NASA, CNES Warn SpaceX of Challenges in Flying Reusable Falcon 9 Rocket


In theory, the SpaceX Falcon 9 v1.1 booster can be reused more than three-dozen times.

That's because the rocket's LOX/Kerosene Merlin 1D engine -- nine of which power its first stage -- has a cycle of 40, according to Stella Guillen, SpaceX director of business development.

"It's not obviously the entire system or the entire stage," she said during a space conference in Paris last month. “We don't know how many times we can fly the first stage. But the engines have a cycle of 40.”

As part of the company's goal to dramatically lower the cost of space launch, Hawthorne, Calif.-based SpaceX is slowly moving out on a strategy to test the reusable Falcon 9 core stage with the goal of eventually recovering and reflying the entire rocket, Guillen says.

But even without first-stage flyback, the Falcon 9 is already the lowest-cost rocket on the commercial market today. Able to throw 4,850 kg of payload to supersynchronous transfer orbit at roughly $60 million per launch, it bests even the Chinese Long March 3B at $70 million a pop.

Noting that the cost of fuel, oxygen and other expendable liquids in the rocket amounts to just 0.3% of the cost of a Falcon 9 mission, SpaceX CEO and founder Elon Musk says there is potential to further lower the cost by a factor of more than 100, assuming a high launch rate and the ability to fully and rapidly reuse the entire rocket, including its first and second stages.

“This is a difficult thing to achieve,” he told an annual U.S. Export-Import Bank conference in Washington April 25. “A lot of people in the aerospace industry think it's not possible, and most in industry have given up on it. But we think it's possible.”

Among the doubters is NASA Deputy Associate Administrator Dan Dumbacher, a former Space Shuttle engineer who leads the agency's exploration systems development. Dumbacher says the agency learned a lot from its experience with the orbiter's reusable Space Shuttle Main Engines (SSMEs).

“We tried to make the engines reusable for 55 flights,” he said in Paris last month. “Look how long and how much money it took for us to do that, and we still weren't successful for all parts.”

While NASA is using SSMEs to power its new heavy-lift Space Launch System (SLS), Dumbacher says the cost of refurbishing the engines means the agency has no plans to reuse them.

“All we're going to be doing is working on the manufacturability and getting the cost down,” Dumbacher said. “We're going to retain the reusability that's in there, but I'm actually going to throw them away.”

Dumbacher said unlike commercial airliners, rockets have limited flight opportunities, presenting a challenge to engineers in terms of measuring and understanding the environments in which launch vehicles operate.

“In the airline industry, you have lots of flights, lots of repetition, lots of opportunity to go learn, figure out what broke, and fix it,” he says. “With launch vehicles, we've launched a lot, but not to the order of magnitude we have with airplanes.”

Dumbacher said the Space Shuttle solid-rocket boosters were likewise problematic, requiring too much overhaul to justify the cost of reusing them.

“We'd pick them out of the ocean, but we spent a lot of time cleaning those things up when we got them back, inspecting the hardware to make sure that it was still good to go,” he said.

In addition to physics challenges, Dumbacher said the economics of launching reusable rockets is a key factor to consider in justifying the cost.
“You have to understand the economics of the situation and how that translates into flight rate and production rate to make the overall economic calculations for the system worthwhile,” Dumbacher says.

Christophe Bonnal of the launcher directorate at French space agency CNES, agrees.

“If you reuse, you stop producing, depending on the level of reusability,” he says. “So you end up with a permanent prototype, and to keep costs down you need to have a high rate of production.”

Barry Matsumori, SpaceX vice president of commercial sales and business development, says Falcon 9's 10 Merlin 1D engines, including nine on the core stage and a vacuum version flying on the second, mean SpaceX is already manufacturing a lot of motors.

“It doesn't take very many flights for us to hit rate production,” Matsumori said in April. “We are popping out engines at a pretty high rate.”

Bonnal said CNES has looked at “optimistic” flight rates of around 50 per year, assuming a single payload, but that, “Under the best conditions we could save 10% of the launch costs, plus or minus 15%,” he quipped.

One of the most challenging aspects of reusability, he said, is the weight penalty added by hardware and propellant. He says the latter means reserving 30% of first-stage fuel in order to return a booster to the launch site.

“You end up designing much larger vehicles, with landing gear, with legs or wings, so it's heavier and you need more propulsion, at least 25-30% more propulsion on the stage,” Bonnal said, adding that a previous study by CNES and Russian space agency Roscosmos looked at the feasibility of making the Ariane 5 solid-rocket boosters liquid-fueled and reusable, but scrapped the idea after the hardware grew too large.

“The thing that shocked me was that at the beginning, this reusable flyback booster was just a cylinder with engines and little wings, just a turbo fan in the back,” he said. “And three years later these were complete Airbuses in terms of size with four engines in each of them.”

SpaceX President Gwynne Shotwell says Falcon 9's reusability is already designed into the rocket's first stage, including the weight of the landing legs that would otherwise detract from the rocket's performance. She also said Falcon 9 retains 30% performance margin over the company's advertised mass-to-orbit capability of 4,850 kg to GTO – margin SpaceX is using to conduct operational trials of a reusable Falcon 9 first stage.

“The mass of the recovery hardware is not a dramatic impact on payload performance,” Shotwell said in a February interview. “Conservatively, if I put 1,000 lbs on the first stage, I'm only losing 100 pounds of payload or so to orbit. What impacts is the fuel we need to reserve to execute the reentry and the landing burns.”

However, Musk says he sees a Falcon 9 with a reusable core stage targeting the small- to medium-size weight category for missions to GTO, with the larger Falcon Heavy now in development expected to do the heavy lifting.

"Where I basically see this netting out is Falcon 9 will do satellites to roughly up to 3.5 tonnes with full reusability of the boost stage, and Falcon Heavy will do satellites up to 7 tonnes with full reusability of all three boost stages," Musk told Aviation Week in February. "Now Falcon Heavy could double its payload, almost; if, for example, we went expendable on the center core, we could do 14 tonnes to GTO."

But beyond performance, says Bonnal is the impact of rocket reusability on ground installations. As an example, he said CNES has found that safety requirements would make return of a boost stage problematic at Europe's South American space port in Kourou, French Guiana.

“I will be very interested in seeing the three Falcon Heavy boosters coming back to Vandenberg with propellant sloshing,” he said, referring to SpaceX plans to start flying a heavy version of the Falcon 9 from the U.S. Air Force's California launch installation next year. “In terms of safety, it must be quite challenging.”

Paul Eckert, manager of budget, policy, and international affairs at the FAA Office of Commercial Space Transportation says regulators are prepared to work with SpaceX to ensure the company can proceed with its reusability initiative.

“There are always risk factors, there are certain criteria, and the company and the government are both well aware, so we don't anticipate any problem,” he said in Paris last month. “I'm sure we can work it out.”

With telemetry gleaned from a simulated soft-landing of a Falcon 9 first stage in the Atlantic Ocean last month, SpaceX says it is chipping away at its goal. But as Musk has emphasized, the economics of reflying launch vehicles are most promising if the entire rocket is reusable.

“The most fundamental breakthrough is a rapidly and fully reusable rocket,” Musk said in February. “It doesn't help if it's partially reusable.”

Shotwell concedes that second stage-recover presents a different set of challenges, notably the performance offset required for additional propellant.

“The second stage mass trade is pound for pound,” Shotwell says. “Whatever incremental mass I add, the same has to come off the payload.”

When asked if SpaceX has the technology in hand to achieve second-stage flyback, Guillen declined to comment, though she said “it is in our plans to recover the entire rocket.”

Matsumori agreed that recovering a Falcon 9 upper stage is more difficult than returning a core booster.

“The thing goes to the other side of the world; bringing it back is another challenge,” he said.



Discuss this Blog Entry 9

on May 5, 2014

The lack of an existence proof isn't proof of nonexistence. If this was easy, someone would have done it already.

on May 5, 2014

I wish Elon all the best in his endevor, but I think he needs to do it first before he shakes up the industry which is long overdue. It would also be a good idea to start designing a space tug to remove all the dead stuff in orbit, because when Reusable does come around everybody will jump onboard and space junk will take on a whole new meaning.

on May 5, 2014

If returning and landing the booster at the launch pad is too risky, then simply use a barge out in the ocean and tow it back. This isn't rocket science . . .

on May 6, 2014

It doesn't take a rocket scientist to know that landing a booster in salt water is a bad idea if you want quick turn-arounds. The havoc that stuff can play on precision engine parts and other components is harsh to say the least. One could never seal things well enough to have 100% confidence in there being no leakage.

I imagine that the folks who restore the Shuttle's boosters know the issues full well. Refurbishment duties will be high if he cannot land it on solid (dry) ground. Just the possibility of salt water getting into the wiring or the electronics via an unseen pinhole leak puts up barriers to fast re-use. This is why only the casings of the Shuttle's boosters were re-used.

And those engines of his will require a major (costly and time consuming) inspection after use before NASA let's any of it's people fly on them again. Likewise with the rest of the bird.

Manned rockets will never reach airline schedule-like flights in our lifetime because space flight (and space itself) is a much more severe and unforgiving environment than the one airliners fly in, and orbit-bound passengers can still only get up there by riding on top of huge fuel tanks that have one end on fire.

on Nov 15, 2014

"And those engines of his will require a major (costly and time consuming) inspection after use before NASA let's any of it's people fly on them again. Likewise with the rest of the bird."

NASA would probably require SpaceX to use new never used rockets and motors when launching NASA's astronauts. At least initially. These recovered rocket stages and motors could then be re-used repeatedly for launches of cargo or satellites.

SpaceX might charge a premium launch price for customers who insist that a "virgin launch vehicle" be used.

on May 6, 2014

SpaceX is really pushing the envelope but it's time for someone other than NASA (or their cronies) to do some pure rocket science and move the industry forward. When Musk re-uses the first booster, every other rocket manufacturer is going to try as well.

It's only impossible to someone makes it possible.

on May 7, 2014

I'm surprised Mr. Dumbacher would conflate the complexity of the RS-25 with the Merlin engine, though to be fair to him nothing in article shows Mr. Dumbacher to have been specifically addressing SpaceX or the Merlin, with the article possibly "re-purposing" his statement's original context.

Nevertheless, the RS-25 SSME is a dual pre-burner, staged combustion, liquid hydrogen engine meant for massive performance. It was designed in the '70s on the absolute bleeding edge of the decade's technology, and so it's no real surprise it wasn't quite as cheaply reusable as intended. The Merlin, in contrast, is an extremely basic gas generator, designed for modern manufacturability and low cost über alles, incorporating simple pintle injector technology straight out of the 60's Apollo-era lunar lander. Cutting edge design, it is not.

Historical experience with RS-25 did, however, provide valuable lessons to SpaceX, even if it seems primarily in what to avoid.

It seems to be likewise with CNES' experience. Ignoring the unmentioned fact that CNES is intimately related to SpaceX-rival Arianespace, and that a few grains of salt are thus required before swallowing they might say about each other, it seems quite obvious that adding wings and jet engines to a rocket will markedly increase its mass, complexity and expense.

Fortunately, it appears that SpaceX took this "lesson" to heart, however, through either learning from CNES' experiments or through simple common sense, and its Falcon 9 sports neither wings nor jet engines. In fact, simplicity seems to be SpaceX's watchword as its reusable configuration differs from an expendable configuration only by a beefed up cold gas RCS system, lightweight and fairly cheap carbon fibre and aluminum honeycomb landing legs, and additional landing control software in its avionics core.

The latter, of course, was tested extensively not just in simulation, but on SpaceX's real world Grasshopper and new F9-R Dev1 vehicles, allowing as many cheap "flight opportunities" as SpaceX's engineers would like for "measuring and understanding the environments in which launch vehicles operate."

If Mr. Bonnal's plan for reusability netted his organization no reduction in costs, then there's yet another roadmap that SpaceX can learn from -- and not follow. Fuel is needed for boostback and landing, of course, but fuel is cheap. Rocket cores, on the other hand, are expensive. As is refurbishing sea-water drenched equipment, so SpaceX aims to return the rocket cores to their launch site, touching down on water being only a temporary test regime until they can prove sufficient control over the returning stages to land safely.

I suspect Mr. Bonnal's organization would have had more luck with Ariane 5 reusability if it had been designed in from the start. I find it surprising that reusability is not a goal for Ariane 6 -- now would be the time to bake in the capability. If SpaceX is successful in its reusability goals, when the A6 debuts it'll be in an even worse competitive spot that A5 is in today with F9 -- and betting that the competition will fail is not wise business strategy.

Since SpaceX mostly just followed the KISS principle, any group of bright engineers should be able to replicate their success. However, from clean sheet of paper to boosting rockets back to their launch site will still take several years, years in which any non-reusable offering is non-competitive. I suppose they could modify their expendable rocket for reusability, but CNES already found that didn't work too well, so they might wish to take their own lesson to heart.

Reusability does reduce a rocket's performance, but "[SpaceX President Gwynne Shotwell] also said Falcon 9 retains 30% performance margin over the company's advertised mass-to-orbit capability of 4,850 kg to GTO."

Though the article's author has argued otherwise recently, unless SpaceX's president doesn't know her own company's products, that statement strongly implies that with no reserved performance margin, a fully expendable F9 can lift 4.85t * 1.3 = ~6.3t to GTO.*

Using all the less-than-successful or outright failed reusability plans as a guide, it seems SpaceX has decided the way to make a medium-class reusable launcher is to build an extremely simple and low cost 6t heavy launcher, designed from the ground up for reusability, and then market it at a reduced 3-4t capacity. And, as a follow up, they decided the way to make a heavy launcher is to take a 30-50t superheavy, and market it as a heavy launcher with a capacity of 7t -- returning all three of the boost cores -- or 14t -- returning only two of the cores.

To badly paraphrase Edison, it could be said that the concept of reusability hasn't failed -- it's just that 10,000 ways that won't work have been found. After observing the fates of others' efforts, in a few months or a year we shall know whether SpaceX's own plan for reusability will, in fact, finally work -- and lead to the success that has eluded all others.

* This further implies that the 5.3t SES satellites the F9 has been contracted to launch into sub supersynchronous GTO are not, in fact, beyond the limit of what F9 can throw into GTO. Hence there must be another reason that they're being launched to less-than-full SSGTO.

Perhaps 5.3t to SSGTO is merely beyond the limit of the F9's capabilities *with* first stage recovery, and the sub-SSGTO was required to attempt recovery of the stage? It'll be interesting to see if those F9 flights sport legs or not.

on May 8, 2014

As if we should respect the opinion of an engineer from Nasa, who promised a $7 million per flight 'cheap, safe, reliable access to space' shuttle then delivered a $1.6 billion per flight boondoggle which killed 2 crews and had chronic and multi-year service outages... Nasa's shuttle was the most unaffordable, dangerous, unreliable space vehicle in history....
Then while Nasa blew $20 billion on it's miserably failed/cancelled Constellation, SpaceX produced far superior/advanced boosters/capsules for only $300 million.... now Nasa's shameless, unneeded, unsustainable, unaffordable multi-$billion/flight SLS...
The paper-pushing fools at Nasa need to just shut up and get out of the way.

on May 10, 2014

Until very recently, all space-faring activity was funded and undertaken by governments. The state-funded designers and builders of flight hardware had no real incentive (competition) to reduce the cost of mass to orbit. Mr. Musk, more than anyone else, threatens to dump the world’s launch paradigms on their collective heads. Understandably, the priests and acolytes at the core of the world’s incumbent launch beliefs oppose the radical Mr. Musk and fear the possibility of his success. The right call is to give Mr. Musk the ball; let him run with it. Win, lose or draw, everyone must now follow him down the path of real reusability and reduced launch costs or risk irrelevance and extinction.

Please or Register to post comments.

What's On Space?

On Space

From The Archives

Aviation Week is approaching its 100th anniversary in 2016. In a series of blogs, our editors highlight editorial content from the magazine's long and rich history.


Sep 18, 2015

The U.S.-Russian Moon Landing That Never Happened (1963) 7

U.S. President John F. Kennedy is well known for the 1961 speech to Congress in which he made the Apollo program a national goal. Less remembered is his controversial offer two years later to cooperate with the Soviet Union on a manned lunar landing....More
Sep 10, 2015

Airbus A380 Is Your Hotel In The Sky (2001) 5

In a September 2001 advertisement, Airbus promised to bring glamour back to passenger travel with its new A380, comparing the double-decker airliner to a hotel....More
Sep 9, 2015

Pratt & Whitney's Record Behind the DC-6 Record (1947) 2

Pratt & Whitney takes to the cover of AVIATION to credit its Double Wasp engines for the DC-6's record transatlantic flight....More
Aug 27, 2015

Aviation Week Lifts Veil On Boeing B-52 Bomber (1952) 22

In 1952, Aviation Week provided the first details on the new Boeing B-52 bomber....More
Blog Archive
Penton Corporate

Sponsored Introduction Continue on to (or wait seconds) ×