Aerobraking vs Propulsive Braking
A lot of commenters in the last thread asked why I like aerobraking over propulsive braking for return to LEO. The answer is simple. Propulsive braking requires just as much delta-V going out as it does coming in. Which gives you a whopping 8.4km/s end-to-end delta-V requirement. That's practically the same delta-V range as an SSTO vehicle. The K-1 upper stage couldn't even fly its own weight (empty!) to lunar orbit and back using propulsive breaking. I checked the numbers assuming something like LOX/subcooled-propane, since that mixture tends to get 20s better Isp, and has the same relative volumes as LOX-Kerosene, so the K-1 OV could conceivably be upgraded to use it, possibly just require requalification, with few if any physical modifications to the stage. Problem is that the most payload I'm getting is in the 1-4 tons to Lunar Orbit range, based on a 150 ton stage in LEO with about 131.5 tons of propellant. That really sucks. Even if LEO prices got ridiculously cheap (say $50/lb in LEO), then you're still talking about $7000/lb in lunar orbit. If you use more realistic numbers, it climbs to prices that make Pegasus and the Shuttle look economical. It's a nonstarter.
Now, you could switch to LOX/LH2, but now you're talking about a whole new vehicle, probably with a lot worse of a drymass. The numbers for that actually look like they might work, but you've now pigeon-holed yourself into a single-propellant combination, that just happens to have the worst bulk density, worst boiloff issues, etc. And you don't even get as good of performance as a LOX/Kero stage with aerobraking.
Maybe someday in the future where you have massive amounts of ISRU derived propellants available in lunar orbit, you might just be able to get such a system to work, but why not just figure out how to aerobrake. Is it really that much tougher? I don't think so. Not by a longshot. Aerobraking is one of those technologies whose payoff-to-risk ratio is so high for longterm space development, that you may as well consider it a neccessity.
Now, you could switch to LOX/LH2, but now you're talking about a whole new vehicle, probably with a lot worse of a drymass. The numbers for that actually look like they might work, but you've now pigeon-holed yourself into a single-propellant combination, that just happens to have the worst bulk density, worst boiloff issues, etc. And you don't even get as good of performance as a LOX/Kero stage with aerobraking.
Maybe someday in the future where you have massive amounts of ISRU derived propellants available in lunar orbit, you might just be able to get such a system to work, but why not just figure out how to aerobrake. Is it really that much tougher? I don't think so. Not by a longshot. Aerobraking is one of those technologies whose payoff-to-risk ratio is so high for longterm space development, that you may as well consider it a neccessity.
posted by Jon Goff at 2:01 PM
17 Comments:
That really sucks. Even if LEO prices got ridiculously cheap (say $50/lb in LEO), then you're still talking about $7000/lb in lunar orbit.
That's what the mass drivers and tethers are for, dear.
But long before then, an active aero braking system seems like a very good idea. There will need to first be considerable transport to and fro long before tethers or mass drivers take over, and it is the here and now with which we are most concerned.
Aero braking need not be that hard. For example, envisage a nose cone four sides of which actively hinge out like a shuttlecock or four petals on a flower. By actively controlling how much each flap hinges out, L/D and frontal area can be independently controlled. If one goes too deep just fold the flaps in a bit – easy abort. This general approach should allow one to actively fly the aero braking manoeuvre, which is what is required, and this can also mostly avoid the need for a circularisation burn.
I was watching Serenity recently, noticed the early scene of a fiery re-entry, and thought "how retro" (as it were). If you've got the delta-v to hop between planets in days, you've got more than enough to float/fly in without all the fireworks.
That said, I agree: as long as we're stuck with these Isps, and have been forced to invest so much in technologies and techniques for unpowered re-entry, we might as well refine them for the special case of aerobraking (aka "re-entry interruptus.")
I have been looking at some solar thermal power stuff of late and I suspect that a very aggressive system might ultimately get near 10 kW/kg. (Thin plastic film inflatable radiators, single stage closed cycle gas turbine with a xenon working fluid, a very high speed generator, perhaps super conducting, etcetera.)
A one ton power system might have a 10 MW electrical output. If one could develop a rocket engine to match, perhaps one that even used insitu LOX, going to the moon and back, or mars and back, without aero braking might be practical. 10 MW equates to 2000 N thrust at 0.2kg/s, (assuming 100% efficiency). Travel times might hopefully be similar to chemical rockets.
> Now, you could switch to LOX/LH2, but now you're
> talking about a whole new vehicle,
At that point, since you're no longer talking about the K-1 anymore, it makes sense to consider other architectures.
For example, do you really want to bring those great big propellant tanks back with you? If the tanks can be productively reused (in lunar orbit, on the lunar, or at L1), the answer is probably "no."
For the near future, at least, there will be many more tons of outbound payload than return payload. So, a tug could be sized with just enough internal propellant to make the return trip with a small payload (e.g., a few passengers and a hundred kilos of rock samples), while all the propellant for the outbound leg is carried in drop tanks that are themselves part of the payload.
> you could switch to LOX/LH2, but... you don't even get
> as good of performance as a LOX/Kero stage with aerobraking.
One reason to doubt Mike Griffin's claims that he would reuse the Earth Departure Stage, if an orbital propellant depot suddenly appeared.
Ed,
At that point, since you're no longer talking about the K-1 anymore, it makes sense to consider other architectures.
Well, there are plenty of ways that you could more finely optimize a lunar transfer stage. However, the nice thing about the K-1, is that between NASA COTS funding and private sources, they've got another $600M committed to the project, which means that there's a pretty high probability that K-1 will actually make it into service. While the OV may not be the absolutely perfect LTV, it sure looks pretty darned good for something that may soon be on-the-shelf.
For example, do you really want to bring those great big propellant tanks back with you?
Yup. That way you don't have to ship up replacements, and replumb them in orbit. Even if I were to design a new LTV from scratch I'd still use unitary tankage. Keeps things much simpler, eliminates failure modes, makes maintenance and operations easier, reduces the amount of zero-G rocket plumbing to a reasonable low....
I'd rather just figure out how to aerobrake.
~Jon
Randy,
It's NOT a snap, and we NEED to test out the idea, but it's NOT that hard to do, nor more complex than propulsive braking. And, (agreeing with you) it will be the ONLY economical way to 'reuse' a lunar transport system for a long time.
Yeah, figuring out aerobraking is worth doing even if you find a good way of making a workable transportation system without it. It just makes everything more economical. If you had aerobraking figured out, as well as propellant depots in LEO and propellant depots in L-1 or lunar orbit (with the lunar depots supplied by ISRU stations on the moon), you could greatly increase the amount of cargo or people flown per flight, while greatly dropping the price per pound of cargo/passenger delivered.
Thanks for the postings
You're welcome. Thanks for the comments.
Ed,
One reason to doubt Mike Griffin's claims that he would reuse the Earth Departure Stage, if an orbital propellant depot suddenly appeared.
To be fair to Griffin, refeuling the EDS in orbit would allow you to ship a lot more cargo per lunar shot. But, I do agree that that doesn't do anything to help with the main cost-drivers of lunar missions--the massive standing armies for Orion, Ares I/V, the EDS, and LSAM.
The good news is that if RpK doesn't botch things again (they have blown through a lot of money in the past without flying anything) K-1 may well be flying long before NASA can even start spending money on Ares V, EDS, or LSAM. If commercial space can figure out on-orbit refueling quick enough, a round-the-moon K-1 OV flight might very well be doable before Orion actually goes into service, at which point NASA will be very hard pressed to explain why not to use the K-1 and its OV in lieu of EDS and Ares V.
~Jon
Jon, I hope you don't tell Dave that $50 per lb in LEO is ridiculously cheap. With some luck you guys should be able to hit that price point.
Anonymous,
$50/lb in LEO is ridiculously cheap at this point. I think that at some point in the future, it'll probably be doable, but that's a 3rd or 4th generation thing. $250-500/lb while still very aggressive is a lot more realistic for the medium term.
~Jon
Jon Goff wrote:
"...a round-the-moon K-1 OV flight might very well be doable before Orion actually goes into service..."
That would be an interesting space race, NASA vs. the private sector.
> For example, do you really want to bring those great big propellant tanks back with you?
> Yup. That way you don't have to ship up replacements, and replumb them in orbit.
Not having to ship replacements may not be a big savings. How many pounds of propellant do you need to expend to recover one pound of tankage to LEO? If it's more than one, you're losing on shipping costs. (Obviously, you need to consider tank manufacturing costs, too.)
Even if recovery were free, based on your figure for tank mass at 7% of the propellant mass, replacement tanks would only add 7% to your launch cost.
On the other hand, if you have a use for tanks at the other end, you can get a "bonus payload" equal to 7% of the propellant mass. That's significant. (The big question is how many tanks you can use at the other end.)
I also think you greatly overstate the plumbing problem. You're going to have to make connections between tanks to do refueling anyway. Adding a drop tank needn't be any more complicated than one refueling connection. In fact, if the drop tanks are filled on Earth, they might actually decrease the number of connections you have to make on orbit.
Ed,
Where'd you get the 7% number? Maybe it was a typo on my part, but the entire K-1 OV stage drymass is about 9% of the fully fueled mass. That includes engine, TPS, recovery systems, structures, payload fairing, and tanks, as well as plumbing, pumps, etc. State of the art aluminum tanks can get down to 0.7% of the mass of LOX they can carry. Getting 3% is pretty easy, with 1.6% doable with a plain-jane 5000 series aluminum and only a very little inexpensive tricksiness. Those last two numbers were with 50psi feed pressure and a 2.0 FOS, which is rather high for the aerospace world. In order to get tanks as heavy as 7%, you'd have to try (or you'd have to be flying DOT rated tanks or something).
But you do have a point--dropping tanks at lunar orbit will boost your payload efficiency a bit, and if you have customers for them, it might make sense...but I'm not really convinced. After 1-2 years, you'd have so much propellant tank capacity in lunar orbit, L-1, and the surface compared to demand that the price you could get for the things wouldn't be worth the extra cost and complexity. IMO at least.
~Jon
Oops. I was reading your post on inflatible tanks and somehow missed the decimal point.
I agree that the demand for tanks is the big question. (Right now, of course, the demand is zero, since the only agency spending significant money on the Moon isn't interested in anything innovative.)
One of the advantages of external tanks is that you can choose to either retain them or drop them, depending on market conditions and the requirements of the particular mission.
They would also provide greater flexibility in customizing the propellant load for various missions, by varying the size and number of tanks. That would allow the same tug to be used for pushing various size payloads to the Moon, servicing satellites in GEO or GPS orbits, or even NEO asteroid missions.
Excuse me for going way off topic -- did you used to post on Chris Whitten's "Free Market Network?" If so, I think I had some discussions & debates with you maybe 8 or 9 years ago there. For some reason I happened to think of this the other day, and suddenly you (if I'm correct) appear on Tom Palmer's blog.
Your post on Palmer was right on target, BTW.
Best,
Charles
Charles,
That would be me. I took off on my mission a little before FMN seemed to peter out. I mostly keep my politics to snarky comments on other peoples' blogs these days. Drop me a line on my gmail account though. jongoff "at" gmail "dot" com.
~Jon
Post a Comment
<< Home