress.
Polishing was carried out using a 1.2 m stressed lap for smoothing and large-scale figuring, and a set of smaller passive
rigid-conformal laps on an orbital polisher for deterministic small-scale figuring. The primary mirror is accurate to about
25 nm rms surface after the active-optics correction, while work continues toward completion of the tertiary.
Segment production for the Giant Magellan Telescope is well underway, with the off-axis Segment 1 completed, off-axis
Segments 2 and 3 already cast, and mold construction in progress for the casting of Segment 4, the center segment. All
equipment and techniques required for segment fabrication and testing have been demonstrated in the manufacture of
Segment 1. The equipment includes a 28 m test tower that incorporates four independent measurements of the segment's
figure and geometry. The interferometric test uses a large asymmetric null corrector with three elements including a 3.75
m spherical mirror and a computer-generated hologram. For independent verification of the large-scale segment shape,
we use a scanning pentaprism test that exploits the natural geometry of the telescope to focus collimated light to a point.
The Software Configurable Optical Test System, loosely based on the Hartmann test, measures slope errors to submicroradian
accuracy at high resolution over the full aperture. An enhanced laser tracker system guides the figuring
through grinding and initial polishing. All measurements agree within the expected uncertainties, including three
independent measurements of radius of curvature that agree within 0.3 mm. Segment 1 was polished using a 1.2 m
stressed lap for smoothing and large-scale figuring, and a set of smaller passive rigid-conformal laps on an orbital
polisher for deterministic small-scale figuring. For the remaining segments, the Mirror Lab is building a smaller, orbital
stressed lap to combine the smoothing capability with deterministic figuring.
Production of segments for the Giant Magellan Telescope is well underway at the Steward Observatory Mirror Lab. We
report on the completion of the first 8.4 m off-axis segment, the casting of the second segment, and preparations for
manufacture of the remaining segments. The complete set of infrastructure for serial production is in place, including the
casting furnace, two 8.4 m capacity grinding and polishing machines, and a 28 m test tower that incorporates four
independent measurement systems. The first segment, with 14 mm p-v aspheric departure, is by some measures the most
challenging astronomical mirror ever made. Its manufacture took longer than expected, but the result is an excellent
figure and demonstration of valuable new systems that will support both fabrication and measurement of the remaining
segments. Polishing was done with a 1.2 m stressed lap for smoothing and large-scale figuring, and a series of smaller
passive rigid-conformal laps for deterministic figuring on smaller scales. The interferometric measurement produces a
null wavefront with a 3-element asymmetric null corrector including a 3.8 m spherical mirror and a computer-generated
hologram. In addition to this test, we relied heavily on the new SCOTS slope test with its high accuracy and dynamic
range. Evaluation of the measured figure includes simulated active correction using both the 160-actuator mirror support
and the alignment degrees of freedom for the off-axis segment.
The primary mirror of the Giant Magellan Telescope consists of seven 8.4 m segments which are borosilicate
honeycomb sandwich mirrors. Fabrication and testing of the off-axis segments is challenging and has led to a number of
innovations in manufacturing technology. The polishing system includes an actively stressed lap that follows the shape
of the aspheric surface, used for large-scale figuring and smoothing, and a passive "rigid conformal lap" for small-scale
figuring and smoothing. Four independent measurement systems support all stages of fabrication and provide redundant
measurements of all critical parameters including mirror figure, radius of curvature, off-axis distance and clocking. The
first measurement uses a laser tracker to scan the surface, with external references to compensate for rigid body
displacements and refractive index variations. The main optical test is a full-aperture interferometric measurement, but it
requires an asymmetric null corrector with three elements, including a 3.75 m mirror and a computer-generated
hologram, to compensate for the surface's 14 mm departure from the best-fit sphere. Two additional optical tests
measure large-scale and small-scale structure, with some overlap. Together these measurements provide high confidence
that the segments meet all requirements.
The Large Synoptic Survey Telescope uses a unique optomechanical design that places the primary and tertiary mirrors
on a single glass substrate. The honeycomb sandwich mirror blank was formed in March 2008 by spin-casting. The
surface is currently a paraboloid with a 9.9 m focal length matching the primary. The deeper curve of the tertiary mirror
will be produced when the surfaces are generated. Both mirrors will be lapped and polished using stressed laps and other
tools on an 8.4 m polishing machine. The highly aspheric primary mirror will be measured through a refractive null lens,
and a computer-generated hologram will be used to validate the null lens. The tertiary mirror will be measured through a
diffractive null corrector, also validated with a separate hologram. The holograms for the two tests provide alignment
references that will be used to make the axes of the two surfaces coincide.
The first of the 8.4 m off-axis segments for the primary mirror of the Giant Magellan Telescope is being manufactured at
the Steward Observatory Mirror Lab. In addition to the manufacture of the segment, this project includes the
development of a complete facility to make and measure all seven segments. We have installed a new 28 m test tower
and designed a set of measurements to guide the fabrication and qualify the finished segments. The first test, a laser-tracker
measurement of the ground surface, is operational. The principal optical test is a full-aperture interferometric test
with a null corrector that includes a 3.75 m spherical mirror, a smaller sphere, and a computer-generated hologram. We
have also designed a scanning pentaprism test to validate the measurement of low-order aberrations. The first segment
has been cast and generated, and is in the process of loose-abrasive grinding.
The solar submillimeter-wave telescope (SST) is the only one of its kind dedicated to solar continuous observations.
Two radiometers at 0.740 mm (405 GHz), and four at 1.415 mm (212 GHz) are placed in the Cassegrain focal plane of
the 1.5-m dish at El Leoncito high altitude site, San Juan, Argentina. The aperture efficiencies are close to design
predictions: 20% and 35% for 2 and 4 arcminutes beam sizes at 405 and 212 GHz, respectively. The positioner absolute
pointing accuracy is 10 arcseconds. Spectral coverage is complemented by ground-based mid-infrared telescopes
developed for high cadence observations in the continuum 10 micron band (30 THz), using small apertures and room-temperature
microbolometer cameras. Using the system, a new solar burst spectral component was discovered,
exhibiting fluxes increasing for smaller wavelengths, separated from the well known microwave component. Rapid sub-second
pulsations are common for all bursts. The pulsations onset times of appear to be connected to the launch times of
CMEs. Active regions are brighter for shorter submillimeter-waves. Mid-IR bright regions are found closely associated
with calcium plages and magnetic structures near the solar photosphere. Intense and rapid 10 micron brightening was
detected on active centers in association with weak flares. These results raise challenging difficulties for interpretation.
Under contract from the Cornell-Caltech Atacama Telescope Project (CCAT), Composite Mirror Applications, Inc.
(CMA) has undertaken a feasibility design study for the use of Carbon Fiber Reinforced Plastic (CFRP) panels in
forming the primary mirror surface. We review some of the past projects using CFRP panel technology for
millimeter and submillimeter wavelength radio astronomy telescopes. Pros and cons of the technology are
discussed. A particular panel configuration was proposed and computer modeled with finite element analysis
(FEA). The technology of replicated CFRP panels for short wavelength radio astronomical telescopes is mature and
cost effective. For shorter wavelengths into the IR and visible, it is becoming a very attractive alternative to
traditional, heavy glass or metal technologies.
The design, manufacture and support of the primary mirror segments for the GMT build on the successful primary mirror systems of the MMT, Magellan and Large Binocular telescopes. The mirror segment and its support system are based on a proven design, and the experience gained in the existing telescopes has led to significant refinements that will provide even better performance in the GMT. The first 8.4 m segment has been cast at the Steward Observatory Mirror Lab, and optical processing is underway. Measurement of the off-axis surface is the greatest challenge in the manufacture of the segments. A set of tests that meets the requirements has been defined and the concepts have been developed in some detail. The most critical parts of the tests have been demonstrated in the measurement of a 1.7 m off-axis prototype. The principal optical test is a full-aperture, high-resolution null test in which a hybrid reflective-diffractive null corrector compensates for the 14 mm aspheric departure of the off-axis segment. The mirror support uses the same synthetic floatation principle as the MMT, Magellan, and LBT mirrors. Refinements for GMT include 3-axis actuators to accommodate the varying orientations of segments in the telescope.
AMiBA consists of a 90 GHz interferometric array telescope with dishes ranging in size from 0.3 to 2.4 meter in diameter, mounted on a 6-meter fully steerable platform. The dishes are attached to the receivers, which are mounted on a platform controlled by a six degree of freedom hexapod mount. The hexapod mount is a parallel connection manipulator also called Stewart Platform. The basic reference for this mechanism is a paper by Stewart. The Stewart Platform is a unique kinematically constrained work platform. It can be manipulated through the six degrees of freedom. The hexapod also provides better accuracy, rigidity, load to weight ratio and load distribution than a serial manipulator or traditional manipulator. The advantages of the hexapod shows that it is a great choice for the AMiBA project. Vertex Antennentechnik GmbH fabricates the hexapod. Testing has started in Germany. The telescope will be delivered in the summer of 2004. The 6m in diameter hexagonal platform is made of carbon fiber reinforced plastics (CFRP) and consists of seven pieces of three different unique types. The platform can be disassembled and fits in a container for transportation. The mounting plane flatness is an important issue for the platform assembly. The deflection angle of the mounting plane relative to any other mounting position must be less than 20 arcsec. Meanwhile, the platform must endure a loading of 3 tons. The platform has been built by Composite Mirror Applications, Inc. (CMA) in Tucson, and mounted on the Hexapod in Germany. This report describes the design and testing of platform and mount for the AMiBA telescope.
For the international ALMA project’s prototype antennas, we have developed a high performance, reactionless nutating subreflector (chopping secondary mirror). This single axis mechanism can switch the antenna’s optical axis by ±1.5’ within 10 ms or ±5’ within 20 ms and maintains pointing stability within the antenna’s 0.6” error budget. The light weight 75 cm diameter subreflector is made of carbon fiber composite to achieve a low moment of inertia, <0.25 kg m2. Its reflecting surface was formed in a compression mold. Carbon fiber is also used together with Invar in the supporting structure for thermal stability. Both the subreflector and the moving coil motors are mounted on flex pivots and the motor magnets counter rotate to absorb the nutation reaction force. Auxiliary motors provide active damping of external disturbances, such as wind gusts. Non contacting optical sensors measure the positions of the subreflector and the motor rocker. The principle mechanical resonance around 20 Hz is compensated with a digital PID servo loop that provides a closed loop bandwidth near 100 Hz. Shaped transitions are used to avoid overstressing mechanical links.
We present a unique hexapod platform array for the new Taiwanese AMIBA project. AMIBA is a 90 GHz radio interferometric array consisting of 19 elements mounted on a roughly 10 m diameter platform. A hexapod mount is used to steer this platform. The resulting design is lightweight in comparison to a more conventional mount. The design goals of pointing stability, platform accuracy and reduced cost can be met with this design. A metrology system for pointing is proposed for inclusion in the design.
Several submillimeterwave astronomical telescope projects have recently employed the use of cast aluminum, machined panels for the reflector surface. Although the resulting surface has several advantages, there are also some drawbacks. In particular, the weight per area is relatively high since it is difficult to make elaborate casting details in the backing ribs and there are quality control concerns in the casting process. To address these problems, we have developed an alternate method of forming the metal reflector blank prior to machining. We have used a high grade, proprietary cast aluminum sheet to form over a mold by slumping. Light- weightedbacking ribs are then welded to the rear. The particular application discussed here is a complete 1.5 m submillimeter wave reflector. The technique is of interest for smaller size panels typically used with large, submillimeter wavelength reflectors.
The prospect of a joint Millimeter Array development effort between the U.S. and Europe has led to various antenna designs. This paper describes a new 12-m antenna design that has many new features which are not widely used among existing millimeter wavelength antennas. These include: light-weight machined aluminum panels; feedlegs with triangular roofing for reflecting scattered rays to the sky; double-layered carbon- fiber reinforced plastic (CFRP) trusses on large radius supports; rotating counterweight for reducing the moment of inertia; a yoke incorporating CFRP trusses and a steel beam structure; and a displacement-measuring metrology system. A design incorporating these features could achieve a combination of high performance and reasonable overall cost. The paper also discusses in detail a number of key issues of interest for future millimeter wavelength antenna development. The design is influenced by the large number of antennas required for the Millimeter Array.
A number of possible antenna designs have been considered for the Millimeter Array. This paper presents two designs which were studied in detail but are no longer being considered, as well as three of the designs currently being studied. The major performance goals for the antennas are listed. Two of the most challenging performance requirements, getting sufficient pointing accuracy and achieving fast position switching, and some of the possible approaches to meeting these challenges are discussed.