Analog Science Fiction & Fact Magazine
"The Alternate View" columns of John G. Cramer 

Previous Column  Index Page  Next Column 

Mega-Projects & -Problems; The Hubble in Trouble

by John G. Cramer

Alternate View Column AV-44
Keywords: Hubble Space Telescope, HST, NASA, Perkin-Elmer, Kodak, mirror aberrations
Published in the February-1991 issue of Analog Science Fiction & Fact Magazine;
This column was written and submitted 7/29/90 and is copyrighted 1990, John G. Cramer. All rights reserved.
No part may be reproduced in any form without the explicit permission of the author.

    Last week I listened while an astronomer friend, a principal investigator for one of the five instrument packages aboard the the Hubble Space Telescope (HST), described in agonizing detail the problems with the Hubble's optics and their impact on his plans. He is angry, defensive, frustrated, cynical, and relieved that the damage wasn't worse, all at the same time. He has invested 13 years of his professional career in an instrument package for this $1.5 billion science mega-project, the most expensive telescope in the history of astronomy, our new "Big Eye" for viewing the cosmos in the visible and near-ultraviolet regions of the spectrum. The Hubble Space Telescope, 7 years late, $700 million over budget, shaky and suffering from myopia, is not dead, but for the next three years it will have to function in a state of severely reduced capacity.

    In this column I want to focus on three things. First, the nature of the HST problem, its origins, and its cure. Second some observations on the difference between scientific funding agencies (in particular, DOE and NASA) in dealing with scientific megaprojects (i.e., any scientific project costing over $100 million). Finally, I want to offer a modest proposal for fixing our ailing national space program.

    The emerging consensus on the HST problem is that its primary mirror was manufactured improperly and is distorted by "spherical aberration". The primary mirror's surface deviates by a distance of about half a wavelength of visible light from its required hyperboloid shape toward the shape of a sphere. Half a wavelength doesn't sound like much of an error, only about 2 10-7 m.  But the specification of the HST mirror requires an accuracy of 1/70 of a wavelength or better, so the mirror is a factor of 35 out of spec.

    What went wrong? At this writing, it is not clear. The primary mirror, until recently touted by NASA's PR staff as their "Crown Jewel", was made under contract to NASA by the optics division of the Perkin-Elmer Corp. located in Danbury, CN (and now owned by Hughes Aircraft which in turn is owned by General Motors). The current best guess is that the fault lies with the "null corrector", a system of one lens and two mirror used by Perkin-Elmer to deflect light for testing the HST primary mirror. During the fabrication of the HST mirror, a discrepancy was noted between the optical properties of the the precision null corrector and a simpler null corrector made of lenses and used for the initial rough cutting of the mirror. Unfortunately, this discrepancy was attributed to inaccuracies in the smaller lens system and ignored. The null corrector still exists and is now being checked.

    Perkin-Elmer had considerable experience with precision mirror fabrication and had previously made mirrors somewhat like the HST primary for military reconnaissance satellites. The specifications and performance of these, however, remain classified. The military units were optically tested after assembly. The HST was not tested in this way, a "saving" that permitted Perkin-Elmer to submit the low bid for the mirror contract. Such a test would have unambiguously revealed a spherical aberration of the magnitude present in the HST.

    Eastman Kodak, another bidder for the mirror contract, lost in the competition but was subcontracted by Perkin-Elmer to construct a backup mirror. Astronomers, if they had been in charge of the HST project, would certainly have compared the two mirrors and selected the one of best optical quality for the telescope (and in the process, discovered the aberration problem). However, Perkin-Elmer, the prime contractor for the primary mirror, chose not to make that comparison and to use the mirror they had constructed. Corporate pride took precedence.

    The spherical aberration of the primary mirror causes the image of a point-like star to appear in the focal plane of the instrument as a bright spot surrounded by a larger ring or halo containing most of the light of the image. The halo diameter is about 3-4 times larger than the HST design specification, so only about 10% of the collected light can be delivered to most of the instrument packages. In cases where a dim object is close to a brighter one, the light from the dim object will be swamped by halo light from the brighter object, compromising or defeating many planned measurements.

    There are five instrument packages aboard the HST. The Wide Field/Planetary Camera (WFPC) was given non-competitively by NASA to its own Jet Propulsion Laboratory for development. The other four instrument packages were selected from competitive proposals submitted by university-based consortia of astronomers. The packages were specified by the consortia, which then received grants from NASA and used the money to make contracts with aerospace firms for the actual construction of the instruments. Each university instrument package is roughly the shape of a telephone booth, with the light from the HST entering at one corner, to be reflected, analyzed, and imaged by the remainder of each instrument.

    The spherical aberration has a devastating effect on the Wide Field/Planetary Camera, the JPL-designed instrument that was supposed to produce close-up pictures of the cosmos. Only about 10% of the planned measurements with that instrument are uncompromised. The four university packages are impacted in varying degrees. My friend's instrument package is an ultraviolet spectrometer, an instrument for studying ultraviolet spectral lines from astronomical objects. Ultraviolet is high-frequency high-energy light that is invisible to the eye and does not penetrate the Earth's atmosphere. About 1/3 of the planned high-priority observations to be made with this package are now impossible, and another third are seriously compromised. Another experiment, a photometer package that measures the brightness of astronomical objects, is scarcely impacted at all. But the Faint Object Camera, another package, has had 1/6 of its experiments eliminated and another 2/3 compromised. The slapping sound you hear is that of astronomers rapidly reshuffling their priorities.

    Now I'll answer some frequently-asked questions about the HST problem:

Q: Why not send astronauts up with the Kodak replacement mirror and install it in space?

A: Changing a telescope mirror is not like changing a tire. The installation is far too complicated to be done in orbit. It requires precision alignment jigs that must position the mirror to distances of a fraction of a wavelength of light. The HST was not designed for such on-orbit repair.

Q: Then why not go up with the Shuttle, fetch the HST back, install the Kodak replacement mirror under laboratory conditions, and put it back in orbit?

A: Nobody in charge wants to do this because (a) it's too expensive, (b) the Shuttles are, as I write this, grounded due to hydrogen leaks and have a huge backlog of other instruments to transport, (c) no Shuttle has ever landed with a load as heavy as the HST in its cargo bay, and the shuttle pilots are not anxious to "break new ground" in this way, and (d) the astronomers, having waited 13 years already, want the opportunity to use the HST for those measurements which are within the present capabilities of the instrument.

Q: How will they fix the aberration problem?

A: The present instrument packages were intended to be used for only about three years before they were replaced with updated packages. The HST is constructed for easy in-space replacement of the instrument packages by space-walking astronauts. It is even equipped with hand-holds and rings to aid in this operation. When the light enters the instrument package "telephone booths", it is still about a meter from its final focus and has a diameter of a few centimeters. Therefore, the "bad" light from the aberrated outer edges of the primary mirror is sufficiently separated from "good" light from the central region that it can be operated on optically to correct the aberration effects. Astronomers are confident that corrective lenses and mirrors can be designed into new instrument packages that can eliminate most or all of the aberration problem. But first, on the ground detective work must reveal the true character and origin of the aberration.

Q: When will the problem be fixed?

A: The old instrument packages are scheduled for replacement in 1993. It is unlikely that this schedule can be accelerated, either from the point of view of the Shuttle workload or that of the instrumentation teams that must redesign their instruments to compensate for the aberration. Even with the reduced number of priority experiments, the HST will not stand idle for lack of good measurements to be made. The astronomers who have been associated with the HST from its inception have had 13 years to learn from NASA the virtue of patience. Another three years at this point does not seem so long.

Q: Will there be other problems?

A: Probably. Another "economy" that NASA made was that there is no software simulation of the HST that can be used to test command instructions before they are transmitted to the instrument for execution. Instructions are checked by "eyeball" as time permits. Each instruction transmitted to the HST has some non-zero probability of error. A number of erroneous instructions have already been sent to the instrument, many of which have caused the HST's rather primitive control computer to dump the instrument into hibernation mode, a state from which it must be coaxed with days of effort. If this goes on, the proverbial "halt and catch fire" computer command could one day go out to the HST unnoticed, bringing its limited usefulness to an end. Let's hope this doesn't happen.

    Now let me turn to the other topic of this column, the federal funding of large science projects. In my not completely unbiased opinion, the best federal agency for the funding and management of large-scale science is the Department of Energy (DOE) and the worst is the National Aeronautics and Space Administration (NASA). The difference between the two agencies is where scientific excellence and scientific priorities rank in the overall priorities of the organization. The DOE is a huge organization with many parts and many missions. The part of the DOE that supports basic scientific research, mainly in atomic, nuclear and particle physics, had its genesis in the Manhattan Project, the World War II organization that, against all odds, developed and produced the first nuclear weapons, moving from a vague scientific concept through a whole series of scientific and technical breakthroughs to a working device in just four years.

    The scientists and military people working on the Manhattan Project were driven by the fear that Hitler would develop the fission weapon first. In that environment there was no alternative to putting scientists in charge and allow them to make most of the major decisions as directly and as rapidly as possible. This tradition of scientific leadership carried over to the peacetime outgrowth of the Manhattan Project, the Atomic Energy Commission, perhaps the only federal agency that ever had a Nobel Laureate as its Chairman. The tradition of scientific leadership persisted in later incarnations of the same scientific organization as it became the Energy Research and Development Administration and finally a part of the Department of Energy.

    The DOE tradition of scientific leadership, if somewhat muted by the bureaucracy of a cabinet-level department, has not been lost and is in fact alive and well. The DOE has an outstanding record, perhaps the best of any U. S. scientific agency, for managing science megaprojects and bringing them to fruition on time and under budget. This string of successes started with the Manhattan Project, and has continued with the Bevatron, the Brookhaven Cosmotron and AGS, SLAC, and FermiLab. Even aborted DOE mega-projects like Isabelle can be viewed as demonstrations that when hard budget decisions must be made, scientific priorities come first, not somewhere down the list.

    If the scientific research part of the DOE had its origins in the linear priorities of WWII, NASA was the product of the convoluted issues of the Cold War. It was born in the wake of Sputnik. Initially it also mounted a string of major technical triumphs from the Mercury and Gemini Programs to the lunar landings of the Apollo Program and SkyLab, the first U.S. space station. But NASA carried within its structure a tragic flaw. The organization was recruited from Department of Defense personnel who had little enthusiasm for or understanding of scientific priorities. The focus of the agency was technical excellence and quality control, with scientific considerations placed in a secondary role. Tom Wolfe's The Right Stuff can be read a a chronicle of NASA's systematic rejection of scientific considerations while providing expensive toys for the test pilots and spectaculars for the media.

    Nevertheless, NASA attracted teams of highly talented and dedicated experts that made Apollo possible. These key teams are now long gone. They have been replaced with many new layers of paper shufflers. And there is evidence that the dedication to technical excellence has been subverted to a dedication to the appearance of excellence. Few will now argue that our national space program is not in serious trouble.

    So what can be done to repair our ailing space program? In a recent interview reported in Physics Today, Dr. D. Allen Bromley, the Presidential Science Adviser, commented that the turn-down in defense-related activities has created a transition now in progress at DOE laboratories. Bromley said that Secretary of Energy Watkins is aggressively promoting new missions for the Department of Energy, and that groups at the two DOE weapons laboratories, Los Alamos National Laboratory and the Lawrence Livermore National Laboratory are investigating "some of the new technologies that might make a significant difference in the economics of our space activities, ... technologies for the space station, the Moon-Mars initiative, and the commercialization of space."

    I regard Dr. Bromley's remarks as wonderful news for our national space program. Competition is the driving force of our economy. Its appearance within the federal government, if unusual, is perhaps the one thing that could revitalize our space program. If the best talent from Los Alamos and Livermore, masters of technical innovation and the development of new technologies, turn their attention to cheaper and better ways of getting into space and doing science there, NASA must respond to this competition or be left standing at the starting gate. This domestic space race is going to be very interesting to watch.


Bromley Interview:
"Washington Reports", Physics Today 43 #7, 49 (July-1990).

HST Problems:
"Sleuths Zero In on Cause of Telescope Flaw", William J. Broad, New York Times, p. B5, (July 31, 1990).

SF Novels by John Cramer:  my two hard SF novels, Twistor and Einstein's Bridge, are newly released as eBooks by Book View Cafe and are available at : .

AV Columns Online: Electronic reprints of about 177 "The Alternate View" columns by John G. Cramer, previously published in Analog , are available online at:

Previous Column  Index Page  Next Column 

Exit to Analog Logo website.
This page was created by John G. Cramer on 7/12/96.