ESA, ArianeGroup Sign Contract for Prometheus Engine Demonstrator


Prometheus rocket engine (Credit: ArianeGroup Holding)

PARIS, 14 December 2017 (ESA PR) — An ultra-low cost reusable rocket engine, Prometheus, using liquid oxygen–methane propellants, is set to power Europe’s future launchers.

Today, ESA and ArianeGroup signed a contract to develop a full-scale demonstrator to be ground tested in November 2020.

Prometheus demonstrates the systematic application of an extreme design-to-cost approach, new propellant and innovative manufacturing technologies.

It lowers costs to a tenth of those for Ariane 5’s Vulcain 2 engine.

Additive layer-by-layer manufacturing of engine parts enables faster production, with fewer parts.

Prometheus casing prototype (Credit: ESA)

Key characteristics of Prometheus include a computer system enabling realtime adjustment and immediate diagnosis for potential reusability.

Methane propellant is widely available and brings high efficiency, standardisation and operational simplicity, making it a perfect candidate for a reusable booster engine demonstration.

By 2020, technical knowledge of liquid oxygen–methane propulsion gained through the Prometheus project will allow fast and informed decisions to be made on useful applications.

Prometheus provides a nominal 1 MN of variable thrust, is suitable for first- and second-stage applications, and is reignitable. It will propel a range of next-generation launchers, including future evolutions of Ariane 6.

The Prometheus contract, worth €75 million, was signed by ESA Director of Space Transportation, Daniel Neuenschwander, and Alain Charmeau, CEO at ArianeGroup, at ESA headquarters in Paris in the presence of ESA Director General Jan Wörner.

The project is part of ESA’s Future Launchers Preparatory Programme.

“Prometheus will power Europe’s future launchers, forging a path of continuous improvement in competitiveness,” commented Mr Neuenschwander.

“This contract paves the way for the future of Europe’s space transportation, and the development of European propulsion technology of tomorrow,” added Mr Charmeau.

The project benefits from significant synergies with other launcher demonstration projects within ESA, national agencies and industry.

  • So, they’re going to have a methalox gas generator engine that might have a vehicle associated with it somewhere between 2022 and 2024, and it’s gonna compete with two different methalox staged combustion engines, both with higher thrust and higher specific impulse, that’ll probably both have vehicles deployed 2-4 years earlier.

    When I first heard about this, I joked that they’re calling it Prometheus not because it brings fire from heaven, but because it’ll spend eternity chained to a rock with a bird or prey eating its liver. I don’t see much reason to retract the joke.

  • Mr Snarky Answer

    And both those other engines will be manufactured at a higher rate. They will, in all probably, be cheaper unit price adjusted for performance. Too little, way too late.

  • Enrique Moreno

    What an ugly engine!

  • Hypx

    Should be pointed out that the F9 is powered by a gas generated kerolox engine and is beating rockets powered by staged combustion engines. If Prometheus is cheap it may be a good engine.

  • It’s gonna have to be really cheap, and highly scalable, to compete in a market where Raptor and BE-4 have been deployed. Prometheus is never going to be suitable for anything other than a medium-lift launcher. Maybe the SpaceX and BO vision of reusable heavy lift won’t work out, but if it does, there’s really not going to be a reason to favor a medium-lift expendable vehicle over a reusable heavy-lift one.

  • Hypx

    It says in the article Prometheus is ultra low cost, 1/10th that of Vulcain. From what I’ve read BE-4 is going to be $8-10M. Raptor will probably be even more expensive due to its complex FFSC design. We shouldn’t just assume BE-4/Raptor will succeed when they have serious technical and cost challenges of their own. In the right scenario, a Prometheus powered rocket could be cost effective.

  • redneck

    Actually you are assuming difficulty with FFSC that may not exist. With the full flow of both propellants driving turbines, temperatures and pressure drop can be quite low relative to gas generator. And gas-gas injection in the thrust chamber should lead to relatively stable combustion and a short L*.

  • Note that I specifically assumed that Raptor and BE-4 would be successful, and you are of course right that, if SpaceX and BO screw up, then Prometheus might be the greatest thing since sliced bread. But “maybe my competitors will screw up” isn’t exactly the best basis for a business model.

    As for expense, it doesn’t matter how expensive a reusable engine is if you can amortize the cost over enough flights. I’m also pretty skeptical that Raptor will be as expensive as you think: It’s substantially smaller than a BE-4, and, while FFSC has the disadvantage of a lot of new technology and an extra preburner, the turbines and seals themselves are considerably simpler than driving two turbopumps off of a single shaft with a higher-power preburner.

  • Hypx

    Since nothing is launching now, all of this is speculation. Assuming one engine will succeed and another will fail is simply guesswork. Just pointing out that in the real world, stage combustion has not shown itself as magic and gas generators do win sometimes.

  • As I said, I wouldn’t want to bet my business plan–and a few hundred million euros of public money–on SpaceX and BO both messing up.

  • Hypx

    You do realize both SpaceX and Blue Origin can “succeed” with their new engines, and still fail if their engines cost too much?

  • If they succeed with the kind of reusability they’re talking about, the manufacturing cost of the engines is almost irrelevant.

  • Hypx

    That could also fail. If everyone is trying reusability, the one with the cheapest engines can legitimately have a real advantage.

  • Michael Vaicaitis

    Both Raptor and BE-4 work in testing, so failure has already been avoided…barring late some extremely unlikely late development disaster.
    Musk has said that Raptor is designed for manufacturability (i.e. cheap as f**k), and reliability, (i.e. lasts for ages and reduces amortised cost even further).
    The cost of BE-4 (and NG) is irrelevant due to being made from Bezosium.

    With that said, Prometheus is perhaps the first forward thinking response of the Ariane project. At least it is an experience building route to reusability, that A6 is not but should have been. As long as European governments keep faith (and funding) with the idea of a European launcher then this is at least a move in the right direction.

  • Michael Halpern

    Maybe a reusable light lift, if it can get small enough,

  • Aerospike

    You keep writing about business plans… This is a government project, commercial success is almost irrelevant here.
    The only goal is to be “cheap enough” so you can justify launching on more expensive but “domestic” rockets.

    Europe is still a long way behind the US “new space” sector. There won’t be a truly commercial launcher in Europe until 2025 at the earliest.

  • Jeff2Space

    “In the right scenario, a Prometheus powered rocket could be cost effective.”

    ULA said a similar thing about Falcon 9 and their launchers for several years (mostly harping on ULA’s self proclaimed perfect launch record). ULA did so at their own peril.

  • Jeff2Space

    Form follow function. I’d be more worried about the fact that this engine is getting a *very* late start compared to not one but two US LOX/methane engines.

  • Ignacio Rockwill

    Heh. Bezosium. That’s good, I might have to steal that one.

  • I don’t think that’s really true. This is more like the ESA’s version of a public-private partnership. Yes, the ESA is coughing up some of the funds for development, but ArianeGroup is doing the work (no clue if Airbus and Safran are matching funds or not), and will ultimately operate the finished product for both ESA and commercial missions. ArianeGroup isn’t a government entity: they’re in business to make money.

    I disagree that Europe doesn’t have a “truly commercial” launcher. Ariane 5 is, as are the Soyuz launchers that ArianeGroup operates. Whether they can compete with the other international ventures is a different story.

  • Aerospike

    Of course the lines are really blurred here, but I really would not consider the European launcher sector (and the current developments) as “truly commercial”.

    Yes, SpaceX also got a lot of money from government development contracts (Falcon 9 and Dragon for the most part), but imho ArianeGroup still operates more like ULA does and less like SpaceX. They do what government entities tell them and pay them to do, not what they want out of their own incentives.

    And Ariane5? That is an ESA design as far as I know.

  • …but imho ArianeGroup still operates more like ULA does and less like SpaceX.

    The vast majority of Ariane and Soyuz launches out of Kourou are for commercial payloads–even more so than what SpaceX launches. ULA, on the other hand, is rapidly becoming an arm of the Air Force.

    I agree that the Ariane design parentage is heavily ESA-influenced, and is even more incestuous than ULA, due to the need to pass out pork to more constituents. And I’m not even going to pretend to understand the current iteration of the corporate divisions between ArianeSpace, ArianeGroup, EADS, Safran, Airbus, and the rest of the cast of characters. But the net result is highly commercial.

  • Jan Bach Andersen

    One thing i am wondering about when speaking about these new engines , Eleon said ind his latest presentation that they have reduced the engine size due to thottle.. Blue origins new engine is now much bigger than the raptor and the rocket is smaller than the BFR ..but they dont seem to that the engine is to big thottle down to land ..

  • duheagle

    With all due respect, anyone still holding out “hope” that reusability will fail is nearly on a par with Flat-Earthers and Young-Earth Creationists these days.

    As to engine economics, that depends not only on what the incremental variable cost of manufacturing one additional unit is, but also what the projected total production volume is going to be.

    SpaceX designed the Merlin to be as cheap as possible to manufacture from the get-go, then incrementally continued to improve the design and take out costs. It also figured on building Merlins in then-unheard-of quantity – and has done so. The total number of Merlins produced to-date is probably somewhere north of 500. That number will likely have doubled as little as two years hence.

    The current production cost of a Merlin is somewhere south of a million dollars. Given its modestly larger size and greater complexity, a Raptor might cost twice what a Merlin currently does, at least initially. Long-term, Raptors could well cost even less per unit than Merlins do now.

    Raptor is the beneficiary of all the designment, refinement and continuous improvement experience accrued in building Merlin. But Raptor’s production volume will, in all likelihood, vastly exceed that of Merlin over time. BFR will need, per unit, four times as many Raptors as F9 does Merlins – half again as many as even FH.

    The number of BFR boosters built, over time, will also exceed the number of F9 1st stages if Elon’s point-to-point transport scheme takes hold. Assuming BFR boosters prove capable of the claimed 1,000 launches, the boosters used to dispatch deep space missions would constitute only a minority of the total in service and might last decades apiece. Point-to-point service, though – assuming once-a-day round trips between any given pair of served city pairs – will exhaust a BFR booster’s 1,000 lives in less than 18 months. Cranking out replacements for the entire P-to-P fleet every year and a half is going to require a Raptor production rate that would make even the Merlin’s numbers look cottage-industrial by comparison.

    At the moment, no one who doesn’t work at either ULA or Blue Origin knows what a BE-4’s “retail” price is going to be. But, over time, it too will depend on production volumes. “Bezosium” or no, I don’t see Jeff B. looking to book perpetual losses.

    Given Blue’s announced intention of building only a handful of New Glenn 1st stages and the rather modest incremental production of BE-4’s that will be required to supply ULA, I simply don’t see where BE-4 is likely to be less expensive, on a fully-burdened per unit basis, than the Raptor – unless Jeff B. decides to give Elon some competition in the terrestrial P-to-P market. Now that would be fun!

    As for Prometheus, it is roughly a match for Vulcain 2 in terms of sea-level thrust and is going to be air-startable, deep-throttleable as well as reusable. Three good solid steps forward.

    But to be economically reusable, Prometheus must also be part of a vehicle architecture that will have to differ radically from Ariane 5 and 6. In particular, the solid boosters have to go and the 1st stage has to have enough engines to allow it to land on the throttled thrust of just one.

    A single-stick configuration with 13 Prometheuses in the booster stage would yield lift capacity, in fully expendable mode, as good or better than the Ariane 5 or 6. But one would want to allow for the propellant reserves at MECO and the landing legs necessary for RTLS or at-sea recovery. That would likely require a total of 17 engines – one center engine, eight in an inner ring, ala SpaceX F9, and eight more in an outer ring.

    A design of this sort would still not likely be truly competitive with SpaceX or Blue on a truly level playing field, but it might be – literally – close enough for government work.

  • Hypx

    With all due respect, it’s pretty clear you’ve drank the reuse kool-aid. Reuse has never been the silver bullet, and has failed miserably before. At best, it’s going to be a small cost saver. If you’re going to try implement reuse, it should be with a rocket that’s relatively cheap to build. Prometheus may fit that bill. The only way Raptor or BE-4 works out is if reuse really does become that silver bullet, a highly risky bet.

  • redneck

    You’re right that reuse will just be a small cost saver and should be with cheap rockets only. If the choice is between an expendable at 100 million and a 1,000 use vehicle that costs a billion, it should be obvious that a million a flight for the reusable is unsustainable.

  • Raptor doesn’t work at scale yet, and the actual life of the engine (in thrust minutes) has yet to be determined. Same thing with the BE-4 actual life.

    Beyond that, there’s any number of ways to have problems with the entire launcher system, not just the engines. I’m pretty confident in both SpaceX’s and BO’s ability to deploy something, but how cost effective they are and when they go into service are still highly uncertain.

    The best I can muster for Prometheus is that it’s sorta like Step 1 in an AA 12-step program. But, like Roscosmos, Europe seems to be shooting for where the market is now, not where it’ll be in 6 or 7 years.

  • duheagle


  • duheagle

    Reuse isn’t “kool-aid,” it is now Standard Operating Procedure at SpaceX and Blue Origin. Even ULA is looking to do some reuse, though their approach is deficient and Rube Goldberg-ish IMHO. Shuttle reusability failed to be economical because of poor engineering choices. SpaceX and Blue have declined to follow in NASA MSFC’s infamous footsteps by designing vehicles that do not require weeks-long partial rebuilds between missions by a small army of touch laborers. Reuse allows operations costs to asymptotically approach those of propellant alone. Nothing the least bit “risky” about it.

  • Hypx

    Neither one has proven reuse to be worth it just yet. Blue Origin hasn’t even launched a payload into orbit. For all we know, reusing a F9 could require a small army of touch laborers too to refurbish one back to working form. It’s pretty obvious you should start with a cheap rocket that could potentially be profitable with minimal reuse rather than an expensive one that requires reuse to work perfectly for hundreds or thousands of times.

  • duheagle

    What, exactly, would you regard as suitable “proof?”

    No, Blue hasn’t been to orbit yet, but it has launched and recovered a single suborbital vehicle five times and a second one once. The turnaround cost for New Shepard 1.0 was said, by Bezos, to be “a few tens of thousands of dollars.” NS 2.0, which just flew, is evidently even more maintenance-free. Blue will have more applicable experience with reuse when New Glenn flies than SpaceX had when it started converting the F9 to reusability.

    As for what is known about F9 reuse, Shotwell said, after the first reuse mission for SES back in March, that SpaceX had deliberately done a lot of disassembly and inspection of the candidate booster and took six months to do so. Even so, the cost was less than half what a new booster cost. Since then, the reuse processing flow has gotten much less labor-intensive. When the Block 5’s fly, turnaround labor costs will take another huge drop.

    And SpaceX did “start with a cheap rocket.” An analysis I initially conducted several years ago – and which has been buttressed by additional data that have subsequently come to light – led me to conclude that it costs SpaceX about $16 million to make an entire F9. I still catch flak for that conclusion, but you may take to the bank the fact that there is no one who thinks the F9 is cheaper to make than I do.

    Even so, reusing F9’s is a solid economic win. The 1st stage accounts for roughly $12 million or so of that $16 million cost total. So even if SpaceX was still treating all reused stages like it did the first one – which it isn’t – reuse would still make financial sense.

    Given that every other rocket currently in service costs multiples of what an F9 costs SpaceX – despite all of them also being completely expendable – I can only attribute the have-to-be-dragged-kicking-and-screaming attitude of legacy aerospace to the whole reusability idea to sheer bloody-mindedness. Any one of the legacy firms that implemented reuse would profit even more from doing so than SpaceX does, but apparently those particular mules haven’t yet been hit on their heads with a sufficiently large clue bat yet.

  • Hypx

    This is pure speculation on your part. What we do know is that SpaceX raises external capital, on the tune of $100 million and $350 million this year. Obviously, something at SpaceX costs a lot more than anticipated. We’ll have to see internal numbers out of SpaceX before we can make any judgments. But it’s pretty obviously they’re not printing money from reuse.

  • duheagle

    Speculation, but hardly “pure.” It’s based on evidence – statements by SpaceX officials over a period of years and the posted prices of F9 and FH launches. This is quite unlike your assertion that SpaceX may be harboring some secret army of touch laborers to refurb its rockets.

    SpaceX raising more money is hardly a sign of unanticipated difficulties no matter how “obvious” you may allege to find it. The money from the latest financing round will, along with the considerable sums it earns from ongoing operations, allow the company to make a full court press on building BFR.

    And we have seen some “internal numbers out of SpaceX.” SpaceX didn’t release them on its own, they were purloined in some fashion by a Wall Street Journal reporter and published in that paper back in January of this year.

    Unfortunately for the SpaceX haters in the crowd, the numbers did not show that the company was in any sort of financial difficulty. The numbers confirmed claims by SpaceX that the company had been profitable for a number of years, losing money only in 2015 and 2016 when accidents occurred. The numbers also buttressed my “speculation” about SpaceX’s costs of F9 production and their profit margins on missions.

    Your chosen “handle” in these forums – Hypx – makes it pretty clear you are one of the dwindling cadre of cynics who imagine, to varying degrees, that SpaceX is all, or at least substantially, smoke and mirrors. The past few years have been frustrating for those of your evident persuasion, especially the one now nearly past. The coming few years show every sign of being even worse.

    If the worst of your mistaken beliefs anent SpaceX is simply that the company has yet to “prove” the economic benefits of space vehicle reusability, then then you are just stubborn. If you are actually of the opinion that SpaceX has some secret army of helots refurbing their 1st stages, then you have crossed the border into Tinfoil Hat Land.

    Fortunately, you can probably still see the border and, thus, can easily find your way back should you so choose. Others, progressively less rational than yourself, have gone so far into the interior of THL that they now seem hopelessly lost there.

    One fellow, for example, is convinced that the U.S. government is somehow or other supplying SpaceX with a $400 million subsidy for each of its launches. Another believes that Elon Musk is, simultaneously, a radical libertarian acolyte of Ayn Rand and is also besties with Barack Obama.

    Progressing up the scale of mental dysfunction, we find people who think all of SpaceX’s achievements are fake and all the copious video evidence of same just exercises in CGI. How these folks explain the sonic booms whenever SpaceX does an RTLS stage recovery to LZ-1 I haven’t taken the trouble to find out.

    Next up are the people for which SpaceX denial is just a modest addendum to their belief that the entire U.S. space effort has been a scam and a fraud. Moon landing denial seems to go back almost as far the landings themselves, much as Holocaust denial appeared shortly after the end of WW2.

    The absolute peak of lunacy is occupied by the self-proclaimed believers in either or both a flat Earth and a geocentric universe.

    I would urge you to climb back down from your apparent current place low on this particular pyramid of crazy. Continuing to climb will, in the end, get you nowhere.

  • Michael Halpern

    Fast, high quality or cheap, they chose fast and cheap, traditional choice for ELVs

  • Hypx

    FYI, those WSJ leaks suggested SpaceX only had a small gross profit in its best year, not net profit. Realistically, if were publicly traded and had to report numbers using GAAP methods, they would show a loss every year. They also have a huge workforce (7000 workers), a number that doubled in the last 3 years. If there is a huge laborer requirement for reuse, SpaceX is so far proving it to be true. The evidence suggests they are not saving much money from reuse, and certainly worlds away from your claims of the F9 being many times cheaper than anything else.

    Anyways, I’m increasingly unenamored by your condescending debating style. Especially when you use ad hominems or strawman tactics. It’s clearly impossible for you to grasp that some people actually have reasoned disagreements against SpaceX that aren’t crazy. There’s not much point in continuing this discussion more further if this what I have to put up with.

  • duheagle

    The small profits are deliberate. SpaceX spends most of its considerable profits from operations on more R&D and capital expenditure. That is typical of closely-held startups. SpaceX is no longer a startup, but Elon finds it very useful to continue operating like one. In addition to saving many millions per annum on government reporting, Elon, personally, saves time for more worthy pursuits by not having to spend it soft-soaping Wall Street.

    SpaceX does have outside investors, though, so its books are definitely audited. The outside investors providing recent funds are the same small group who provided SpaceX’s earliest outside funds a decade ago, except for Google and Fidelity a couple years back. SpaceX’s investors are all down with Elon’s vision – especially his vision for growth through aggressive capex – or they wouldn’t have gotten in in the first place.

    As to the WSJ-pilfered financial numbers, one of them was the quarter billion loss SpaceX took in 2015 pursuant to the CRS-7 failure. Absent said failure, SpaceX would have done roughly six more launches that year. Given my contention that SpaceX makes $40 – 45 million per list-price launch, that 2015 loss is just a bit less than what SpaceX would have made had it gotten to do those launches in 2015. If CRS-7 had succeeded, SpaceX would have kept launching and booked the usual modest yearly profit it did in at least the three years preceding 2015.

    SpaceX’s workforce is certainly large, but not in comparison to what it does with that workforce. Sure SpaceX’s headcount has more than doubled in the past three years, but it is also building and launching a lot more rockets now than it was three years ago too. ULA’s workforce is about 40% that of SpaceX, but it also outsources much of its vehicle work, notably the engines. SpaceX is famously very vertically integrated. ULA has a huge “tail” of contractor employees that is not reflected in their nominal workforce stats.

    If it comforts you to believe that SpaceX is overstaffed and much of said staff is busy pressure-washing used rockets, well, different strokes. You at least have the distinction of putting forth a novel criticism. Many of SpaceX’s detractors fault it for having too few people to properly pursue all the projects it has on its plate.

    As for ad hominem and strawmen, I plead not guilty. I went out of my way, in fact, to indicate I do not regard you as being in anywhere near the same class as quite a number of other SpaceX detractors. Nor did I engage in ad hominem or the construction of strawmen even in relation to them. When someone actually states in a public forum that he believes, for example, that the Earth is actually flat and that there are no such things as satellites, it is not ad hominem to call that person a Flat-Earther, it’s just an accurate shorthand description.

    You believe neither SpaceX nor anyone else has yet “proven” the economic benefits case for space vehicle reusability. I think you are seriously, even profoundly, mistaken in that belief, but I don’t think that makes you a nut.

    And I agree that this exchange has probably reached a point of nearly completely diminished returns. This thread is now so far down in the weeds that no one except we two are, I suspect, any longer keeping up with it. So I will accept your offer to go our separate ways – at least anent this comment thread.

    I will, of course, feel free to comment, or perhaps even give you a thumbs-up click, on any posts I encounter under your nom-de-guerre elsewhere, either on this forum or any others we may mutually frequent. I find that there is pretty much no one with whom I disagree absolutely all the time.

  • Hypx

    I agree this discussion is over. I will finish by saying I’m confident SpaceX will someday go public, and it will do so because they flat out need the money. If or when that happens, I think it will disprove at least one of your firmly held beliefs.