Even More Random Ideas About Lunar Transportation

Ok, now that I’m home watching little Jonny, I think that I have two more subtopics relevant to the Lunar Transportation discussion that I wanted to post about. The first involves propellant depots, and the second revolves around how to get people to and from the lunar surface.

To Depot or not to Depot (or Better Yet, When to Depot)

One of the most common complaints about anyone who has a lunar transportation strategy that doesn’t involve spending billions on an uber-sized rocket built by the
steely-eyed missle men at BloMart and ATK, is that you’ll then have to build a huge propellant depot before you can implement your plan. Quite often comparisons to ISS are made, with the obvious implication “well, your plan requires so much infrastructure to be done first that we’ll never be able to pull it off. Better settle for what worked in the past.” But the interesting question that goes unasked (even if it is begged) is do you really even need a propellant depot in order to do on-orbit refueling? And if so, then when does it really make the most sense to build

To get to the bottom of that question, one needs to ask what having a propellant depot does for you. I can think of a few benefits:

  • The bigger a cryogenic tank is, the slower boiloff occurs, due to the square-cube law. So having a station with big storage tanks in it will reduce the amount of natural boiloff during storage.
  • Active propellant cooling systems are much easier to use if you don’t have to pay the propulsion penalty of shipping them around each time you fly. That allows for even more efficient storage of propellants
  • A sufficient sized propellant storage facility can be designed such that its natural gravity gradient provides sufficient milli-gees to settle propellants, thus vasly simplifying the plumbing and pumping.
  • A propellant depot serves as a buffer, kind of like a capacitor or an UPS for a computer. This allows you to have frequent smaller propellant shipments that then get transfered on a less frequent basis to cis-lunar and other tugs.
  • A propellant depot allows you to have multiple providers and multiple customers. The group running the depot doesn’t even need to be one of the end users or suppliers. This makes both the demand and supply side more stable, since problems with a tug or problems with one company’s delivery service are less likely to throw things off.

I could probably think of other benefits, but you get the general idea. However, while it is pretty obvious that a propellant depot does at some point become desirable, it turns out that it really isn’t near term critical.

For instance, even though it simplifies propellant settling, there are ways to do that settling that don’t require a propellant depot. I discussed several ways in a post last year, but in brief review, you could deploy a tether and a mass to create your own gravity gradient, you could dock the two vehicles then spin them just enough to create a couple milli-gees worth of settling acceleration, you could use paramagnetics on the LOX side to attract the LOX to one side, or diamagnets on the fuels, for storable propellants you could use elastomeric diaphragms in the tanks, and the list goes on.

While not having active cooling systems and the square-cube law on your side will increase your boil-off loss problems, there are options there too. The first one is to not use hydrogen. It might very well make sense to use a hydrocarbon fuel at first just to avoid the worst of the boil-off issues. There’s no rule that says “Thou shalt use Hydrogen for Lunar Transfer Vehicles”. Another option would be to just eat the losses. For instance, since you’re likely going to be launching LOX first, and the LOX tanks only weigh about 1% of the weight of propellant in them, oversize them by enough that even with boiloff losses over the course of the fueling process that you’ll have enough left over at the end. If you have too much left, you can either vent down a little, or take it with you. In a well insulated tank on all but the smallest of vehicles, boiloff losses should be in the single digit percentages per month for hydrogen, so if you have a system that can handle a lunar flight per month, handling the boiloff issue shouldn’t be a big issue. Having a depot would be nice, but isn’t critical.

The one area where not having a depot-type facility would hurt is with vehicle maintenance between flights. Lander maintenance could be done on the lunar surface fairly easy, but without some sort of “dry-dock” in LEO or at L1, maintaining and repairing a lunar tug will be more difficult.

The other benefits of a depot station aren’t show-stoppers, but they definitely show why such a thing is long-term desirable.

So, in summary I think a good case can be made that you can at least start refueling operations even without a depot in place already. It isn’t as convenient, and makes at least maintenance and repair substantially more difficult, but there aren’t any total show-stoppers. You’d still want a depot eventually, but once you already have refueling traffic going on, it’ll be a lot easier to raise money for such a depot.

How to Travel
Which leaves the last thought, which was actually spurred by one of Henry’s earliest comments in the original discussion. Once you have reusable tugs, low-cost and probably reusable earth-to-orbit transportation, and a reusable lander, does it still make sense to have people carted around in a huge reentry module like the CEV? Does that really make sense when you’re trying to develop a transportation system that’s cheap enough to be long-term sustainable at robust levels of lunar flight demand?

I don’t really think so. Many of the subsystems wanted in the CEV already are needed for the other components. If you can return the habitat portion safely to LEO after the flight, why not ride down on whatever style of craft you rode up in? Why do you even need a capsule at all? Direct return to earth may be the simplest route, but it’s unlikely to be the cheapest. If you can come down from orbit in a reusable vehicle for instance, then you don’t need to ship the habitat module up and down with every flight. If that hab module is meant for in-space use only, it will also likely be designed in a substantially different manner from a capsule that has to survive reentry, steer, land on land or at sea, be watertight, have components that won’t corrode with sea-water, have bear-proof latches on the doors…etc.

But I’m still chewing on this particular thought, and was wondering if any of you had comments?

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.
This entry was posted in Uncategorized. Bookmark the permalink.

7 Responses to Even More Random Ideas About Lunar Transportation

  1. Ben Reytblat says:

    Hi, Jon,

    IANARS (I am not a Rocket Scientist), but I had to start with a strawman architecture for the ISDC paper I’m submitting. I’m looking at it from the business point of view: Development Costs, Cash Flow, Balance Sheet, Investment Risk, ROI.

    I’m still only just a little bit into the modeling, but my starting point is a multi-modal architecture:

    1. Surface-LEO: one or many of the current crop of alt.space launchers or their progeny

    2. A Space Port @ LEO, containing:
    ….a. Tourist destination
    ….b. Crew & Passenger lay-over space
    ….c. Pressurized and un-pressurized cargo transhipment space
    ….d. Vehicle maintenance facility, including a spare parts inventory
    ….e. Fuel and other consumables depot

    3. LEO-L1: via one of two TransLunar Vehicle types (one optimized for people, including a human crew, one optimized for cargo – remotely piloted)

    4. A Space Port @ L1 similar to #2, but potentially different volumes of storage and maintenance capacities (remote vs. depot level maint.)

    5. L1 – Surface: via a dedicated, single stage, reusable Lander

    6. A Space Port on the Moon, similar to #2, growing down the line into an ISRU processing facility and storage depot, and whatever else people will pay for

    7. Surface – L1: via the same vehicle as in #5

    8. L1 – LEO: via the same vehicle as in #3, using (I originally started with full power braking, but will switch to partial aerobraking, as you suggested in the previous message) partial aerobraking

    9. LEO – Surface: via one of the several vehicles as in #1

    All of these constructed by vehicle integrator companies (one or more per segment) and operator companies (several per segment).

  2. Ed says:

    I agree completely. There is no point in hauling a lunar lander to the moon, only to use it once and then discard it. There is no point in hauling an ablative shield all the way to the moon and back when it is only used for reentry. It is much better to have a craft that makes the journey from the earth’s surface and back, a different design for a craft that makes the LEO to L1 and back journey, and another design for a reusable L1-moon-L1 craft.

    Fuel is cheap. Send it up at as high an acceleration as possible and store it in depots at LEO and L1; have multiple providers.

    NASA has to get out of the mindset that they are the only ones who can do anything in space. They have to change to paying for services rendered, rather than the cost-plus system that stifles innovation (by concentrating contracts at Boeing and LockMart) and boosts prices (as efficiency is punished: the more efficient you are the less money you get).

  3. murphydyne says:

    My thought is that the LEO-L1-LEO vehicle doesn’t necessarily need to be different from the L1-Moon-L1 vehicle IF you make a few assumptions.

    1) the purpose of the Hab vehicle is to keep X number of crew alive for Y period of time in space.

    2) you have the ability to attach and detach components to the vehicle at each of the 3 major cislunar destinations.

    Thus, a caplet style Hab could be fitted with landing legs at L1, which could be detached and the caplet placed in a tracked or wheeled cradle. For a trip from L1 to GEO there would be no need for landing legs, but there would be a need for grappling waldoes. A trip to a NEO might go for landing legs AND waldoes.

    Of course, there are other alternatives. Homer Hickham envisioned a Bigelow-stle inflatable on a landing platform for ‘Back to the Moon’. I’m all for tinker-toying together a couple of caplets w/attached propulsion units with a universal docking node with a couple of Bigelow balloons stuffed with supplies and tools for a trip to a NEO.

    As for the depot – who said we needed to use fixed tanks? What’s wrong with bladders? These can be fitted with straps that use servomotors to tighten/loosen as needed depending on the extent to which the bladder is filled.

    I’m in favor of LH largely because it serves other purposes as well, in fuel cells and as part of the water cycle. Everyone seems to hate the idea but why not store it as water? It’s not like traffic is going to be a complete unknown. If you know you have a tug that’s going to be coming through in a week you get started on cracking the water. I’m not necessarily against hydrocarbons or other types of fuels, I just think that hydrogen is so fundamental that we might as well master it for our purposes. Then again, I’m not against aluminum as propellant, or even Ed Gibson’s proposal of silicon in a LOX slush. The main point is to get out there and start figuring it out.

    Also, two things of note:
    -Sunshields are our friends
    -I once heard talk of some kind of hydrogen superslush that’s more compact than your normal LH.

  4. Tom Cuddihy says:

    The whole fuel depots /propulsive return question has me intrigued.
    I started running the numbers myself & got some slightly different numbers than you did. I got 3.7 km/s just to get to L1, then another 500 m/s to go into L1 orbit.
    For the return from L1 you have the same effect, for a total LEO-L1-LEO delta V of around 8.4 km/s. That’s a lot for a single vehicle. Close to an SSTO.

    Boeing evidently did these trades for a similar architecture, and came up with even bigger numbers, although Mark Wade didn’t post the actual sources, and I’m not sure if that maybe that includes margin:


    anyway, it’s a big difference when you start applying a minimum payload and a realistic payload to dry mass fraction. The LEO-L1-LEO leg is the ugly one.

    One thing that would work is adding another depot at GEO. You don’t really gain all that much delta-v wise. But there are 3 principle reasons to do it that way.
    1)radiation environment is different than L1, although it’s outside the plasmasphere(usually) it’s inside the magnetosphere & stays out of the solar wind. It’s a good baby step before going for L1.
    2)rather than building a station that provides no benefit except for NASA exploration, a GEO station could do lots of things, including repairing comms sats, telescope stuff,etc. (without having to worry about the SAA or reboost). The cost would be a lot more palatable. And it would never come down like Skylab or MIR.
    3)At the beginning, for fuel stocking, you could just use expendable GTO vehicles. Every launch provider makes these and it’s easy to get a wide market of initial supply. Reusable GTO vehicles could come later. The LEO-GEO-LEO leg is also the most demanding for this architecture.

  5. Jon Goff says:

    The whole fuel depots /propulsive return question has me intrigued.
    I started running the numbers myself & got some slightly different numbers than you did. I got 3.7 km/s just to get to L1, then another 500 m/s to go into L1 orbit.

    That’s really weird, because all the other sources I’ve seen have claimed a Delta-V to get to a station in L1 to be only 3.7-3.8 km/s. 4.2 km/s is the same delta-V as getting to lunar orbit, and it just doesn’t make sense that getting to L1 would take as much energy. I could be wrong, but the sources I used claimed to be getting those numbers from reference texts on the subject. How did you come up with the 4200m/s number? I’m just curious, because I don’t have a good analytical feel for how to calculate these things myself.

    Boeing evidently did these trades for a similar architecture, and came up with even bigger numbers, although Mark Wade didn’t post the actual sources, and I’m not sure if that maybe that includes margin.

    Those were the numbers from their final report, but it wasn’t entirely clear what each delta-V number was for. I’m at a loss for why those numbers are the way they are. I’ll do some asking around and searching to see if I can dig up more evidence one way or another. If it really takes the same amount of delta-V to go to L1 as LLO, and a lot more to get from there to the lunar surface and back, all the sudden L1 looks like a really lousy place for a staging base.

    I’ll have to look into it a bit further.


  6. Tom Cuddihy says:

    The 3.7 is a TLI, or more correctly, a TL1-I delta v.
    It’s all geometry. picture the earth-moon plane from north, with the earth rotating counter-clockwise (and the moon around the earth same direction). Put the earth in the center and the moon at 3:00. When you go into a hohmann tranfer orbit from 400 km circular to L1, when you get to apogee (at L1), your velocity vector is pointing straight up with ~500 m/s. But the moon’s velocity (62000 km farther out, but circular) relative to the earth’s is closer to 1000 m/s. To stay in L1, you have to add the moon’s delta to your velocity vector, matching the moon’s rate around the earth.
    It comes out to about 450 m/s to circularize into L1.

  7. murphydyne says:

    Hi Tom, thanks for diving into a topic that has driven me nuts on more than one occasion. My innocent question to you is what you are assuming the orbital velocity of the L-1 point?

    My first mistake was trying to derive it using traditional Keplerian methods, which when I solved for the period proved to be entirely incorrect. My second attempt was to try to fit a sidereal month onto the geo-centric orbit of L-1, but I wasn’t comfortable with that number either. It does have a impact on the delta-V numbers.

    I generally use the ones from Human SMAD (which I will admit assume aerobraking for Earth-orbit returns), which shows 3.77 km/s dV from LEO to EML-1 (4.02 km/s dV for polar orbits because of J2).

    This recent paper:
    notes a dV1 (Earth-departure) of 3.07 km/s and dV2 (L-1 capture) of 0.64 km/s, for a total of 3.71 km/s, which is consistent with Human SMAD.

    Of note is that the great chart in the paper shows a LEO to Lunar surface dV of 6.26 km/s (6.3 km/s with a stopover in Low Lunar Orbit), with which a stopover at L-1 compares favorably at 6.48 km/s dV (6.29 km/s in Human SMAD).

    The point is to take it a step further and see what the delta-Vs are from L-1 to other Destinations of Interest, especially GEO. L-1 isn’t about unlocking the Moon, it’s about unlocking ALL of near-Earth space.

    Oh, and for fun, there are also trick trajectories to explore , like a Trans-LEO burn that actually ends outside the L-1 point, but slides up over the gravitational saddlepoint through L-1 (like Genesis and Stardust), passes in front of the Moon in its SOI (slowing it down), looping around out to L-2, coming back behind the Moon (speeding it up a bit) and rolling into L-1 with a minimum of delta-V change required. Probably take over a week, but a nice way to introduce the Moon to new visitors. Like a bi-ellptic transfer, you do the dVs where it’s most advantageous to do so.

Leave a Reply

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