LAST May, the space shuttle carried an unusual payload into orbit. Made up of
little more than a box packed with polymer sheeting, bottles of compressed
nitrogen, cameras and a large LED display, it was what NASA calls a class D
experiment鈥攁 venture that has a high risk of failure but offers huge
potential gains if it does prove successful.
On the second day of the mission, with the shuttle flying with its bay open
towards Earth, the gamble began. The crew released the box, then manoeuvred the
shuttle a hundred metres or so away to watch. As the experiment began, the box
popped open and the carefully folded sheeting spilt into space. Next, nitrogen
gas began forcing its way into the material and a complex structure started to
take shape. First, three struts began waving clumsily, before forming a tripod
with its apex still attached to the box. Around the free ends of the tripod
legs, a huge ring unfurled within which sat an empty inflatable discus-shaped
balloon 15 metres across. The side of the balloon closest to the box was
transparent, but the inner surface of the other side, a carefully-shaped
parabolic dish, had a metallic coating. The shapeless bundle that had emerged
from the box had turned itself into a giant radio antenna the size of the space
shuttle.
Bad omen
Advertisement
The plan was to check for irregularities in the parabola using patterns of
light generated by the LEDs mounted in the box. Cameras inside the box would
record the images of these patterns as they were reflected from the balloon, and
store the results. The shuttle was to collect the box, and back on Earth,
scientists would study any distortion in the reflected images to work out how
true the parabola was.
But something was going wrong. Strange ripples began to appear on the surface
of the uninflated discus. Then, to the astronauts鈥� astonishment, the entire
structure began to tumble end over end. It was a bad omen come true: before the
launch, a NASA engineer had told colleagues of a dream in which the antenna
wrapped itself around the shuttle like an octopus. Quickly the astronauts moved
the shuttle to a safe distance.
From then on, the only part of the $10 million experiment that went as
planned came 90 minutes later, when explosive bolts cut the antenna loose. The
shuttle returned to retrieve what was left鈥攖he empty spacecraft with its
LEDs, onboard cameras and measuring equipment, and whatever data had accumulated
during the test. Less than a day later, the antenna burnt up as it re-entered
the Earth鈥檚 atmosphere.
Nobody yet knows exactly what went wrong. 鈥淲e鈥檙e still analysing the data,鈥�
says Robert Freeland, the physicist who coordinated the project from the Jet
Propulsion Laboratory, near Los Angeles. Freeland has his own ideas, but
confesses: 鈥淲e鈥檒l probably never know for sure.鈥�
The Inflatable Antenna Experiment or IAE was designed to test the
technologies that could build a new generation of giant, lightweight inflatable
structures in space. Among the priorities are larger dishes for remote sensing
satellites and orbiting telescopes.
Big dishes
The resolution of optical telescopes is limited by the size of their mirrors,
and the same applies to the dishes of radio antennas. Bigger dishes also allow
antennas to collect weaker signals and to broadcast signals over a smaller
鈥渇ootprint鈥� on Earth. In telecommunications this would allow the same set of
frequencies to be used again nearby.
One project favoured by Freeland would use a large antenna to tune into the
1.4-gigahertz frequency reflected by liquid water several metres beneath the
Earth鈥檚 surface. A structure very similar to the IAE could be used with the
parabola facing Earth and with radiometric sensing equipment at the antenna鈥檚
focus to map the moisture content of the ground. Such a map would give valuable
insight into the way moisture is transported around the globe. Carried on a
journey to Mars, such an antenna might even be able to map the water or ice
content of the Red Planet鈥檚 surface. 鈥淭he potential is enormous,鈥� says
Freeland.
Inflatable structures in space are hardly new. The earliest was a 30-metre
balloon known as Echo, which served as a reflector for communications
experiments in the 1960s. NASA used the balloon to bounce radio signals between
locations many hundreds of kilometres apart, and American radio hams used it to
make the first satellite-based radio links with the Soviet Union. During the
Cold War, inflatable decoys were put aboard intercontinental missiles for
deployment before re-entry, in an attempt to disguise the number of warheads
they were carrying.
The IAE was more complex altogether. To design and build the antenna,
Freeland commissioned a small Californian company called L鈥橤arde which has more
than twenty years of experience building inflatable decoys for intercontinental
missiles. Even with this experience, designing the far more complex inflatable
antenna was not easy. The designer has to work out its eventual shape, how best
to construct it, what materials to use and pressure is required for each part of
the structure.
L鈥橤arde decided to build the antenna in sections. For the dish itself, the
company came up with the large discus-shaped inflatable, which was made from a
transparent polymer known as Mylar, in a film only 0.3 millimetres thick. Mylar
is often used for the novelty balloons with a metallic coating sold at carnivals
and funfairs. A similar metallic coating was used for the reflective inner
surface of the antenna which formed the all-important parabolic dish. The discus
was held firmly in place by the large inflatable ring, which was itself attached
to the three inflatable struts that held the dish a fixed distance from the box
and the LED array.
Working out how to make this three-dimensional structure from two-dimensional
sheets is done by a computer, says Freeland. Starting from a 3D model of the
fully inflated shape, the computer works backwards to calculate the shape of the
sheets that must be bonded together to create it and the pressure required to
maintain it. Too great a pressure and the Mylar would begin to stretch, too
little and the dish would flop about. For this the computer must take into
account the physical properties of the materials and their thickness. Even then,
small but unavoidable variations in the thickness of the Mylar sheets can put
paid to the best calculations. 鈥淚mproving the uniformity of the material is one
of the challenges for the future,鈥� says Freeland
Assembling the pieces is the most difficult part of the entire project. Each
half of the discus, for example, was made up from 62 pie-slice segments. The
pieces, made with clingfilm-like Mylar, had to be bonded together by hand with
an accuracy of a few millimetres. The material is a real headache to work with,
says Freeland. 鈥淎ny errors cause bumps or wrinkles on the surface that could
have ruined the experiment. We鈥檙e still learning how to do it better.鈥�
The final challenge was to work out how to pack the structure in such a way
that it would unfurl easily. It was during this stage that the experiment
eventually went wrong. Without the budget to test how different packing
strategies could affect the deployment, Freeland settled for a best guess, based
on experience with inflatable structures on Earth. He folded the struts
accordion-style, while the ring and discus were folded like an inflatable life
raft. 鈥淭he deployment dynamics turned out to be crucial,鈥� says Freeland.
To prevent the legs looping behind the box during the inflation, Freeland
planned to eject the folded structure using a spring. Inflation would begin only
when the uninflated ring and discus were some 30 metres away and the struts were
more or less straight.
But when the experiment began, the bundle of Mylar tubes and balloons was
forced into space by a small amount of air that had been trapped in the box and
by the natural springiness of the sheeting. It came to rest some four metres
from the box鈥攖oo far away for the ejector spring to work but too close to
guarantee a stable inflation. 鈥淲e didn鈥檛 anticipate that,鈥� says Freeland. The
ungainly waving of legs was the result.
Although the legs and ring did inflate successfully, says Freeland, 鈥渋t鈥檚
obvious from the video footage taken by the shuttle crew that the dish didn鈥檛
inflate,鈥� he says. Although the experiment carried three times the volume of gas
required to fill the antenna, post-flight analysis shows that all the gas was
used up during the 90 minute flight. 鈥淔or me, the most disappointing thing was
that we had a leak somewhere,鈥� he says.
The antenna was never designed to make the return journey, so working out
exactly where any leaks occurred is bound to be difficult. But Freeland has some
theories. One of the most likely explanations is that the relatively violent
waving of the struts may have torn the material or even ripped loose the pipe
that supplied nitrogen to the discus. The expulsion of gas may be the reason the
antenna tumbled out of control.
The price is right
Despite his team鈥檚 failure to perform a controlled deployment or verify the
precision of the parabolic reflector, Freeland says the experiment was well
worthwhile in other ways. 鈥淭he basic objective was to show that we can build a
big inflatable structure with a small amount of money. On that basis it was a
success,鈥� he explains. Given the hundreds of millions of dollars it often costs
to build conventional satellites, $10 million spent testing a technology
that could make them much more powerful is good value for money.
The IAE experiment has left spacecraft engineers eager to press on with their
work. During a meeting last month at the Jet Propulsion Laboratory, they came up
with options for future missions that include inflatable arrays for radar
imaging devices, sunshields to protect large telescopes from variations in
temperature and large radio telescopes that could be linked with ground-based
arrays. The US National Oceanographic and Atmospheric Administration, based in
Boulder, Colorado, is even interested in inflatable solar sails that could push
spacecraft into orbits where they could study the Sun and the magnetic storms it
produces. Even the commercial world could benefit, using larger dishes to
improve the performance of telecommunications satellites.
Freeland believes he can iron out the problems that the IAE revealed, and is
already studying the physics of deployment in preparation for his next
mission鈥攅ven though no decision has been made about what it will be or
when it might fly. He has started work on a computer model to simulate what goes
on. 鈥淭he folding has a major effect on the deployment. If we can understand the
physics, we can perfect the packaging,鈥� he explains.
New materials may also help with reliability. Freeland says that certain
polymers that harden in sunlight could be used to create rigid, durable
structures. Other materials might be coated in some kind of solvent to keep them
pliant while they are folded, and prevent their springiness from causing the
kind of premature deployment suffered by the IAE. After deployment the solvent
would simply evaporate.
Freeland believes the most promising project for the next inflatable mission
is the 1.4 GHz Earth-looking radiometer that would map the planet鈥檚 surface
moisture content. He is confident that it will happen. 鈥淚f we can get funding,
it could be flying within five years,鈥� he says. He may not have long to wait.
NASA and the US Department of Defense, which is also interested in inflatables
for classifed space projects, expect to make a decision on what should fly next
within a year.