The Hubble Space Telescope, to which we owe the most extraordinary photographs of the Universe, has taken images again after spending more than a month in blind due to a computer error. The veteran observatory resumed its scientific operations last Saturday, just two days after being subjected to a complex and risky repair from the ground, 547 kilometers away. A faulty power regulator was ‘unplugged’ and replaced with a backup copy.
Hubble’s payload computer, which controls and coordinates the science instruments aboard the observatory, came to a sudden stop on June 13. When the main computer could not receive a signal from the payload equipment, it automatically placed the science instruments in safe mode. That meant the telescope couldn’t do science.
Engineers at NASA’s Goddard Space Flight Center in Greenbelt, Maryland, began to investigate what the problem was. It was not easy. Hubble was built in the 1980s and has been in space for more than 31 years, so it needed someone who knew its ins and outs well. Retired engineers who helped build the observatory returned to help out. Documents from thirty or forty years ago were even used.
At first, the team thought the most likely problem was a degraded memory module, but switching to backup modules did not solve the problem. Then other bugs were looked for and tests were run that involved turning on Hubble’s backup payload computer for the first time in space, but it didn’t work either. So the engineers went on to explore whether other hardware was to blame, including the science data command unit and the power control unit, which is designed to ensure a constant voltage supply to the payload computer hardware. However, this meant taking a greater risk to the telescope and modifying other components.
The change required 15 hours. “The main computer had to be shut down, and a backup computer in safe mode temporarily took over the spacecraft. Several boxes that had never been turned on before in space also had to be powered up, and other hardware needed their interfaces to be changed, ”says Jim Jeletic, Hubble’s deputy project manager at Goddard. “There was no reason to believe that all of this would not work, but it is the team’s job to be nervous and think about everything that could go wrong and how we could make up for it. The team planned and meticulously tested every little step on the ground to make sure they got it right, ”he adds.
Over the next two weeks, more than 50 people worked to review, update, and examine procedures for switching to backup hardware. Simultaneously, previous test data was analyzed and the findings pinpointed the power control unit as the possible cause of the problem. On July 15, the engineers made the planned change successfully.
The science instruments were put into operation and Hubble returned to work two days later. Among its targets, an unusual galaxy with outstretched spiral arms and the first high-resolution look at an intriguing pair of colliding galaxies. These new observations will continue to change our understanding of the universe.