by Tim Fernholz
While the money was helpful, it wasn’t really the point. Indeed, when Blue hit the first milestone of its partnership agreement, one of the team members called up Dennis Stone, the NASA project manager, and asked him to make the appropriate payment. “We said, ‘Give us an invoice,’ and they went ‘Huh?’” Stone told me. “They never invoiced anybody; they never had to.”
The bigger gain for Blue was access to NASA’s repository of experimental and operational rocket data, its experts in every field of space technology, its best practices, and its test facilities. This kind of “soft” exchange was immensely valuable to all of the new space companies, which found themselves calling on the collected institutional wisdom of the US space program more often than they expected.
In 2011, NASA requested a second round of CCDEV proposals, with a much bigger pot of money. It would be the first time, but hardly the last, that Blue Origin and SpaceX would compete head-to-head. Both companies put forward plans to continue leveraging their human transportation strategies—for SpaceX, the Falcon 9 and the Dragon; for Blue Origin, the New Shepard.
This time around, NASA was starting to think broadly about how to seed a variety of ways to transport people into space. They found that SpaceX and Boeing were clearly the furthest along toward completing plans to fly people into space in capsules, with hardware development under way. Sierra Nevada won a share of the contract because of its unique spacecraft, called a lifting body because it had an aerodynamic shape that would allow it to glide back to earth. These three companies received significant development awards of $75 million, $92 million, and $80 million, respectively.
Blue Origin was the fourth player in the game. It, too, had a unique design, and a different business plan to “walk before you run” with suborbital flights ahead of more ambitious missions to carry astronauts into orbit. It hadn’t been easy to win over the NASA selectors; the initial evaluation of Blue’s technical plans found a significant lack of detail, down to a “lack of understanding of NASA’s draft human certification requirements.” The company improved its proposal with the agency’s feedback, but even then, the evaluators noted a “failure to identify long term development risks” and said that “investments may not accelerate development of a [crew transportation system].”
The company had one big advantage: in NASA-speak, Blue Origin “demonstrated realism in future markets through diminished dependency on early revenues for sustainability, demonstrating commitment to a long-term strategy that was unique among all proposers.” In plain English: Blue knew it was not going to make any money soon, but had enough of Bezos’s capital behind it to keep going longer than the other competitors. NASA gave Blue Origin $22 million to work on the New Shepard—less than they asked for, but arguably more than they needed. That year, a prototype New Shepard booster was destroyed in its second test flight when it veered off course and Blue’s engineers had to shut off its engines to ensure it didn’t leave the test range. The company released a video of the first test, which saw it rise to 450 feet before settling back down again; no longer aping the pyramidal DC-X, the booster had become a rounded-off cylinder, with four stubby wings protruding from the aft section. The next year, Blue preformed another key test, demonstrating that its passenger-carrying capsule could escape a failed launch with its own abort jets.
In 2012, NASA rolled out the next round of the program. This time participants needed to demonstrate a proper road map toward an integrated system—rockets, spacecraft, ground operation, mission control, the whole shebang—that could fly astronauts to the space station. Blue didn’t throw its hat in the ring, most likely because it knew it could not feasibly claim, even by the loose standards of aerospace project management, the capacity to fly astronauts to the station by a reasonable date. This left Boeing, SpaceX, and Sierra Nevada splitting more than $1.1 billion in development funding ahead of the final selection for the Commercial Crew program.
Blue wasn’t out of the space race; it was just out of NASA’s commercial partnerships. Boeing would always come after these contracts; its raison d’être was to do engineering work for the government. SpaceX had already come as far as it had thanks to NASA’s support, and knew that their fruitful partnership could smooth the path toward space because Musk’s wealth was not unlimited. By contrast, Blue didn’t need to worry about the US government’s money or its scheduling needs.
Just before the final space shuttle mission took off, in 2011, mission commander Chris Ferguson was approached by Jerry Ross, another astronaut, who was in charge of prepping crews for their missions. Ross handed Ferguson a small American flag.
“No kidding: orders of the White House, this has to go up to the station, and you’ve got to leave it there,” Ferguson told me of Ross’s explanation. The flag had flown on the first-ever shuttle mission. Now it would go back to the ISS to wait for the first of the Commercial Crew vehicles to come collect it as a symbol of America’s return to human spaceflight. Then the flag would be put on Orion to go into deep space. (That’s assuming, Ferguson notes, that they can track it down in the station more than seven years later.)
He didn’t realize that he would be one of the architects of those commercial vehicles, playing the first game of capture the flag in space. In 2014, NASA chose from among the three finalists it had identified two years before: SpaceX and Boeing were tapped to build human transport systems to the ISS. Boeing was seen as having the strongest proposal, according to Bill Gerstenmaier, but SpaceX had proven itself through COTS and offered a lower price, $2.6 billion. Boeing’s program cost $4.2 billion. Each company was eager to be the first to take astronauts to the space station.
Ferguson now works at Boeing, managing the aerospace giant’s first step into commercial space services. Boeing’s vehicle is dubbed the CST-100 Starliner. The biggest difference Ferguson noticed between NASA and Boeing was that the space shuttle “was a cost-plus environment. If we thought we needed it, by God, we went and did it.” Now, under fixed-price Commercial Crew contracts, those “ornaments” are stripped away to develop the safest vehicle that is “an order of magnitude” cheaper. The former pilot set to work building a new vehicle and ignored questions about whether he would be one of the first Boeing test pilots to fly it.
SpaceX, meanwhile, dove into the task of upgrading its cargo capsule to carry humans. The vehicle, which he showed off at a flashy Washington event in 2014, had all the styling of a Musk product: futuristic chairs and fold-down touchscreens made it seem as if the spacecraft would be controlled by a giant iPad. (SpaceX also has an in-house former astronaut, Garrett Reisman; the company declined to let him speak with me.)
NASA appointed four veteran astronauts with backgrounds as test pilots—Robert Behnken, Eric Boe, Sunita Williams, and Douglas Hurley—to be the first astronauts to fly private to space. They shuttled back and forth between the two companies, evaluating designs and offering insight—“SpaceX in the morning, Boeing in the afternoon,” as Behnken told me. They also began testing flight software and training for flight operations as the vehicles’ designs were finalized, bringing more than sixty years of collective space experience to bear on the challenges faced by the companies. “The providers don’t have a large pool of astronauts to draw on,” Behnken says. “They need us to do those evaluations and provide that data to them.”
In the meantime, US astronauts were stuck flying the Russian Soyuz to the ISS until the two companies got their designs off the ground, and delays were building up. “In a perfect world, we would have stepped off one rocket, onto another,” Behnken noted ruefully.
Politics was part of the problem. The compromise worked out in 2010 between the Obama team and the powerful lawmakers who controlled space funding had been, at best, incomplete. During the negotiations, Garver told me, “we never said at what level [the programs would be funded]—we put in these budgets, and they’d cut it in half.” Later, investigators found that between 2011 and 2013, the Commercial Crew program had received only 38 percent
of the expected funding, which led to two years of delays.
But the engineering challenges were serious as well. Putting humans on the rocket demanded far more care and planning; the eventual requirements document put together by the space agency for Boeing and SpaceX reached 297 pages. Much of the focus was on the basics learned during the previous commercial program: how much mass the vehicles could carry, how they would approach the ISS. But human factors were also at play. One requirement was that the spacecraft provide a supply of drinking water to the astronauts on board, which needed to be tested to ensure that it was not contaminated by bacteria or fungus. Others mandated that the vehicle be protected from extreme spins and g-forces, uncomfortable vibrations, and noises loud enough to cause injury.
Sometimes the document reminds you that astronauts, for all their elite training and dedication, are people just like you: they needed a private line to their doctor on earth when discussing health issues. Privacy concerns had already cropped up on the cargo missions: Pettit, the astronaut who caught the Dragon by the tail, told me that NASA had negotiated with SpaceX about microphones in the vehicle to monitor it during flight. Pettit wanted to ensure that whatever it overheard on the station would be kept private; after all, “when we are unloading the vehicle, we are the equivalent of orbital longshoremen.”
Above all, the space agency was concerned with safety. The rockets that the two vehicles would fly on, the ULA Atlas V and SpaceX’s Falcon 9, had to be completely reliable. For SpaceX, that meant addressing cracks found in its turbopumps. The company began working toward a final version of the Falcon 9, called Block V, that would meet NASA’s standards. Boeing had to deal with vibration problems and cut mass from its vehicle, but another big question mark was ensuring that the Russian engines in the Atlas V were safe for humans. Access to data about their design was restricted by agreements between the United States and Russia, but Boeing urged NASA to use performance data gained during flights.
When it came to the capsules themselves, both companies faced a magic number called the “loss of crew” metric, which expressed the probability of astronauts being killed by a failure. By the end of its tenure, the space shuttle’s loss-of-crew number was about one in a hundred. At first, engineers went big, looking for a vehicle ten times safer than the shuttle—one in a thousand.
“You have goals, and then you have engineering reality,” Kathy Lueders, the respected NASA manager in charge of the Commercial Crew program, told me. The biggest problem was micrometeoroids and orbital debris from old spacecraft. When the shuttle maneuvered in space, it always flew backward—engine first—to protect the crew; a minuscule piece of space junk impacting at orbital velocity could threaten even a well-protected spacecraft. To make that one-in-a-thousand standard, Lueders told me, “you would probably have to have a spacecraft that had so much tile on it that you would never get it off the ground.”
The Constellation program was able to reach a loss-of-crew metric of one in 270, and even that was “really, really tough.” For Boeing’s and SpaceX’s capsules, it seemed increasingly likely that NASA would have to accept a vehicle below that standard. Lueders pushed the companies to meet it nonetheless, as well as figure out operational techniques, like on-orbit damage inspections, that would give them a bigger margin of error.
Ultimately, however, it was another case where pushing the limits of technology and physics makes assessing risk with any certainty difficult. The metric itself conveyed a false sense of reassurance, simply because the amount of data we have about spaceflight—compared with something like passenger aviation—is very limited. When NASA redid its risk models for the first space shuttle flight with the full program’s data, it found that the LOC number had been one in twelve.
“We may have done ourselves a disservice in the agency where we don’t talk about how risky this environment is,” Gerstenmaier told me. “One in 275, the general population sees that as an absolute value, and they don’t see the uncertainty in that number.” In an editorial about the “interminable management of risk” in human spaceflight, he noted that “no human spaceflight mission will be absolutely safe by any reasonable definition of that word,” but that, “in order to demonstrate to everyone that the benefits outweigh the risks, we must be allowed to perform.”
I asked Behnken, who would be among the first astronauts to fly on these vehicles, how he thought about the risks involved.
“It is important for the risk taker, and at some point, that’s my job: to figure out what they are going to accept,” he told me. “There is always more time, and if you spend more money, someone will always communicate that there is a way to make it safer, [but] space is a very unforgiving environment, and that’s not going to change.”
Behnken, Ferguson, Reisman, and many of the other astronauts in the program had lived through the Columbia experience. “That’s an important piece that we all share,” Behnken said. “It’s not hypothetical.” But the challenge of returning human spaceflight to the United States was too attractive for them to pass up. “There’s no more exciting time than where we’re at right now,” he said, noting that there were three human-rated spacecraft in development, including one, Orion, intended to push beyond low earth orbit.
All this risk was met by a commensurately larger NASA bureaucracy. Where the COTS program had been run by a dozen people at Johnson Space Center, a bit outside the mainstream, the Commercial Crew program, as formally instituted, comprised three hundred NASA employees at Kennedy Space Center. Some early COTS proponents feared that the program was migrating back to the traditional NASA approach. “This is how NASA does business,” Alan Marty, the COTS program’s in-house venture capitalist, told me. “There is a natural culture within NASA that says if something like COTS Cargo was done with $500 million and thirteen people, surely, if we put two or three billion dollars into the program and have hundreds of NASA people, surely it will be better. It’s the complete antithesis to Silicon Valley thinking, and a complete antithesis to what we tried to build at COTS Cargo.”
Mike Griffin, the NASA administrator who had kick-started the COTS program, had also become a critic of NASA’s approach to flying astronauts on privately owned spacecraft. Though it was a fixed-price contract, he viewed the Commercial Crew development rounds as a subsidized approach, akin to the old traditional contracts, but without government control. “NASA is supplying every dime of money to both Boeing and SpaceX,” he told me. “All that money for development, it’s not being done in a way that allows NASA or other government managers to direct the contractor what to do. As best anyone can tell, companies aren’t kicking in anything. When they’re done, they own the design.”
This wasn’t how NASA would portray the program; before it signed the final crew contracts, it had worked through detailed requirement sets with the participating companies, and afterward NASA was able to mandate additional tests and send inspectors to witness hardware manufacturing. Still, the two companies had far more leeway in how they approached design than prior NASA contractors. And Griffin’s critique homed in on a real challenge in designing the program to leverage commercial incentives. While the space taxi program had funded the development of rockets that could do more than simply take cargo to the space station, Griffin argued in a 2012 speech that “human spaceflight in particular is, for the present and near future, one of numerous ‘products’ not furnished by the marketplace, one of those things which, if we desire it, can only exist if government pays to build what is desired.”
Others disagree. Bretton Alexander, a NASA official who worked with Griffin on COTS and later joined Blue Origin, sees human spaceflight as far more commercial than cargo, pointing to the burgeoning interest in space tourism. “The first market really is people, and until you get there, it’s just a government infrastructure,” he said in 2013. He also suggested that Griffin might have other reasons to criticize the commercial programs—they meant the end of Constellation. “If you had existing launch vehicles that you
could put people on top, then why did you need to build Ares I? The whole program would fall apart.”
Neither NASA nor the two competing companies were able to provide me with a specific estimate of how much private investment went into the Commercial Crew vehicles. Griffin’s concerns resonated with many who wondered if NASA, rather than finding a low-cost way to reach space on behalf of the American people, was simply subsidizing the space dreams of already wealthy dreamers. “The only outcome of such behavior that can possibly occur is that a technical, operational, or business failure will occur—and NASA will be held accountable for the failure, because public money was expended,” Griffin warned in his speech.
Lueders told me that the cargo program gave NASA faith in its approach to crew. “In 2008, you know, we thought cargo vehicles were a Hail Mary,” she told me. “It’s the same thing with crew: we’re betting on industry to come through for us. I’m pretty sure in the next few years we’re going to be sitting here—and I don’t know if it’s Boeing or SpaceX, and don’t ask me exactly the day they’re going to fly—but I know that it’s going to be an amazing thing because of what we bet on in 2011.”
As competition between Boeing and SpaceX geared up, Blue Origin wasn’t forgotten by the US space community.