SpaceX’s Riskiest Business

Launching astronauts into space means being responsible for their safety.

A SpaceX rocket blasts off
Bill Ingalls / NASA

SpaceX’s first attempt to fly astronauts to space and back was, from start to finish, a success. The launch into orbit—seamless. The spacecraft’s arrival at the International Space Station—smoothly done. On return, the capsule, buffeted by billowy parachutes, coasted through the sky, toward the waters off the coast of Florida—a vision of a new era of American spaceflight.

But later, when technicians inspected the capsule up close, they saw something they didn’t like. On the astronauts’ way home, the heat shield, the hardware that protects the capsule as it plunges through the atmosphere, eroded more than SpaceX had expected.

When Nicole Jordan, an engineer at NASA, heard about that, she shivered. Her colleagues at NASA did too. “The hair stood up on the back of their neck,” Jordan told me. “Everybody understood the potential seriousness of a problem.”

Jordan was an intern at the agency in 2003, when the space shuttle Columbia broke apart during reentry and its crew was killed. Now she’s a mission manager on the program that supports SpaceX’s astronaut flights, and the accident report compiled after Columbia was required reading for her job at Mission Control in Houston. She works alongside other engineers who lived through the tragedy—and, 17 years before that, through the Challenger explosion. For many engineers at NASA, these disasters define their work as much as the successes. Every astronaut launch involves risk; every launch puts lives at stake. America is poised to launch humans into space on a regular basis again, but the people behind the effort can never act as if a safe flight is a given.

These days, SpaceX, Elon Musk’s rocket company, is doing the work the space shuttles once did. Tonight, if weather conditions are good enough, SpaceX will launch astronauts from Cape Canaveral, Florida, again—three from NASA, and one from Japan’s space agency.

As NASA ended the longtime shuttle program nearly a decade ago, it began another: an effort to use private companies to transport astronauts to and from space on the agency’s behalf. The successful flight of two NASA astronauts, Doug Hurley and Bob Behnken, earlier this year marked the most significant milestone in the program so far. It was the first time since 2011 that NASA astronauts launched from the United States, in an American-built vehicle, instead of from Kazakhstan, in a Russian-built vehicle.

SpaceX is now responsible for astronaut safety to a degree no private business has ever experienced. And as the company moves toward a future of regular astronaut flights, the lessons of the space-shuttle era are at the forefront. SpaceX employees have received briefings about the aftermath of the two space-shuttle disasters, which killed 14 astronauts, as well as a launchpad fire during the Apollo program that killed three. They’ve heard from veteran NASA employees about the mistakes the agency made and how to avoid them. Jordan said that when new engineers joined the project, they were given a tour of a room at the Kennedy Space Center, closed to the public, where shuttle debris was displayed, so that they wouldn’t forget the perilous nature of their work.

SpaceX has said that its first passengers were not in danger during their descent, that there was “nothing to be concerned” about. But the company decided to redesign part of the heat shield for future missions. The company also made a tweak to the capsule’s parachutes, which deployed closer to the water than engineers had anticipated. And it replaced two engines on the booster of its Falcon 9 rocket after a problem on a different, uncrewed mission shut down the flight right before liftoff.

Like the NASA engineers, SpaceX employees are acutely aware of their responsibility to these astronauts. When Benji Reed, the senior director for human spaceflight programs at SpaceX, comes to the Cape, he visits a set of seven oak trees—one for each astronaut who died on the Challenger—arranged in a circle on the grounds of the Kennedy Space Center. For Reed, who was in sixth grade when Challenger exploded, the small memorial reminds him of the weight of his work.

“We hold the lives of people in our hands; we transport them into space, stay there with them to be a lifeboat if they need it, and then bring them home to their family,” Reed told me this week. “We take it very, very seriously.”

In the early 1980s, after the country had put a dozen men on the moon, America turned its attention closer to home. NASA debuted space shuttles meant to carry astronauts into orbit as often as every week to conduct research, deploy satellites, and, eventually, assemble the International Space Station. Within a few years of the shuttle’s debut, astronauts were soaring into the sky every couple of months—not at the pace the agency had envisioned, but frequently enough that launches eventually stopped making front-page news. To renew interest—and to show Congress the program was worth the budget—NASA decided to fly its first non-astronaut, a teacher.

To the astronauts, the highly publicized effort made the shuttles seem safer than they really were. “We knew it was risky business,” George “Pinky” Nelson, a retired NASA astronaut, told me recently. Nelson flew three times in that decade, including the mission before the doomed Challenger flight, and the mission that resumed the program two years later.

The mission known as “the teacher flight” ended in tragedy just over a minute after liftoff in January 1986. Thousands of spectators, including the crew’s families and friends, stared in confusion and horror at the twisty contrails hanging in the sky. The cause of the disaster was eventually revealed to be rubber rings inside the rocket boosters, which turned brittle in very cold temperatures. Engineers at Morton Thiokol, a NASA contractor, had known about this design flaw for months, had seen it over and over in the boosters when they were recovered from launches and inspected. Engineers told managers at NASA, who warned their higher-ups that damage to the rings could lead to “catastrophic failure.” Program managers even raised concerns about the rings at a meeting the night before the launch, one of the coldest days that month. But top NASA officials overruled them, deciding the risk wasn’t as serious as the engineers’ data suggested.

After Challenger, astronauts launched in pressure suits instead of jumpsuits, and officials vowed not to repeat their mistakes. Years passed without incident. Wayne Hale, a former space-shuttle program manager and flight director, remembers the flight-readiness review—a formal meeting where managers decide whether a mission is ready to launch—before Columbia’s final flight. The atmosphere in the room, he recalled in an essay, was “oppressive.” “The general impression was that people were not to question topics outside their area of concern or expertise,” he said. As with Challenger, the flight was felled by a well-documented issue that NASA had failed to address, believing the risk wasn’t serious enough.

The flight-readiness review for tonight’s launch, held earlier this week, unfolded quite differently—a deliberate effort to avoid the mistakes of the past. “We had some very healthy discussions about very complex technical issues we’ve been dealing with recently,” says Ralph Roe, the chief engineer at NASA who oversees the technical readiness of all of the agency’s programs. “We’re asking each other questions; we’re challenging each other.”

Hale and Roe are among the NASA veterans who worked at the agency during Challenger and Columbia and who have provided SpaceX workers with lectures about what the agency learned from the shuttle disasters. “You don’t want other people to go through what you went through,” Roe says.

Although the space agency helped fund the astronaut-transportation system, SpaceX is not a traditional NASA contractor. Musk’s engineers designed the spacecraft from top to bottom, including the flight software on the computers and the toilet inside the capsule. Mission Control is in Hawthorne, California, at SpaceX headquarters, where the company’s technicians guide astronauts through their mission from start to finish. For NASA, adjusting to this arrangement—ceding control over operations, learning to trust SpaceX—wasn’t easy. Agency officials felt they “knew how to do spaceflight better than anyone,” Phil McAlister, NASA’s director of commercial spaceflight development, told reporters recently.

But NASA judges whether the SpaceX systems are safe, and its personnel can intervene. NASA engineers work closely with their SpaceX counterparts, and they have access to the company’s most detailed plans. “You can literally see work orders and steps as they get checked off electronically; we can see all of their issues as they are written; we can see all of their risks,” Jordan, the NASA mission manager, said.

The SpaceX system is far less complicated than the shuttle. And in the case of an emergency right after liftoff, the SpaceX astronaut capsule can push itself away from the rocket and toward safety. The space shuttle didn’t have that option. “I think they’re on a safer vehicle than the space shuttle was,” Nelson, the retired shuttle astronaut, said. “If something went wrong [on the shuttle], you were stuck. There was no way out.”

But human spaceflight is still risky, and many of the pressures that plagued the space-shuttle program remain. “You’re never going to have enough money, you’re never gonna have enough time, and there’s going to be more technical challenges than you’ve ever dreamed of,” Roe says. “It’s how you deal with those pressures that’s important. And looking back now, we didn’t handle those pressures very well.” When engineers in 1986 urged management to delay the Challenger launch because of the cold temperatures, one frustrated official spat out, “When do you want me to launch, next April?” Today, leaders at NASA and SpaceX say often that they will launch only when they’re ready. And Musk has continued his tradition of sending a brief company-wide email before a launch encouraging anyone who doesn’t think the rocket should launch to tell him.

“I have heard our managers repeatedly emphasize the importance of speaking up, and that we’re not going to launch until we’re ready,” Jordan said. “I truly believe if somebody really thought there was something wrong, and they were really worried, that they would say something. I just can’t imagine somebody not listening to that. And even if they didn’t, I can think of a lot of ways that that person could go around and get that message out.”

As SpaceX notches more launches, the company will need to preserve that sense of urgency. The shuttle disasters revealed that complacency can be deadly. Each successful flight bolstered NASA’s confidence that the next one would unfold just as seamlessly. Engineers had noticed, in many flights, that some foam insulation on a shuttle’s external fuel tank peeled away during launch; it didn’t seem to matter until some of that foam damaged the orbiter’s heat shield on Columbia’s last flight. SpaceX is the beginning of what could be a years-long program, flying not only astronauts, but private citizens too. “I’m not so worried here at the beginning, where people are very focused on safety,” Hale told me. “But as we go forward in time, we need to make sure that the focus on safety doesn’t evaporate.”

It might be tempting to believe that, 17 years since NASA’s last tragedy, the work of sending people to space is easier, safer, more reliable. After all, humankind sent people as far as the moon 50 years ago, and surely reaching the International Space Station, just 260 miles above Earth, is a simpler task, especially with someone as brilliant and driven as Elon Musk in charge.

But this is a false reality. Spaceflight is just about as risky as it was during the moon program. And although space agencies and rocket companies have to believe they can push the boundaries of human accomplishment, there’s a fine line between encouraging that confidence and letting it cloud judgment.

Musk himself knows that. “I’m not very religious,” he told reporters the day Hurley and Behnken returned to Earth. “But I prayed for that one.” At the end of this decade, SpaceX missions might seem routine, perhaps not even very newsworthy. But for those in the space business, a tragedy can seem like not a matter of if, but when.

“I don’t know whether it’ll be going to the space station, or when we try to go back to the moon or to Mars,” Hale said. “As much as we would like to prevent them, there will be another bad day. And we need to make sure that we believe what we’re doing is worth that price.”