The Future of Space Launch Is Near
justatinker.com
justatinker.com
Ariane 6 seems to be in response to the SpaceX challenge, at least; they aim to halve the cost per kg through cheaper and common upper/lower stage components. Which is good, but still not competitive with F9, much less a reusable version. At one point they almost made it all-solids in the first two stages, plus boosters, which would have totally doomed any future attempts at re-usability. The final design uses a F9-style common component liquid core, so they could maybe do the same thing as SpaceX later. They are still stuck with solid boosters, though, in a pork-provision move for Italy.
I've got to imagine that ULA is quietly looking into it. But it's quite possible they're just hoping SpaceX will eventually implode of its own internal contradictions and they can go back to their nice little government-launch monopoly. Or maybe they've decided they can't do it without absurd expense, based on their internal development costs. Or maybe they just hope to coast on their existing contracts for a while, and deal with it later once SpaceX has taken all the lumps.
NASA, via SLS, sure isn't looking at re-usability anymore. Lower cost to orbit via research should be their main task, if you ask me, but Congress would rather have them fly giant rockets a couple times a decade, apparently.
The Russians, well, are not much into changing things up. (They've been running Soyuz and Proton, their current launch families, since the sixties!) And their recent aerospace development programs have been less than impressive. But they have a real industry to protect, so one would hope they have a strategy.
I think that everyone's waiting for SpaceX to make it work, and then they'll get to copying. My long-term guess is that this'll kill several of the incumbents, a few will make the transition eventually, but most of SpaceX's competition will come from new entrants. In the meantime, SpaceX will be free to reap the profit from the margin between what it costs them and what their old-fashioned competitors can do. And/or capture the whole darn market. And/or grow it with lower prices. Maybe all of the above.
Ariane 5 went expendable specifically to make it feasible to build, where Sanger and similar SSTO re-usable spaceplanes were too far-fetched to build.
The EELV and Ariane were a reasonable approach at the time to make rockets a bit more affordable. Any way to do reusability at the time would have required a novel and exotic architecture, with a matching exotic budget. And that just wasn't in the cards.
But now it's not that impossible or expensive to ask a rocket stage to drop out of the sky and sit on its tail. (The hovering rocket is a garage project at this point, though a difficult one--but mostly because of the "rocket" part and not the "hovering" part.) Any existing stage could be modified to do the same with modern avionics, with the addition of an RCS and restartable engine (if it doesn't already have those). Sure, SpaceX's choice of 9 throttlable engines makes it easier, but you can do the "hover-slam" with just one giant one.
Any incumbent who can't manage to do so will rightly be done within a decade. And as hide-bound as they all seem to be, that might be most of them. They'll still have to deal with SpaceX's manufacturing efficiency, which might be even harder, as manufacturing inefficiency is part, for example, of Ariane's DNA. But at least they'll stay in the race.
After stage separation, it detach the entire engine block (where most of the cost is), and a helicopter captures the parachute mid-flight! (See figure 2 from the PDF)
Airbus wanted to use SpaceX but the French flipped out so they have to stick with Ariane.
You see, there are two ways to build reusable rockets. One way is to build a purely reusable vehicle from the ground up, but this is extraordinarily expensive because of all the testing involved before it can be used commercial. The other is to build an expendable/reusable hybrid. A rocket that can be used in an expendable configuration but has all of the needed elements to allow for reusability down the road. This is vastly easier because it means you can bootstrap development on the back of commercial launch business, and do testing that is largely subsidized by paying customers. And this is what SpaceX has been doing.
However, there's a problem. You have to head into reusability from the get go, you can't just modify any old rocket for reusability, you need to design from a place of suitability for reuse from square one. And while that's not an enormous difficulty it is very much non-trivial, and imposes a substantial overhead on R&D and manufacturing costs. This is because most of the optimizations for building a purely expendable rocket drive the design away from the optimizations you'd want for a reusable rocket. In an expendable design it's common for there to be a mix of propellants and a diversity of engine designs between the first and upper stages, it's also common for the upper stage to be more expensive than the first stage, because that stage is more critical for payload performance and launch precision. Also, it's common to minimize the number of engines on the first stage. Even the Saturn V had only 5 first stage engines. And generally there'd be no reason to add first stage engine restart capability on an expendable vehicle.
All of those things results in an expendable vehicle that is completely unsuited to reuse. It leads to a first stage that is very difficult to modify for reuse, and an overall rocket design that does not reward the easiest type of reuse (of the first stage) because most of the cost is in the upper stage anyway. SpaceX went to a great deal of effort to ensure that they built a rocket that could serve the expendable launch market immediately (and pull in hundreds of millions of dollars in revenue) while also being suitable for reuse with comparably moderate modifications. The first stage has 9 engines, so throttling down to a thrust level suitable for precision landings of a mostly empty stage is possible by using only one engine. The first stage use restartable engines (using onboard TEA/TEB igniters). And the first stage is about 3/4 of the hardware cost of the vehicle, since the 2nd stage uses only one engine and is also LOX/Kerosene fueled.
That foresight has allowed SpaceX to be in the launch business for several years, proving the capability of the company and gaining much needed experience in building and flying rockets while also poising the company to bring to market a reusable launcher. In comparison, every other established launch provider in the world could only do reusability by designing a new launcher from scratch.
Why wouldn't they use radiation hardened computers, though? Are they that much heavier and more expensive than the more commonly available components? Out of a $40 million spacecraft, the difference of a few thousand dollars or even a hundred thousand dollars seems rather trivial, in exchange for a more reliable avionics system.
It's not about cost at all:
> It's really not the expense that drives it. We're committed to having the best possible parts in all of our designs. So if it cost a lot and we needed it, we'd go get it. We were already required to have all this redundancy in the computers to meet all the different safety requirements. Then we started looking at what parts do we want to use and what is appropriate for this design. And what really is more important to us than the cost of the parts is the capability of the parts – how much power do they use, how much memory do they hold, how much do they process, and how physically big are they. That's the first thing.
> The second thing is what tools they come with. We run the Linux operating system, we program everything in C++, and that enables us to tap into a huge pool of very talented people and find the absolute best people in the computer and software industry to work with us. If you go into the radiation hardened parts, they are very limited in terms of what languages you can work in, what support packages there are for them, who knows how to program in them. It really limits your ability to work with the parts. And the other thing it really does is they all take a little longer time to get and they're a little harder to come by.
Also, there is something about using the same engine on every stage to gather more data per launch, but iirc different engine profiles are used for different velocities and air densities long the launch. I wonder what's the efficiency loss of using a single engine profile for the whole launch.
No, it reduces payload to orbit by ~30% or so. Fuel is insanely cheap in comparison to the cost of building the 1st stage.
The Merlin 1-D Vacuum used on the upper stage does have a wide vacuum optimized nozzle, but the other parts are almost the same, to save costs during manufacturing and testing.
The article is ok, but says nothing new to people who have followed SpaceX for years. There is a lot of info online about the company and its products.
It's probably not even as bad as that, since they certainly fly with some performance reserve already, at least several percent, and most payloads are probably not max-weight. The landing propellant in many cases may be entirely free, mass-budget-wise. They do add some additional mass for the legs and fins, in addition to any extra propellant they need, so it's not free, but the first stage is not as performance-sensitive as higher stages (see: rocket equation), so they can get away with it. I don't see 'em getting second stages back any time soon, though, and I think they've given up on that, at least for F9.
WRT engines, the efficiency based on air density is mostly a function of the nozzle, and the upper-stage Merlins have a nozzle extension to deal with this. Higher-efficiency fuels (i.e LH2) are historically preferred for upper stages, since they are more performance-sensitive, so SpaceX is losing some performance there. But LH2 is super-cryogenic and a beast to work with, and an entirely different engine. SpaceX's thing is that they are willing to sacrifice maximum efficiency for reasonable cost, so they decided to make that tradeoff. It seems like a good plan so far.
It does reduce payload capacity (30-40%, IIRC) but the expected upside is drastic cost reductions. If I can launch several payloads for the price of one as a result, it becomes well worth it, and if I've got a large payload I can either use the upcoming Falcon Heavy or we'll start doing more in-orbit assembly.
The weight of the legs is probably quite low so this might have been the easiest solution.
Maybe they just wanted to keep complexity down? Or they simply can't land with sub meter resolution to garuntee they get within the claws reach?
Also I think they want to use the same system on Mars where there won't be any infrastructure.
Thus the landing profile is the "hover-slam": the engine is started at the exact point in time necessary to reach a "landable speed" (from 0-several m/s) at the ground, and is cut off when it gets to the target.
While modern avionics are quite good, it's still coming in pretty fast and from far away and will have some variance in location and speed, just because that's the real world. Probably they don't quite know how much yet, but it could easily miss the target by multiple meters and a few meters/s. And that's a heck of a condition to build some sort of giant grabbing apparatus for.
Perhaps once they're routinely landing on land, they may be able to determine they can hit the target well enough to build a grabbing contraption. Or maybe it's too hard. Or maybe they'd rather keep their infrastructure to a minimum: a concrete pad is a lot cheaper to build and easier to operate, and SpaceX wants to do a lot of flights.
Besides, the 4000lb of the legs is on a 40,000lb rocket.
It's been said to be capable of 60% (or maybe 40%, depending on how you interpret a three-character tweet). But even that is still not enough: the stage is 40k lb empty and a Merlin D at sea level is 165k lb thrust. You'd need to get it to 25-33% to hover nearly-empty.
(Keep in mind most SpaceX numbers involve a fair amount of guesswork and Kremlinology, and they're always changing, but they should be ballpark accurate.)
The F9v1.1 cannot hover when landing as the TWR with an engine at 60% thrust (the minimum) is almost 2.
Even if unsafe could it still make sense for unmanned launches, supplies, cargo?
This means you need fire safety everywhere, and heavy ground equipment too for lifting and attaching and assembling the solid rockets.
Meanwhile, you can fuel a liquid rocket at the pad, meaning it's empty and light and can not explode or burn during transport and handling.
So solids cause heavier rocket ground infrastructure and more danger.
Of course, with liquids you need the fueling infrastructure on the pad. It probably comes to a question, not only of scale, but also of launch rate.
In theory you could have mobile trucks fueling missiles here and there. That's how you do it at ship ports where you don't want to invest in ground infrastructure.
An earlier design for the Ariane 6 was 4 identical solid boosters, 3 on the 1st, and one for the 2nd stage--before they came to their senses.
Solids are convenient and reasonably simple and good at high thrust and smaller scale, but don't scale up as well as liquids. (You try casting a fuel grain the size of an office building!) So they're not well suited for launching anything particularly large into orbit. And unfortunately they have nowhere near the mass fraction to do SSTO. Liquids, though, are tantalizingly close.
What exactly were these "Air Force requirements"?
1) The payload bay grew in size from 12 ft wide to 15 ft wide, in order to accommodate future unspecified military needs.
2) There were new cross-range requirements specified for reentry.[2] Basically, the military wanted to be able to launch satellites into polar orbits (useful for surveillance) and retain the capability to land after one orbit of the Earth. However, this meant that the Shuttle also needed to be able to -reach- the landing sites, which are normally not in the direct orbit path, but are divert 1000 miles away laterally (the cross part of cross-range).
[1] http://en.wikipedia.org/wiki/Space_Shuttle_design_process#Ai...
[2] http://yarchive.net/space/shuttle/shuttle_crossrange.html
Current gas turbines are not optimized for pure thrust-to-weight or else why are military engines outperforming civilian in thrust-to-weight? Turbofan-to-turbo-jet for high Isp in atmospheric flight and then fly everything back on its own. Firefly's rocket has a max thrust of 90klb on the main while a big gas turbine pulls 100klb easy. While we're talking about Isp and re-usability, the entire first stage of a rocket could be gas-turbine powered and the engines just peel away and fly home after they're done. The engine needs gas. The gas takes space. Put the gas in a wing. Wing glides the engine home after launch. Easy as 1,2,3-a-bunch-of-engineering. Go!
It has been successfully tried - http://en.wikipedia.org/wiki/Pegasus_%28rocket%29 - but my understanding is payloads are pretty severely limited. You can see the size of the Pegasus launcher is much, much smaller than a Falcon 9 - it's certainly not going to work for SLS/Falcon Heavy/MCT-sized payloads.
edit: For ground-launch with jets as the first stage, I really don't think that's going to work.
http://en.wikipedia.org/wiki/Thrust-to-weight_ratio#Jet_and_... indicates the SR-71's engine had a thrust-to-weight ratio of 5.2. SpaceX's Merlin has a thrust-to-weight of 150. There might be room for improvement, but 30-fold? I doubt it.
More modern F119 has a T:W of about 8. With ceramics and pure focus on re-heat and specific thrust, you can get over T:W of 10 likely. There may be other tricks on fuels besides jet fuel that might even allow more focus on re-heat, making the T:W even higher without necessarily affecting fuel consumption.
Going COTS with F119 is interesting, but ceramics might make some game-changers. Ceramic turbine blades and afterburner pipes will drastically lower the weight vs Nickel superalloy parts while lower cost and manufacturing complexity. Turbine blades are really impressive, but it's all cost.
Main obstacle there is that optimum staging altitude and speed are very high by aircraft standards. 100km and Mach 10. Not much air up there. So you couldn't replace an F9-1 like that. You could perhaps use it as a 0.5-stage to get an otherwise SSTO rocket over the hump, but it may not be worth it compared to a standard architecture: still requires a separate giant lower-stage airframe, and one with much different technology from the rest of the system.
Max q is what makes it seem right. An STS flight profile can't even use max thrust again until 50k ft and Mach 2. Turbojets peeling away and glide back from around 60k ft and Mach 2.5 seems right. This is easier than a sea recovery, and because the burn duration is so short, it almost doesn't make sense to do anything but go full reheat and target the highest thrust in the lightest package. As more ceramic matrix composites replace Nickel, I would expect the T:W to get over 10, so it's really about what the smallest airframe is that you can wrap around a gas turbine and still land it on its own.
The airframe size of the 0.5 stage (1st stage?) is the biggest departure from normal. You can get ten F119's around the base of an F9 without thinking too hard, but that's only ~400k lb thrust for 40k lb of engine and almost doubling the diameter at the base. Could get weird.
Ceramic composites will be pretty critical. If you can replace Nickel parts with ceramic, you can get rid of bleed air and raise the temperature, which is what you need when you're sacrificing a little extra NOx for vastly higher T:W.
A smaller 1st stage fuel tank might mean that heat-shielding or higher temp materials is more of an option, which means you can raise the 1st stage separation velocity closer to what is "optimum" in the rocket sense and still recover the 1st stage. This puts the 2nd stage closer to its natural optimum design point.
Overall size might not be a show-stopping problem now, but for building space colonies or Mars missions, overall vehicle size does eventually become concerning. Think of trying to build a jig to friction-stir weld a 180m long, 14m wide fuel tank. It's cool but also a bit of a distraction to have to build and handle everything at unusual scale.
It would be super cool to see big packs of ten F414's at the base separate off, curve back towards the launch site, flip out some wings, and then cobra into a vertical landing. GE should step in right now to have the perfect platform to develop ceramic parts without the same service life requirements as aviation gas turbines.
I wouldn't build a 180 m long 14 m wide fuel tank. I'd launch lots of smaller missions and do space refueling, assembly or docking. None of the components of any mission are really credibly over 20 tons dry. Look what the air forces do - they don't build massive aircraft with excessive range - they do aerial refueling.
Space launch is an acceleration mission. Your goal is over 8 km/s. The air breathing turbine engines can help perhaps a few hundred meters per second, for a lot of dry weight, and that's nice. With wider speed ranges they get even more complex with finicky inlets and produce diminishing results.
There's been some cool engines in the history that could provide some interesting capabilities http://en.wikipedia.org/wiki/General_Electric_GE4 I don't know what a jet engine designed for acceleration would look like.
I think the most potential in air breathing lies in air launch of relatively small rockets. Dropping the rocket from a large subsonic airplane. It brings a lot of flexibility with launch sites and times, plus abort options for the dropped vehicle.
Including the rest of the dry mass fraction of the vehicle, (the tank, the structure to hold the tank etc), each Merlin's dry thrust-to-weight ratio drops to 43.05.
(654kN * 1000 N / kN / 9.8 (kg * m/s2)) / (Inert mass / 9.0).
The GE F414 in the high-thrust configuration is the dry-mass comparison using COTS. Let's say it's 8.0 when factoring in some control system and inlets.
Isp of a jet engine burning liquid methane is ~2000 even in full reheat. Isp of rocket engine is ~330 for methane-lox in the newer full-feed Merlin. The impulse to fuel weight ratio is almost 10x inverse in favor of the air-breathing system.
The dry mass thrust-to-weight is decisively in favor of the rocket engine. The wet mass thrust-to-weight is decisively in favor of the jet engine. Affect of these factors on cost is inconclusive from the napkin.
How can air-breathing engines potentially lower the TCO?
46 reusable RS-25 were flown on 135 shuttle missions, so the average lifetime of each engine was...8. The lifetime of a gas turbine is measured in thousands of hours, so let's say 200 missions and without so many major overhauls. Even if Space X is pulling out some cool voodoo for Merlin reuse-ability, gas turbines are in a different league in terms of both reliability and lifetime.
While a 1st stage rocket body has to re-enter, land, and be transported back to the launch site, an air-breathing 0.5 stage can fly itself back from its short distance downrange, thereby incurring zero transport cost back to the launch site.
An extreme application of air-breathing might include ram-jet inlets that opened after mach ~1.8 and took over air-breathing duties up to Mach 4.0 as the 414's idled. Even in that case, the whole system is not that far downrange and can use altitude to make up quite a bit of the journey back and use some vertical landing system if it saves weight. The fuel efficiency lets you do this with 414's in spite of their lower T:W.
Does using air-breathing assist pay off by making a lighter 1st stage or heat-shielding the 1st stage to make a faster release point? Only extensive analysis of the various options can answer these questions. I believe the mechanisms are there, and that's what justifies the design research.
"The trade-off with upgrading the engine to produce 26,400lb-thrust is a considerable hike in maintenance costs. Running the F414 EDE at the higher thrust setting reduces turbine life to 2,000h, Caplan says. This is just one-third of the current 6,000h interval."
"GE also has the option of switching to a more heat-resistant material to make the blades. The company is considering designing the first-stage turbine blades of the GE9X with silicon carbide-based ceramic matrix composites (CMC), which are lighter and more heat-resistant than metal. Indeed, the company tested CMCs in the low-pressure turbine of the F414-GE-400 for a 2011 demonstration."
Turbines designed for rocket assist need to live <100hrs to service 200 launches. GE has basically already said they can push the engine to > 11:1 T:W, not including the advantages of ceramics or LNG combustors, which they also know how to make plenty of for utility gas turbines.