By David A. Ord
…… Hubble sees double; toil and trouble.
The 25th anniversary of the launch of the Hubble Space Telescope (HST) was celebrated on April 24th and there can be no finer example of success snatched from the jaws of failure. The eventual success of the project is unique in that not only has science been greatly advanced but it also captured the imagination of the public with its utterly amazing images of deep space.
Over the years, we seem to have been treated to a vast number of Hubble images, each one more stunning than the last and the image produced for the 25th anniversary does not disappoint.
But, what Hubble has achieved for astronomy could so easily have gone another way. Much has been written about the technical flaw in the primary mirror; but if you ask only technical people to find out what went wrong, they will generally find a technical reason for failure which everyone can get behind. But, many people who were an integral part of the project have suggested that the ‘technical failure’ was only a symptom of something more pervasive within NASA. Similarly, the ‘O-ring’ technical failure which caused Challenger to disintegrate at launch was only a symptom of the real problem. The physicist Richard Feynman, who led the inquiry into the Challenger disaster did not have to tell engineers about the sealing properties of ‘O-rings’ at low temperature; these characteristics were well documented. But what was wrong, was the culture created within NASA which was so biased to a launch decision that any objections or potential safety concerns had to be very major indeed to be considered; even to the extent that some engineers felt that any doubts they had were discouraged from being verbalised.
The funding of projects by Congress for NASA was historically based on the visionary demand of President Kennedy – ‘a man on the moon within the decade, whatever it takes’. NASA would generally lie about how much a project would cost and would request further funding as a matter of course. This was not guaranteed, but generally the argument that the cost to cancel a project was so high anyway, that the extra funding to continue was often a cheaper option, was accepted. But this method of initially ‘low balling’ project budgets created an unnecessary angst about cost right from the off.
The Hubble Space Telescope was no exception in this regard. NASA had requested $430 million at the outset and ended up spending £1.8 billion up to the point where Hubble was aboard the Shuttle for launch. The degree of pressure being somewhat proportional to the overruns which were vast in time and money, it is no surprise that an environment was created whereby there was ‘no inclination to address anything that was not a critical and obvious problem.’ The review board looking into Hubble failures referred to not only NASA project staff being under extreme stress, but that the relationship with contractors was positively hostile – “tell us about any problems at your peril!”.
The primary and secondary mirrors of the Hubble Telescope were the key technologies. NASA had tenders for the mirrors from Perkin-Elmer Corp. and Kodak. The former quoted $70 million, some $35.5 million dollars less than Kodak. Just like NASA and congress, Perkin-Elmer could ‘low ball’ the budget to get the contract. They had been assured that once Congress had approved the project, they could demand more due to ‘unforeseen technical challenges’. A key official is quoted as saying, “Perkin-Elmer had to lie to get the contract, NASA had to lie to get the funding; it was a fable from the very start.”
But what they had not bargained for was that the mood for indiscriminate ‘visionary’ spending was changing and Congress was showing the beginnings of a backbone with regard to NASA.
Given his ‘low-balled’ budget, it was probably with mixed feelings that the Perkin-Elmer engineer responsible for the mirrors took delivery of the first 2.4m round of glass from Corning Glass Works. To make the space telescope’s main mirror, as well as its 12.5-inch secondary minor, Perkin-Elmer devised a system at its Danbury plant that substantially advanced the science. Lying on a bed of 134 titanium nails to simulate the gravity-free environment of space, the glass would be polished by a spinning abrasive pad attached to a swivelling arm. For the first time, computers would control the speed, pressure and direction of the arm.
After polishing runs lasting from six to seventy hours, the mirror would be trundled on rails to an adjacent room so it could be determined how much closer they had come to the desired shape. For that purpose, the company had designed an optical measuring system so precise that they used it only in the middle of the night when there were no vibrations from tractor-trailers rumbling down Route 7 outside the plant. The system, consisting of an instrument called a null corrector and a special camera, was so sensitive the company had to pull out the speed bumps in the parking lot. They even had to shut down the air conditioners when they used it.
Even before polishing began, issues had arisen with grinding the blank glass and it arrived for polishing some 9 months behind schedule. As the mirrors were the critical component, this had a huge cost impact as several hundred engineers at Lockheed-Martin who were to build the 12 tonne satellite sat idle, collecting their pay-checks.
Perkin-Elmer placed the polishing on a 24 hour operation and with limited resource, the entire team were being pushed to the limit. The master optician, a man called Richard Geissler took residence in the plant to be on-call whenever required. After about a year of mind-numbing polishing, one morning, perhaps through fatigue, Geissler tapped in ‘1.0’ instead of ‘0.1’ into the polishing computer. To everyone’s horror, the whirring polishing tool began digging a groove near the inside edge of the mirror. It would have been much worse but for the actions of a technician who pressed the kill switch on the motor. The groove was smoothed during further polishing, but ‘Gessler’s groove’ as it became known never went away completely.
Perkin-Elmer had gone to NASA with a new cost estimate of $272 million, four times the original figure. NASA refused to accept the new figure. The company felt let down and the hostile environment between NASA and the contractor was complete.
As the polishing continued, some questions were asked about testing the mirrors by means other than the new ‘null corrector’ instrument, but given the cost pressures there was ‘no inclination to address anything that was not a critical and obvious problem.’
There were even some data measurements which suggested there might be a problem and something was going on. But, no one had any time or incentive to investigate properly. No one really questioned the accuracy of the state of the art ‘null corrector’.
Even in May 1981, as the mirror reached completion, although the project engineers had no serious doubts that they had produced the perfect mirror, they concluded their technical appraisal with 5 final recommendations before releasing to NASA. Number 3 on the list was a check of the ‘null corrector’; to determine any ‘gross error’. As we now know, the state-of-the-art ‘null corrector’ was the source of the error. But senior management at Perkin-Elmer had no inclination to address anything that was not a critical and obvious problem. The recommendation was rejected.
Unusually, for a project of this size, NASA had very few of its own people deployed at the Perkin-Elmer site. This was partly due to NASA’s budget pressures and interestingly pressure from the Department of Defense for whom Perkin-Elmer were doing classified work and that the Pentagon was nervous about outsiders moving around the site. Of the few from NASA who were allowed on site, the quality specialist had no knowledge of optics and was briefed to only safeguard the mirror’s safety, not its quality.
NASA required that one of its quality assurance officials give final approval to each piece of equipment destined for space. Of all the approval documents on the thousands of parts that went into one of the most complicated scientific instruments ever assembled, only one is known to lack a signature from the quality office. The quality controller refused to sign off on the main mirror.
Perkin-Elmer also manufactured other components for Hubble and by 1984 had managed to obtain $350 million from NASA, almost 5 times the original tender.
In January 1986, NASA announced that the Hubble would be launched in October of that year. But on Jan. 28, the Challenger exploded, killing the seven people aboard and halting the space shuttle program for more than 2 years. Storing the Hubble telescope at Lockheed—along with the continuing testing, safety checks and engineering work—cost NASA more than $7 million a month until shuttle flights resumed. The launch schedule was so congested then that a planned 1989 Hubble launch had to be postponed until 1990.
So, it was on April 24th 1990 that Hubble finally launched. The first light image was fuzzy to say the least; but no one was overly concerned and it was several weeks later that the British Physicist Christopher Burrows working at the Space Telescope Science Institute of Baltimore determined that the only explanation was that the primary mirror was the wrong shape.
For those involved in the project the news was sickening. No one could believe it. The one saving grace was that Hubble was designed to be serviced in situ. Once calm science prevailed over hysteria and the blame game, a solution was found. The failure was systematic, Hubble had a perfect flaw. All that was required was something akin to a pair of glasses to correct. The repair was completed on Shuttle mission STS-61 in December 1993.
In all, 5 service missions have considerably added to the capability of Hubble. Congress was persuaded to keep the funding going to around $2.5 billion for an operational Hubble and current estimates suggest that accumulated investment, including its running costs is closer to $10 billion.
As with all major projects which go awry, the inquiries come up with ‘lessons to learn’. Some observers believe that they have missed the point when they focus only on technical failures. The cause of technical failure can often lie elsewhere.
In 1996 NASA proposed what is now called the James Webb Space Telescope; a replacement for the Hubble. The budget was to be $1.8 billion and its launch would be in 2011. As was the norm NASA requested a further $2 billion from Congress ‘for unforeseen technical challenges’; however, the project was voted out by the House of Representatives. This was a first.
Not surprisingly (there are many jobs at stake in various states) in 2012 the US Senate restored the project, but capped it at $8 billion and accepted a NASA launch date in 2018. As of December 2014, the cost has risen to $8.83 billion. The main issue apparently is with the mirror contractor. Its déjà vu all over again.
Happy birthday Hubble Space Telescope!