Genius at Work

Part 2: Genius at Work

The Story Behind Fixing Hubble's Vision

By Todd Neff


Continued from Part 1

NASA asked some of the nation’s top optical minds for ideas to salvage what was already in orbit. The Space Telescope Science Institute leading the Hubble astronomical effort formed the HST Strategy Panel, including famed astronomer Lyman Spitzer, astronaut Bruce McCandless and more than a dozen others. Ball’s Murk Bottema was among them.

The group vetted dozens of ideas. Scientists and engineers proposed wrenching the faulty mirror into the correct shape; sending an astronaut or, more likely, a robot spelunking into the school-bus-sized barrel of Hubble to replace the secondary mirror; recoating the edges of the flawed mirror in space; and bolting a massive optical corrector lens into the inlet of the telescope to reshape light before it entered.
Murk Bottema, the brilliant Ball optical engineer who was integral to correcting Hubble's hazy visionBottema, coming out of retirement to tackle the Hubble problem, had another idea. He suggested using relay mirrors similar to those planned for JPL’s next-generation main camera. The mirrors would pick off incoming light, remold it, and then bounce it to Hubble’s scientific instruments. Just how one might arrange a bunch of coin-size mirrors inside the orbiting telescope came from an inspiration in a German hotel shower.

While at a two-day Hubble Strategy Panel meeting in Garching, Germany in September 1990, James Crocker, a Space Telescope Science Institute engineer on the panel, was soaping up when he noted the adjustable, articulating arms holding the shower head in place. Inspiration struck. “I could see Murk Bottema’s mirrors on the shower head,” he later explained.

Back home, Crocker built a mockup using foam board, his son’s toy Ramagon construction set, and a few metal discs from the local hardware store. He presented his creation to colleagues at the strategy panel’s final meeting in Baltimore. The idea was to sacrifice one of Hubble’s four scientific instruments and replace it with a new device whose sole job was to fix the light entering the other three. Crocker was soon leading the team charged with creating such a thing. Ball Aerospace, as Ball Brothers Research was now known, started the work of designing and building it in January 1991.

Bottema went to work. The Dutchman’s depth in optics was akin to Pete Bartoe’s skill in mechanical design, involving a comprehension that plumbed the depths of the underlying physics. They shared an ability to visualize in three dimensions and then convert their mental imagery into mathematics.

Bottema scratched out pages of integrals and partial derivatives and Bessel functions, sketching light paths darting from mirror to mirror. He used computers, but mainly to confirm what his mind and hand had already established. Bottema was among the last of a breed. Modern optical engineers do some algebra to set boundaries, then pour their rough outlines into optical-design software with names like Code V, Zemax and Oslo, which churn out a range of shapes and surfaces and positions and suggest optimal configurations.

With computers, there is little need to understand optics the way Bottema did. He intuitively knew why a lens or mirror behaved in a certain way, how it related with all the other optical surfaces, and how it shaped the final image. A computer can arrange and optimize, but Bottema could observe a blurred image and calculate why it was blurred. Then he could work the problem forward again to figure out what sorts of optics would correct it.

Bottema spent his days on the Hubble problem, but his best hours were at night. After tea, reading the newspaper and a late dinner, he sat at the dining-room table with pencil and paper, writing equations. When things were going well, he whistled. When they went less well, he paced back and forth through the living and dining rooms. He talked through nettlesome technical details with his wife, who sewed on the couch under a window that looked out on the southerly portion of the Flatirons formation. She didn’t understand him, knew he was answering his own questions, and stitched away.

He designed a pair of mirrors on a telescoping arm. The first would reach out and snatch Hubble’s light before it could feed an instrument, redirecting it to a second mirror. The second mirror would be deformed in a way precisely opposite to the flaw of Hubble’s primary mirror but 200 times smaller. He solved the equations and showed that a few strategically placed mirrors could deliver corrected light to three Hubble instruments.

Ball's John Troeltzsch, asleep on top of COSTAR, spring 1993.This work provided the basis for the Corrective Optics Space Telescope Axial Replacement, or COSTAR, which a team of about 400 workers built at Ball Aerospace, in a crash 28-month program. From Crocker’s vision in a German shower and Bottema’s optical mastery evolved a $50 million masterpiece of space hardware. Built into a spare Hubble instrument box the size of a voting booth, the business end of COSTAR would reach out into the body of the telescope itself. The optics package squeezed ten mirrors, four telescoping arms, a dozen electric motors and various heaters, wiring and sensors into a four-foot-long triangular prism with a cross section the size of a slice of apple pie.    

NASA astronauts installing COSTAR during the first Hubble servicing missionIn December 1993, the space shuttle Endeavor made the first house call to Hubble. Astronauts replaced the main camera with JPL’s new one fitted with internal “eyeglasses” and sacrificed a high-speed photometer to make room for COSTAR. When its mirrors extended, it was as if COSTAR’s three neighboring instruments had undergone laser eye surgery. Aside from light lost along the way—an inevitable by-product of relay mirrors—the new JPL camera and COSTAR restored Hubble’s vision remarkably. Less than six months later, scientists announced a COSTAR-corrected Hubble instrument had brought home hard spectral evidence of a supermassive black hole at the heart of a galaxy 50 million light years away, confirming the existence of the long-theorized cosmic vacuum cleaners once and for all. Weiler considered it “the single biggest observation that COSTAR enabled. And it’s slightly important, because it changed something from Star Trek fantasy into scientific reality.”

Bottema had not lived to see the day. In July 1992, as his creation took shape, he succumbed to cancer. His Ball colleagues mounted a plaque in his memory on COSTAR. Bottema had had no doubt the instrument carrying it would work.

“You don’t have to worry,” he had assured his wife.

Excerpted from the book  From Jars to the Stars: How Ball Came to Build a Comet-Hunting Machine. © Todd Neff, published by Earthview Media. Used with the author’s permission.