If it were possible to label one day as "the worst" of spaceprobe Galileo's career, most scientists would choose Thursday, April 11, 1991.

At that time, the spacecraft was 18 months into a six-year trek to Jupiter and seemed to be running smoothly. Following a flawless launch by space shuttle Atlantis, it headed first for Venus, then back to Earth, picking up two vital gravitational slingshots to help it on its way. Galileo's fortunes, however, were about to change.

An artist's concept of Galileo launching from the shuttle. Credit: NASA/JPL

In what should have been a routine procedure that fateful Thursday, engineers instructed the spacecraft to unfold its high-gain antenna. Galileo would use this device at Jupiter to send data and pictures home at a rate about equal to a slow Internet-link. To ensure it did not get too hot during the Venus flyby, the antenna was kept safely folded for the first leg of the mission. Meanwhile, a smaller low-gain antenna was being used as Galileo's phone line home.

No one expected the nightmare that followed. The procedure, tested on Earth years before, called for a motor to drive 18 antenna 'ribs' out from a central mast, like opening an umbrella. When fully open, the antenna would snap into place and Galileo would report success to the ground. It didn't. As engineers watched the incoming telemetry, they noticed that, although electricity was flowing to the motor, it was not turning. The high-gain antenna - lynchpin of the mission - was stuck, partly open and useless.

Today, a full-size replica at NASA's Jet Propulsion Laboratory in California reveals the likely result. After examining the telemetry, engineers concluded three or four ribs remained stuck to the mast by their pins and stubbornly refused to budge. The cause can ultimately be traced back to another bad day in Galileo's career: Tuesday, January 28, 1986, when it was at Cape Canaveral awaiting a planned mid-May shuttle launch.

Galileo's date with destiny was abruptly cancelled at 11:40 a.m., when Challenger and her crew were lost in a horrific explosion shortly after lift-off. The inquiry into the tragedy inevitably refocused NASA on improving flight safety and one key issue applied to the liquid-fuelled Centaur rocket that would boost Galileo towards Jupiter. It was feared such a rocket was too hazardous to ride a manned Shuttle and a smaller, 'safer', solid-fuelled Inertial Upper Stage (IUS) replaced the Centaur.

New route
Although the deletion of the Centaur appeased the safety critics, it forced NASA to confront the reality that an IUS as not powerful enough to heave Galileo towards Jupiter. Instead, it was decided to send it on a more roundabout route through the inner Solar System. After launch, it would flyby Venus once and Earth twice to pick up sufficient gravitational impetus to reach Jupiter. The trip would be longer - six years, compared to just two with the Centaur - but the results would be worth it.

The astronauts' patch for shuttle mission STS-34 that deployed Galileo. Credit: NASA

In response to this more complex flight profile, Galileo had to be redesigned to accept new components. One of these was a sunshade to protect its sensitive instruments from anticipated high-temperature solar heating in Venus' neighbourhood. For three years, Galileo underwent these changes in California, before heading back to Florida, snugly packed on its side in a cross-country flatbed truck.

It was later found that this journey across the southern United States probably caused the high-gain antenna fault. As the truck bounced its way along thousands of kilometres of rutted, potholed freeways, pressure exerted on one side of the antenna somehow caused lubricant on a few rib pins to wear away. When Galileo reached Florida, this lubricant was left unchecked.

Losing the high-gain antenna was a severe blow, but engineers knew enough time remained to correct it before reaching Jupiter. A team was formed to identify the cause and assess NASA's options. Its findings were grim. Although telemetry showed most ribs had unfolded properly, the renegade ones would probably remain stuck. This did not, however, deter several fruitless attempts to free them over the next three years.

Efforts
Engineers tried revving the motor, spinning the spacecraft at up to 10 rpm to shake the ribs open and alternately freezing and cooking Galileo by rotating the antenna towards and away from the Sun. It was thought this might warm whatever lubricant remained just enough to free the stuck pins. One mad hat even suggested 'dipping' Galileo into Earth's atmosphere during its second flyby in the hope that air friction would do the job! Luckily, this idea was not taken seriously.

An artist's concept of Galileo shows the stuck antenna. Credit: NASA/JPL

During the 1980s, from its coast-to-coast road trips until the final weeks on its seaside launch pad, Galileo had been a focus of controversy as anti-nuclear groups debated the safety of its small onboard plutonium power unit. The wounds of Challenger were still raw and protesters pointed to the very real chance that another explosion could spew radioactive material across Florida.

To be fair, nuclear power was NASA's only realistic option when Galileo was planned in the 1970s. Ordinary chemical engines would be little use on such a long, arduous journey - over a billion kilometres - and even solar panels were hopelessly inadequate. Indeed, the energy available in sunlight at Jupiter is just 4% as much as in Earth orbit, far too small to represent an efficient power source.

Not only was Galileo saddled with the nuclear hot potato from its genesis, it was also designed with a specific launch vehicle in mind - the Shuttle - which, in the 1970s, seemed a problem-prone white elephant that wasted billions of taxpayers' dollars. The Shuttle's first flight, targeted for 1978, fell further behind schedule and would not leave the launch pad until 1981.

Most of the Galileo team had wanted their spacecraft to ride the expendable Titan rocket, a reliable workhorse with a proven track record. However, as NASA languished in the post-Apollo doldrums, the Shuttle offered a new future, promising to routinely ferry people and payloads into low-Earth orbit and build a permanent space station.

It seems extraordinary with the benefit of hindsight that NASA dropped so many of its eggs into the Shuttle basket, at one stage even considering halting its use of expendable rockets altogether in favour of the manned launcher, but the Galileo team were pushed to accept it. Unfortunately, as the Shuttle's first launch was delayed, Galileo's own ticket to space slipped from 1982 to 1984 and finally 1986. Then, just as the Shuttle began to prove itself, Challenger exploded and Galileo was grounded again.

Achievements
The spacecraft that reached Jupiter in December 1995 was thus battle-scarred and much changed from the mission sketched out 20 years earlier. Indeed, it was quite different from the Galileo sent aloft in 1989. The high-gain antenna failure forced NASA to extensively re p rogram its onboard computers to better handle and compress its data and pictures, beef-up the capabilities of ground tracking stations and relay Galileo's findings to Earth through the much slower low-gain antenna.

That victory was snatched from the jaws of defeat and the spacecraft went on to accomplish one of the most brilliant missions ever undertaken in the annals of planetary science all stand testament to the dedication and ingenuity of the scientists, technicians and engineers who formed the remarkable Galileo team.

Ben Evans is a schoolteacher and freelance astronomy and space exploration writer, based in Warwickshire.