It has been through some tough times, but now the James Webb Space Telescope (JWST) project has pulled up its socks and is slogging through one of the most complex space engineering efforts ever attempted.

Some 1,000 engineers and technicians in the U.S.—and more in Europe and Canada—are working on the massive instrument with a new approach that managers believe will pull everything together for a 2018 launch, without the overruns and schedule slips that threatened to kill the program two years ago.

“Basically Congress said 'get it right; you've got one more shot at this,'” says Scott C. Willoughby, prime contractor Northrop Grumman's vice president and JWST program manager.

The 6.5-meter (21.3-ft.) infrared telescope has adequate funding now, he said during the recent 28th National Space Symposium here. The money supports cost and schedule reserves that will make it possible to turn up the heat when problems arise in one area by bringing on more engineers, while maintaining pace in others.

That's a far cry from past practice on the flagship NASA astrophysics mission, when work was deferred to stay within budget caps. Ultimately it added so much cost to the program that an outside panel that was convened to diagnose the faltering project and recommend a cure termed it “non-executable.”

“Without the reserves needed to fund the exigencies normally experienced in the course of development, the project simply continued the practice of deferring some of the work planned in the current year to a future year,” stated the JWST Independent Comprehensive Review Panel. “Due to the inefficiencies created when deferring already planned work, this led to escalating increases in the life-cycle cost of JWST and continued erosion of the schedule.”

Ultimately, the group known as the Casani panel after chairman John Casani, an experienced Jet Propulsion Laboratory program manager, found NASA was $1.4 billion short in its estimated cost to launch the JWST, and 14 months behind schedule.

After incorporating the Casani panel's recommendations for adequate reserves and better management at Goddard Space Flight Center and NASA headquarters, the space agency concluded it would cost $8.7 billion to complete the telescope and launch it in 2018, a four-year delay (AW&ST Aug. 29, 2011, p. 36; Nov. 15, 2010, p. 50).

Congress appears to have accepted that grim conclusion. In its fiscal 2013 markups, the House and Senate appropriations subcommittees that oversee NASA both met the agency's $628 million request for the telescope, making it likely Congress as a whole will go along.

Results of the program restructuring are starting to come in. ATK has completed the flight composite structure that will hold 12 of the telescope's 19 hexagonal beryllium mirrors in close enough alignment that actuators and software developed by Ball Aerospace can tweak them into position to act as a single mirror.

Lockheed Martin has completed assembly of the telescope's main infrared camera, known as the Near Infrared Camera. It consists of twin optical imaging modules with focal plane assemblies designed to operate at 35K with 40 million pixels.

To keep the instrument that cold, NASA and the European Space Agency will use an Ariane 5 ECA launcher to send it to the Sun-Earth L2 lagrangian point, where it will use the Earth and a tennis-court-sized sunshield made of five layers of coated Kapton to block the Sun's heat. Folding all of that hardware into the rocket's payload fairing, which has a diameter of 4.57 meters and 16.19 meters of usable length, would challenge an origami master.

Workers at ManTech International Corp.'s NeXolve facility in Huntsville, Ala., are hard at work developing the sunshield layers, using a proprietary “welding” process to connect 10,000 in. of seams, and are just beginning to punch holes in the material to accommodate the stays and wires that will hold the layers in place after it unfolds almost 1 million miles from Earth.

Given that kind of development complexity, the program is using pathfinder components across the entire spacecraft to find problems before starting work on flight hardware. The work at ManTech is no exception.

“You don't want to do that the first time on flight [hardware], because there's not much of a go-back on that,” says Willoughby. “So we're doing the full-scale article, and we're already done manufacturing two of those, testing their shape and comparing them to what our models predicted, and what we needed. It was great. The first one came out about 0.75 in. of error across that. The second one came out at 0.35. And that's good. On an 80-by-40-ft. assembly, if we can hold within an inch we're happy, so we've basically proven our manufacturing technique.”

Still, ManTech will build the other three pathfinder membranes before starting work on the flight versions next year. The overall JWST program is carrying 13 months of critical path reserve to get to launch, but when Northrop Grumman starts putting all of the pieces together at its Redondo Beach, Calif., facility there will have been plenty of chances to consume it in fixing unexpected problems.

The entire “cold section”—the telescope and instruments—will be tested in a large Apollo-era thermal vacuum chamber at Johnson Space Center, modified to use helium instead of nitrogen to chill the hardware to operational levels.

The sunshade will be folded and deployed three times, and all of the pieces will go through the standard acoustic and vibration testing to ensure they can withstand the rigors of launch. Willoughby says the most worrisome stage of the development will come as the main elements are integrated and tested to ensure they will work together as planned at the L2 point, far beyond the range of human servicing that rescued the Hubble Space Telescope after it was launched with a flawed main mirror.

“Any time you get to the next level of integration you're doing something new for the first time,” Willoughby says, explaining that the program is conducting an unusual amount of testing after integrating subcomponents to avoid unpleasant surprises that would require extensive rework.

“As soon as two pieces come together, test it,” he says. “Don't wait for the third and the fourth and the fifth.”

The results, if it all works, will be unprecedented. The JWST is designed to peer back to the very dawn of time, when the first stars and galaxies began lighting up after the “cosmic dark ages” that followed the Big Bang. It will also be able to analyze the atmospheres around extra-solar planets for the signatures of life (see p. 26). And if it is like the Hubble instrument, it is likely to raise as many new questions as it answers old ones.

James Webb Space Telescope Budget ($ millions)
Prior years Fiscal 2011 Actual Fiscal 2012 Appropriation Fiscal 2013 Request Projected Fiscal 2014 Projected Fiscal 2015 Projected Fiscal 2016 Projected Fiscal 2017 Projected Cost to complete Projected Total
$2,992 $476 $518 $627 $659 $647 $622 $571 $1,650 $8,763
Source: NASA budget documents.