Repairing Hubble

Posted on Wed, April 23 2014

Soon after the Hubble Space Telescope was launched in 1990, images and data from its instruments revealed that its main mirror was optically flawed. It suffered from spherical aberration—not all portions of the mirror focused to the same point. The mirror’s shape was off by less than 1/50th the thickness of a human hair, but this tiny flaw proved devastating to the quality of the Hubble’s images and to the efficiency of all of its instruments.

Deployment of Hubble Space Telescope
This photograph was taken by the STS-31 crew aboard the Space Shuttle Discovery and shows the Hubble Space Telescope being deployed on April 25, 1990, from the payload bay. Credit: NASA.


This was a serious, but not fatal flaw.  If the Hubble was like all other astronomical instruments lofted into orbit on rockets, it would have had to live out its operational life with that flaw, working at a fraction of peak efficiency.  But Hubble was not like any other space telescope.  It was designed to be serviced by astronauts visiting it on the space shuttle.  That’s one reason why it was placed in a low earth orbit accessible by the shuttle.

\"Repairing Hubble"
Front and center is the Wide Field Planetary Camera 2 (WFPC2) tilted on a wedge to reveal its inner components. To the left is COSTAR and behind COSTAR is a developmental model built at the Ball Aerospace Corporation to demonstrate how COSTAR employed mirrors on stalks that could be inserted into the primary light beam from Hubble's flawed mirror, to correct the flaw. On the wall at the back are classic images from Hubble's cameras, and the Structural Dynamic Test Vehicle used to test the Hubble design is seen at the upper right. Credit: Eric Long, National Air and Space Museum.


The question now became, how could corrections be made? One option involved bringing it back to Earth and replacing the mirror with a backup (now on view in our Museum, in the Explore the Universe gallery).  But NASA, encouraged by the expertise at the Space Telescope Science Institute in Baltimore, and the Ball Aerospace Corporation in Boulder, Colorado, chose a different approach. One instrument, the Wide-Field/Planetary Camera (WF/PC), already had an upgraded replacement available. Its engineering and science team at NASA’s Jet Propulsion Laboratory knew how to adjust the optics within WFPC2 to compensate for the aberration in the primary mirror. For the other instruments, engineers created an optical box called COSTAR (Corrective Optics Space Telescope Axial Replacement). It contained a set of five pairs of small mirrors on deployable arms that corrected the light beams entering the Hubble’s Faint Object Camera, Faint Object Spectrograph, and Goddard High Resolution Spectrograph.  Fitted within a standard axial instrument enclosure, the small mirrors would deploy after launch and checkout, enter the reflected optical beam from the main mirror, and counteract its flaw, sending the corrected light to the other instruments.

Detail of the deployed mirrors in COSTAR, protected by a plex vitrine. The WPFC2 radiator can be seen in the background. Credit: Eric Long, National Air and Space Museum.


In the foreground: the developmental model for COSTAR showing how the mirrors could be deployed on hinged arms. The color coding is for each separate instrument. Credit: Eric Long, National Air and Space Museum


COSTAR contained 10 optical elements, 12 motors, and over 5,000 individual parts. After being installed in the Hubble, each of its five optical channels had to be precisely aligned.  In the end, COSTAR’s performance exceeded the original specifications.  Given its complexity, the real challenge was to make it strong enough to withstand launch, and yet delicate enough to insert tiny mirrors into the Hubble’s optical field without disturbing any of the other components.  A Ball Aerospace engineer came up with the solution while taking a shower in a German hotel, which was equipped with ingenious articulated shower heads.

Hubble \"Witness Sample"
A "witness sample" from the moment in the early 1980s when the Hubble mirror was vacuum electrocoated with a reflective layer of aluminum and overcoated with a protective layer of transparent magnesium fluoride. Even with this protection is was standard practice to include small samples with the main mirror. The small samples could be tested from time to time to keep track of the coating on the main mirror. This sample was flown into space on the deployment mission as a commemorative gesture and was presented to the popular magazine Sky & Telescope. Rick Feinberg, former editor of S&T, donated it for the exhibit. Credit: Eric Long, National Air and Space Museum.


After several more servicing missions through the 1990s, all the new instruments onboard Hubble had their own corrections for the flaw in the main mirror. Therefore COSTAR was no longer needed, and, given the rapid advance of solid state detector technologies through the decade, WFPC2 was no longer state of the art.   NASA therefore planned another servicing mission to replace them with new more powerful cameras and detectors.  But the shock of the Space Shuttle Columbia accident in February 2003 was deeply felt worldwide, making NASA cautious about flights that did not go to the International Space Station. Therefore, in 2004 NASA cancelled Hubble’s fourth servicing mission. Without it, the telescope’s life was projected to end by 2007. The decision incited uproar from scientists, the public, and Congress. Twenty-six former astronauts signed a petition in favor of keeping the Hubble alive. The fifth and final Hubble servicing mission took place in May 2009 and was the most complex and demanding yet. During five spacewalks, Atlantis astronauts installed two new instruments, repaired two others, and performed extensive maintenance. They removed COSTAR and WFPC2 and installed the new Wide Field Camera 3 (WFC3), which included greatly upgraded CCDs and some important reusable hardware from the original WF/PC.

The radiator section (rear end) of WFPC2. During the various servicing missions astronauts noticed tiny dimples and dents in the radiator - results of impacts from space debris. After over 15 years of exposure to space, this surface became a record of the accumulation of such debris in low earth orbit. Naturally, NASA wanted to evaluate the amount and nature of this debris, and so after the camera was returned to earth the impact sites were cored out for analysis. The core samples are far larger than the original debris, because the splatter pattern of these impacts was far larger than the original debris particles themselves. The largest core samples left holes about 30 mm in diameter, but the debris particles were less than a mm in size. The analysis is ongoing. Credit: Eric Long, National Air and Space Museum.


Astronauts brought the two old instruments back to Earth and they were soon shipped to NASA’s Goddard Space Flight Center in Greenbelt, Maryland. Technicians at Goddard and then at the Johnson Space Center examined WFPC2 for effects from prolonged exposure to space. Its radiator, the curved white section that formed part of the Hubble’s outer skin, absorbed more than 15 years’ worth of impacts by micrometeoroids and orbital space debris. Scientists measured the chemical composition of these small impactors to help shed light on the nature of space debris, a danger that affects all space missions. In order to make the analysis, NASA had to core out all the impacts, cutting holes far larger than the debris itself.  That’s why there are so many large holes in the image of the radiator above.