|
At the University of Arizona's Steward Observatory Mirror Laboratory, a team of
scientists and engineers is making giant, lightweight mirrors of unprecedented
power for a new generation of optical and infrared telescopes.
These mirrors represent a radical departure from the conventional solid-glass
mirrors used in the past. They have a honeycomb structure on the inside; made out of
Ohara E6-type borosilicate glass that is melted, molded and spun cast
into the shape of a paraboloid in a custom-designed rotating oven. Honeycomb mirrors offer the advantages of their solid counterparts -
rigidity and stability - but they can be significantly larger, and
dramatically lighter. With their lightweight structure, air can be
circulated through the honeycomb structure forcing the glass to reach
thermal equilibrium with the air temperature in a relatively short time,
on the order of 20-30 minutes.
The Mirror Lab team has also developed a revolutionary new method to
polish the honeycomb mirrors with a deeply curved, parabolic surface
that results in much shorter focal lengths than conventional mirrors.
Such fast mirrors not only improve telescope performance, but they can
fit into a much shorter telescope body that requires a smaller, less
expensive enclosure. The typical focal ratios are of order f/1.25 to
f/1.14.
The pioneering work being done today at the Steward Observatory Mirror Lab
had its beginning around 1980 with a backyard experiment by Dr. Roger Angel,
the lab's founder and scientific director. Curious about the suitability of borosilicate
glass (the kind used in glass ovenware) for making honeycomb structures, he
tested the idea by fusing together two custard cups in an improvised kiln.
The experiment was a success and led to a series of bigger kilns and small
furnaces and, eventually, the spin casting of three 1.8 meter mirrors.
 By 1985, with financial support primarily from the US Air Force, the National
Science Foundation and the University of Arizona, Roger Angel (in photo at left) and a talented
Mirror Lab team moved to the current facility under the east wing of the UA
football stadium. A large, rotating furnace was built and a series of mirrors
as big as 3.5 meters in diameter were successfully cast.
By 1990, the rotating furnace was expanded to its current size, and a new wing
was added to the Mirror Lab to house two mirror polishing stations and a test
tower. The new furnace, which is large enough to cast mirrors up to 8.4 m in diameter, was first used in 1992 to
make a 6.5-m mirror. In January 1997 the first
8.4-m mirror
for the Large Binocular Telescope (LBT) was completed.
As part of the technology development process, the Mirror Lab has
successfully produced fourteen mirrors with diameters of 1.2, 1.5, 3.5, 6.5, and
8.4 m. Nearly all of these mirrors are now
operating in telescopes including: the SAO 1.2-m f/1.9 on Mt. Hopkins,
the Lennon 1.8-m f/1.0 on Mt. Graham, the ARC 3.5-m f/1.75 on Apache
Point, NM, the WIYN 3.5-m f/1.75 on Kitt Peak and the Phillips Lab 3.5-m f/1.5
at Starfire Optical Range, NM. The 3.5-m mirrors and larger ones have been
polished at the Mirror Lab using the stressed-lap polishing technique to produce
a surface figure ~ +15 to +20 nm rms.
The Mirror Lab continues its impressive history of successful, ground
breaking mirror castings. After a series of 1.2m and 3.5m mirrors proved
the casting and polishing techniques, the oven was enlarged for the
casting of 8-m class mirrors in 1991. In April 1992, the first 6.5-m f/1.25 honeycomb blank was successfully cast. This mirror went
into operation in May 2000 as the MMT Conversion Project on Mt. Hopkins. In February 1994, the
second 6.5-m f/1.25 honeycomb blank was cast. This mirror, for the Magellan
Project, was installed at Las Campanas Observatory, Chile in 2001.
Construction of the mold
for the casting of an 8.4-m f/1.14 honeycomb blank was completed in the fall of 1996.
Furnace modifications were performed to accommodate the geometry of the mold on the
furnace hearth. This mirror, cast in January of 1997, is the first of two mirrors
for the
Large Binocular Telescope Project
on Mt. Graham, AZ. The faceplate was remelted in June 1997 to replace a small
amount of glass that leaked out during the casting and bring
the faceplate back to its optimum thickness.
The image (at left) shows three large honeycomb mirrors in the casting lab at the
end of 1997. In the foreground, the first LBT 8.4-m mirror is still in the mold on
the furnace. In the center, the second 6.5-m mirror for the Magellan Project hangs in the turning
ring. Before polishing the mirrors are lifted with a fixture glued to the front
surface with silicone rubber sealant. At the far end of the lab, the first 6.5-m mirror
for the MMT is being lifted off of the polishing cell with a vacuum lifting fixture.
After polishing the mirror is lifted with vacuum pads to avoid marring the precisely
polished surface.
The large mirrors are generated (by machining with diamond grinding tools) on a machine
called the Large Optical Generator (LOG) which is a high precision vertical milling
machine. Then the mirrors are polished on the same machine by lapping with the stressed-lap
polishing tool. The image (at left) shows the second 6.5-m mirror being polished on the polishing
machine in 1998 (photo by Peter Wehinger).
The dark reddish frame below the turntable is an air cart used to
transport the mirror and its polishing cell between the polishing
machine and the test tower.
The  image
(below left) shows the completed
Magellan I 6.5-m mirror in the foreground, and the LBT I 8.4-m mirror in
the background under the test tower (photo by Steve Miller).
After polishing has been completed, the mirrors are lifted with a
special vacuum- lifting fixture. The lifting fixture is shown in the
image (below left) with 36 pads which
 attach to
the mirror surface and three pairs of redundant vacuum pumps. The
polished surface of the mirror is coated with a layer of blue plastic
(opti-coat) to protect it from scratches during shipping. The frame below the
mirror is the base of the transportation box (photo by Peter Wehinger).
A facility for polishing and testing the highly aspheric secondary
mirrors required for these various large telescopes is now nearing
completion. Stressed-lap polishing and testing with computer-generated
holograms will allow the production of secondary mirrors up to 1.8 m in
diameter. A 1.2-m secondary has been completed for the wide-field
telescope of the Sloan Digital Sky Survey in New Mexico.
Steward Observatory's honeycomb mirrors represent the next technological
leap that will allow astronomers to press even farther into space, while
keeping down the cost of doing big science.
|
