One of the nice things about blogging is that, even for print journalists, it provides an outlet for information gathered that may be of interest, but for which there wasn’t room in the publication. Here’s a good example: interview notes from Rob Coppinger’s discussion with Phil Sumrall on Ares V performance issues.
As noted, the vehicle has come a long way from the originally advertised “Shuttle-derived” system that was supposed to save us so much money and time, and utilize the existing Shuttle infrastructure (though the latter was always a politically-induced pork-driven bug, not a feature, if one wanted to actually lower launch costs). It (like Ares I) is now essentially a new vehicle, including components, though if Ares I ever comes to fruition, Ares V will probably be at least in part derived from it.
Of course, this part is what really has me grinding my teeth (and it’s probably what I’ll be talking about on Jon Goff’s propellant depot panel at Space Access):
…once the EDS and Altair were in orbit there was a 95-day loiter in Earth orbit for the concept of operations. That was changed from 95-days when Griffin said it was not acceptable. Instead the new target date was four-days and this may also assume a launch of the Orion CEV prior to Ares V
Reasons for the four-day change are propellant boil off and electrical power requirements. For four-days fuel cells are sufficient and solar arrays not needed. Less than four-days and batteries could be used for EDS power. During Apollo they had 15% boil off over 3h so over several days Ares V would lose a lot of propellant. To stop boil off the choice is a passive system and “we have to eliminate heat leaks”. The solution to boil off is seen as multi-layered insulation as they want to reduce the boil off losses to 1-2%, but MLI is very expensive in terms of money, not payload margin.
So, they’re going to launch the Orion, with crew, on an Ares I, and hope that they can get a successful Ares V mission off within four days, because they can’t afford the duration. They build this mondo grosso launch vehicle to avoid having to do multiple launches, and yet, they not only have dual launch, but it’s one on a tight window. And if they can’t get the launch off on time, the lunar mission is scrubbed, and the crew comes back home from LEO, having wasted the cost of an Ares I launch (and an Orion, if they end up not making it reusable).
This is an affordable, resilient, sustainable infrastructure?
All of these issues go away if you use orbital infrastructure. The propellants are brought up over a period of time, with a number of different vehicles, and vehicle types. The propellants are stored on orbit with a combination of passive and active thermal control systems, eliminating boil off completely. If MLI is expensive, that’s OK, if you only have to manufacture/lift it once and then continually reuse it at the depot. If you have power at the depot, you don’t have to worry about battery life at the vehicle (note: the next Shuttle mission will set a record for duration, because it doesn’t have to rely on its fuel cells for power–it will draw power from the new solar arrays at the ISS while docked, allowing it to stay up for two weeks). And the same system will scale to a Mars mission (perhaps based in L1 instead), obviating the need to develop Ares XI.
Put the power/propellant/other-utilities infrastructure up once, and continually reuse it, instead of making each vehicle have to be a self-contained Winnebago, like the Shuttle. Even if the moon remains a wilderness, there is no longer any excuse for LEO to be so.
Why not just start at EML1 from the beginning and get it over with? That gives you a useful infrastructure that can be used to get to anywhere else.
I’d love to see somebody cobble together some numbers using EELV/F9, Dragon for transport, and, say, a Centaur, a 330 (hab for visiting crews, just add more for permanent staffing), and a tanker solution (Murphy Bags, modular tanks, something derived from Dragon or Centaur, whatever) to build a EML1 depot. I suspect it would cost less than building the stick.
All of these issues go away if you use orbital infrastructure.
No, not really. The main problem about any LEO-based rendezvous architecture, be it Ares, ISS, or propellant depot, is proper targeting of the translunar injection.
For a given inclination, there is a particular right ascension of the ascending node for the desired translunar injection. Roughly speaking, the launch site will rotate under that node about once per day (assuming only one of the two launch azimuths is available, like in the case of a launch from KSC).
If you were going to execute a TLI from the ISS, assuming it was all decked out with cryo propellant storage, you would still have to wait for the time when nodal regression (about 7 degrees per day) brought the ISS into the correct orientation for TLI. That can be a long wait.
The situation is somewhat better for Ares, but still not very attractive. When either the CEV or the LSAM stack is launched into LEO, a particular nodal orientation has been established. Odds are that you picked a node that will drift into the proper orientation in some future time–let’s say four days. That means you have four days to get the other element launched, rendezvous, docked, and checked out before TLI, otherwise you will miss the injection.
Let’s say four days isn’t enough time, so like Michael J. Fox in Back to the Future, you will just go back earlier, or in this case, set up your initial node with more “drift time” to the proper orientation. You can dial in whatever drift time you want (at the expense of boiloff) but when your node is right, it’s time to go. Otherwise you’re looking at weeks before the nodal orientation is right again and you can inject.
This problem is inherent to ANY LEO-based rendezvous architecture. There is a good paper by Gordon Woodcock on the subject written back in the 1980s called “Missions and Operations Modes for Lunar Basing” that discusses these constraints in much more detail.
The solution to these problems, in my opinion, isn’t a propellant depot. It’s to move the first rendezvous location from LEO to the lunar vicinity, preferably the L-points. This can be done by launching the CEV and LSAM on separate rockets and doing independent translunar injections to the first rendezvous point. That way, each vehicle can independently launch at the optimum time (to target the node needed for TLI) and execute a TLI shortly after attaining orbit. Apollo didn’t have this problem because the whole stack was launched together, and the launch time could be chosen to target the optimum node for TLI, which was executed shortly after launch by the S-IVB.
No, not really.
Yes, really. It may introduce other issues in terms of scheduling as you note, but the boiloff/power issues go away.
And the constraints aren’t that bad, as long as you’re not trying to go directly to the moon from LEO. I certainly wasn’t proposing doing it from ISS–that would be insane, from a cost standpoint.
Ideally, the depot would be in an equatorial orbit, so that the delta vee to L1 (not the moon) would be essentially constant, with an opportunity to go every two weeks, and from L1, you can go anywhere on the moon you want, any time. L1 would be the place for the second depot. It’s less propellant efficient, but if we can get the cost of propellant delivery down (by doing earth launch smarter, and perhaps getting it from extraterrestrial sources) it’s a much more cost effective and resilient way to go.
Of course, that implies that vehicles won’t operate out of Florida (unless they can self ferry down to the equator prior to launch), so it’s probably a political non-starter. From 28.5 degrees, opportunities will be more restricted, at least with reasonable propellant requirements (though plane change doesn’t really cost that much out at L1).
What’s the disadvantage of skipping LEO, anyways? Just the payload per launch for a given size of rocket?
What’s the disadvantage of skipping LEO, anyways? Just the payload per launch for a given size of rocket?
Yes, but that’s a big “just.”
Big enough that a LEO depot is pretty much a requirement for launching with EELV-sized boosters?
That’s why I’d really like to see somebody (open invitation, y’all 🙂 crunch some numbers and figure out what’s feasible/practical for us non-engineer/rocket scientist types…
What’s the disadvantage of skipping LEO, anyways? Just the payload per launch for a given size of rocket?
I have a hard time seeing the “advantage” of LEO operations. Lots of orbital constraints and a bad debris environment getting worse. Look, if you plan to go to the Moon, GTO, or any high-energy orbit then you’re going to be carrying the propellant to orbit to do that maneuver. Stopping in LEO doesn’t much change that basic issue.
If I want to send a 10 tonne CEV to TLI and a 20 tonne LSAM to TLI, I can either launch 10 tonnes of CEV and 10 tonnes of LH2/LOX and 20 tonnes of LSAM and 20 tonnes of LH2/LOX, or I can launch 30 tonnes of CEV/LSAM and then 30 tonnes of LH2/LOX, or I can launch 60 tonnes of CEV/LSAM/LH2/LOX altogether. Or I can launch six 5-tonne LH2/LOX carrying missions to an orbiting propellant depot. But nothing materially changes. In all cases I’m launching roughly 1 kg of propellant for each kg of TLI payload.
To “refuel” an orbiting propellant depot, I have to launch a tank of propellants. Stick an engine on it and you’ve got a stage. So what does the propellant depot buy you except orbital constraints.
If you’re not going to LEO, launch to LEO and get out of it as quick as you can. No sense hanging around.
To “refuel” an orbiting propellant depot, I have to launch a tank of propellants. Stick an engine on it and you’ve got a stage. So what does the propellant depot buy you except orbital constraints.
The propellant depot buys you scalability and economy of scale.
After launching a bunch of small propellant payloads to the depot, a large booster can then fill up and go a lot further than many other possibilities.
A well-insulated and cooled depot also allows your launched stages to be less insulated. Assuming a quick rendezvous, the boiloff trades become much more sane.
Also, through a mix of active/passive cooling, you can produce water for storage on orbit, another useful but heavy component.
>>>>If I want to send a 10 tonne CEV to TLI and a 20 tonne LSAM to TLI, I can either launch 10 tonnes of CEV and 10 tonnes of LH2/LOX and 20 tonnes of LSAM and 20 tonnes of LH2/LOX, or I can launch 30 tonnes of CEV/LSAM and then 30 tonnes of LH2/LOX, or I can launch 60 tonnes of CEV/LSAM/LH2/LOX altogether. Or I can launch six 5-tonne LH2/LOX carrying missions to an orbiting propellant depot. But nothing materially changes. In all cases I’m launching roughly 1 kg of propellant for each kg of TLI payload.
If you look at the Direct concept, that is what they are proposing–sticking a fuel depot at one of the L points. You can get the europeans to pay to stock the fuel tank in exchange for a ride on the rocket. I also got last years article from Boeing on a space fuel tank. If we really want to get to the moon and be serious we really need a fuel stop. No-one takes all their fuel when travelling across country, it is ALOT more efficant to have a fuel stop on the way. That is the business of govt. to build that infracture. Once there, NASA and commerical players will say why did we not build it 30 years ago! Just ask yourself why do most satilaties come down that are not in GEO. Sometimes because of age, but alot of the time since they don’t have enough fuel to stay up so they do a controlled burn into an ocean.
Vanilla,
The whole point of doing operations in LEO is that you no longer need a heavy lift vehicle to go anywhere outside of LEO. A system that uses propellant depots at LEO as well as a smaller depot in the vicinity of your target location (L1/L2/LUNO for the Moon, or Martian Orbit for Mars) really can scale up substantially both in size and in scope a lot better than an architecture that depends on Heavy Lift vehicles.
Most importantly, if you ever want to have a truly reusable transportation architecture (which is a must if the prices are ever to become low enough that humanity can ever become a spacefaring society), you emphatically have to have LEO infrastructure.
If all you care about are flags and footprints stunts, sure, L1/L2-only rendezvous has some performance benefits. But if you actually want to do anything real in space, you need infrastructure.
~Jon
~Jon
Jon, I never disputed infrastructure. I just don’t think LEO is the place for it. Launch all your same pieces directly to the first rendezvous point at the L-points. Skip LEO. It doesn’t change the architecture measurably. You will still have the same mass-on-target in each case.
Vanilla,
Actually it does affect things. Things don’t scale down very well past a certain point. While an RLV with a 5000lb payload to LEO can be quite useful with an LEO propellant depot, in order to get anything to L2 you pretty much need an additional transfer stage now. But without LEO rendezvous that transfer stage is now either expendable, or you have to have it reenter and recover it on the ground. Packing all the necessary subsystems into a 5000lb package, while still having any sort of reasonable payload isn’t realistic.
And for sending people, you’re pretty much limited to launchers in the Atlas V 55x and Delta-IVH class if they don’t stop in LEO for refueling. While I’ll admit that is better than I thought the numbers would come out to (I thought you were going to need an HLV to pull it off), it still doesn’t look very good as far as sustainability goes. You still can’t take advantage of lower cost RLVs for the personel or large equipment transfers, because you’re launching directly to L2 without any EOR. 25ton to LEO RLVs are not going to happen any time soon.
So, yeah, I’d rather take the astrodynamics restrictions and open up a much larger and more capable transportation market than to have the first rendezvous point be at L1/L2, and pretty much restrict manned exo-LEO flight only to heavy EELV class launches.
~Jon
Jon, so you don’t want an additional transfer stage for the RLV? Well, what do you think is departing the propellant depot for L1/L2/LLO? What did you “fill up” at the propellant depot? A (probably expendable) transfer stage. You don’t get around it. If you want to make the transfer stage reusable you’ll be using a lot more of the precious propellant you bring up to the depot.
For a propellant depot to make economic sense, you would have to have some launcher, somehow capable of making it to LEO with a big tank of cryogen but somehow unable to put an engine on that big tank of cryogen to send it beyond LEO, at a much lower price point than another rocket that could. Don’t see that anywhere. Any cheap rocket that can loft a big tank of cryogen to LEO can use half that cryogen (for a ~3100 m/s TLI) to send the rest of it to another rendezvous point.
Vanilla,
A couple of points:
1-A tanker delivery to an LEO depot, and a fully capable transfer stage are two totally different beasts. “Just” adding an engine isn’t enough. You’re talking about a whole additional spacecraft. And many of the subsystems on such a spacecraft do not scale down very well. Which means that for RLV launches, you end up getting a lot less actual mass delivered to L1/L2 compared to stopping at LEO to aggregate payloads somewhat.
2-The transfer stages don’t need to be expendable. For early flights where most flights are dropping off cargo or propellants, your “return” payload to LEO is very light compared to your outbound payload, so propulsive braking doesn’t cost you very much propellant (do the math). For manned flights, this might require for a while having the transfer stage deliver the people to L1/L2/LUNO, and on the homeward bound insert the personnel carrier into a TEI trajectory, separate, and only the transfer stage does propulsive braking (with the manned capsule doing an Apollo style reentry). But eventually if aerobraking can be made reliable enough, you can fly the whole thing back without paying a big penalty.
3-You still haven’t addressed my issue that you have to have a Delta IVH class vehicle to send people to an L-point station, thus entirely eliminating near-to medium term RLVs from the mixed for manned flights.
4-I still see only one advantage to having all of your flights go direct to an L-point other than that you get more flight opportunities per month at the minimal propulsive cost. And all of that at the cost of shutting yourself out from the benefit of RLVs for manned flight and requiring lots of additional expendable transfer stages for propellant delivery flights (which also wipes out a lot of the benefit for using RLVs for the propellants.
Basically I still think that such an approach might be better than the Constellation approach–but only if your goals are very limited and you think RLVs will never happen (or that if they do that they’ll be monster Shuttle-class RLVs).
~Jon
Jon,
I think you need to hit a bit harder on the point that the propellants can be prepositioned by someone else. Vanilla can refill the upper stage he used to reach LEO at the depot, and use it to reach his destination with the same upper stage engine and tanks he already paid for. Prop customers do not have to worry about several launches to place fuel at the same time as the hardware launch. They only have to worry about getting their main hardware into the minimal LEO orbit of the depot.
It is not clear enough to most that the depot operator is there to facilitate what they do, not tell them how to run their mission. If there is excessive boil off, or it cost more to deliver prop, or depot operations are too difficult, it is not a problem they have to address. It is the depot operators problem. All they have to worry about is the price at the pump. They can do their mission in one launch, whatever it is. And if they want more mission margin, they don’t have to redesign anything, just write a bigger check at the depot.
And while they are there, a preliminary checkout of their vehicle systems can take place at a location that has frequent deliveries. The replacement part can be delived by the next tanker for marginal mass cost plus markup, not by a dedicated launch by someone that knows he has you in a bind.
I know I am repeating things you, Rand, and others have been writing for years, it is just frustrating that the message does not seem to be clear enough to all interested parties.
Rand, John, vanilla, etc-et-all;
I’m late on commenting on this post so I’m guessing I won’t probably get a response, but here goes anyway:
“Sustainability” and the idea of LEO/L-X infrastructure as well as “expanded” manned presence in space seems to be the theme of most of the discussion on how the VSE/ESAS should be run by NASA, but I’m a bit confussed as to how this ‘applies’ the argument since it wasn’t what Congress has been telling NASA to do and much as I loved the concept of the VSE Congress has never been really ‘open’ to advancing manned space flight and even less so allowing NASA to do it.
I’ve often heard that NASA has “failed” the VSE and “defied” the Presidents mandates etc. My difficulty with this though is while the President can (and did) present suggestions for goals and programs for NASA he has no power what-so-ever to actually assign NASA tasks, approve or disapprove financial or materials resources or programs all that being Congress’ duty being overall in charge of line-by-line authorization and mission direction of NASA.
Since it is Congress that has specifically forbidden any development of technology or programs that can “directly” relate to sending people to Mars it would stand to reason that they would also oppose NASA participation in any activity that would expand manned space flight into LEO or beyond wouldn’t it?
If someone can give me a more indepth explination or set me straight I’d appreciate it, but right at this moment I have to agree with:
“If ‘pro’ is the opposite of ‘con’ then is Congress the opposite of Progress?”
Randy