In a stunning landscape of jagged limestone hills in southwestern China, engineers are putting the finishing touches on a grand astronomy facility: a half-kilometer-wide dish nestled in a natural depression that will gather radio signals from the cosmos. The world's largest radio telescope will catalog pulsars; probe gravitational waves, dark matter, and fast radio bursts; and listen for transmissions from alien civilizations.

PHOTO: XINHUA/ALAMY STOCK PHOTO

An otherworldly karst landscape made getting materials and equipment to the site fiendishly difficult. Far left, a worker on the giant truss encircling the dish. PHOTO: VCG/CONTRIBUTOR/GETTY IMAGES

Yet the architect of the tour de force is blasé about what his telescope might capture. "I'm really not very interested in science, I'm sorry to say," says Nan Rendong, chief scientist and chief engineer of the Five-hundred-meter Aperture Spherical radio Telescope (FAST) here. Colleagues insist he is joking, but there is no question that what has consumed 2 decades of his life—and is now wowing other astronomers—is engineering. "As a civil engineering feat, FAST is obviously amazing," says Fred Lo, former director of the U.S. National Radio Astronomy Observatory (NRAO) in Charlottesville, Virginia.

It is not just FAST's sheer size—it has more than twice the collecting area of the runner-up, the 305-meter dish in Arecibo, Puerto Rico. FAST is also breaking new ground in radio astronomy with a design that pulls a section of the spherical dish into a gradually moving paraboloid to aim at and track cosmic objects as Earth rotates, bringing the benefits of a tilting, turning antenna to a fixed dish. This innovation "is absolutely unique, nobody has ever done this before," says Dick Manchester, a radio astronomer at Australia's Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Sydney.

There's another reason that Nan isn't worrying about science just yet. Although dignitaries celebrated FAST's completion at a 25 September ceremony here in Guizhou province, the telescope isn't yet working as planned. Nan is focused on solving a recently uncovered technical glitch that's hindering the dish-shaping capability and limiting the early science. To get FAST to live up to its scientific promise, he says, "we have a lot to do."

"There is no question, Nan is the main driving force for FAST, from beginning to end," Lo says. Nan's engineering bent dates from his undergraduate days studying ultrahigh-frequency electronics at Tsinghua University in Beijing in the mid-1960s; during China's tumultuous Cultural Revolution he spent a decade working at an electronics factory. He then returned to academia, earning a Ph.D. in astronomy and astrophysics from the University of Science and Technology of China in Hefei.

FAST's long gestation began in the 1990s, when Nan served on China's delegation to the international working group that eventually proposed the Square Kilometre Array (SKA) as a next-generation radio telescope. Astronomers were counting on advances in interferometry to combine radio waves from dozens or even hundreds of dishes, thereby creating a collecting area far bigger than any existing telescope. In the early days of planning, China vied to host the SKA, proposing to build several large dishes in the limestone depressions that dimple its southwestern provinces. Chinese astronomers even did preliminary work on FAST as a prototype.

Instead, the SKA's backers opted for a design featuring thousands of small dishes. China was dropped from the list of candidate sites in 2006; construction of the first phase of the SKA, in South Africa and Australia, is expected to begin in 2018. Swallowing their disappointment, Nan and his colleagues pushed China to build FAST anyway (Science, 19 June 2009, p. 1508).

Single dishes excel at observing point sources like neutron stars and at scanning a multitude of frequencies in the search for extraterrestrial intelligence, says astronomer Li Di, a FAST project scientist, who previously worked at NASA's Jet Propulsion Laboratory in Pasadena, California. Another advantage is that, compared with the multiple dishes in an array, single dishes are "relatively cheap and relatively straightforward to upgrade," says George Hobbs, an astronomer at CSIRO. "You just keep building better receivers."

After an international review panel endorsed FAST, China's National Development and Reform Commission in 2007 provided $90 million to launch the project; additional support from several agencies topped up the funding to $180 million. Nan had led much of the preliminary work on FAST, and in 2008 he was anointed chief scientist and chief engineer. "Traditionally, that's two jobs," Li says, but Nan's practical experience and scientific training, he says, allowed him to speak to funding agencies “with both his engineering and science hats on.”

Nan already knew where to build the dish—and he knew it wouldn't be easy. In the 1990s, satellite surveys had identified 400 candidate depressions in China's southwest provinces. That number was winnowed down until Nan personally inspected a few of them. He recalls hiking into the Dawodang depression and looking up to see a nearly circular view of the sky that “was like being at the bottom of a well.”

Radio astronomy writ largeCREDIT: C. BICKEL/SCIENCE

"The site is the best for FAST, [but is] the worst for construction," Nan says. It is hours from the nearest highway, down bone-jarring, narrow roads that wind through towering pinnacles and tiny villages. The FAST team built a 7-kilometer road linking the depression to the nearest town. The difficult access curtailed use of heavy equipment. "Almost everything had to be done manually," Nan says. "Very devoted laborers," he says, shouldered 100-kilogram loads in temperatures topping 40°C. By comparison, he says, "Building a bridge in Beijing or Shanghai is a piece of cake."

The team modeled FAST after the Arecibo telescope. Both are built in karst depressions and supported on steel cable meshes slung like hammocks from supports fixed to the limestone pinnacles. And like the Arecibo Observatory, FAST takes an unusual approach to focusing incoming radio waves.

Radio telescope dishes are typically parabolic because that shape focuses waves from astronomical objects in line with the parabola's axis to a point above the dish. The telescope's receivers—or a subreflector—are positioned at that point. But a parabolic telescope must be steerable and able to point at astronomical objects and track them as Earth rotates, because parabolic reflectors distort waves from off-axis targets.

An astronomer with an engineering bent, Nan Rendong led much of the early work on the radio telescope and is its chief scientist and chief engineer. CREDIT: NATIONAL ASTRONOMICAL OBSERVATORIES, CHINESE ACADEMY OF SCIENCES

It's impossible to steer Arecibo and FAST because their enormous dishes are anchored to the ground. So both have a spherical shape, allowing them to collect and concentrate waves from off-axis sources without focusing on a point. Arecibo aims at cosmic objects by shifting the position of the receiver platform to catch the reflected waves. Within the platform, a complex mirror system brings them into focus. But that limits the slice of observable sky to about 20° from zenith; farther from the zenith, the distortion is too great. The correction system also results in a platform weighing about 900 tons. "That mirror system and the whole platform are very big for Arecibo, and it would have been huge for FAST—and heavy," Manchester says.

Instead, Nan and his team designed a system that pulls a roughly 300-meter-diameter section of FAST's spherical reflector into a subtle parabola, while positioning receivers along the parabola's axis (see graphic, above). "It's like forming a smaller bowl within a big wok," Li says. The position of the parabola can be shifted in real time, so that the parabolic axis always aims at a cosmic object of interest as Earth rotates, just as a steerable radio telescope does. FAST can observe up to 40° from zenith. And because it does not need a complex corrective mechanism, its receiver platform can host more instruments than Arecibo's.

Making an active surface reflector "was a bold move," Manchester says. To deform the reflector, FAST has 2225 actuators, essentially high-tech winches, anchored into rock beneath the dish. The actuators tug the dish into a parabola by pulling on tie-down cables connected to the dish's supporting mesh. The mesh's natural springiness restores the dish's spherical shape when the actuators relax the tension. Lasers mounted on small posts protruding through the dish check the coordinates of 1000 points on its surface, allowing fine-tuning of the shape.

To make this system work, the engineers solved a host of challenges. For starters, the actuators emit radio interference that is "many times stronger than the signal from the sky," Nan says. There was no suitable commercially available shielding, so they developed their own.

Another problem was caught during the final design review before construction, when engineers belatedly realized that repeatedly stressing and relaxing ordinary steel cables could lead to fatigue failure, the phenomenon familiar to anyone who has bent a paper clip back and forth until it snaps. The FAST team solved this problem by turning to a Chinese-developed cable that's fatigue-resistant for up to 2 million stress cycles, far more than the 300,000 cycles the FAST cables will endure over the telescope's 30-year design life.

IN JANUARY, with most of the dish's 4450 triangular reflector panels in place but none of its receivers available, Nan had his crew rig up something resembling a spindly fishbone TV antenna and suspend it over the dish. As radio receivers go it was primitive, but FAST's enormous collecting area enabled it to pick up signals from the Crab Pulsar, a radio source at the heart of the Crab Nebula. “It was amazing that they could do this with a simple antenna,” CSIRO's Hobbs says.

For Nan, the test was FAST's first light, the initial scientific observation that marks the start of a telescope's working life. In focusing on the Crab Nebula, Nan turned one of China's earliest astronomical observations into one of its latest: A millennium ago, Song dynasty observers noted a transitory "guest star," later pegged as the supernova that created the nebula.

But as the dish's construction entered the home stretch, another glitch cropped up. "The actuators are breaking down at a higher rate than anticipated," Li says. The team is investigating the cause and possible fixes.

Getting big telescopes working at full potential is always a challenge, says John Ford, formerly in charge of electronics at NRAO's Green Bank observatory in West Virginia. On Green Bank's 100-meter steerable dish, actuators used to counter gravitational effects were "failing on us way more than we thought they would," he says, because water was leaking in and freezing. Waterproofing the actuators was a headache because they are 100 meters in the air. FAST engineers should have an easier time because their actuators are accessible from the ground, Ford says.

For now, Li says FAST operators will use their working actuators to hold part of the reflector in a parabola, point it at the sky, and simply catch whatever signals they can as Earth rotates. It will take 200 days of such drift scan observations to survey the full northern sky, and he expects they will discover as many as 1000 pulsars, adding to the roughly 2500 now known. Astronomers use these radio beacons, powered by spinning neutron stars, to study the interstellar medium and detect gravitational waves. "We will conduct the world's best survey for pulsars," Li says.

"Eventually we will get the telescope working perfectly," vows Zhu Ming, a FAST project astronomer. Manchester agrees: "They're very close to achieving an absolutely remarkable feat." (Science)