by Michael Koller (

On Oct. 18, 1989, Space Shuttle Atlantis headed into space with the National Aeronautics and Space Administration’s Galileo spacecraft in its cargo bay. Hours later, the spacecraft was deployed in space and sent on a five year odyssey that would take the spacecraft past Venus once and Earth twice in route to study the giant planet Jupiter, its moons, rings and other phenomena for at least two years.

The Galileo spacecraft performed well during its encounter with Venus and its first encounter with Earth. It’s various instruments provided plenty of new information about the structure of Venus’ lower atmosphere, the ozone hole over Antarctica, and mapped the far side of the Moon in high resolution for the first time.

On April 11, 1991, just months after Galileo’s first pass by Earth, NASA controllers sent commands to Galileo that would open the spacecraft’s 4.8 meter diameter high-gain antenna. The device, which transmits science data back to Earth at rates up to 134 kilobits per second, would be needed for the remainder of the spacecraft’s journey and mission at Jupiter.

The spacecraft executed stored computer commands designed to unfurl the antenna, but something was wrong: The antenna had failed to open. For the hundreds of scientists and their graduate student assistants scattered across the United States, Canada, and Germany, news of the antenna failure was their worst nightmare come true.
“I didn’t believe it,” says Dr. Michael J.S. Belton, leader of the team responsible for the operation of Galileo’s cameras and an astronomer for Tucson’s branch of the National Optical Astronomy Observatory “I went over to the project manager and asked what we could do. He said, ‘Pray.’ “

A “tiger team” of over a 100 technical experts was assembled at NASA’s Jet Propulsion Laboratory in Pasadena CA, which built and operates the spacecraft, to diagnose the problem.

“The motors worked fine,” says UA scientist Alfred S. McEwen, an interdisciplinary scientist with the Galileo mission, “The antenna just didn’t open.”
The antenna is useless unless it is unfurled in a perfect parabolic shape. Extensive analysis of the spacecraft’s telemetry, along with ground tests on a spare antenna, suggested that the pins on three of the antenna’s 18 ribs were stuck to the antenna mast; the result of friction between the standoff pins and their sockets. Further analysis of data from Galileo’s sunlight detector revealed “shadows” of the hung antenna ribs, an indication that the device had only partially deployed.

“Because of all the delays, the spacecraft was transported cross country several times,” said Dr. Belton. “During that transport, certain parts of the antenna were warped.” A dry lubricant was applied to the pins during the antenna’s manufacture. Apparently, this lubricant was knocked off during the spacecraft’s overland transport back and forth from JPL and Cape Canaveral. The problem went undetected during pre-launch checkout and went unnoticed until the attempt to deploy the antenna was made. “This was very frustrating because we could fix the problem with our little fingers,” says Dr. McEwen.

“That is just one more reason why the space shuttle program was not good for Galileo. Part of the reason we had a collapsible antenna to begin with is that Galileo had to fit in the shuttle’s cargo bay,” he said with a hint of anger. “Everyone knows that the more deployable parts you have the greater the chance of having a failure.”
Originally approved by Congress back in 1977 for a 1982 launch onboard the Space Shuttle, the Galileo project was held hostage by delays in the development of the space shuttle program. Indecision over what launch vehicle to use to move Galileo from Earth orbit, and finally, the 1986 Challenger disaster resulted in further postponing of the launch of the spacecraft. Initially, Galileo was to have cost only around $600 million. By 1989, the costs of maintaining the spacecraft had topped $1.5 billion.

Yet in NASA’s Jet Propulsion Lab in 1991, while the spacecraft sailed through space, the scientists and engineers were focused only on the antenna’s stuck ribs. They tried turning the spacecraft’s antenna in and out of shadow numerous times in the hope that thermal expansion and contraction would “walk” the pins of the their sockets. This was a no go. Another technique, called “hammering”, involved turning on and off the antenna’s deployment motors to create additional forces that would move the pins and free the ribs. Engineering readings showed that the additional force had been generated, but the ribs remained stuck. Still more methods were tried, but were ultimately unsuccessful.

“Unfortunately, the forces involved in shaking, hammering, etc.. were just too small to do much good,” Belton said. While the engineers sat brainstorming, Galileo continued its science observations, using its low-gain antenna, which transmits data at a painfully slow 10 bits per second. The craft’s historic first encounter with the asteroid Gaspra went on without a hitch, as most of the data was recorded on the spacecraft’s tape recorder and played back when Galileo passed by Earth in 1992 for the second and final time.

Meanwhile, Galileo’s complement of scientists and engineers were beginning to realize that it was likely that the high-gain antenna’s ribs would never come unstuck. This pointed to severe consequences. At the worst, it meant that all the science objectives would have to be abandoned except for the data gathered from Galileo’s atmospheric probe, which would be dropped into Jupiter’s turbulent clouds.

“We went through so many stages. At first NASA headquarters wanted us to retrieve the probe data and then turn the spacecraft off, “ said McEwen, shaking his head. The project scientists and engineers, many whom have devoted nearly their entire lives to this project, were not going to give up that easily.

Galileo sailed by the asteroid Ida in the summer of 1993 during its transit through the asteroid belt. The spacecraft used improvised data compression techniques that squeezed more data into single bits to relay a limited number of pictures and other data back to Earth for study. But Galileo would need to compress even more data in order to fulfill even a portion of the rest of its science objectives. It fell on the computer programmers to come up with a scheme that would work. It was a daunting task. “Galileo is an extremely archaic spacecraft, “ McEwen explains, “The spacecraft’s’ main computer was a radiation hardened model from the 1970s. Actually, its design was even earlier than that.

“It has an extremely dumb computer, “McEwen laughs as he leans back in his chair. “I don’t think you could find a person today with a notebook computer with such little memory.” “Galileo already had the data compression codes in it, “explains Dr. Belton. Those codes weren’t enough. They could give you maybe 2.5 times the compression, but we were looking for a factor of 10.” While JPL’s engineers continued to grapple with the spacecraft, NASA in turn beefed up the communications antennas of its Deep Space Network. NASA uses those antennas, positioned around the world, to communicate with its planetary probes. This created even bigger “listening ear” with which to receive Galileo’s feeble signals.

By 1995, NASA had given up on freeing the antenna and instead started loading the first phase of the new computer software into Galileo’s main computer. Still, even with the new software, Galileo’s overall mission had to be drastically de-scoped. Instruments that generated megabits of data, such as the cameras or the Near Infrared Mapping Spectrometer, were forced to reduce the number of observations. Galileo was to return over 500,000 photos. Now it would only return, at most, 2000 pictures.
This had a severe impact on some experiments. For example, meteorologists who wanted to study the circulation of Jupiter’s atmosphere had planned to use the photos to create movies of the atmosphere. Since many of the photos were needed for exploring the surface geology of Jupiter’s four largest moons, scientists could only produced short two-frame movies.

All the various science teams would have to make due.

“The loss of the antenna forced us to really plan ahead,” says McEwen, “We had to plan our various observations very carefully, taking the most important objectives first.”
On Dec. 7, 1995, Galileo was successfully inserted into orbit around Jupiter. The deployment of the atmospheric probe operated perfectly. For the next 23 months, the spacecraft weaved 11 orbits around Jupiter, each time passing by one of its large moons, snapping pictures and collecting other data.

Among the many surprising discoveries made include that Jupiter’s moon Ganymede, has its own magnetic field. Galileo’s cameras revealed what look like glacial flows and broken ice sheets that resemble the Ross Ice Shelf in Antarctica, a discovery that suggests the possibility of a warm, subterranean ocean underneigth the visible ice crust. Galileo’s atmospheric probe found that Jupiter’s atmosphere was more water rich than anyone had expected. And Galileo also monitored volcanic eruptions on Jupiter’s innermost large moon, Io. This last discovery intrigues McEwen, who is a planetary geologist.

“What we saw on Io was widespread high-temperature volcanism.” McEwen says, “I just published a paper to the journal Science on the topic. We are now convinced that there are lavas being erupted on Io that are hotter than any erupting lavas here on Earth.”

Presently, Galileo is in the midst of its extended mission at Jupiter. Also called the Galileo Europa Mission,, or GEM for short, the spacecraft will turn its instruments on Europa to try and find whether or not an ocean exists under the ice.

For Dr. Belton, whose team assembles 4-5 page computer sequence plans for Galileo’s camera system, the success of the mission is especially sweet.

"This really was a miracle. The last of the flight software was sent up during Galileo’s first orbit around Jupiter. During that time, we wrote, configured, moved in the new software and removed the old, “ Belton says from his desk, shaking his head in wonder, “You know, spacecraft have been lost in the past because people fooled around with the flight software. As a rule, that is just not done.If there ever was an engineering triumph with this mission, then that was it.”

“It’s been a roller coaster, “ McEwen reflects with a bit or mirth. “But it’s been a great journey overall. The Jupiter science really has exceeded my own expectations.”