Starliner repairs
Technicians attempt to repair valves in the propulsion system on Boeing's CST-100 Starliner that have forced an extended delay in the launch of the spacecraft on an uncrewed test flight. Credit: Boeing

Updated 2:45 p.m. Eastern after media teleconference.

WASHINGTON — A test flight of Boeing’s CST-100 Starliner commercial crew vehicle will be delayed for at least several months to fix a problem with valves on the spacecraft.

Boeing announced Aug. 13 that it will remove the Starliner spacecraft that was to launch this month on the Orbital Flight Test (OFT) 2 mission from its Atlas 5 rocket and return it to the company’s Commercial Crew and Cargo Processing Facility at the Kennedy Space Center for additional work.

“This is, obviously, a disappointing day,” Kathy Lueders, NASA associate administrator for human exploration and operations, said in a call with reporters Aug. 13. “This team is going to go figure this out and we will go fly when we are ready.”

Boeing scrubbed an Aug. 3 launch attempt after discovering problems with what the company later said were 13 valves in the spacecraft’s propulsion system that were unexpectedly closed. After being unable to resolve the problem while the spacecraft was on the pad, Boeing and United Launch Alliance rolled the Atlas 5 back to its Vertical Integration Facility to give technicians access to the spacecraft.

As of Aug. 12, Boeing said it had fixed 9 of the 13 valves “after the application of electrical and thermal techniques” to open them. Four other valves remained closed and were still being worked on.

John Vollmer, vice president and program manager of Boeing’s commercial crew program, said on the media call that the leading cause of the valve problem is that nitrogen tetroxide (NTO), the oxidizer used for Starliner’s thrusters, permeated Teflon seals in the valves. That NTO interacted with moisture on the “dry” side of the valve, creating nitric acid. The acid corroded the valves, causing them to stick in the closed position.

He said the nine valves they were able to open then worked normally, but the company ran out of options to fix the remaining four valves while still stacked on the Atlas 5. “We made the decision that we were just out of runway and we had to come back to the factory to complete anything else we needed to do,” he said.

Boeing tested the valves when it fueled the spacecraft about five weeks before rolling out to the launch site in mid-July. “All of them operated perfectly,” Vollmer said. The valves, provided by Aerojet Rocketdyne, are the same as those used on the original OFT mission in December 2019 as well as a pad abort test in November 2019, which did not experience similar problems.

Vollmer said it’s not certain where the moisture came from, but could have happened during assembly of the spacecraft in the factory or exposure to humid conditions on the pad just before launch. He ruled out water intrusion from a thunderstorm the night before the scheduled launch as a factor, although he said the storm did cause some erroneous sensor readings that took time to sort out.

Once Starliner is back in the factory, he said engineers will take a deliberate process to study the valve problem, including determining what portions of the spacecraft need to be disassembled to correct the problem. He added it’s too soon to say if all the valves, including those working properly, will need to be replaced.

The decision to remove Starliner from the Atlas 5 means an extended delay for OFT-2. Even if the problem is quickly solved, Boeing will have to wait until after the mid-October launch of NASA’s Lucy asteroid mission on another Atlas 5 before beginning preparations for another launch. That launch would wait until after the launch of SpaceX’s Crew-3 commercial crew mission, scheduled for the end of October, and the return to Earth of the Crew-2 spacecraft, freeing up the docking port for Starliner.

“It’s pretty early to speculate on when the flight might end up,” said Steve Stich, NASA commercial crew program manager. “When we get to a point where we understand the cause and the fixes, we will work hard to find that slot and get the vehicle flown as soon as we can.”

The launch of OFT-2, they acknowledged, could slip into 2022, or more than two years after the original, flawed OFT mission. “It’s probably too early to say whether it’s this year or not,” Vollmer said. “If we could fly this year, it would be fantastic.”

Boeing said in January 2020 it would fly the OFT-2 mission at its own expense after correcting the software and communications problems found on the first test flight. However, when asked if Boeing would cover the costs of this latest delay, Vollmer declined to confirm it. “I don’t know that I have the complete answer to that,” he said.

Lueders said she was not frustrated by the extended delay. “You might be hearing us sound a little tired, because teams have been working really hard, but we’re not frustrated,” she said. “I’m a little sad, because I wanted to go fly, but I’m also very proud of the team.”

Jeff Foust writes about space policy, commercial space, and related topics for SpaceNews. He earned a Ph.D. in planetary sciences from the Massachusetts Institute of Technology and a bachelor’s degree with honors in geophysics and planetary science...