PASADENA, Calif. (AP) - A $75 million NASA spacecraft designed to study solar flares was heavily damaged when engineers mistakenly shook it 10 times harder than intended during a preflight test.
The shaking cracked at least two of four solar panels on the High Energy Solar Spectroscopic Imager, and tests were planned to find internal damage. Launch, which had been scheduled for July, will be pushed back at least to January.
It's the latest embarrassment for the space agency and Jet Propulsion Laboratory, which last year lost two high-profile missions to Mars. The cause of the mishap was still under investigation, JPL spokeswoman Mary Beth Murrill said Thursday.
JPL engineers were performing tests on a shake table Tuesday to ensure the probe could withstand twice the force of gravity, which it would experience during launch. Instead, it was subjected to 20 times the force of gravity for about 200 milliseconds.
''The folks who were involved in the test are mystified at this point,'' said Larry Dumas, JPL's deputy director. ''There's no obvious reason that's presented itself (for the unexpected shaking).''
The 850-pound HESSI probe is designed to explore the basic physics of particle acceleration and the energy release of solar flares from an orbit of 360 miles above Earth. It's not clear whether any of the scientific instruments were damaged.
HESSI's engineers were confident that the spacecraft can be saved, said Mark Hess, spokesman for NASA's Goddard Space Flight Center, which is managing the mission.
''It continued to function even through the test,'' he said. ''We know there are structural and other elements of the satellite that are still working.''
NASA will appoint a review board to investigate the mishap - just as it did twice last year after the Mars Climate Orbiter and Mars Polar Lander losses.
Investigators later found Climate Orbiter mission failed because critical navigation units were not translated into metric. Polar Lander's review board is expected to make its findings public next week.
HESSI's mishap disappointed dozens of scientists who have been working on the project for more than two years.
The satellite was being tested at JPL because of the lab's proximity to Gilbert, Ariz.-based Spectrum Astro, where it was built, and the University of California, Berkeley's Space Sciences Laboratory, where its primary science team is based.
''To have it damaged in a test at JPL is a bitter blow for us,'' Dumas said. ''And I'm sure even more difficult for the folks at Goddard, Berkeley and Spectrum Astro who have worked so hard on this mission.''