LOS ANGELES — NASA and Boeing are preparing for a June 1 launch of the company’s CST-100 Starliner on a crewed test flight after analysis of a helium leak led to the discovery of a “design vulnerability” in the spacecraft’s propulsion system.
At a May 24 briefing, officials said they believe a helium leak detected in a reaction control system (RCS) thruster in the spacecraft’s service module is caused by a defective seal in a flange that is itself an isolated problem. None of the other thrusters have shown evidence of helium leaks after extensive testing.
Engineers discovered the leak shortly after the May 6 launch scrub for the Crew Flight Test (CFT) mission caused by an unrelated valve problem in the Atlas 5 rocket’s Centaur upper stage. Testing in the days after the scrub caused the leak to worsen.
Steve Stich, NASA commercial crew program manager, said engineers believe that a seal, a rubber ring about the size of a shirt button and the thickness of 10 sheets of paper, in the flange had a defect that grew worse during tests. He said other possibilities included an error installing the seal or foreign object debris that rubbed against the seal.
NASA and Boeing concluded that Starliner can fly with the leak as-is. “If we were to remove the seal completely,” said Mark Nappi, Boeing vice president and commercial crew program manager, “the leak rate would not exceed our capability to manage that leak. That made us comfortable that, if this leak were to get worse, it would be acceptable to fly.”
The lack of leaks in other thrusters led NASA to conclude that there is no systemic problem. “We have an analysis that backs that up, that says we don’t expect the others to leak,” Stich said. “That’s the confidence that we have that we don’t have a common-cause failure mode.”
“This is really not a safety-of-flight issue for ourselves, and we believe that we have a well-understood condition that we can manage,” Nappi concluded.
While the study of the helium leak was ongoing, Stich said engineers performed a review of the rest of the propulsion system “just to make sure we didn’t have any other things that we should be concerned about.”
That review did turn up something he called a “design vulnerability” with Starliner’s propulsion system where, in a rare circumstance, the spacecraft would not be able to perform a deorbit burn if two adjacent “doghouses” that contain RCS and larger orbital maneuvering and attitude control (OMAC) thrusters failed. That failure, he said, would knock out enough OMAC and RCS thrusters to prevent existing backup plans for carrying out a deorbit burn from being implemented.
Engineers developed a new deorbit reentry mode that would use two burns of four RCS thrusters in that scenario. Such a failure mode would be “very remote,” Nappi said, appearing in less than one percent of potential combinations of failures.
“It’s a pretty diabolical case,” Stich said.
It raised the question, though, of why the design vulnerability was not found in reviews during Starliner’s lengthy development. “The helium leak itself caused us to look a little more in detail” about how elements of the propulsion system interact, Stich said. “It just took us a little time to figure this out now. The question is, should we have seen this earlier? Maybe in a perfect timeframe we might have identified this earlier.”
Officials emphasized this was part of the knowledge gained during a test flight program. Nappi said Boeing was looking at several permanent fixes to the problem, using some combination of hardware changes and software modifications, for later Starliner missions.
After two weeks of intensive work on Starliner, teams are taking a break over the Memorial Day holiday weekend before resuming final preparations for the launch, scheduled for June 1 at 12:25 p.m. Eastern. That includes a new flight test readiness review on May 29, followed by rollout of the Atlas 5 carrying Starliner to the launch pad May 30.
There are backup launch opportunities for the mission June 2, 5 and 6. However, limited-life items on the Atlas 5 rocket could result in a much longer delay if Starliner does not launch much longer after that.
Gary Wentz, vice president of government and commercial programs at United Launch Alliance, said the company regularly tests ordnance used in flight termination and other systems on the rocket, but those “expire” later in June and July and would have to be replaced.
He suggested that work could lead to a longer delay as ULA reshuffles its launch manifest. “Some of it will come down to a prioritization of missions throughout the rest of the year,” he said, with the company working with NASA and the Space Force on determining which missions should fly in what order.
The two NASA astronauts who will fly on Starliner on the CFT mission, Butch Wilmore and Suni Williams, are back in Houston but remaining in pre-flight medical quarantine as they await their next launch opportunity. Stich said they were taking the delay in stride. “They’re in good spirits,” he said. “I think they worry about us sometimes more than we worry about them.”
