Random Thoughts: Lunar One-Way-to-Stay (For a While)

If you’ve read enough space blogs, forums, and usenet groups, you’ve probably heard of the one-way-to-stay Mars mission, but what most people don’t realize is that in 1961 and ’62 such an approach was also seriously suggested and investigatedfor early lunar missions as well. While this may sound like a historical curiosity, I think the idea may have more merit today than it did when it was originally proposed.

One of the big challenges in lunar exploration, settlement, and economic development is our profound ignorance about our closest celestial neighbor. We have no idea if economically interesting concentrations of important metals or elements exist, or if so, where. We have ideas for dust mitigation and ISRU techniques, but really have no idea for sure if they will work.  More importantly, we have no idea  how many iterations it’s going to take to make it work.  We have no idea if lunar gravity is high enough to avoid the effects of microgravity on the human body.   There have been lots of lunar manufacturing ideas floated, from fusing regolith with microwaves based on nanophase iron embedded in the regolith to bulldozers, pneumatic excavators etc.  The problem is that it’s impossible to simulate enough of the lunar environment here on earth to really know that there aren’t subtle but important effects we’re ignoring.  Really for most of this stuff, the only way to find out is to go there and try it.  And fail.  And figure out why you failed, and try again.  The whole idea that we can make complex aerospace hardware work the first time in an environment we do not fully understand and can’t adequately replicate here on earth is hubris.  We might get lucky, but more likely than not, it’s going to take work to sort a lot of this out.

There is a reason why the Technology Readiness Level scale exists.  It doesn’t matter how pretty your calculations are, because in reality you’ve made at least one or two incorrect assumptions.  It doesn’t matter how good your bench test went, because when you test it in the right environment in a flight-weight integrated system there are environmental effects and subsystem interactions you couldn’t adequately test on the ground.  That doesn’t mean you can’t eventually make the system work–it’s just that the only way to know a technology is ready for primetime is if it has been adequately tested in the actual environment.  It’s only then that you’ll have figured out all the weird and quirky second-order effects that can’t be ignored.  It’s only then that you can know that a system is actually dependable.  That may sound harsh and unfair, but it’s probably sugar-coating the case.

What would really be useful is to find a way to break out of this knowledge-poor environment we’re in, and lower the cost drastically of getting lunar technology through the try-fail-fix cycle.  The earlier in the design process you can get correct information, the better the decisions you can make.  One-way lunar missions are an interesting way of achieving that goal of early information.

When you think about it, you actually only need a single piece of new transportation hardware for a one-way lunar mission–the lander itself.  You don’t need a crew capsule return capsule capable of reentry from lunar trajectories or capable of autonomous operations in lunar orbit for months at a time.  You don’t need an ascent stage that has to be hyper-weight-optimized because it drives the mass of the whole system.  You don’t even need an HLV or an Earth Departure Stage–a stock Atlas V 551 or Delta-IVH would be more than able to softland two people on the moon in a single launch.  You don’t even need depots, long-duration cryo storage (though it would help), tugs, RLVs or anything else.  In fact, you probably don’t even need a “man-rated” booster, a crew capsule, or a launch escape system unless they’re already available (the odds of dieing on a mission like this are substantially higher than the odds of an EELV failing during the 80% or so of the mission that the lander couldn’t do an abort-to-orbit).

You’d probably want to prelaunch some supplies and extra living quarters, and be ready to launch regular supply missions.  Call it three flights to get setup, and the fourth carries the crew.  You might be talking about $1-2B total to develop the lander and get the crew and their supplies there, and another $500m-2B per year to support them (assuming current EELV prices–it’d be cheaper if Falcon 9 pans out, or if you can use foreign boosters).   I’m probably being overoptimistic here, but we’re in all likelihood talking a very tiny fraction of the cost of any of the typical lunar architectures, and you could have the project going within only a few years.  More importantly, the project is cheap enough that you could start getting your initial information from the lunar surface while simultaneously preparing the systems needed for a more robust round-trip transportation system (depots, transfer vehicles, crew return vehicles, reusable landers, etc).  In comparison, the Constellation PoR would cost about $200B to get to the first boots on the ground, and cost about another $1-2B per set of boots thereafter.  For the first few years, you’d probably be talking about mostly short sorties, so the $/boot-day factor would probably be ridiculously high for a long time.

The risk is a lot higher of course, since you can’t easily get back to earth in a hurry, but the situation isn’t anywhere near as bad as a one-way Mars mission.  First off, even without any of the other systems developed beyond the lander, you can actually get the crew back from the Moon.  It would be a complex, inefficient, sub-optimal, and more risky approach, but you could do it.  You could just send enough cargo landers with propellant to refuel a lander to take them back to lunar orbit, and from there do a TEI burn.  Instead of direct reentry, you could do an aerobraking maneuver designed to minimize the number of passes through the van Allen belts.  You would need access to some sort of crew return vehicle like Soyuz to get you down from orbit, but you could do all the hard work of coming back.  So the crew isn’t really stranded with no hope of return until a bunch of other stuff gets developed.  Also, if something critical breaks down, it’s realistic that you could launch a resupply mission fast enough that the crew could just sit-tight in their spacesuits for a few days.

Is something like this still beyond the pale politically?  I’m not sure.  You’d definitely be accepting risks much higher than CxP (probably in the 25% range), but many of the worst risks from the Mars one-way option would be gone, and a lot of them are front-loaded.  If you can make it through the first month or so, it’s likely you can keep them resupplied with stuff frequently enough to prevent catastrophe.  It would definitely be a more nitty-gritty and therefore interesting mission.

Anyhow, I know I’m leaving out a lot of details and thoughts, but I wanted to throw the idea out there for discussion.  I’m not suggesting that this should be done in lieu of getting a real lunar transportation system done right (ie one with RLVs, depots in LEO and L1 or L2, private crew transportation to orbit, reusable transfer stages and landers, etc)  I’m just suggesting that this is a way to get information quickly, and quickly figure out what ISRU options are really viable, and therefore how difficult it would really be to start doing medium-scale settlement of the Moon.

The following two tabs change content below.
Jonathan Goff

Jonathan Goff

President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
Jonathan Goff

About Jonathan Goff

Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
This entry was posted in Lunar Exploration and Development. Bookmark the permalink.

48 Responses to Random Thoughts: Lunar One-Way-to-Stay (For a While)

  1. Ian Sales says:

    On the subject of historical one-way lunar missions, you might the following of interest:

    A review of The Pilgrim Project, Hank Searls – a 1960s potboiler based on the concept – see here.

    A piece of flash fiction by yours truly, also based on the concept – see here.

  2. We need to send people to the Moon to stay for at least a year anyway in order to find out if a 1/6 gravitational environment is deleterious to human health. We also need to see how well agricultural animals and crops grow and reproduce at a lunar habitat. We also need to know how efficiently oxygen can be produced from the lunar regolith.

  3. Martijn Meijering says:

    Or you could avoid the extra risk entirely and use noncryogenic lander propellants, which need not be toxic. This would accelerate the benefit to commercial development of space by a number of years and remove some risk. By my count that gives us two advantages to what you propose above, which I’ll agree is still a good idea.

    You’ve been less than enthusiastic about this idea before, but I still don’t think I understand why. It seems as if you would rather have extra risk and a longer delay than see noncryogenic propellant transfer deployed for exploration missions before cryogenic propellant transfer and full depots. You don’t have this point of view when it comes to development of RLVs. The difference in approach puzzles me. It can’t be because you’re a closet SDLV proponent 🙂 , but I’d love to know the reason.

  4. Um Martijn,
    I never said anything about cryogenic or non-cryogenic landers. How does using a non-cryogenic lander reduce any of the risks inherent in this mission at all (other than at the margins)? For an affordable two-way lunar transportation system, you *have* to have the ability to handle cryogens in orbit, even if the lander is storable. I know this is your pet hobby horse, but please try responding to what I actually wrote.


  5. Doug Graham says:

    I like the idea. Heck, that’s kind of how North America was settled. THere was the odd hiccup (Roanoke Island) but in the end, the (surviving) settlers of Massachusetts and Virginia adjusted and things started moving along.
    I don’t think there would be a shortage of volunteers for the mission. Heck, I would go. But the current Euro-American fetish with avoiding risk of any kind, leads me to believe that the political elites would be unwilling to face a possible drawn-out soap opera involving astronauts stranded and left to die.

  6. Martijn Meijering says:

    Jon, I didn’t mean to annoy you and I’m sorry I did.

    I meant to 1) agree early precursors are valuable and 2) to state that to the degree compromises are necessary to the capabilities of the manned vehicles by the absence of cryogenic propellant transfer, that can be remedied by storable propellant transfer. Another advantage would be the earlier benefits to commercial development of space due to earlier propellant launches.

    You described a way of avoiding the need for more than existing EELVs, the need for a a highly mass-optimised lander and even the need for a very capable capsule. I agree those goals are important and deserve more attention. What I wanted to point out is that the minimum required throw weight of an upper stage used as EDS can be reduced by using storable propellants for the lander even before you had the capability of refueling cryogenic upper stages. Perhaps this is what you meant by the phrase other than at the margins.

    So, to answer your question: to the degree mass is the sticking point, you could still have the capable capsule without needing heavy lift or cryogenic propellant transfer. If it was the cost of the capsule that was the concern, then storable propellants will do nothing to address that. In any event, your proposal remains a good one.

    It’s not surprising that storable propellants will look like a hobby horse to you and others, but it’s really just the combination of wanting to follow the principle of incrementalism and the urgency of commercial propellant launches that drive me to it. I would expect you to agree with those considerations, in fact you seemed to be doing just that in your post. So I don’t understand, and I’d like to. Of course you don’t owe me any explanations, but I believe we have the same goals. I’ve learned a lot from reading your blog and intelligent replies by fellow readers. I’ll be happy if the pollination works only one way, because I feel enriched by it. It would be even nicer if it could be cross-pollination and work both ways. That presupposes I have something of value to add and I’ll agree that’s not a safe assumption to make.

    Anyway, cheers and may you long continue to write interesting and valuable blog posts!


  7. Will McLean says:

    The biggest problem I see is that you don’t eliminate the cost of round trip capability, you just defer it.

  8. Neil H. says:

    I wasn’t sure if you implied this in your original post, but since you’d be launching living space separately anyways, one could probably use something like the open-cockpit lunar lander design from the barebones 2001 “Human Lunar Return” study to further reduce needed equipment:


    Also, it seems like a total development/operations cost in the few billion dollar range also potentially puts it in the range of the random retired billionaire who wants to fund a project to make themselves the first “lunar settler.” I wonder what the legal issues around such an endeavor would be…

  9. Thomas Matula says:


    Actually you wouldn’t really need to get back to Earth orbit to get a ride on a Soyuz. It was originally intend to be their CSM for their lunar architecture. I am sure they would be more then happy to dock with the lander in lunar orbit for a price if given some lead time.

    And it would be a small step from that to simply contracting for the Soyuz to be your return vehicle. Then the lander could simply be become a lunar shuttle once it reaches the Moon, making quick trips to lunar orbit to first dock with a Soyuz then a refueling depot placed in lunar orbit to allow it to refuel for lunar landing.


  10. Robert Horning says:

    Contrary to popular belief, America isn’t a risk averse society. There are a great many people willing to take risks and go into the unknown and find out what is there.

    The U.S. Congress, on the other hand, doesn’t want to be seen as the entity responsible for sending somebody to their death (usually). This unfortunately includes very risky occupations like a soldier or marine that has to be standing in harm’s way by design in front of somebody else’s gun that wants to do real damage. This is no different, therefore, with astronauts and that a congressman would be willing to spend that extra billion dollars (that isn’t really theirs in the first place, so it is easy to spend it) on safety and protection on the off chance that astronaut might die because of a shortage of funds. That is a billion dollars extra per astronaut that I’m talking about… which is one of the things that drives up costs of manned spaceflight.

    The problem with this approach of going with the safest possible approach for going to the Moon is that the cost of doing so is prohibitive in nature and can’t be justified except for some monumental national accomplishment (such as Neil Armstrong’s original trip up there).

    No, I don’t have any special solution to this, other than to note that I think it will take private individuals willing to take the risk of going to the Moon on their own dime, and hopefully taking advantage of substantial reductions in launch costs from start-up “new space” companies that can get them at least to LEO in the first place. As is often said, LEO is halfway to the rest of the Solar System. If you can put a fairly large group of private citizens into low Earth orbit, I don’t see how anything short of Navy or Air Force ships with guns explicitly shooting down anything leaving LEO is going to stop people from going to the Moon or Mars. Somebody is going to try it out someday, and the budget for doing that is going to be considerably less than the proposed Constellation architecture.

    For space advocates, the question I would raise is how would you get from where we are at now to a sky filled with private citizen-astronauts that are willing to tinker with vehicles that can get to the Moon or elsewhere in the Solar System? You know it will be close when regular private circum-lunar trips are being made. That day is much closer than most of us would think.

    The first settlements on the Moon may very well be something like is being discussed with this thread, but I don’t see that happening with a government entity picking up the tab.

  11. David Gadbois says:

    Jonathan, I think you are on the right track if the underlying quesiton is “how can we get boots onto the moon for, say, under $10 billion”. It is a good question because you only need, say, 2 or 3 eccentric billionaire-adventurers to put this kind of program together. I assume you are not thinking that this kind of thing would be NASA-run.

    Your one-way ticket to the moon idea taps into two principles that could keep the budget in this realm. First is minimizing the IMLEO mass. Second, and more importantly, is to avoid and minimize development costs. That means

    1. develop 1 spaceship, not 2 spaceships. Most folks overlook this, but developing a CM and a lander means you double (or more) the spacecraft development costs.

    2. an obvious one – don’t develop a new LV.

    Your plan follows these rules, but I’d add:
    3. don’t develop a launch escape system. It would probably be cheaper to just pay the Russians for a ride to LEO in their Soyuz (or use Dragon if it becomes available), and dock with your lunar spacecraft. Transfer astronauts over, undock, and make your TLI burn.

    4. Also, don’t develop new rocket engines.

    The main drawback I can see with your scheme is that you’d have to have a pretty substantial lunar habitat that you would have to develop – and it would weigh a lot. “Substantial” in that it would have to be designed to operate through lunar night. You could use a nuke, flywheel batteries, or fuel cells for power during lunar nighttime, but all of those (except perhaps the fuel cells) would require development programs, and they all represent substantial mass to transport to the lunar surface.

    I think a better approach that would follow the above rules would be a direct-ascent (or perhaps a 2-launch EOR) in the vein of the old lunar Gemini concepts. Keep the crew cabin super-spartan, no larger than a Gemini or Soyuz reentry module, with only 7 days of consummables. Land near a modest lunar habitat that you previous launched and landed on the moon, stay for a lunar day (14 days), get back in your spaceship and depart for earth. So you are only developing ONE spaceship, not to mention cutting out the risk and complexity of lunar orbit rendezvous. As far as IMLEO mass, you take a hit for dragging your heatshield and TEI propellant all the way to the lunar surface, but this can be mitigated if you are using LH2/LOX – which you can do if your stay on the surface is limited to the 14 days.

    Incidentally, this is why I posted that question on the NSF forum regarding Centaur that you responded to. One could consider using a Centaur for the TLI push in a 2-launch EOR scenario.

    (longtime lurker of your blog, aerospace engineer by day for an unmanned aircraft outfit near Victorville)

  12. Bob Steinke says:

    I’m going to take a stab at sorting out the discussion between Martijn Meijering and Jon.

    What the original post proposes is basically to do Apollo with a long duration stay and no earth return. This allows you to avoid building three things:

    1) ascent stage that must survive a long stay on the lunar surface
    2) earth return stage that must survive a long stay in lunar orbit
    3) return capsule capable of lunar re-entry

    I think Martijn’s point in comment 3 is that if you have a refuelable lander then your lander can be your ascent stage and you don’t have to give up earth return to avoid designing a separate ascent stage. And a refuelable, long-stay lander/ascender will be easier with storable propellants. Martijn, please correct me if I misunderstood.

    I think some confusion occurred at this point because to avoid the risk of having no earth return you really need to have solutions for all three missing parts, and using a refuelable lander as the ascent stage is only one. Martijn may have been thinking that the earth return stage would be provided by the earth departure stage refueled with storable propellants, but that wasn’t said. Something would still need to be done about the re-entry capsule as Martijn acknowledges in comment 6, “If it was the cost of the capsule that was the concern, then storable propellants will do nothing to address that.”

  13. Bob Steinke says:

    Space Adventures talks about an around-the-moon Soyuz flight for $100 million per seat. I expect SpaceX also has similar capabilities planned for Dragon. If you had a couple billion burning a hole in your pocket it wouldn’t be unreasonable to design a refuelable lander/ascender and a habitat, and then go to the moon for an indefinite stay expecting that return from lunar orbit to earth could be arranged when it is needed. Multiple redundant habitats would provide lifeboat capabilities while you are on the lunar surface.

  14. A_M_Swallow says:

    In-Situ Resource Utilization (ISRU) methods can be tested using small robotic landers. A kilogram of say aluminium will be useless to a manned mission but would be sufficient to prove the techniques. LOX and lunar fuel can be tested by firing a rocket 100 yards upwards. By make the small rocket land and refuel three times permits testing of reusability.

  15. Pete says:

    A single iteration is a stunt not a development program. Probably the first lunar settlement needs to happen in the New Space mold. I am not even sure nowadays whether a single stunt is cheaper than a serious sustainable iterative development program. Though if a stunt, and the publicity that comes with it is what one is after…

    Probably one should develop a small, cheap, modular lunar station that could be reasonably landed on the moon and first test a version in LEO.

    Probably one should assume the development of lunar LOX from roasting regolith. This enables a couple of small reusable landers (say 25:1 LOX/LH2).

    I like the idea of a lunar station that could be iterated onsite, continually send up new an improved versions of different components as required.

    I am reminded that SpaceX ended up building a lot of stuff themselves because available sub systems were unsustainably expensive.

  16. A_M_Swallow says:

    How much mass can each of the new landers leave on the Moon:
    a. as an expendable lander?
    b. as a reusable lander?

    With Falcons, Atlas V and Delta IV significant amounts of mass can be carried to low lunar orbit.

  17. Paul Spudis says:


    Your idea has much merit. Incremental build up of capabilities and emplacement of assets can be done under the “one-way ticket” paradigm. The difficulty is that no Congress, Administration or incarnation of NASA would ever try it with people.

    The solution is (as A_M suggests above) is to do most of it robotically. A cryogenic RL-10_based lander launched on an Delta-IV H can put 1.5 mT on the lunar surface; it can carry deoployable solar arrays and use SEP to spiral out to the Moon, then propulsive landing to set down on the surface. Each lander then becomes a 25 kW part of a surface SV power system; placed at a strategic continuously illuminated location near the poles, it is available for power generation at a future outpost. The lander delivers a variety of payloads, including resource rovers, prospectors, processors, construction and propellant production equipment. All payloads can be operated telerobotically from the Earth or directly by crews on the Moon. These assets emplace and build a turn-key surface facility and become available for use by the crew when(ever) they arrive.

    Such an architecture was scoped out by an internal study team in early 2004, right after the Vision was announced. It got us on the Moon robotically while the Shuttle program wound down and then transitioned to the human program afterward. ISRU, self-sufficiency and propellant production are the principal components of the surface “mission.”

    No magic, no voodoo, no mega-budgets or mega-boosters. That path is still open.

    I won’t hold my breath.

  18. Seer says:

    Jon, how is the habitation module landed? Constellation has the Ares V to land cargo. Also, would the lander be able to break into a high earth orbit, which it would need to before aerobraking?

  19. Jonathan Goff Jonathan Goff says:

    You’d have a basic lander design that either carried a crew cabin/airlock, or a hab module/cargo load.

    As for braking into HEO from a lunar return trajectory, wouldn’t that be a really small maneuver? Actually, coming back from the moon, aren’t you typically already in an earth-orbiting trajectory?


  20. Jonathan Goff Jonathan Goff says:

    Paul, and AM,
    One of the main reasons I was suggesting this idea is that I don’t think ISRU is actually going to work the way people expect on the first try. Doing iterations when you have a person on the ground who can make modifications to the equipment is going to be a lot cheaper than having to do an SEP-based cryogenic robotic lander every single time you find something needs a change. Not to mention that for things like lunar rovers, having a habitat where they can periodically go to for maintenance, or overnight protection should allow you to make them last a lot longer, and get a lot more done.

    I just don’t have that much faith in robotics for doing debugging of complex systems in only partially understood environments. If I really thought we could do things so well that the first try would just be a “demonstration” as opposed to an honest-to-goodness “will this thing work at all?” experiment, I might be more willing to buy the “we’ll just do it on a robot” argument.

    That said, the political questions for a human mission are real, but I have the beginnings of an idea of how to deal with that.


  21. Jonathan Goff Jonathan Goff says:

    DG in comment #11,
    Good comments, I have a couple thoughts of my own:

    re: #1) You’re actually developing one spacecraft vs. 4 spacecraft (a descent stage an ascent stage, an EDS, and a reentry capsule).

    re: #3) I was actually suggesting just launching the crew inside the lander itself. That way it doesn’t need docking ports, etc. Sure, your odds are higher of losing a crew, but it may be an acceptable crew–depending on where lander and engine technology go by then, it may be possible to get feasible aborts over most of the risky parts of the flight without a capsule/LAS combo (the only exception is low-altitude flight near max-q). If the trades show that as being unacceptably risky, you could fall back on sending the crew up on a Soyuz or something.

    re: lunar nights and just doing a direct ascent return.

    Doing a direct ascent system would require orbital assembly or depots/propellant transfer. Feasible, but that’s extra tech dev and/or infrastructure and/or cost and complexity you have to deal with. A better solution would be to have the initial setup at one of the lunar poles. That way lunar night durations are fairly brief (or possibly even non-existent for a few peaks near the north pole), and you can get away with a less substantial setup.

    I wasn’t just trying to avoid development costs–I was also trying to keep a mission cheap enough that you could afford to do the exploration on a small budget. If you have to bring the crew back every two weeks, it’s going to lose most of its advantage vs a more conventional lunar program.


  22. publius says:

    I’ve been arguing for this for years!

    It’s a case of what Sir R.A. Watson-Watt called “third best today”. I have been trying hard to explain to people that a robotic-ISRU architecture has too many holes in it, too many places where one reverse could stall the whole project forever, whereas once you actually have boots on the ground with an intention to stay, it’s at least “game”, & probably “set & match” as well.
    A DELTA IV HEAVY can put 10 tonnes into an Earth-escape trajectory (about 10250 kg after you count out the mass of the payload adaptor), which with a direct-landing trajectory & RL10 engine translates to about 5500 kg on the surface. That’s a great deal of life-support supplies, or more than one human-tended pilot industrial plant, especially if you design the lander on the L59 principle (eg, cargo capsules retrofittable into living space, cryogen tanks demountable, structural aluminum either disassemblable for reuse or suitable as feedstock for one of the pilot plants).
    That mass is also much more than enough for a GEMINI capsule, & it so happens that the “Americans in Space – 50” people are working on building flight GEMINI-derivatives right now. It’s enough more than sufficient that you could provide an oversized adaptor-module section between the common lander & the capsule with living space in it, to serve as an interim habitat, with the GEMINI-B/MOL hatch in the heat shield for access (which allows using the capsule as an airlock). Alternatively, for sortie missions you could use the extra mass for additional propellant, & do lunar-surface rendezvous to bring on enough to make up the difference.

    The “enabling technology”, if you will, is not technical : we have solutions good enough to start with already (since, unlike zero-gee, we can use a lot of existing, well-understood techniques), & we will never develop optimal ones if we don’t get out there & get experience. It’s financial, or organizational. The money Bernie Madoff stole could have financed a whole One-Way Manned Moon Program! I have become convinced that the way to do it is on a grass-roots basis, direct action by the people who actually want to see it accomplished ; it’s certain neither politicians nor investment-bankers will do the job, so WE are the ones left with it.

  23. Sam Dinkin says:

    I’m in.

    Let’s try to get the cost down to $500M all in. At this level, it could be funded by one quirky billionaire, one Senator’s ear mark, a movie studio, a small nation, a state government and many other organizations.

    With a couple of years lead time, one might be able to get about 5-10 heavy launches for half that price.

    With aggressive commercialization, one could probably get mainstream corporations to donate research and equipment in exchange for having their product flown and logo prominently displayed.

    Bigelow might be convinced to lease most of a hab for cheap if it gets to retain ownership of it, get access to data and doesn’t have to pay the freight.

    Might an oxygen tank of a second stage make a good backup hab?

    Can 8 heavies deliver two times everything needed for a crew of two to live on the Moon for a year? Could a 5-year mission be accomplished with high probability using 10 heavies?

  24. publius says:

    A polar landing site is a poor choice for ISRU development, because the mineralogy is not very diverse. There are a lot of mid-latitude sites which have a good variety of major soil types (=feedstocks) close at hand. This is important since, with the One-Way architecture, your initial landing site is probably going to be the main place where you’re building up infrastructure for a long time.

    The advantages of a polar site are (a) the “permanent” sunlight (most of which isn’t, thanks to the swift precession of the lunar axis), & (b) the anytime-access capability to an Earth-return vehicle in polar orbit. (b) is irrelevant for the One-Way Mission, & (a) isn’t a necessity. A fuel cell running on surplus propellants, & after the first night products of wastewater electrolysis during the day, should provide ample keep-alive power (dumping boiloff gas through a fuel cell & dissapating the power with a load resistor, rather than blowing it off through safety valves, when pressures rise to dangerous levels for the tank, since that way you can keep the water).
    There is a loss of productivity due to down-time, but there may be useful activities which go better under Earthlight when the waste-heat loads are smaller ; if you can’t get a nuclear power unit, a solar-power satellite parked at L1 is a possibility. I need to get around to doing the numbers on that one, but one DELTA IV HEAVY payload into medium Earth orbit, boosting itself the rest of the way with ion engines powered by its solar cells (think the DEEP SPACE 1 probe) could probably provide at least as many kilowatts as the Russian RORSAT nukes, either by laser to a frequency-matched photovoltaic (heavy, high power density) or by microwave to a large, low-density mesh array.

    First Landing at a mid-latitude site, near 0 degrees longitude, just before local sunset may be desirable from a radiation-protection standpoint. You’re out of the Earth’s magnetotail, with the mass of a small planet between you & the Sun, for a good two weeks, which appears to provide the minimum likelihood of being cooked by a relativistic proton storm while you assemble & bury your semi-permanent habitat. The directional dispersion of solar protons is high enough that simply hiding behind a crater rim at a polar site is unlikely to do you much good.

  25. Will McLean says:


    You need to subtract the dry mass of your descent stage from payload on the surface.

    Jon at 21:

    Direct ascent could be done without orbital assembly if you land the ascent stage dry and land the propellant with another lander or landers.

    Also, I think you are far too optimistic about aerobraking. At the current state of the art it requires hundreds of drag passes. And technically you could start in an elliptical orbit coming back from the moon without propulsive braking, but it’s an immensely high elliptical orbit with an apogee going out to the moon’s orbit.

    Also, I think doing a two launch mission with lander on one EELV and upper stage on another other is much more practical, since limiting your hardware to pieces small enough to fit on a single lander is going to seriously complicate your surface infrastructure.

  26. Jonathan Goff Jonathan Goff says:

    True, but now you need to do 2.5x as many flights as for a one-way to stay mission. That may be fine if you decide down the road that this isn’t going to work, and want to bring the crew back to throw in the towel, but that’s a lot of hardware to waste on a single mission if you’re not staying for a long time.

    As for aerobraking, the loads they’re assuming are very low, for structures that aren’t actually designed with thermal protection in mind. Cutting that down from hundreds of passes over six months to a survivable number of van Allen belt crossings is probably doable with a little extra thought. You’re still talking about far less heat load than an orbital reentry, and working at much higher altitudes/lower densities. I think it can be done.

    Lastly, sure you could do it two launches, but at some point you’ve made stuff big enough and complicated enough that it’s no long a cheap and near-term option. Big cryogenic landers that can put 10tonne chunks on the lunar surface would make the lunar surface stuff easier…but it would also drive the development cost and time through the roof, because now your lander has to be able to do prox-ops, rendezvous and docking, it has to be able to handle boiloff during loiter, etc. It’s not impossible, but I think at that point you start losing a lot of the benefits of this approach compared to a more traditional approach.


  27. Jonathan Goff Jonathan Goff says:

    You could theoretically do a non-polar site for your base using regenerative fuel cells, but those fuel cells are now a highly critical piece of hardware, because if they fail, you have to do an emergency replacement delivery (on short notice), do an emergency evacuation, or you die. It’s a science project that has to work right off the bat. With a polar system, while there may not be any place permanently sunlit, the nighttime duration may be short enough that you can just go to battery power. We should have more info on that soon, AIUI, due to the Lunar Reconnaissance Orbiter.

    Also, I notice your comments imply that we would just be processing bulk regolith for materials. While it’s possible to do it that way, the key to making the Moon interesting is going to be finding the abnormal concentrations of minerals and elements (nickel-iron meteor cores or fragments, polar frozen volatiles concentrations, etc), and exploiting those. Bulk regolith operations are still interesting, but if we don’t find any “ore bodies” on the Moon, the odds of it becoming an economically interesting place are definitely lower.

    That said, I’m not opposed to doing a non-polar site, I’d just rather have the first such setup be at one of the poles. Once you have successfully done such a site for a while, the risk of doing another one-way mission (and hopefully the cost as well, since technology won’t be standing still during this time) will be substantially lower, and interest higher.

    That’s my take on it.


  28. publius says:

    @Will McLean
    Yes, of course, ~5500 kg is the gross landed mass — but quite aside from the possibility (I would almost say necessity) of a lander design which makes payload out of some of that deadweight, even after cutting out 1500-2000 kg of landing stage mass the available payload is enough to really DO something with.
    I have a copy of the Cord & Seale paper here, which assumes a SATURN IB type launch architecture, with a landed mass of 2190 lb EXCLUSIVE of the descent propulsion & landing system itself, on an escape payload of 6000-9000 lb ; of that 2190 lb, for the cargo version, 1280 lb is deadweight, leaving only 910 lb real payload. We can do far better now, but even at that (& with several other very cautious assumptions, based on the total lack of experience with lunar conditions or long-duration space missions at the time) they estimated a requirement for 16 launches per year for one man (plus an additional 6 in the first year), less than that per man-year for a multi-man mission.

  29. publius says:

    @Jonathan Goff
    Yes, of course, the fuel-cell/photovoltaic/storage combo becomes mission-critical (although it wouldn’t be impossible to provide emergency batteries for several Earth-days of absolute minimum requirements, because unless the habitat thermal design is very strange your metabolic heat will keep you from freezing), but that is a low-risk element.
    ISS currently uses wastewater electrolysis to reclaim breathing oxygen from metabolic water (I know they’ve been doing condensed atmospheric moisture for a long time, & I believe that’s also what they’re currently doing with urine), & throws the hydrogen overboard. You’ve got tanks suitable for H2 & O2 there on your stripped landing stages — no need to liquefy the gas, with all that volume just cooling & compressing to some margin below the burst pressure should suffice. If you’re willing to take a weight penalty on the fuel cells (which aren’t a big fraction of the system mass), you can get very high reliability, & there is no reason why the pioneers can’t, as a last resort, tinker on them the way they will on other systems.

    As far as resources are concerned, there’s a basic principle that you don’t go somewhere for what’s rare there, but for what’s common. Oxygen by itself is above 80% of life-support requirements, & there are a variety of techniques which can get it out of bulk regolith either anywhere or in many locations. Water ice & other volatiles are a nice bonus, but hydrogen, carbon, &c are the SMALL fraction of what you need, the thing you take care of AFTER you handle the big mass items.
    Again, aluminum, magnesium, titanium are everywhere you look. Nickel-iron meteorite fragments are marvellous, providing a variety of substances which will be very useful in small quantities, but there is meteoritic dust in the bulk soil, & certainly no reason to believe that the bigger fragments (which one might find with a metal-detector, or a variation of the Ninninger magnetic rake used to hunt meteor debris here on Earth) would be concentrated toward the poles.
    In other words, the processing techniques which are going to be important for life & industry in Luna are precisely those which handle the bulk material. You’re focusing on the hard-to-achieve last 15% of closing the loop, rather than the low-hanging fruit of the first 85%.

  30. Seer says:

    How much payload can a Atlas V heavy land on the moon? Isn’t it about a tonne or two? So how are you going to land a hab with that?
    From what I can tell of the 1960’s scheme, no one took it seriously.

  31. publius says:

    Addendum : I’ve spoken to engineers with some applicable experience who’ve studied the problem, & none of them thinks that extracting ices from polar cold traps is going to be at all an easy or quickly-solved problem. In fact I’d go so far as to say that’s the problem we’re farthest from a solution for at the present moment. On the other hand, there are (I am reliably assured) processes which could be implemented now for recovering oxygen, solar-wind volatiles, & some metals from bulk soil.
    A polar site would be an attractive location for a mining camp subsidiary to the main settlement after the initial startup phase, but at the beginning it adds complications (such as being out of touch with Earth for 2 weeks at a time, finding landing sites for several vehicles in very rugged terrain, & the aforementioned sparse mineralogy) which are probably out of proportion to its advantages.

    Really, the only really striking advantage is continuous sunlight, which does have its drawbacks (not only the radiation thing, but eg the fact that the light is coming in parallel to the horizon, & moving around the compass at about 20 degrees per day, requiring either heliostats or a lot of extra fixed collector area). The overnight power problem is going to have to be solved for any site beyond those few patches at the poles ; in fact the power system is mission-critical during daylight as well as darkness, & there’s no guarantee a breakdown couldn’t happen even in a pure photovoltaic system with constant illumination.
    You wouldn’t undertake a mission like this without a minimum of days or weeks (I’ve figured on months for some things) of backup supplies for all kinds of contingencies — water & oxygen for when the regenerative life-support system or the ISRU oxygen producer breaks down, carbon dioxide absorbers for when the separator breaks down, &c. Making sure some of that kit would operate during outages, & including emergency batteries for absolutely critical things, does not appear to impose much of an additional burden, at least of a kind which could be avoided by going to a polar site.

  32. publius says:

    According to the Payload Planner’s Guide, ATLAS V HLV with Wide-Body Centaur would have an escape payload about 16 tonnes, & gross landed mass would be about half that. Of course that means a couple of years’ lead time for WBC development.

    A major reason that the One-Way Manned Space Mission proposal of the 1960s wasn’t pursued was uncertainy of a kind which doesn’t now exist. When it was proposed, no American had spent as much as a day in space, & no spacecraft had made a soft landing on Luna. After 6 APOLLO landings, with the experience they accumulated on the surface, & the samples they returned ; after decades of operating manned spacecraft, some for years at a time, in the far more challenging environment of microgravity ; after individual humans have spent years in weightlessness ; those uncertainties have diminished enormously. At the time it was guesswork, but now we have knowledge.

  33. Jonathan Goff Jonathan Goff says:

    How big does the Hab have to be for spartan quarters for two people? If nothing short of a Bigelow module will do, then yeah, that changes things drastically.

    Just ran a bunch of numbers, and depending on assumptions (whether or not the Centaur or equivalent stage did the LOI burn as well as TLI, what propellant used on the lander, etc), I’m getting in the 2-3 ton range with a stack that could be launched on an Atlas V 552, Delta-IVH, Ariane V, Proton, Angara, with a slightly lower payload on a Zenit or HII-A. If you actually maxed-out an Atlas-V or Delta-IV Heavy, you could probably increase that by another bit. That was assuming a fairly simple lander design. I’m glossing over some nuances, but a payload in the 4500-6500lb range per landing isn’t too shabby for a single-launch system. That means you could deliver a crew cabin that weighed ~1-1.5x the dry loaded weight of the Apollo LM ascent stage to the lunar surface. For cargo, you’d need a few landings, but with just 2-3 deliveries you’re talking about something comparable to a Bigelow Sundancer module worth of pressurized volume.

    Anyhow, that’s all the time I have for now,


  34. Subsurface habitation is the only sensible option for long-term lunar stays. If you’re not launching a bulldozer then you need to be targeting natural formations like lava tubes.

    As for ISRU, there’s speculation of active volcanic vents with continuously escaping gases. Depending on their composition, that’s much more workable than regolith processing on the small scale.

    We’ll only know enough about the Moon to plan a long duration mission once robotic probes have identified the available natural resources. If the GLXP is actually won by the end of 2014 we might see some affordable lunar science in the following years.

  35. Jonathan Goff Jonathan Goff says:

    Subsurface habitation is the only sensible option for long-term lunar stays. If you’re not launching a bulldozer then you need to be targeting natural formations like lava tubes.

    Well, I’d change that to say “if you’re not launching some means of burying the habitat…” A bulldozer may not only be not necessary, but not desirable for the purpose. I’ve written previously about some other methods that seemed particularly promising (like the pneumatic excavator).

    The challenge is that a lot of the interesting questions are really hard to answer from orbit. Realistically speaking, for something on the scale of what I’m talking about, the most likely answer is a compromise of some sort. It may not be the perfect or most interesting site, but at least one that’s interesting enough and that makes your life easier–at least enough to get a toehold. Once you’ve reached that point, you can always diversify to other sites, and as I mentioned in other comments, hopefully while all this is going on, you’re working out depots, good reusable ETO and in-space transportation options, etc.


  36. Mike Lorrey says:

    Robert Horning
    “Contrary to popular belief, America isn’t a risk averse society. …
    The U.S. Congress, on the other hand, doesn’t want to be seen as the entity responsible for sending somebody to their death (usually). ”

    As a military veteran, IMHO if congress, or the president can send us in harms way for sake of oil industry profits, they sure as heck can do so to ensure the future of mankind.

  37. Will McLean says:

    Jon @26:

    The number of extra flights you need for direct ascent return will depend on how long you stay and how often you come back. For illustration, suppose you need 7 tonnes of logistic modules a year, your two men stay for ten years and you ultimately deliver 30 tons of habs, power systems, rovers, etc. A direct ascent stage and reentry module, fully fueled, is 14 tonnes. That’s a small increase in landed mass for the ability to come home, and riding to the moon in the reentry module greatly improves your abort options.

    In reality, your crew is probably going to be pretty toasty after ten years of surface operations and you may want shorter missions for that reason, but that’s also an argument against the one way mission.

    Also, direct return isn’t the best way home from a mission mass standpoint, but it could be done.

  38. A_M_Swallow says:

    Jonathan Goff in comment 20 said
    One of the main reasons I was suggesting this idea is that I don’t think ISRU is actually going to work the way people expect on the first try. Doing iterations when you have a person on the ground who can make modifications to the equipment is going to be a lot cheaper than having to do an SEP-based cryogenic robotic lander every single time you find something needs a change. Not to mention that for things like lunar rovers, having a habitat where they can periodically go to for maintenance, or overnight protection should allow you to make them last a lot longer, and get a lot more done.

    I want a manned Moon base but it is 10 to 20 years away, where as a robotic outpost could be 3 to 4 years away. No need to waste 17 years.

    Without tested ISRU the manned Moon base will have to be 100% supported from Earth, that will require many tons of supplies. To obtain the repair and enhancement facility of a human the base will need living quarters, a work shop, a full set of tools, life support equipment, food, air, water, human waste processing and astronauts. Where the prototype ISRU machines are the payload that is a very poor payload to structure ratio.

    With humans 1 tonne will not get you very far but with robotics 1 metric ton on the surface will give you a central controller and 9 off 100 kg ISRU machines.

    A second landing a few years later will give you better machines and a start on building the Moon base. The actual life support ISRU machines can be a third much larger landing.

    The machines can be tested and improved on Earth. This can go on in parallel with development of the large launch vehicles and landers.

  39. Pingback: How To Go To The Moon On The Cheap « Tai-Chi Policy

  40. Karl Hallowell says:

    publius, you wrote:

    First Landing at a mid-latitude site, near 0 degrees longitude, just before local sunset may be desirable from a radiation-protection standpoint. You’re out of the Earth’s magnetotail, with the mass of a small planet between you & the Sun, for a good two weeks, which appears to provide the minimum likelihood of being cooked by a relativistic proton storm while you assemble & bury your semi-permanent habitat. The directional dispersion of solar protons is high enough that simply hiding behind a crater rim at a polar site is unlikely to do you much good.

    The problem with this approach is that we don’t have experience working in the environment of a lunar night (which rapidly gets to near deep space temperatures). That’s a difficult thermal environment to work in. I’m not sure what the record for working in darkness in space is, but it’s probably on the order of 30 minutes (with the Earth, another heat source nearby). You propose to extend that to two weeks. I think it can be done, but it’s a big hurdle.

    IMHO, it’s much more likely that all near future human activities will occur in lunar day despite the radiation risks.

  41. Pete says:

    With ~1/6th Earth gravity it is possible to make towers and catenaries ~6 time longer (or lighter) than on Earth.

    Place four poles around a given area and hang a tool head from them with cables and it might be possible to landscape a sizable lunar base – perhaps including sintered/buried habitats and regolith processing. The moving machinery is mostly above the abrasive regolith and may require little maintenance. Power to the tooling head can also be hard wired (no mobile power source required), and the system might be relatively easily automated/teleoperated.

    The obvious way to do night energy storage on the moon is to use solar concentrators to heat up regolith – thermal energy storage. Perhaps in old propellant tanks extracting volatiles in the process. Cooling radiators are actually a hard part – another good use for old propellant tanks.

    Still need a sustainable way of delivering stuff to the lunar surface at flight rates which may well favor reusability. Short prototyping cycle times are imperative to some developments (when failure is not an option success becomes very expensive…) and this will be limited to the time between deliveries. Prototyping onsite would not generally seem to be an option in the short term.

  42. Pingback: Random Thoughts: Lunar One-Way-to-Stay (For a While) « The Four Part Land

  43. Ed Wilson says:

    I’ll qoute someone else:

    “I tell you hold a wake for us, though we are not dead . We will never come back from where we’re going. We are dead to you.” T. Atkins in Urban Highest Adventure – Ed Buchan (In story sales).

    He refers to the Irish practice of holding a wake for the people leaving for America in the 1800’s, so few of them ever returned (less than 10% ever visited back). This was NOT the situation elsewhere, more than 40% of most OTHER european immigrants to America (both the US & Canada) retruned years later to visit what was left of the family.

    On the comment that “No Congress or NASA” would ever do something like this – I agree.

    The moon will not be colonized;

    It Will Be Stolen.

    Fait Lux.

  44. Jim Gagnon says:

    Cool idea for a shoestring moonshot. It’s actually quite similar to how the Chinese might do one. Not sure if you saw it, but John Tkacik wrote a recent article about how it could all happen when the 25 tonne to LEO Long March V is ready:

    I personally would like to see us do the things necessary to start off a land rush in space. I’ve written about it in various places, but it’s the sort of thing that would attract the quirky billionaire and corporations as well as governments. If there were a profit to an adventure like you’ve proposed, then it would happen; with the realities of space travel, I’m afraid the only profit is either geopolitical in nature or the long term value of holding huge chunks of the Moon, Phobos, and asteroids.

  45. John Fornaro says:


    I liked your posting; I would suggest a few changes. I think one of the pre-positioned resources should be an escape rocket, sufficient to get back to the ISS. Yeah, it costs a “bit” more, but I think the political fallout for a one-way mission might be too much. As to volunteers, sure, let Mickey do it, but that volunteer will have to be a well trained individual. The random billionaire might very well pay his own way all the way, which would further the enabling technology at the expense of, say, lunar prospecting. That would be ok, I’d say, as long as the technology were in the public domain. By “enabling technology”, I mean every piece of hardware and software for the mission; the lander, the hab, the EDS, etc. Since our patent system is broken, I’d include the trajectory as well.

    At the same time, a “grass roots” effort would be good too, but it would require significant seed money to get the corporate structure and facilities going. $500M is just not going to do it. Corporate sponsors would be good to have.

    The suggestion about using existing rocket engines and so forth is also good. Use “trailing edge” technology. I like the idea that a Delta 4H has the oomph to do the job.

    About surviving the lunar night: Certainly one of the pre-positioned elements would have to be a reactor. Even if you were stationed at one of the poles, you’d still need power 24/7. An orbiting solar power sat would be a more “leading edge” technology, I think. In any case, its development, construction, launching and operation would add substantial costs, arguing firmly against the $500M supposition.

    About ISRU. I don’t think that it could happen in this mission, at least not at first. for starters, we don’t have any workable ISRU factories small enough to launch, so this technology would have to be developed outside of the budget for the proposed mission. It’s true that a couple of well trained technicians could troubleshoot certain problems with a lunar ISRU plant, but they would be limited in their tool selection. It points to the fact that it’s not a problem about the number of volunteers for the mission, it’s a problem about the capabilities and training of those volunteers.

    I think the two ISRU projects are propellant and solar cell production. Which should be first? The benefit by staying is that you don’t have to do everything in two weeks. With such a long stay supposed, it starts, I think, the argument for sending two more crew members and another hab, say, six months into the mission. Which suggests eventual crew rotation, to me. Nobody can stay there for ten years.

    There’s also the mission of getting the water ice from the shadowed craters at the poles.

    Finding the “abnormal concentrations” of minerals suggests prospecting on a global (would that be lubal?) basis. Whether this prospecting is done robotically or not, it would also add to the costs of the mission. Could it be that you decide to work with what you have at the selected site, or should there be more study on that site? In all cases, the amounts needed for “proof of concept” of the ISRU techniques is small, but, other than unfiltered regolith, how easy are they to acquire? An excavating rover needs to be included in the mass estimates.

    Anyhow, the first landing would start a land rush.

  46. A_M_Swallow says:

    If you want ISRU propellant on the Moon try magnesium and LOX. Both easily available all over the planet. They just need mining, refining and moulding/cooling.

    Heat engines made from local materials are an alternative to solar cells.

  47. Tony Harris says:

    I have entertained an idea like this for some time, only in my speculation the lander would be sent ahead to lunar orbit by a slow but fuel-efficient weak-stability boundary trajectory, with the manned capsule (a Dragon?) rendevousing only after the lander had arrived and “checked out.”

    Personally, it would amuse me no end if the lander could be derived from either Masten’s or Armadillo’s designs. Could anyone tell me whether a “bare bones” 2-3 person lander, using stored LOX/LNG, could be feasibly built under, say, 10,000 pounds fully fueled, (4500 kg or so) for launch on an existing medium booster (or Falcon IX?) on a WSB trajectory to the Moon?

    Just wondering,

    Tony Harris

  48. Pingback: On Avoiding Some of the Mistakes of Apollo | Selenian Boondocks

Leave a Reply

Your email address will not be published. Required fields are marked *