July 22, 2013
For more than a decade NASA's most expensive science mission, the James Webb Space Telescope (JWST), has suffered cost growth and schedule delays owing to poor management and inadequate budgets. But until recently, technical progress on the enormous space observatory appeared sound.
Conceived in the late 1990s as a follow-on to the Hubble Space Telescope, JWST was projected to cost just $1 billion to build and launch an observatory so advanced it would revolutionize scientific understanding of star and planet formation and identify galaxies in the early universe.
By 2011, however, the program had seen almost a decade of cost overruns and schedule delays. Under pressure from lawmakers, NASA rebaselined the program with a revised cost estimate of $8.8 billion, a new launch date of October 2018, and a healthy amount of schedule margin to maintain both. At nearly nine times the original cost, and more than a decade behind schedule, JWST was finally on track.
Since then, the program has entered a critical phase; myriad technical concerns have emerged, including mass issues on the spacecraft, delayed delivery of two instruments and technical problems with key subsystems, one of which required the addition of a third round of lengthy cryo-vacuum testing to the Integrated Science Instrument Module (ISIM). Combined, these issues have cost 18 of 26 months of schedule reserve on the ISIM, the heart of the telescope that houses JWST's four instruments, designed to detect light from distant stars and galaxies.
One of the late instruments is the Near-infrared Spectrograph (NIRSpec), a 200-kg spectrometer designed to observe up to 100 celestial bodies simultaneously at various spectral resolutions being supplied by the European Space Agency (ESA) and built by Astrium GmbH of Ottobrun, Germany.
According to ESA, in July 2011 three cracks were found in the part that holds the optics components for NIRSpec. After a failure review board in January 2012, ESA had to reassemble the instrument using a flight spare optical bench.
During the rebuild and test, however, ESA encountered additional problems with NIRSpec, including failure of the NASA-supplied microshutter arrays to close. The project also suffered from slower-than-planned progress on the reintegration on the part of Astrium, according to the Government Accountability Office (GAO), which conducted an in-depth review of JWST in 2012.
ESA is reworking the optical bench, but will not deliver the instrument to NASA in time for ISIM's first cryo-vacuum test, which gets underway in August. The agency says NIRSpec is now reassembled and has passed the first cryo-performance and vibration tests, with a final cryo-test now underway. Shipment to NASA is planned for mid-September, more than a year late but in time to incorporate NIRSPec—still fitted with the faulty microshutter arrays—into the ISIM's second round of cryo-vacuum tests next year.
Meanwhile, NASA is making changes to a spare microshutter array that will be less sensitive to the acoustic environment. The goal is to have ESA swap the array assembly ahead of the ISIM vibration test and third cryo-vacuum test in 2015.
Together with the Mid-Infrared Instrument (MIRI) delivered to NASA in May last year, NIRSpec is one of two instruments Europe is contributing to JWST. Combined with launch atop an Ariane 5 ECA rocket in October 2018, and post-delivery and operational mission support, Europe's JWST contribution comprised €370 million ($500 million) in 2004. “Today, ESA says the figure is around €600 million ($790 million), though that appears to exclude launch costs of around €140 million.
NASA is also awaiting delivery of a second instrument delayed by almost 11 months, the Near-infrared Camera (NIRCam) built by Lockheed Martin Advanced Technology of Palo Alto, Calif. NASA determined the need to electrically ground a mirror on the instrument, a process company spokesman Buddy Nelson says is complete, with delivery expected at the end of July.
In addition to instrument troubles, NASA is grappling with a problem on a key subsystem—a cryo-cooler designed by NASA's Jet Propulsion Laboratory to cool MIRI.
In 2010, NASA realized an essential valve in the cryo-cooler was leaking at rates that exceeded requirements, according to GAO. NASA says the MIRI cryo-cooler is particularly complex because it spans approximately 10 meters—or 33 ft.—through the entire JWST observatory. Following the results of a failure review board, a new valve was designed, but it also did not meet leak-rate requirements. NASA says yet another new valve will not be manufactured in time for use in the first ISIM cryo-vacuum test, and is concurrently developing alternatives.
Problems with a second subsystem, also discovered in 2010, involved the degradation of Teledyne-built infrared flight detectors used by three of JWST's four instruments. As a result, approximately $42 million and 15 months of schedule reserve to replace the detectors were included in the rebaseline of JWST cost and schedule, according to GAO. These additions covered the cost of manufacturing the detectors, fabrication, assembly, and test of new focal plane assemblies, and changing the detectors on three of JWST's four instruments.
Since the manufacturing takes some 30 months, the detectors cannot be delivered until after the second round of ISIM cryo-vacuum tests in 2014. As a result, $2 million of the $42 million in the replan was used to add a third round of cryo-vacuum tests for ISIM.
Another challenge JWST has had to address recently is reducing mass on the spacecraft bus to account for the mass of its electrical wiring harnesses, which turned out to be larger than expected.
NASA has been concerned with JWST's mass since its inception in 1999, due to the telescope's size and the payload-carrying capacity of available launch vehicles. Mass constraints were allocated for each subsystem, including the spacecraft, limited to 1,754 kg. But as of September 2012, its estimated mass was 1,960 kg, a 12% overage.
NASA spokesman J.D. Harrington says design changes were insignificant, and included finding lighter-weight components and low-mass means for meeting spacecraft requirements.
“We also received additional mass margin from the launch provider,” he said, refering to Arianespace, which manages Ariane 5 launches. “All of this is completely normal as the observatory design matures.”
NASA says despite these concerns, JWST remains on cost and schedule under the rebaselined plan, asserting the project has overcome several challenges in the past 18 months alone: Mitigating an increase in the amount of heat expected on the instruments, accelerating optics work that added a month of funded reserve to the schedule, and completing 18 segments of the primary mirror—the project's highest technology risk—six weeks ahead of schedule.
“We are managing to the 2013 budget, and in the replan there is a plan for 2014 and each year after, so we're on track,” said Jonathan Gardner, deputy senior project scientist for JWST, at a NASA Advisory Council astrophysics subcommittee meeting July 17. Gardner warned, however, that any changes to the rebaselined plan—including a proposed $74-million funding cut to JWST in the House Appropriations Committee's version of NASA's 2014 budget—would give the program an out. “The project is committed to the budget and schedule, the profile that's in the replan, and if we don't have that profile then we're no longer committed to maintaining the same budget and schedule.”
Under the rebaselined plan, NASA has excluded the ISIM—one of JWST's four primary components that will eventually be integrated with the optical telescope, mirrors and spacecraft bus—from the so-called 'critical path' of milestones that could delay JWST's launch. Although the ISIM has burned through 75% of its schedule margin and added a third round of thermal vacuum tests, “since the ISIM is not on the critical path, there is no change in the launch date,” Harrington says.
Once ISIM test and integration is complete, JWST will have 14 months of funded schedule reserve before it launches in 2018. Only half of that time is allotted for the final three of five complex integration and test efforts, during which problems are commonly found and schedules tend to slip.
“For now, when they run into problems they can shift things around to accommodate late deliveries and other issues,” says a congressional source familiar with the project. “Once you get farther along in integration and test, you lose that flexibility, so any delays could delay the launch.”
Tap the icon in the digital edition of AW&ST for a timeline of JWST integration and test efforts, or go to AviationWeek.com/jwst