SpaceShipTwo’s PF-04: A High-Risk Flight

Mike Alsbury

Part 1 in a Series

By Douglas Messier
Managing Editor

Mike Alsbury’s day began with a 3 a.m. wake up at his home in Tehachapi, Calif. He showered, dressed and ate a breakfast that likely consisted of an apple and a granola bar.

Alsbury rarely awoke at so early; but this Oct. 31 was a flight test day. That meant a lot of people were getting up early for the latest milestone in the Tier 1B program. At least that’s what they called it at Alsbury’s employer, Scaled Composites. The rest of the world knew it as WhiteKnightTwo and SpaceShipTwo – the foundation of Sir Richard Branson’s suborbital space tourism program. Scaled built and tested the vehicles for the British billionaire’s spaceline, Virgin Galactic.

Alsbury’s job this day would be to fly SpaceShipTwo as co-pilot for the craft’s fourth powered flight (PF-04). It would be the ninth flight aboard SpaceShipTwo for the 39-year old test pilot, who had joined Scaled Composites in 2001 after graduating from California Polytechnic State University.

Pete Siebold

Alsbury gave his wife of 12 years, Michelle Lynn Saling, a big hug and a kiss before he headed out into the chilly fall morning at 3:30 a.m. His destination – the Mojave Air and Space Port – lay a half hour’s drive over the Tehachapi Mountains in California’s High Desert.

At about the same time, Alsbury’s good friend, Pete Siebold, was sitting down to a breakfast of scrambled eggs and coffee a few miles away. Siebold would command this latest flight test of SpaceShipTwo, which he had flown 15 times in glide tests. The 43-year old pilot had joined Scaled Composites as a design engineer in 1996, working his way up to his current position as director of flight operations.

Siebold finished breakfast and left the house for Mojave. It was 3:45 a.m.

+++++

SpaceShipTwo was the successor to SpaceShipOne, a small three-seat, air-launched vehicle that had made history in June 2004 by becoming the first privately-funded manned vehicle to reach space. Scaled Composites test pilot Mike Melvill had flown the ship just above the Karman Line – the 100 km (62 mile or 328,000 foot) boundary of space.

White Knight and SpaceShipOne in flight

Four months later, SpaceShipOne won the $10 million Ansari X Prize for accomplishing the same feat twice within a two-week period. The prize was won on Oct. 4, 2004, the 57th anniversary of the first spaceflight by Sputnik. The 10th anniversary of the prize-winning flight had been celebrated in Mojave just four weeks earlier during a luncheon attended by Branson and other luminaries.

SpaceShipOne had been the brainchild of Scaled’s legendary founder, Burt Rutan. A genius at aircraft design, Rutan had built the Voyager aircraft that his brother, Dick, and Jeanna Yeager had flown around the world non-stop without refueling in 1986. Five of Burt Rutan’s flying vehicles – including SpaceShipOne – are displayed in the National Air and Space Museum in Washington, D.C.

For SpaceShipOne, Rutan had teamed with Microsoft co-founder Paul Allen, who bankrolled the prize entry with $25 million. SpaceShipOne was a small experimental vehicle that was hardly ideal for flying passengers. Rutan wanted to build something bigger.

However, Allen wasn’t interested in doing anything beyond SpaceShipOne; when he received an offer to donate it to the Smithsonian, he took it. SpaceShipOne never flew again after its prize-winning flight that October morning in 2004. By that time, however, Rutan had already found another billionaire to fund his dream of flying regular people into space for fun and profit.

Sir Richard Branson and Burt Rutan pose for a photo in front of the SpaceShip2 resting under the Mothership WhiteKnight2 inside a hangar in Mojave, Calif.

Like many of his generation, Richard Branson had developed an interest in space while watching the Apollo 11 moon landing in 1969. Or, if what he told radio shock jock Howard Stern is correct, his interest in space was sparked by the arousal he felt watching Jane Fonda in the 1968 space fantasy, Barbarella. His accounts varied.

Whatever the case, Branson had found out about Rutan’s little spaceship project and agreed to fund the follow-on system. SpaceShipTwo would evolve into a eight-person vehicle the size of a business jet with two pilots up front and six space tourists in the back with plenty of room to float around. Branson’s new company, Virgin Galactic, would fly the vehicles. The billionaire promised flights would begin in three years.

If SpaceShipTwo seemed like a dream pairing of Rutan’s design savvy and Branson ‘s marketing genius, it soon turned into a nightmare beset by years of technical delays and cost overruns. With only 17 flights, the SpaceShipOne program proved to be too short to fully test and evaluate all the systems and technologies required for the successor vehicle. Rutan was a genius at vehicle design, but a novice at rocket engines.

In 2007 – the year flights were supposed to begin – an explosion on a test stand killed three Scaled Composites engineers and left three others hospitalized with serious injuries. The accident revealed that Rutan’s choice for an oxidizer, nitrous oxide (laughing gas), was not nearly as stable as the famed designer had believed. Work on SpaceShipTwo was stopped for many months while the cause of the explosion was investigated.

+++++

SpaceShipTwo made its first glide flight October 2010. But, it would take more than 2.5 years before the engine was ready for use.

Powered flight tests of SpaceShipTwo finally began on April 29, 2013, more than eight years after Branson had announced the program. SpaceShipTwo dropped from its mother ship and lit its hybrid nitrous oxide/rubber motor for 16 seconds. Alsbury and pilot-in-command Mark Stucky zoomed out to Mach 1.3 and up to 56,200 feet before landing safely back on Runway 30 at Mojave.

Boom camera shot. (Credit: Virgin Galactic)

Branson was in Mojave to watch the flight along with Rutan. Branson soaked up the sun and attention from the media, congratulated the crew, and announced he would be shortly jacking up ticket prices from $200,000 to $250,000.

The billionaire was exuberant over the flight and, tossing caution to the Mojave wind, predicted that he and his family would fly to space on the first commercial SpaceShipTwo flight at the end of the year. He might even dress up as Santa Claus and fly on Christmas Day. It was vintage Branson: only he would be bold enough to try to upstage old Saint Nick and Jesus Christ on his birthday.

In truth, there was no chance of that happening in 2013. As with most everything else tossed into the Mojave wind, Branson’s prediction ended up stuck in the airport’s perimeter fence along with any number of plastic bags, tumbleweeds, drink cups, and greasy lunch receipts from Mojave’s quartet of fast food restaurants.

SpaceShipTwo would have to fly a lot in the months ahead to complete the flight test program in time. Virgin would need a launch license from the Federal Aviation Administration (FAA). And the company would have to transfer operations down to Spaceport America in New Mexico, where Branson’s flight would take place.

All that would have been a tall order even if everything had gone perfectly. They didn’t. It would take more than four months – until Sept. 5 – for SpaceShipTwo to conduct its second powered test. It didn’t do much to expand the ship’s flight envelope. The engine burn was for only 20 seconds.

A third flight test – again with a 20 second engine firing – would follow four months later on Jan. 7. The ship topped out at Mach 1.3 and 71,000 feet – well short of the 328,000-foot boundary of space. Virgin Galactic and Scaled had tried to get the third flight off in mid-December to end the year on a high note. But, bad weather and the upcoming holidays intervened. Christmas was safe, at least for another year.

Virgin Galactic reset its schedule with the goal of completing the flight test program and beginning commercial service as early as possible in 2014.

+++++

Following the third powered test, SpaceShipTwo flew a glide test one week later on Jan. 17. The spacecraft then disappeared into the hangar for six months, not flying again until an un-powered glide flight at the end of July. The long gap and the slow expansion of SpaceShipTwo’s flight envelope raised questions about why things were moving so slowly.

SpaceShipTwo on final approach during a glide flight on Jan. 17, 2014. (Credit: Douglas Messier)

The reason was the nitrous oxide/rubber engine. It couldn’t get SpaceShipTwo to where it needed to go. This motor, which was a longer and more powerful version of one used a decade earlier on SpaceShipOne, had proven to be difficult to scale up. It was a challenge to get the rubber to burn properly, and to dampen the serious oscillations and vibrations the engine caused in the ship.

By January 2014, the Sierra Nevada Corporation (SNC), which was developing the engine, thought it had found a solution that used helium to stabilize the burn. Engineers would need to install tanks and new plumbing in SpaceShipTwo’s wings to accommodate the helium.

There were some issues, however. The extra equipment would add weight to a spacecraft that had already come in very heavy. It also reduced the maximum altitude the ship could reach to below 100 km (62 miles) and the number of passengers from six to four. That would slice revenues by $400,000 to $500,000 per flight. There was another issue: the deal with SNC was costing Virgin Galactic a substantial amount of money at a time when funds were tight.

In late May – on the Friday before the long Memorial Day holiday in the U.S. and a bank holiday weekend in Britain – Virgin Galactic dropped a bombshell: it would be switching to an alternate engine developed by Scaled Composites that used nitrous oxide and nylon. Virgin Galactic officials said the nylon engine performed better in ground tests than the rubber one, providing increased thrust.

Nitrous nylon engine test on Jan. 16, 2014. (Credit: Ken Brown)

Sources say Sierra Nevada employees were blindsided by the announcement; they ended up scouring media reports for information. Sources also indicated that Scaled’s nylon engine was a cheaper option for Virgin, a factor that played a major role in the change.

Virgin Galactic officials misled the media and its own ticket holders as to the complexity of the switch, characterizing it as primarily a change in fuel grain. Virgin Galactic CEO George Whitesides told Space News:

The vehicle’s hybrid engine, manufactured by Sierra Nevada Corp., will need to undergo what Whitesides called “minor tweaking” to accommodate the plastic propellant.

“We just need to tweak the plumbing line in the motor to go with the plastic, which is sort of a minor mod,” he said.

In a May 23, 2014 email to ticket holders, Virgin Galactic Commercial Director Stephen Attenborough made similar claims:

Both fuels were designed to be interchangeable with the hybrid motor, with only minor plumbing modifications, meaning that we have been able to remain objective and unbiased in our assessment of the alternative fuel choices. Both have also been tested extensively.

As we have entered the qualification phase of the commercial hybrid motor in preparation for the final series of powered test flights, it has been critical to finalize the fuel that we’ll use for this and into the start of commercial operations.

Having made this decision, we will now be working hard to make the minor modifications to the spaceship systems in order to continue powered flights with the chosen fuel type as soon as safely possible.

In reality, the nylon engine would use the tanks and other equipment installed for the upgraded rubber motor, with some key differences. The nylon fuel needed a little bit extra to get started; if it didn’t light properly, the engine would suffer a hard start that could blow the ship apart in mid-air. Sources said that methane would be used to help the fuel burn at the beginning. Helium was also needed to ensure a smooth shutdown. There were concerns about new valves leaking and a new post-flight process to safe the vehicle after landing.

SpaceShipTwo’s propulsion system using the nylon engine. (Credit: Scaled Composites)

Tom Martin, a FAA safety engineer who reviewed the modifications as part of the renewal of Scaled Composites experimental permit to test SpaceShipTwo, told the National Transportation Board (NTSB) that the changes were anything but minor. (The specific details were redacted from his comments for proprietary reasons.)

“They’ve made some major modifications,” Martin said. “They’ve shortened the engine. They’ve changed the fuel, the shape, and they’ve added [redacted] – [redacted] plumbing and tanks, [redacted] psi tanks in the wings and changed the structure of that.”

There was also the question of the pre-flight ground certification and analysis of the new nylon engine. Virgin Galactic pilot Michael Masucci described the process for NTSB investigators.

“Asked about a risk analysis on the propulsion system, he indicated there was a qualification program that required 3 good burns on the motor,” Masucci said, according to an interview summary. “Asked if VG [Virgin Galactic] was kept in the loop on the RM [RocketMotorTwo] developing and testing, and he said [Director of Operations] Mike Moses was in the loop. A rocket motor risk analysis and qualification firings had been carried out, in which VG and TSC [The Spaceship Company] were included.”

In interviews with NTSB investigators, Scaled officials said they believed the engine had been thoroughly tested on the ground before the fourth powered flight. Virgin Galactic echoed those sentiments in statements made after the accident, saying they were not rushing the program or taking any unnecessary risks.

Others disagreed. A safety expert familiar with the program told Parabolic Arc that three full ground tests were not sufficient to properly qualify the nylon engine. Sources in Mojave expressed similar concerns about the new motor. One source described people in the program being on “pins and needles” over how well the motor would perform.

+++++

When SpaceShipTwo returned to flight test following the modifications, Scaled did not immediately start with powered flights. The spacecraft flew a glide test at the end of July, followed by another glide flight a month later that featured an in-flight dump of nitrous oxide. A third glide flight followed on Oct. 7.

With the modifications tested, the next step was a powered flight. Flying with a brand new engine for the first time was a tall order for a flight test, one that would entail a significant number of new risks. However, this flight would go further.

SpaceShipTwo is dropped over Koehn Lake. (Credit: Virgin Galactic)

After being dropped from the WhiteKnightTwo mother ship at about 50,000 feet, Siebold and Alsbury would light the ship’s hybrid engine for 38 seconds and soar up to 135,000-138,000 feet. Both were nearly double the previous burn times and maximum altitude reached. Another key objective was to increase the ship’s maximum speed from Mach 1.4 to Mach 2.0. The pilots would also attempt SpaceShipTwo’s first supersonic reentry at Mach 1.2.

As Siebold later told NTSB investigators, it was a “high risk” flight test that involved “doing a significant envelope expansion that day. Flying an unproven rocket motor in an unexplored aerodynamic regime… classic test hazard assessments would categorize that as a high-risk flight.” They also were testing “a propulsion system that history has shown can be unreliable, or much less reliable than a turbine or reciprocating engine.”

Scaled Vice President Cory Bird told the NTSB that the flight was originally scheduled for Oct. 23. A flight readiness review on Oct. 3 resulted in about 40 open issues that engineers would have to address before SpaceShipTwo could fly under power. Due to the change in engines, new trajectories were required that placed different loads on the vehicle.

Bird said the new nylon motor’s “weight was a little different so the tail loads were different,” the summary of Bird’s interview states. “They needed to go study this and concluded that to give them more margin, they would not do a 1.5 factor of safety. They said they had to meet a 1.9 factor of safety and that was what they did.”

Aerodynamicist Jim Tighe worked up the new trajectories before leaving Scaled Composites on Oct. 9 to take up a new job with a Silicon Valley startup called Zee.Aero. (He would continue as a consultant to Scaled, and he returned to Mission Control for the fourth powered flight.) As the Oct. 23 flight date approached, Program Manager Matt Stinemetze still had concerns about the loads on the vehicle. The flight was delayed eight days to Oct. 31.

Stinemetze told the NTSB that Virgin Galactic was not happy with the postponement, saying the flight had been scheduled for a month and that the issue should have been dealt with earlier. He said he had the full backing of Scaled’s management in postponing the flight test.

Ben Diachun, Scaled’s vice president of Engineering, admitted there was pressure on the company.

“There was pressure to have PF04 completed on October 31, but not undue or unreasonable pressure,” he told investigators. “There was also pressure to perform the flight safely and pressure to meet the program requirements, so they were constantly balancing all of those pressures. If there was a reason to not meet a schedule, everyone had a voice to speak up, and they all had spoken up at some time, and if that was the case they would adjust the schedule.”

+++++

This would be one of the last flight tests that Scaled Composites would conduct. If things went well today, another more ambitious flight test with a longer engine burn of about 50 seconds was scheduled in about three week’s time. A full duration burn, probably lasting about a minute, would follow that would get the ship up to its maximum altitude – whatever that turned out to be with the new engine.

In December, Scaled Composites would turn SpaceShipTwo over to Virgin Galactic, as it had done earlier in the year with WhiteKnightTwo. Virgin Galactic planned to make modifications to the ship and to move flight operations down to Spaceport America, where the company was the anchor tenant. Branson and his son, Sam, would fly within a few months, followed by commercial passenger flights.

Branson kisses SpaceShipTwo. (Credit: Virgin Galactic)

Virgin had begun 2014 with the goal of completing all these tasks by the end of the year. As delays dragged on, the Dec. 31 goal was pushed back to the end of the first quarter of 2015. Sources say the company was under pressure from its main financial backer, Aabar Investments. The sovereign wealth fund had committed $390 million to Virgin Galactic over the previous five years, and it was anxious to see results. Further financing would not be forthcoming without significant progress.

The small number of remaining flight tests was one of the biggest concerns floating around Mojave in the months leading up to the fourth powered flight. Virgin Galactic had originally promised investors that they would conduct up to 30 powered tests before beginning commercial service. Now, there would be only a handful of additional flights with a new engine that had only three qualification tests on the ground.

Sources were concerned the remaining flights would not be sufficient to fully test the spacecraft. Scaled Composites had never intended the first SpaceShipTwo, which was a proof-of-concept vehicle, to enter commercial service. Sources feared Virgin Galactic was taking shortcuts with safety because of schedule and financial pressures. The company would later vigorously deny that it had ever compromised safety.

Siebold and Alsbury couldn’t worry about such matters as they arrived at the Mojave spaceport in the pre-dawn darkness. Their sole focus was on successfully completing SpaceShipTwo’s most ambitious and challenging flight to date. After 8.5 months of preparation, they felt they were ready.

The Series

• Part 1: SpaceShipTwo’s PF04: A High Risk Fight
• Part 2: SpaceShipTwo Pilots Faced Extremely High Work Loads
• Part 3: A Good Light, Then a Fatal Mistake
• Part 3.1: SpaceShipTwo Powered Flight No. 4 Flight Transcript
• Part 3.2: The Breakup of SpaceShipTwo Frame by Frame From the Tail Boom
• Part 4: Pete Siebold’s Harrowing Descent
• Part 4.1: SpaceShipTwo Emergency Response Timeline
• Part 5: Shock, Tears & Spin: The Aftermath of the SpaceShipTwo Crash