Hubble Space Telescope

• 1. Introduction

  ›

  The Hubble Space Telescope launched into low Earth orbit on the 24th of April 1990, aboard the Space Shuttle Discovery as part of STS-31. Within weeks, scientists realized something had gone terribly wrong. The images coming back were blurry. A mirror polished to an accuracy of 10 nanometers had been ground to precisely the wrong shape. Comedians made jokes. Politicians questioned NASA's competence. For a brief moment, the most expensive science mission in NASA history appeared to be a very costly failure.

  How did one of the most precisely crafted optical instruments ever made end up wrong? Who spent decades fighting to get it built in the first place? And what happened once the problem was fixed? Those are the questions this documentary sets out to answer.

• 2. The Dreamers Behind the Idea

  ›

  Lyman Spitzer published a paper in 1946 titled "Astronomical advantages of an extraterrestrial observatory," and in doing so planted the seed that would eventually become Hubble. Spitzer laid out two arguments. First, a telescope above Earth's atmosphere would not suffer from the turbulence that causes stars to appear to twinkle, a phenomenon astronomers call seeing. Ground-based telescopes at the time were limited to resolutions of 0.5 to 1.0 arcseconds, whereas a mirror 2.5 m in diameter could theoretically achieve about 0.05 arcseconds. Second, a space telescope could observe ultraviolet and infrared light, which the atmosphere blocks almost completely.

  Spitzer devoted much of his subsequent career to advocating for this idea. In 1965, he was appointed head of a committee charged with defining scientific objectives for a large space telescope. But arguably as important to Hubble's existence was Nancy Grace Roman, who became known as the "Mother of Hubble." Well before the project was formally approved, she delivered public lectures making the case for the telescope's scientific value. After approval came, she became the program scientist, writing congressional testimony throughout the 1970s to keep funding alive.

  Even further back, in 1923, Hermann Oberth had mentioned in his book "Die Rakete zu den Planetenraeumen" that a telescope could be propelled into orbit by a rocket. By 1962, the U.S. National Academy of Sciences had formally recommended developing a space telescope as part of the broader space program. The vision had been building for decades before a single mirror was polished.

• 3. A Fight for Funding

  ›

  NASA's ambitions for what was then called the Large Space Telescope called for a mirror 3 m in diameter, with a target launch date of 1979. Then came the budget wars. In 1974, Congress eliminated all funding for the project. The telescope seemed dead.

  Noel Hinners, then NASA's Associate Administrator for Space Science, made a gamble. Rather than accept a token $5 million placeholder that NASA Administrator James C. Fletcher had proposed, Hinners pushed to zero out funding entirely. His reasoning: a small sum would reassure the astronomy community that everything was fine, when in fact it was not. A complete cut, he calculated, would energize scientists to fight. As Hinners later recalled: "I figured in my own little head that to get that community energized we'd be better off zeroing it out. Then they would say, 'Whoa, we're in deep trouble', and it would marshal the troops."

  The strategy worked. Astronomers organized a nationwide lobbying campaign, meeting senators and representatives in person, mounting letter-writing drives, and enlisting the National Academy of Sciences to publish a report stressing the telescope's importance. The Senate agreed to fund half the original budget. The mirror diameter was reduced from 3 m to 2.4 m to cut costs and allow for a more compact configuration. A proposed 1.5 m precursor telescope was dropped. The European Space Agency was brought in as a partner, contributing instruments, solar cells, and staff, in exchange for European astronomers receiving at least 15 percent of observing time. Congress ultimately approved $36 million for 1978, and serious design work began with a target launch date of 1983.

  By the time the telescope finally flew in 1990, NASA had spent approximately $4.7 billion in inflation-adjusted 2010 dollars.

• 4. Building the Mirror

  ›

  Perkin-Elmer was commissioned to build the optical telescope assembly, including the primary mirror. Construction of that mirror began in 1979, starting from a blank manufactured by Corning from ultra-low expansion glass. To minimize weight, the mirror was built with top and bottom plates, each 25 mm thick, enclosing a honeycomb lattice. Perkin-Elmer simulated the weightlessness of space by supporting the mirror from behind with 130 rods applying carefully calibrated amounts of force. Mirror polishing continued until May 1981, and by the end of that year the mirror was complete. It was washed with 9,100 liters of hot, deionized water and given a reflective aluminum coating 65 nanometers thick, plus a protective magnesium fluoride layer 25 nanometers thick.

  NASA had demanded that Perkin-Elmer subcontract a backup mirror to Kodak using traditional techniques, just in case the custom polishing machinery ran into trouble. That backup mirror now resides permanently on display at the National Air and Space Museum.

  Throughout construction, NASA reports raised concerns about Perkin-Elmer's management. Schedules slipped at roughly one month per quarter. At certain points, delays accumulated at a rate of one day for every day of work completed. The launch date was pushed from 1983 to 1984, then to 1985, then to March 1986, then September 1986. By then the total project budget had reached $1.175 billion. The Challenger disaster in January 1986 halted all shuttle flights and pushed the launch back further still, though it also gave engineers time to perform additional tests and make improvements. The delay came at a cost of roughly $6 million per month just to maintain the telescope in storage.

• 5. The Flaw and the Fix

  ›

  Within weeks of the April 1990 launch, it became clear that something was seriously wrong. The primary mirror had been polished to extraordinary precision but to the wrong shape. Its outer edge was too flat by about 2,200 nanometers, roughly a fiftieth of a millimeter. This introduced severe spherical aberration: light reflecting off the mirror's edge focused at a different point than light reflecting from the center. Images of point sources spread across a radius of more than one arcsecond instead of the specified 0.1 arcseconds.

  A commission led by Lew Allen, director of the Jet Propulsion Laboratory, traced the problem to a reflective null corrector, the testing device used during the final polishing stage. One lens in that device was out of position by 1.3 mm. During earlier stages of fabrication, two conventional null correctors had correctly reported spherical aberration, but those results were dismissed because the reflective null corrector was considered more authoritative. The Allen Commission placed primary blame on Perkin-Elmer's management failures: the company had not assigned its best optical scientists to the project and had not involved optical designers in mirror verification.

  Astronomers working backward from the distorted images calculated that the mirror's conic constant was -1.01390 rather than the intended -1.00230. That precise characterization made a solution possible. Since replacing the mirror in orbit was not feasible, engineers designed corrective optics with exactly the inverse error, effectively a set of spectacles. The Wide Field and Planetary Camera 2, already planned as a replacement, incorporated corrective relay mirrors built into its design. For the other instruments, a device called the Corrective Optics Space Telescope Axial Replacement, or COSTAR, was designed to intercept light before it reached the FOC, FOS, and GHRS instruments. To make room for COSTAR, one instrument had to be sacrificed: the High Speed Photometer.

  Servicing Mission 1, flown by Endeavour in December 1993, carried out the repair over ten days. Story Musgrave, who had been working on satellite repair procedures since 1976, led a crew of seven experienced astronauts. Training had been unusually rigorous: seven full mission simulations were conducted, the most thorough preparation in shuttle history at that point. Musgrave had discovered during vacuum training, seven months before launch, that spacesuit gloves did not adequately protect against the cold of space. Equipment and procedures were quickly changed after a later shuttle flight confirmed the issue in orbit. On the 13th of January 1994, NASA declared the mission a complete success and released the first sharp images. Hubble's estimated total cumulative cost, including all subsequent servicing, reached about $11.3 billion in 2015 dollars.

• 6. Five Missions to Keep It Alive

  ›

  Hubble was specifically designed to be serviced and upgraded in orbit, and five Space Shuttle missions carried out that work between 1993 and 2009. Servicing Mission 2, flown by Discovery in February 1997, replaced two spectrographs with the Space Telescope Imaging Spectrograph and the Near Infrared Camera and Multi-Object Spectrometer. The NICMOS instrument had an unexpected problem almost immediately: thermal expansion caused part of its nitrogen heat sink to contact an optical baffle, accelerating warming and reducing its originally expected lifetime of 4.5 years to approximately two years.

  Servicing Mission 3A arrived in December 1999 after three of Hubble's six onboard gyroscopes failed, with the fourth failing just weeks before the mission. The telescope had become incapable of pointing accurately enough to do science. All six gyroscopes were replaced, along with the onboard computer, which was upgraded to a 25 MHz Intel-based 80486 processor system that was 20 times faster and carried six times more memory than the previous DF-224 computer.

  The Columbia disaster in February 2003 nearly ended Hubble's serviced life. NASA Administrator Sean O'Keefe ruled that all future shuttle missions must be able to reach the International Space Station as a safe haven in case of trouble; Hubble's orbit made that impossible. Cancellation of the final planned servicing mission drew fierce public opposition, including thousands of letters from school children, hearings led by Senator Barbara Mikulski, and a formal National Academy of Sciences panel that urged NASA to preserve the telescope. The appointment of Michael D. Griffin as NASA Administrator in April 2005 changed the outcome: Griffin authorized a crewed servicing mission and gave final go-ahead in October 2006.

• 7. What Hubble Found

  ›

  Before Hubble launched, estimates of the Hubble constant, the measure of how fast the universe is expanding, carried errors of up to 50 percent. Hubble's measurements of Cepheid variable stars in the Virgo Cluster and other distant galaxy clusters produced a value accurate to plus or minus 10 percent. The universe's estimated age is now understood to be 13.7 billion years; before Hubble, the range ran from ten to twenty billion.

  Astronomers from two separate research teams, the High-z Supernova Search Team and the Supernova Cosmology Project, used Hubble and ground-based telescopes to study distant supernovae. Their data showed that the expansion of the universe is accelerating rather than slowing under gravity. Three members of those teams were later awarded Nobel Prizes for that discovery. The force behind the acceleration remains unknown; scientists call it dark energy.

  Hubble also established that black holes are probably common to the centers of all galaxies, not just some. High-resolution spectra and images from the telescope showed that the masses of nuclear black holes and the properties of their host galaxies are closely linked. On the 3rd of March 2016, researchers announced the discovery of GN-z11, then the farthest confirmed galaxy, observed as it existed roughly 400 million years after the Big Bang. Hubble observations for that discovery took place on February 11 and the 3rd of April 2015, as part of the CANDELS survey.

  In the Solar System, Hubble captured the collision of Comet Shoemaker-Levy 9 with Jupiter in 1994, producing sharper images than any taken since Voyager 2's flyby in 1979. In March 2015, Hubble observations of aurorae around Ganymede, a moon of Jupiter, revealed evidence of a subsurface saltwater ocean estimated to be 100 km deep beneath a 150 km ice crust. In June and July 2012, astronomers discovered Styx, a fifth moon of Pluto, using Hubble data. In April 2022, Hubble imaging determined the nucleus of comet C/2014 UN271 (Bernardinelli-Bernstein) to be the largest icy comet nucleus ever measured, with an estimated mass of fifty trillion tons.

  As of 2025, more than 22,000 papers based on Hubble data have been published in peer-reviewed journals. Only two percent of Hubble-based papers receive no citations, compared to roughly one-third of all astronomy papers in general. A paper based on Hubble data receives on average about twice as many citations as a comparable paper based on other sources. Of the 200 most-cited astronomy papers published each year, about 10 percent draw on Hubble data. The Andromeda galaxy mosaic, combining observations from 1,023 Hubble orbits, resolved 200 million individual stars in an image of at least 2.5 billion pixels, released in full with the southern half completed in January 2025.

• 8. An Open Telescope

  ›

  Anyone, regardless of nationality or academic affiliation, can apply for time on the Hubble Space Telescope. Competition is intense: roughly one in five proposals submitted in each cycle receives a scheduling slot. Calls for proposals go out approximately once a year, with allocated time covering about one year of operations. A class of proposals called "snapshot observations" covers targets requiring 45 minutes or less of telescope time; these fill scheduling gaps that regular programs cannot use.

  Riccardo Giacconi, the first director of the Space Telescope Science Institute, announced in 1986 that he intended to allocate some of his director's discretionary time to amateur astronomers. Thirteen amateur astronomers were eventually awarded telescope time, with observations carried out between 1990 and 1997. One of the earliest results from that program was the observation of the Great White Spot of 1990 on Saturn, discovered by amateur astronomer S. Wilber and then observed by Hubble under a proposal by J. Westphal of Caltech. Later citizen-science contributions came through the Galaxy Zoo project, whose volunteers identified Voorwerpjes and Green Pea galaxies. The object 2I/Borisov, an interstellar comet discovered by an amateur astronomer, was also observed by Hubble. Budget reductions at STScI eventually ended the formal amateur program, but professional-amateur collaborations have continued informally since.

  Hubble is predicted to remain operational until somewhere between 2030 and 2040. Its mid-infrared-to-visible successor, the James Webb Space Telescope, launched on the 25th of December 2021, and the Nancy Grace Roman Space Telescope is due to follow in 2026.