Aerospace and Space Science
Hubble Space Telescope Primary-Mirror Flaw
The Hubble Space Telescope launched in April 1990 carrying a primary mirror ground to the wrong shape, its edge too flat by about 2,200 nanometers because the device used to test the mirror during grinding was itself misassembled. The result was spherical aberration that left Hubble's images badly blurred, a public embarrassment for NASA. Astronauts corrected the flaw with corrective optics during a December 1993 servicing mission, after which Hubble became one of the most productive scientific instruments ever built.
Narrative
The story
The ambition
The Hubble Space Telescope was built to see the universe without the blur that Earth's atmosphere imposes on ground-based observatories. NASA and the astronomical community had pursued a space telescope for decades, and by the time Hubble launched aboard the Space Shuttle Discovery in April 1990, expectations were enormous. Its 2.4-meter primary mirror was ground to a precision described as accurate to within a few dozen nanometers, and the telescope was meant to resolve faint, distant objects far beyond the reach of any instrument on the ground. It was, on paper, the most precisely figured optical mirror ever put into orbit.
The rise
Manufacturing the mirror took years. The Perkin-Elmer Corporation ground and polished the 2.4-meter primary using a device called a null corrector, a precision optical tool that simulates the point of focus and lets technicians check the mirror's shape against the design during grinding. NASA relied on the null corrector's readings as the ground truth for the mirror's figure. A second, cruder refractive null corrector was also used as a check, but when its readings disagreed with the primary reflective null corrector, the discrepancy was attributed to the cruder instrument, and the mirror was ground to match the reflective corrector instead. The mirror shipped, and by every test NASA had performed, it met specification.
The cracks
Within weeks of Hubble's April 1990 launch, the images coming back were visibly blurred. Engineers on the ground worked through May and June to characterize the problem, and on June 27, 1990, NASA publicly announced that the telescope's primary mirror had a defect. NASA formed the Hubble Space Telescope Optical Systems Board of Investigation on July 2, 1990, to determine the cause. The investigation found that the primary reflective null corrector used during the mirror's fabrication had been incorrectly assembled: one of its lenses was mispositioned by about 1.3 millimeters, later traced by NASA to a small, undocumented washer inserted in the device. That tiny spacing error meant the null corrector fed technicians a false reference for the mirror's shape, and the mirror was ground to match a flawed target instead of the design.
The collapse
The consequence was spherical aberration, a defect in which light striking the outer part of the mirror focuses at a different point than light striking its center. Hubble's mirror was too flat near its edge by roughly 2,200 nanometers, an error about ten times larger than the design's tolerance. Instead of concentrating incoming starlight into a point within about 0.1 arcseconds as designed, the telescope spread it over more than an arcsecond. Spectroscopy and observations of bright objects remained workable, but the faint-object and high-contrast imaging that were central to Hubble's scientific mission, including many of its planned cosmology programs, became largely unusable. The flaw was a national embarrassment for NASA, arriving on a flagship instrument that had been sold to the public as an unprecedented scientific tool.
The aftermath
NASA had built Hubble to be serviced in orbit by astronauts, and that design choice made a repair possible. Engineers devised two corrective instruments that used precisely ground mirrors to counteract the aberration, effectively acting as glasses for the flawed primary: the Corrective Optics Space Telescope Axial Replacement (COSTAR), built for roughly $50 million, and a redesigned Wide Field and Planetary Camera 2 (WFPC2) with the correction built into its own optics. Astronauts aboard the Space Shuttle Endeavour installed both during servicing mission STS-61 in December 1993, replacing the High Speed Photometer with COSTAR on December 2 and swapping in WFPC2. On December 28, 1993, ground controllers commanded COSTAR's mirrors into position, and the resulting images confirmed the aberration had been corrected. Hubble's vision was restored, and over the following decades it became one of the most scientifically productive instruments NASA has ever operated, credited with major contributions to cosmology and astrophysics. That later success does not undo the fact that the observatory launched with a defective primary mirror and spent three and a half years operating far below its design capability.
The lessons
The flaw traces to a single miscalibrated test instrument, not to the mirror grinding itself; the mirror was, in fact, ground with extraordinary precision to match the wrong reference. That is the sharper lesson: a measurement system that everyone trusts can be wrong in a way that produces a highly precise, highly wrong result, and a secondary check that contradicts the primary instrument is a warning worth chasing rather than explaining away. NASA had a second, independent test method that disagreed with the null corrector, and the disagreement was resolved by assuming the cheaper instrument was the inaccurate one. The other lesson is about designing for failure. Hubble survived its own manufacturing defect only because NASA had built it to be serviced by astronauts in orbit, a design decision that turned a potentially fatal flaw into a fixable one.
Causal timeline
Failure Anatomy
- 1979-1981
Mirror ground against a false reference
Perkin-Elmer ground Hubble's 2.4-meter primary mirror using a reflective null corrector that had been assembled with a lens mispositioned by about 1.3 millimeters, later traced to an undocumented washer in the device. [1]
Technical failure - 1981
A contradicting check dismissed
A secondary refractive null corrector's readings disagreed with the primary reflective corrector during testing; the disagreement was attributed to the cruder instrument rather than investigated as a possible flaw in the primary one. [2]
Information failure - 1990-04
Launch with a flawed mirror
Hubble launched aboard Space Shuttle Discovery in April 1990 carrying a primary mirror ground about 2,200 nanometers too flat near its edge, an error roughly ten times the design tolerance. [3]
Technical failure - 1990-06-27
- 1993-12-28
COSTAR and WFPC2 correct the flaw
Astronauts aboard Space Shuttle Endeavour installed the COSTAR corrective-optics package and the redesigned WFPC2 camera during servicing mission STS-61 in December 1993; ground controllers deployed COSTAR's mirrors on December 28, 1993, and confirmed the aberration was corrected. [5]
Structured analysis
What Went Wrong
Root causes
Miscalibrated null corrector. The reflective null corrector used to verify the primary mirror's shape during grinding was incorrectly assembled, with a lens mispositioned by about 1.3 millimeters, so technicians ground the mirror to match a false reference. [1]
Contributing factors
A contradicting check overruled. A second, cruder refractive null corrector produced readings that disagreed with the primary reflective corrector, and the discrepancy was resolved by trusting the flawed reflective instrument instead of investigating further. [2]
Immediate trigger
Mirror ground to the wrong shape. Relying on the miscalibrated null corrector's readings, technicians ground the primary mirror's edge about 2,200 nanometers too flat, producing severe spherical aberration once the telescope reached orbit. [3]
Visible symptoms
Blurred images after launch. Within weeks of the April 1990 launch, Hubble's returned images were visibly out of focus, spreading starlight over more than an arcsecond instead of concentrating it within about 0.1 arcseconds as designed. [4]
Warning signs
Disagreement between the two test instruments. The refractive null corrector's readings did not match the reflective null corrector's during mirror fabrication, a discrepancy that in hindsight pointed to the reflective device being the faulty one. [2]
Affected groups
Evidence
Claims & sources
Every numbered marker in the analysis links to the claim it rests on, and each claim to its sources.
- [1]
The reflective null corrector used to test Hubble's primary mirror during fabrication was incorrectly assembled, with a lens mispositioned by about 1.3 millimeters, later traced to an undocumented washer a technician had inserted in the device.
- [2]
A second, cruder refractive null corrector used during mirror testing produced readings that disagreed with the primary reflective null corrector, and the discrepancy was attributed to the refractive device rather than the reflective one.
- [3]
Hubble's primary mirror was ground with its outer edge too flat by roughly 2,200 nanometers, an error about ten times the design tolerance.
- [4]
Within weeks of Hubble's April 1990 launch, its images were visibly blurred; NASA publicly announced the mirror defect on June 27, 1990, and images spread starlight over more than an arcsecond instead of concentrating it within about 0.1 arcseconds as designed.
- [5]
Astronauts installed COSTAR and the redesigned WFPC2 camera during servicing mission STS-61 aboard Space Shuttle Endeavour in December 1993, and ground controllers deployed COSTAR's corrective mirrors on December 28, 1993, confirming the spherical aberration was corrected.
- [6]
COSTAR cost roughly $50 million to build, and the redesigned WFPC2 camera cost roughly $23.9 million.
Sources
Hubble's Mirror Flaw
NASA Science
Hubble Space Telescope
Wikipedia
Fixing Hubble's blurry vision
Spaceflight Now · 2015-04-23