When Boeing’s long-delayed Starliner spacecraft docked at the International Space Station (ISS) on June 6, Boeing executives and NASA leaders must have sighed with relief. But the assertions of mission success might have been premature.
NASA, which currently relies exclusively on SpaceX and its Dragon capsule to ferry U.S. astronauts to the ISS, wants the Boeing ship to become a dependable part of its Commercial Crew Program. Under that cost-saving initiative, NASA pays private companies to fly its cargo and personnel for a set fee. So far, SpaceX is the only firm approved to carry NASA astronauts. For Boeing, a successful crewed Starliner test flight would show that the company can compete as a space transportation supplier. And, after years of safety concerns involving its civilian aircraft—including a Capitol Hill grilling this past Tuesday—the aerospace giant doubtless would welcome some positive press.
Starliner’s first crewed flight to the ISS was marred by “some worrisome hitches,” as I noted after the successful docking. Nonetheless, most observers saw the mission as broadly successful up to that point and believed that NASA would soon approve the vehicle for human spaceflight. That may still be the case, but in the days since Starliner bellied up to the ISS, the problems it experienced enroute have faced growing scrutiny. NASA and Boeing are now conducting a battery of diagnostic tests on the vehicle. To accommodate these unplanned technical reviews, Starliner and its crew—veteran astronauts Suni Williams and Butch Wilmore—have seen their return flight to Earth pushed back three times.
In a press conference on Tuesday, three NASA leaders and the head of Boeing’s Starliner project discussed the reasons for the latest delay. Steve Stich, program manager for NASA’s Commercial Crew Program, said the team needed “a little bit more time to look at the data, do some analysis, and make sure we’re really ready to come home.”
This mission’s problems began even before launch, when the ground crew noticed a small helium leak in the system that pressurizes the vehicle’s reaction control system (RCS) thrusters (tiny rockets that fine-tune the craft’s position in orbit and during docking). One small leak isn’t dangerous in itself—helium is an inert gas—but NASA delayed the launch, giving engineers time to “get comfortable” with the unexplained emission, in the words of an Ars Technica reporter. In the end, they decided the launch could proceed without repairing the leak.
Such decisions aren’t unusual. But to students of NASA history, the idea that engineers should “get comfortable” with a small, poorly understood problem calls to mind a painful lesson from the agency’s space shuttle era. In the early years of that program, NASA engineers gradually learned to tolerate tiny leaks of hot gases from the shuttle’s booster rockets. On January 28, 1986, one of those leaks grew enormous and destroyed the space shuttle Challenger. Sociologist Diane Vaughan, whose 1996 book The Challenger Launch Decision is a landmark in the scientific study of disaster, called this process “the normalization of deviance.”
Whether or not Boeing and NASA engineers were normalizing deviance when it came to the helium leak, they clearly underestimated the problem. Once Starliner was in orbit, four more helium leaks were discovered over the course of a few days. The crew members can turn off the helium flow while docked, but they will have to turn it back on when it’s time to begin their descent. In Tuesday’s press conference, NASA’s Stich said that Starliner has a 70-hour supply of helium at current leak rates and needs only seven hours’ worth to complete the process of undocking and preparing for reentry. It’s a comfortable margin—if nothing else goes wrong. Boeing’s Mark Nappi added that the leaks “are stable” and are less severe than originally measured.
Engineers are also grappling with a potentially more serious problem: during Starliner’s delicate approach to the ISS, five of the ship’s 28 RCS thrusters failed. Docking had to be postponed while the crew and engineers on the ground sorted out the problem. “I think we’re missing something fundamental that’s going on inside the thrusters,” NASA’s Stich said soon after the vehicle docked successfully.
Since then, engineers and the Starliner crew have performed hot-fire tests on four of the five troublesome thrusters. In the most recent press conference, Stich said NASA now feels “very confident” in the system’s performance. (The most problematic RCS thruster will not be turned on for the return trip.) Asked whether Starliner is safe enough to bring its crew back to Earth, Stich said, “We would be very confident in putting Butch and Suni in the vehicle and returning them if we need to.”
Individually, the glitches afflicting Starliner might not be particularly severe. After all, one key objective of a test flight is to find and address hidden issues. But in the context of the program’s many technical setbacks and delays, these in-flight gremlins seem to confirm worries that Boeing’s Commercial Crew Program is technologically challenged. Starliner’s first uncrewed test flight went awry due to major programming errors. The second, while successful, was also plagued by thruster problems. This series of technical woes stands in marked contrast to the reliable performance of SpaceX’s Dragon spacecraft. Dragon has been delivering U.S. astronauts to the ISS since 2020 and has made 42 trips to the station in all. In fact, a Dragon capsule is attached to the ISS today. If Starliner’s issues rule out a safe return, that SpaceX ship would likely be Williams’s and Wilmore’s lifeboat back to Earth.
It is far more likely that Starliner’s glitches will get sorted out, at least enough to ensure a safe flight home. In Tuesday’s press conference, Boeing’s Nappi described the process as one of “learning and additional fine tuning.” He doesn’t anticipate the need for another test flight to receive NASA’s go-ahead to fly astronauts on routine missions. As of today, Starliner and its crew are slated to start their return to Earth on June 25. But then NASA and Boeing engineers will have months of work ahead to determine why so many things seemed to go wrong at once on their maiden crewed flight.
All this must be frustrating for Boeing. But it’s also a headache for NASA, which has always envisioned Commercial Crew as a program that includes multiple vendors certified to fly crewed missions. The Starliner spacecraft will likely get that certification eventually. But it’s hard to see NASA choosing to fly the Boeing craft any more than necessary when it can always turn to the proven SpaceX Dragon instead. In fact, if anything, Boeing’s “successful” flight to the ISS just confirms that SpaceX remains the uncontested global leader when it comes to making human spaceflight affordable, reliable, and routine.
Photo by CHANDAN KHANNA/AFP via Getty Images