Back to business(es)by Jeff Foust
|
| “It was a really surprising problem. It’s never been encountered before in the history of rocketry,” Musk said in November. |
For most of 2016, the attention SpaceX received for its Falcon 9 launches involved what happened about eight to ten minutes after liftoff, when the company attempted to land the rocket’s first stage on either a ship at sea or a pad back on land. The company closed out 2015 with its first-ever landing, on land, and finally started getting into a rhythm of landings on solid ground and ship decks as 2016 progressed.
That ended on the morning of September 1, when a Falcon 9, being fueled for a static-fire test at Cape Canaveral a couple days before the scheduled launch of the Amos-6 communications satellite, suddenly exploded. The Falcon 9 was grounded as SpaceX investigated the costly failure.
Return to flight
In the months that followed, SpaceX released a trickle of information about the failure investigation, implicating by late October a composite overwrapped pressure vessel (COPV) used to store helium pressurant in the liquid oxygen tank of the rocket’s upper stage. But the exact cause of the failure remained unclear.
“It was a really surprising problem. It’s never been encountered before in the history of rocketry,” SpaceX CEO Elon Musk said in an early November interview with CNBC. “This was the toughest puzzle to solve that we’ve ever had to solve.”
On January 2, SpaceX announced the conclusion of its investigation into the failure. That investigation confirmed that one of three COPVs in the liquid oxygen tank failed, leading to the explosion.
Each COPV tank, SpaceX said in its statement about the investigation, consists of an aluminum liner and a carbon composite overwrap. Other COPVs from the vehicle recovered after the accident showed evidence of buckles in their aluminum liners. That buckling alone is not sufficient to cause them to fail, the company said, but would allow liquid oxygen to pool in the void between the liner and overwrap created by the buckles. “When pressurized, oxygen pooled in this buckle can become trapped; in turn, breaking fibers or friction can ignite the oxygen in the overwrap, causing the COPV to fail,” the company stated.
| “We have been working endless hours for the last eight years to get to this day, and to finally be here with ten Iridium NEXT satellites successfully launched into low-Earth orbit is a fulfilling moment,” said Desch. |
SpaceX noted that it did not come up with a single root cause for the failure, but noted that each of “several credible causes” for the COPV failure involved oxygen, in either liquid or solid form, pooling in those buckles. The corrective actions, regardless of the specific failure mode, were the same: “changing the COPV configuration to allow warmer temperature helium to be loaded, as well as returning helium loading operations to a prior flight proven configuration.” In the long term, the company said it will “will implement design changes to the COPVs to prevent buckles altogether, which will allow for faster loading operations.”
That investigation and corrective action plan was approved by the FAA four days later when it issued a launch license for its return-to-flight mission, the first of seven launches of Iridium NEXT communications satellites from Vandenberg Air Force Base in California. That launch, once set for January 8, was postponed to January 14 because of poor weather and range issues at the launch site.
That change in pre-launch operations was evident in the preparations for the launch on Saturday morning. In previous launches of this version of the Falcon 9, SpaceX started loading both RP-1 fuel and supercooled liquid oxygen 35 minutes before launch, a process intended to keep the liquid oxygen chilled to near the freezing point to increase its density and improve the vehicle’s performance. On the Iridium launch, RP-1 loading started 70 minutes before launch and liquid oxygen loading 45 minutes before launch, a slower pace presumably intended to avoid the buckling issues identified in the failure investigation.
In the end, the launch went as planned: the Falcon 9 lifted off in its instantaneous launch window at 9:54 a.m. Pacific time. An hour after launch, the ten Iridium NEXT satellites started deploying from the Falcon 9 upper stage, one every 100 seconds.
There was a tense moment when a ground station problem prevented real-time telemetry of the deployment from reaching mission control, but a playback of data when the stage passed in range of another ground station confirmed all went well. “So it’s a clean sweep: ten for ten!” said SpaceX’s John Insprucker on the company’s webcast of the mission.
The launch was a relief for Iridium, whose original satellites, launched in the late 1990s, are reaching the end of their lives. Failures of some satellites have created a few gaps in coverage in the overall system, which this initial batch of next-generation satellites will help fill.
“Today Iridium launches a new era in the history of our company and a new era in space as we start to deliver the next generation of satellite communications,” Iridium CEO Matt Desch said in a post-launch statement. “We have been working endless hours for the last eight years to get to this day, and to finally be here with ten Iridium NEXT satellites successfully launched into low-Earth orbit is a fulfilling moment.”
And, yes, there was a landing, too: the Falcon 9 first stage landed on the ship Just Read the Instructions in the Pacific, with cameras on the booster capturing an almost uninterrupted view of its descent and landing. It was the seventh successful landing of a first stage, although first time SpaceX had landed a stage after launch from Vandenberg.
Looking ahead to 2017
The launch starts what SpaceX expects to be a busy year for the company. Later this month, it’s scheduled to launch the EchoStar-23 satellite from Florida, in what will be the first Falcon 9 launch from Kennedy Space Center’s Launch Complex 39A. That pad, renovated by SpaceX for Falcon Heavy and crewed Falcon 9 launches, will be pressed into service for other Falcon 9 missions because of damage to Space Launch Complex 40 from the September pad explosion.
Another milestone coming up early this year is the first flight of a reused—or “flight proven,” as SpaceX put it—Falcon 9 first stage. In August, days before the pad explosion, SpaceX and SES announced that the SES-10 satellite would be the first payload to fly on a reused booster. At the time, the launch was planned for the fourth quarter of 2016, but delayed because of the accident. That launch is now expected this quarter: over the weekend, SES announced the satellite’s manufacturer, Airbus Defence and Space, had shipped the satellite to Cape Canaveral in preparation for the launch.
| As SpaceX presses ahead with commercial crew in 2017, it will have to deal with concerns that link back to the pad explosion. |
This year is also supposed to be the year the Falcon Heavy, finally, takes flight. In 2011, when SpaceX formally announced development of the vehicle (after years of suggestions that a heavy version of the Falcon 9 was planned), the company planned a first launch in 2013. That date has slipped, in part because of development issues with the vehicle, and also because of broader delays caused by Falcon 9 launch failures last year and in June 2015.
Now, SpaceX is planning a first launch of a Falcon Heavy, on a demonstration mission, by about the middle of the year. That would be followed by the US Air Force’s Space Test Program 2 mission, likely no earlier than September. Getting Falcon Heavy flying, and flying successfully, will ease concerns from commercial customers, some of whom have in recent months booked alterative launches for their spacecraft because of the extended Falcon Heavy delays.
SpaceX is also grappling with delays in its commercial crew program. Prior to the pad explosion, SpaceX was planning to carry out two test flights of its Crew Dragon spacecraft, one uncrewed and one with two NASA astronauts on board, in May and August of this year, respectively. That would have cleared the way for SpaceX to be certified by NASA to begin regular crewed missions to the station by early 2018.
In December, NASA and SpaceX announced a revised schedule. The uncrewed test flight is now planned for the fourth quarter of 2017, while the crewed flight is set for the second quarter of 2018.
“We are carefully assessing our designs, systems, and processes taking into account the lessons learned and corrective actions identified,” SpaceX said in a statement then, referring to pad explosion investigation. “Our schedule reflects the additional time needed for this assessment and implementation.”
As SpaceX presses ahead with commercial crew in 2017, it will have to deal with concerns that link back to the pad explosion. Because SpaceX loads the rocket with propellants so close to launch, crews will have to board the spacecraft prior to fueling, and not afterwards, as has traditionally been the case for crewed launches.
As far back as December 2015, the NASA International Space Station Advisory Panel, led by former astronaut Thomas Stafford, raised questions about this approach. When the panel met last October, after the pad explosion, Stafford complained that NASA had yet to respond to his concerns.
The latest annual report by another independent group, NASA’s Aerospace Safety Advisory Panel, also raised questions about the so-called “load and go” fueling approach SpaceX uses. In that report, released last week, the panel said that “the dynamic thermal effects on the system associated with loading densified propellants may not be adequately understood, which results in a higher level of uncertainty that must be factored into the risk determination.”
NASA has said it’s still studying the safety issues associated with that fueling approach, and SpaceX has said it’s working with NASA on that and, it said in a statement, “any additional controls will be put in place to ensure crew safety.”
Closing the business case
Because SpaceX is a privately-held company—Musk has repeatedly said he has no plans for an initial public offering of stock for the foreseeable future—there are few hard financial details about how SpaceX is doing. The company can talk about the size of its manifest, or that it has been cash-flow positive, but there are few metrics to use to measure the company’s financial health.
| In short, that means that, in a decade, SpaceX will be a broadband company that also happens to launch satellites, rather than a launch company dabbling in satellite communications. |
That changed on Friday, though, when the Wall Street Journal published an article based on five years’ worth of financial data obtained by the publication. And while the specific numbers were interesting, the trends were less surprising.
According to the data published by the Journal, SpaceX’s revenues from its launch activities grew steadily from 2011, when it was less than $150 million, to 2014, when it his about $1 billion. For all four years the company had a small, but positive, net income.
That changed in 2015, when the company suffered a mid-year Falcon 9 launch failure that grounded the vehicle until late December. Launch revenues dipped by six percent, but the company suffered an operating loss of $260 million. (It’s unclear if the “launch revenue” and income cited in the article include income that SpaceX has received from its commercial cargo and crew contracts, which include payments for vehicle development and services beyond launch.)
While the specific magnitudes of the revenue, and losses, are interesting, it’s also not surprising. Launch has traditionally been a low-margin business, with profits relatively small provided there’s some degree of competition. One thing it does suggest is that, should SpaceX demonstrate the ability to reuse Falcon 9 first stages multiple times, it may not decrease prices that much for reused vehicles: the additional margin the reduced costs of such vehicles provide may be worth more than an incremental increase in business that could be generated if SpaceX dropped prices more steeply.
The real money in space, though, comes from services, not launch. And the Journal article appears to confirm that, as it states projections from both SpaceX’s launch business through the mid-2020s as well as its proposed constellation of broadband satellites. And while launch revenue is set to grow to about $5 billion a year by 2025, revenue from the satellite constellation, first appearing in 2019, grows exponentially to more than $30 billion a year by 2025.
In short, that means that, in a decade, SpaceX will be a broadband company that also happens to launch satellites, rather than a launch company dabbling in satellite communications. And the profit will follow: the same projections call for operating income to grow from essentially zero now to more than $20 billion a year in 2025.
Those projections are perhaps optimistic—$30 billion a year in satellite services revenue would make SpaceX far larger than any existing commercial satellite operator—but also not surprising, either. Two years ago, when Musk announced the opening of a SpaceX office in Seattle to work on elements of that satellite system, he said he was banking on that revenue to fund his aspirations to establish a human presence on Mars.
“This is intended to be a significant amount of revenue and to help fund a city on Mars,” Musk said then. “Looking at the long term, what’s needed to create a city on Mars? Well, one thing’s for sure: a lot of money. So we need things that will generate a lot of money.”