The Space Reviewin association with SpaceNews

Antares failure
An explosion erupts from the aft end of the Antares rocket about ten seconds after liftoff October 28 from the Mid-Atlantic Regional Spaceport in Virginia. Orbital ATK is planning to resume Antares launches, using a new first stage engine, in early 2016. (credit: NASA/Joel Kowsky)

Antares and SpaceShipTwo, six months later

Bookmark and Share

Six months ago, two accidents shook the commercial spaceflight industry. On October 28, an Orbital Sciences Corp. (now Orbital ATK) Antares rocket crashed seconds after liftoff from Virginia’s Mid-Atlantic Regional Spaceport, destroying a Cygnus cargo vehicle bound for the International Space Station. Less than 72 hours later, on October 31, Virgin Galactic’s SpaceShipTwo broke apart in the skies above Mojave, California, seconds into its fourth powered test flight, killing the vehicle’s co-pilot (see “The importance for commercial spaceflight to recover and respond”, The Space Review, November 3, 2014).

Today, both Orbital and Virgin Galactic are recovering from those accidents. Neither the Antares nor SpaceShipTwo have launched again since those failures, as the companies redesign and rebuild their vehicles. Meanwhile, multiple investigations into the accidents are still ongoing, but expected to wrap up soon.

Re-engining Antares

Just days after the Antares failure, Orbital announced it had tracked down the most likely cause of the failure. “[P]reliminary evidence and analysis conducted to date points to a probable turbopump-related failure in one of the two Aerojet Rocketdyne AJ26 stage one main engines,” Orbital said in a November 5 press release. “As a result, the use of these engines for the Antares vehicle likely will be discontinued.”

“The conclusion of that report is that the most probable cause was excessive bearing wear that took place in the turbopump of main engine number one,” Grabe said of the Antares failure.

The primary investigation into the failure has been led by Orbital itself, under the oversight of the FAA’s Office of Commercial Space Transportation. While the company has given no formal announcement of the investigation’s results, an executive recently offered some insight into its work.

“As I understand it, the Orbital-led board will be submitting its report to the FAA within days,” Ronald Grabe, president of Orbital ATK’s Flight Systems Group, said during a panel session on launch systems at the 31st Space Symposium in Colorado Springs April 14.

“The conclusion of that report is that the most probable cause was excessive bearing wear that took place in the turbopump of main engine number one, which led to rubbing between the rotating and the stationary components within that turbopump,” he said.

What’s not clear, though, is what caused the excessive bearing wear. Orbital has suggested the wear was a flaw in the engine itself, and thus the fault of the engine’s supplier, Aerojet Rocketdyne. Grabe, though, did not make that claim explicitly clear in this comments at the conference.

Aerojet Rocketdyne officials, though, strenuously object to that suggestion, arguing that foreign object debris from elsewhere in the vehicle got into the engine and led to the failure. Company spokesman Glenn Mahone said April 15 said that blaming the failure on the engine itself was “inaccurate and could be misleading.” Aerojet’s own investigation into the engine failure was due to be completed in a few weeks, he said.

Regardless of the outcome of the dueling investigations, Orbital has already decided to end use of the AJ26 engine. Orbital plans to replace the AJ26—an “Americanized” version of the Soviet-era NK-33 engine originally developed for the N-1 Moon rocket—with the RD-181 engine from NPO Energomash. Besides using a newer engine, this “re-engined” Antares will also have higher performance, allowing it to carry somewhat heavier payloads.

“We take much higher risk with cargo than we would typically, but we’re not uninformed about the risk level we’re taking,” Gerstenmaier said.

That new version of Antares should be ready for flight early next year. In a presentation to the National Research Council’s Aeronautics and Space Engineering Board (ASEB) in Washington April 21, NASA associate administrator for human exploration and operations William Gerstenmaier presented a schedule of upcoming Cygnus cargo missions that listed the first launch of the re-engined Antares, carrying a Cygnus, scheduled for Mach 30, 2016.

That mission, Gerstenmaier said, will be the first launch of this new Antares version, designated Antares 230. “They are doing a detailed certification activity in Russia now for the RD-181,” he said, adding that NASA was involved in that effort.

“There will also be a hot-fire test, probably in January or February of next year out at Wallops,” he said. “That test firing will also verify the integration activity of the engine and the stage.”

While NASA is involved, Gerstenmaier said the agency didn’t have a role in approving or rejecting the engine. Instead, he said the effort informed NASA’s plans on the level of risk involved with that first new Antares mission, and thus the cargo they plan to place on board it. “We take much higher risk with cargo than we would typically, but we’re not uninformed about the risk level we’re taking,” he said.

Since that presentation, though, NASA may be a little more conservative about the risk it’s willing to accept in transporting cargo to the ISS. The failure of a Russian Progress spacecraft launched April 28—the spacecraft started spinning uncontrollably after reaching orbit, forcing controllers to abandon a planned docking with the ISS—mean that two of the station’s four cargo vehicles are currently unavailable. Only SpaceX’s Dragon, the most recent of which launched last month, and Japan’s H-2 Transfer Vehicle are currently operational.

NASA is not alarmed about the current situation, noting last week that the station has more than enough supplies until the next Dragon launch in June and the next HTV launch in August. Cygnus will return to flight in November launching on an Atlas V, with an option for a second Atlas V launch in early 2016 should the re-engined Antares be delayed.

Rebuilding SpaceShipTwo

As three separate investigations continue into the Antares failure (NASA is conducting an independent review in parallel with the Orbital and Aerojet investigations), the National Transportation Safety Board (NTSB) has been leading the investigation into the SpaceShipTwo crash. That effort, too, appears to be winding down.

“We expect to see their final report in the next month or two,” George Nield, FAA associate administrator for commercial space transportation, said at the ASEB meeting April 21, referring to the NTSB’s investigation. “The final board decision from NTSB will be later on this summer.”

Christopher Hart, chairman of NTSB, offered few additional details about the ongoing investigation in an April 22 presentation to the ASEB and the Space Studies Board. Much of his presentation was an overview of the NTSB and how it functions.

“Everything is going to be very manual. So when I looked at that cockpit, it was very much sort of out of the 1950s, almost,” Hart said of SpaceShipTwo.

In the initial days after the accident, NTSB focused its attention on a decision by the vehicle’s co-pilot to unlock the vehicle’s feathering mechanism prematurely. That mechanism is designed to raise SpaceShipTwo’s tail section for a stable reentry, but unlocking it as the vehicle accelerated though Mach 1 may have caused aerodynamic instabilities that broke the vehicle apart.

“The system was performing nominally during the early parts of the flight, and then at some point the co-pilot did something that was out of sequence with both what was trained and what was written on the test flight protocol,” Will Pomerantz, vice president of special projects at Virgin Galactic, said in an April 30 presentation at the Space Access ’15 conference in Phoenix, noting that the out-of-sequence action was the unlocking of the feather. He added that his comments weres limited to what the NTSB had publicly disclosed.

In those early days of the investigation, the NTSB announced they were forming a group that would study human factors issues associated with the accident. Hart did not mention that effort in his April speech, but did note that one thing that surprised him about SpaceShipTwo was its reliance on manual controls.

“I actually flew the simulator when I was out there to see what it looks like. They basically opted against automation. Everything is going to be very manual. So when I looked at that cockpit, it was very much sort of out of the 1950s, almost,” he said.

One question, Hart said, is how well a pilot could perform those manual actions, particularly during the powered portion of the flight. “So the environment I’m thinking of is, I’m in this cockpit wearing gloves, and all of a sudden I get this three-g kick in my back from this rocket,” he said. “I’m wondering how well is that understood. I don’t think it’s very well understood because we just don’t have that much experience with it.”

Hart didn’t give a timetable for completing the report beyond later this year. While many NTSB investigations result in recommendations that, about 80 percent of the time, are accepted by the parties involved, he said this report may end up with no recommendations.

“I will not be surprised if, by the time we finish this report,” Hart said, “everything we would have recommended was already in process because the parties don’t want this to happen again.”

“There were a lot of paths that we could have gone,” Pomerantz said. “I think we have gone on the best possible one of those paths.”

As the NTSB investigation continues, Virgin Galactic is working on building a second SpaceShipTwo vehicle. That work is taking place on a three-shift schedule, Pomerantz said at the Space Access conference. He said he couldn’t give a percentage estimate of the vehicle’s completion, but that “a lot of major steps” in its assembly had been completed.

“We’re quite close to the first time where we’ll pull the vehicle off the fixture and begin testing it on its own landing gear,” he said. “That will be a nice internal milestone.”

Nearly one year ago, Virgin announced that it was changing the fuel used by SpaceShipTwo’s hybrid rocket motor, abandoning a rubber-based fuel, formally known as hydroxyl-terminated polybutadiene (HTPB), for a plastic fuel similar to nylon. The switch, company executives said at the time, was because the nylon fuel offered slightly better performance.

However, Pomerantz left open the possibility of changing back to rubber fuel when the second SpaceShipTwo resumes powered test flights. “We've had kind of this internal horse race” between the two designs, he said.

“The one we’ll fly is the one that’s best,” he continued. “We’re not flying again yet, so I’m not 100 percent sure. If I had to guess, my personal guess, would be HTPB.”

A switch back to rubber would have nothing to do with any safety concerns about the nylon engine. That engine’s only powered test flight was in the October accident, but the engine hasn’t been implicated as the cause of the accident. “We have an engine that works. We have an engine that we’re confident is safe,” he said.

Last October’s accident “was a punch to the jaw for all of us,” Pomerantz said. However, he said he was heartened by the reaction from industry, its customers, and others in the aftermath of the accident.

“As I sat there on the flight line and was trying to process what I was hearing on the radios and what I was seeing up in the sky, or not seeing up in the sky, it occurred to me that there are a lot of different futures that could happen from this point,” he recalled from the day of the accident. “There were a lot of paths that we could have gone. I think we have gone on the best possible one of those paths.”