NASA’s Starliner decision was the right one, but it’s a crushing blow for Boeing

Ten years ago next month, NASA announced that Boeing, one of the agency's most experienced contractors, won the lion's share of government money available to end the agency's sole reliance on Russia to ferry its astronauts to and from low-Earth orbit.

At the time, Boeing won $4.2 billion from NASA to complete the development of the Starliner spacecraft and fly a minimum of two, and potentially up to six, operational crew flights to rotate crews between Earth and the International Space Station (ISS). SpaceX won a $2.6 billion contract for essentially the same scope of work.

A decade later, the Starliner program finds itself at a crossroads after Boeing learned it will not complete the spacecraft's first Crew Flight Test with astronauts onboard. NASA formally decided Saturday that Butch Wilmore and Suni Williams, who launched on the Starliner capsule on June 5, will instead return to Earth in a SpaceX Crew Dragon spacecraft. Put simply, NASA isn't confident enough in Boeing's spacecraft after it suffered multiple thruster failures and helium leaks on the way to the ISS.

So where does this leave Boeing and its multibillion-dollar contract? Can the company fulfill the breadth of its commercial crew contract with NASA before the space station's scheduled retirement in 2030? It now seems there is little chance of Boeing flying six more Starliner missions without a life extension for the ISS. Tellingly, perhaps, NASA has only placed firm orders with Boeing for three Starliner flights once the agency certifies the spacecraft for operational use.

Boeing’s bottom line

Although Boeing did not make an official statement Saturday on its long-term plans for Starliner, NASA Administrator Bill Nelson told reporters he received assurances from Boeing's new CEO, Kelly Ortberg, that the company remains committed to the commercial crew program. And it will take a significant commitment from Boeing to see it through. Under the terms of its fixed-price contract with NASA, the company is on the hook to pay for any expenses to fix the thruster and helium leak problems and get Starliner flying again.

Boeing has already reported $1.6 billion in charges on its financial statements to pay for delays and cost overruns on the Starliner program. That figure will grow as the company will likely need to redesign some elements in the spacecraft's propulsion system to remedy the problems encountered on the Crew Flight Test (CFT) mission. NASA has committed $5.1 billion to Boeing for the Starliner program, and the agency has already paid out most of that funding.

Boeing's Starliner spacecraft, seen docked at the International Space Station through the window of a SpaceX Dragon spacecraft.

Credit: NASA

The next step for Starliner remains unclear, and we'll assess that in more detail later in this article. Had the Starliner test flight ended as expected, with its crew inside, NASA targeted no earlier than August 2025 for Boeing to launch the first of its six operational crew rotation missions to the space station. In light of Saturday's decision, there's a high probability Starliner won't fly with astronauts again until at least 2026.

Starliner safely delivered astronauts Butch Wilmore and Suni Williams to the space station on June 6, a day after their launch from Cape Canaveral Space Force Station, Florida. But five of the craft's 28 reaction control system thrusters overheated and failed as the spacecraft approached the outpost. After the failures on the way to the space station, NASA's engineers were concerned Starliner might suffer similar problems, or worse, when the control jets fired to guide Starliner on the trip back to Earth.

On Saturday, senior NASA leaders decided it wasn't worth the risk. The two astronauts, who originally planned for an eight-day stay at the station, will now spend eight months on the orbiting research lab until they come back to Earth with SpaceX.

If it’s not a trust problem, is it a judgment issue?

Boeing managers had previously declared that Starliner was safe enough to bring Wilmore and Williams home. Mark Nappi, Boeing's Starliner program manager, regularly appeared to downplay the seriousness of the thruster issues during press conferences throughout Starliner's nearly three-month mission.

So why did NASA and Boeing engineers reach different conclusions? "I think we’re looking at the data, and we view the data and the uncertainty that’s there differently than Boeing does," said Jim Free, NASA's associate administrator and the agency's most senior civil servant. "It’s not a matter of trust. It’s our technical expertise and our experience that we have to balance. We balance risk across everything, not just Starliner."

The people at the top of NASA's decision-making tree have either flown in space before or had front-row seats to the calamitous decision NASA made in 2003 to not seek more data on the condition of Space Shuttle Columbia's left wing after the impact of a block of foam from the shuttle's fuel tank during launch. This led to the deaths of seven astronauts and the destruction of Columbia during reentry over East Texas. A similar normalization of technical problems and a culture of stifling dissent led to the loss of the Space Shuttle Challenger in 1986.

"We lost two space shuttles as a result of there not being a culture in which information could come forward," Nelson said Saturday. "We have been very solicitous of all of our employees that if you have some objection, you come forward. Spaceflight is risky, even at its safest and even at its most routine. And a test flight, by nature, is neither safe nor routine. So the decision to keep Butch and Suni aboard the International Space Station and bring the Starliner home uncrewed is the result of a commitment to safety."

Now, it seems that culture may truly have changed. With SpaceX's Dragon spacecraft available to give Wilmore and Williams a ride home, the decision was relatively straightforward. Ken Bowersox, head of NASA's space operations mission directorate, said the managers polled for their opinion all supported bringing the Starliner spacecraft back to Earth without anyone onboard.

However, NASA and Boeing need to answer for how the Starliner program got to this point. The space agency approved the launch of the Starliner CFT mission in June despite knowing that the spacecraft had a helium leak in its propulsion system. Those leaks multiplied once Starliner arrived in orbit and are a serious issue on their own that will require corrective actions before the next flight. Ultimately, the thruster problems superseded the seriousness of the helium leaks, and this is where NASA and Boeing are likely to face the most difficult questions moving forward.

NASA astronauts Butch Wilmore and Suni Williams aboard the International Space Station.

Credit: NASA

Boeing's previous Starliner mission, known as Orbital Flight Test-2 (OFT-2), successfully launched in 2022 and docked with the space station, later coming back to Earth for a parachute-assisted landing in New Mexico. The test flight achieved all of its major objectives, setting the stage for the Crew Flight Test mission this year. But the spacecraft suffered thruster problems on that flight, too.

Several of the reaction control system thrusters stopped working as Starliner approached the space station on the OFT-2 mission, and another one failed on the return leg of the mission. Engineers thought they fixed the problem by introducing what was essentially a software fix to adjust timing and tolerance settings on sensors in the propulsion system, supplied by Aerojet Rocketdyne.

That didn't work. The problem lay elsewhere, as engineers discovered during testing this summer, when Starliner was already in orbit. Thruster firings at White Sands, New Mexico, revealed a small Teflon seal in a valve can bulge when overheated, restricting the flow of oxidizer propellant to the thruster. NASA officials concluded there is a chance, however small, that the thrusters could overheat again as Starliner departs the station and flies back to Earth—or perhaps get worse.

"We are clearly operating this thruster at a higher temperature at times than it was designed for," said Steve Stich, NASA's commercial crew program manager. "I think that was a factor, that as we started to look at the data a little bit more carefully, we’re operating the thruster outside of where it should be operated at."

In the doghouse

This is the fundamental design flaw that will cause the Starliner test flight to come to a disappointing end. The thrusters are clustered in four doghouse-shaped propulsion pods around the circular perimeter of the Starliner spacecraft's service module. Thermal modeling now shows that these doghouses act like a thermos, trapping heat from the thrusters as they fire over and over in pulses.

“As we look back at OFT-2 now, with this newer lens of what we learned at White Sands, certainly we could have explored OFT-2 in a little more detail, either leading to some redesign of the doghouse to get the thermal environmental lower, or operate the thrusters differently," Stich said.

"It’s easier to do that in hindsight," he said. "If we had [gone] back and thought about the whole integrated problem a little bit more, could we have done some kind of testing? What I would say is it’s very difficult to test the doghouse environment on the ground, where you’ve got thrusters that fire in multiple directions, and it's very hard on the ground to have a test facility, a vacuum chamber, that accommodates thruster firings in multiple directions."

"We thought, obviously, we had done enough analysis to show that the thrusters would be within the temperatures that they were qualified for," Stich said. "Clearly, there were some misses in qualification. We’re going to go through that data in more detail post-flight and then figure out what we can do to go fix them.”

He said NASA is also re-evaluating its qualification and certification processes to determine if the agency should change any of its procedures to reduce the chance of any similar misses in the future.

Scott Hubbard, a former director of NASA's Ames Research Center and a member of the Columbia Accident Investigation Board, told Ars earlier this month that NASA and Boeing should revisit their decision to launch the Crew Flight Test with unresolved technical problems. "Was the decision well-supported, or did someone have 'launch fever?'" Hubbard asked.

Starliner's problems go back much further than 2022. Software woes cut short Starliner's first test flight in 2019 before it could dock at the International Space Station, and they forced Boeing to fly a second test flight, OFT-2, to gain confidence that the spacecraft could safely fly astronauts. NASA and Boeing then delayed the second unpiloted test flight nearly a year to overcome an issue with corroded valves in the ship's propulsion system.

Last year, just a couple of months before it was supposed to launch on the crew test flight, officials discovered a design problem with Starliner's parachutes and found that Boeing installed flammable tape inside the capsule's cockpit. All of that happened before Starliner reached the launch pad for the CFT mission.

SpaceX, which NASA has tapped to rescue the Starliner crew, has now launched eight operational long-duration crew missions to the International Space Station to date, plus an initial piloted test flight of the Dragon spacecraft in 2020 and several more fully private human spaceflight missions. SpaceX has finished all of its work in its initial commercial crew contract with NASA and is now working off of an extended contract to carry the program through 2030, the planned retirement date for the ISS.

What’s next for Starliner?

NASA officials said Saturday that it's premature to decide whether the agency will require Boeing to conduct yet another test flight of the Starliner spacecraft or if the craft could be pressed into operational service after Boeing resolves the myriad problems with its propulsion system.

In either case, don't count on another Starliner crew flight next year. NASA will have to continue leaning on SpaceX, which has shown it is up to the task of launching long-duration crews to the space station every six months. The agency's goal from the start of the commercial crew program has been to fully certify SpaceX and Boeing for operational crew missions, allowing NASA to alternate between Crew Dragon and Starliner missions, with each company flying once per year.

This would give NASA another layer of redundancy for getting its crews to the space station. Right now, the prime route is through SpaceX. NASA continues to fly one astronaut on each Russian Soyuz spacecraft in exchange for a seat for a Russian cosmonaut on each SpaceX crew mission.

NASA Administrator Bill Nelson, Associate Administrator Jim Free, Chief of Space Operations Ken Bowersox, and Commercial Crew Program Manager Steve Stich listen to a question during a news conference Saturday announcing the agency's decision to bring the crew of Boeing's Starliner spacecraft back to Earth on SpaceX's Dragon capsule.

Credit: Mark Felix/AFP via Getty Images

It's unclear if NASA will officially classify the situation with the Starliner Crew Flight Test as a "mishap" or a "loss of mission." Such a determination could trigger a more formal independent investigation, which might trigger longer delays in Starliner's next flight, whatever form it takes.

One thing that could complicate the investigation into the thruster problem is that the control jets are located on the Starliner service module, which jettisons from the crew section of the spacecraft before reentry. The service module will burn up over the Pacific Ocean, so engineers won't have a chance to get their hands on the suspect hardware.

Assuming the investigation doesn't uncover any additional problems and NASA and Boeing return Starliner to flight with astronauts in 2026, there will not be enough time left in the space station's remaining life—as it stands today—for Starliner to fly all six of its contracted missions at a rate of one per year. It's difficult to imagine a scenario where NASA elects to fly astronauts to the space station exclusively on Starliner, given SpaceX's track record of success and the fact that NASA is already paying SpaceX for crew missions through the end of this decade.

Notably, NASA has only given Boeing the "Authority To Proceed" for three of the six potential operational Starliner missions. This milestone, known as ATP, is a decision point in contracting lingo where the customer—in this case, NASA—places a firm order for a deliverable. NASA has previously said it awards these task orders about two to three years prior to a mission's launch.