The Golden Spike Architecture

Here is a technical description. I’m reading through it now, so perhaps comments later.

[Update a while later]

Sorry, link is fixed now.

[Update late morning]

OK, I skimmed it. As one would expect, there’s a trade off between development costs and ops cost with regard to lander reusability. Ultimately, to get to low marginal costs, we have to not throw hardware away. I also wonder how much it would take to make the Centaur reusable over a period of months or years. They’ve got a start on it with the refueling scenario. Eventually, if one is getting propellant from the moon, that would make sense. I would have liked to see a trade between LLO and EML-1 or EML-2, though. It looks to me like they settled on LLO early on.

I’m amused that they have to defend their costs as being “too low.” They look high to me (a hundred million for training?), but I have vastly different expectations about these things.

52 thoughts on “The Golden Spike Architecture”

    1. Instead of complaining, do something original:

      tried to follow link
      but it does not work for me
      link needs to be fixed

  1. Given the number of old NASA hands involved what would you really expect less that $100 million for training?

    And given that the Red Dragon probably has the ability to land on the surface with a few adjustments one wonders if they really do need a lander. Or would some way to refuel the Dragon be cheaper, especially if its only carrying 3 astronauts (2 passengers and a pilot) instead of 7 for a lunar landing?

    1. Dragon doesn’t have retractable gear, nor sufficient delta-v to land from lunar orbit and then return (that will cost 12K fps + about 4K fps to leave orbit). Also, carrying the mass of Dragon to the surface is also somewhat questionable is one wants to stay within the capability of two FHs, which I’m sure is their long term plan. After all, that’s why we invented the LM for Apollo.

      1. Gary,

        The Red Dragon will have landing gear, so that isn’t a issue. Fuel is and I wonder if the 5 folks you are leaving behind, plus a re-sized life support system, could have their mass replaced with more fuel.

        True, the Dragon won’t have enough to return to orbit which is why ISRU will be needed to refuel it. The questions is which is cheaper, building a new LM or developing a ISRU infrastructure for a site.

        Or going skipping the Dragon entirely and building a new version of the “Gusmobile” 🙂

        1. There is no room left in the unpressurized part of the Dragon, so the fuel would have to be in a kit on the nose. The to-be-designed gear may punch through the heat shield (AFAIK) and thus wouldn’t be able to retract unless they have designed to make them somehow reusable (I can think of ways to do this but am not privy to SpaceX insider information to know if they have done so).

          The mass associated with life support isn’t really that great, compared to the propellant mass required for landing and ascent (mass ratio ~4). A custom LM is the best solution, as far as I can tell; I think GS agrees since there is no evidence they are pursuing a Dragon-based lander. On the other hand, their architecture is in flux (disturbingly so, in my view), so “who knows?” 🙂

          1. Although I haven’t read it definitely to be so, the current Dragon, even with 8 Superdracos and landing gear could get to mars but crater on landing. Thus Dragon 2, which I believe will have a greater diameter for better performance in the martian atmosphere. Dragon 2 should have more room for fuel, so would probably make a good lunar lander.

          2. I think SpaceX is already testing part of the technology they will use for landing large mass payloads on the Moon – the Grasshopper. No heat shield needed, so no need to use the Dragon.

            However I like the ACES family that ULA has proposed, especially their horizontal tank lander with clusters of small engines. I think that is the configuration that will be used when non-human cargo needs to be landed. However I haven’t looked at the GS proposal to evaluate their proposed architecture, but it’s not meant to be something that would do more than tourism, so being non-optimal would make sense from a cost standpoint.

  2. I wonder if you could land a Dragon for a surface hab and then rendevouz with it?

    It seems the minimal lander is going to have a very limited surface duration.

    1. The Red Dragon is supposed to be able to place 2,000 lbs on the surface of Mars. Aerial braking helps some, but you also have a gravity well twice as deep and cross winds on Mars, so I expect it would put 2,000 lbs on the Moon. That might make a basic good habitat.

  3. One of my thoughts is that the lander isn’t really capable enough in terms of crew size and mission duration. Like the ISS, their architecture won’t leave much science time. To make such a visit worthwhile to a person, organization, or country, it needs to beat an Apollo J mission in man-hours on the surface. Putting four people on the surface for a week of ten hour a day EVAs (280 man-hours on task) in a single mission is far cheaper than sending two men for one EVA, fourteen times. The mass increment for extending surface stay is very minor, and the mass increment for adding more passengers isn’t very large, either, especially with a partially inflatable lander.

    I would take advantage of our superior computer power and greater experience with unmanned landers to make the system fully automated, with provisions for some human input (but without requiring it). Then split the mission into two landings, the first as debug/flight test of the descent module which also delivers a semi-autonomous rover, which then locates a good landing spot and flags it with a marker beacon that the manned-lander will automatically seek.

    Then the manned lander arrives for an extended stay, using the rover for mobility, solar cells for power, and having a back-up return option because the first lander’s ascent module is still there (and also packed with extra supplies). The crew stows most of their moon rocks in the first lander, which departs the surface shortly after they do so the Dragon can dock with the second lander to get the extra rock samples, so so that a second Dragon can retrieve them from lunar orbit, or depending on the remaining fuel on the first module, somewhere closer to Earth.

    People have spent so many decades thinking of architectures for a more sustained lunar presence that I’m surprised someone proposed an architecture that’s more of a flag-and-footprints trip than Apollo 11 was.

    1. To make such a visit worthwhile to a person, organization, or country, it needs to beat an Apollo J mission in man-hours on the surface.

      Have you done a comprehensive survey of every person, organization, and country on Earth?

      Dennis Tito didn’t have to break NASA’s long-duration record to make the trip worthwhile.

      Putting four people on the surface for a week of ten hour a day EVAs (280 man-hours on task) in a single mission is far cheaper than sending two men for one EVA, fourteen times.

      But are those four people people (or their sponsor) willing to pay as much for a longer mission as 28 people on would pay for shorter missions? That is not rhetorical question, if you’re trying to make a business out of it.

    2. But Dennis Tito spend eight days at his destination, not eight hours. It’s not about breaking the J missions’ record, it’s that it wasn’t until the J missions that we started really exploring.

      Extending the duration would probably add twenty to fifty pounds per man day in an open-loop system (depending on if you want to allow for showers), so the mass penalty of the extended stay is small, as is the added cost. They couldn’t extend Apollo stays any further because the Apollo LM ran on 255 pounds worth of batteries, whereas a lander powered by solar cells (the Dragon is already solar powered) could stay for a very long time. The LM also lacked an airlock, so each EVA carried a large depressurization penalty.

      So the question isn’t whether you conduct one mission with four people versus fourteen missions with two people (four customers versus 28), it’s whether you’ll sell more one-week EXOTIC vacation packages than eight-hour exotic vacation packages. I think everyone in the travel industry knows the answer to that one. For countries or sponsored scientists, that also translates into bang for the buck, giving them a much stronger argument when seeking funding.

      The benefit of the larger crew size is that the pilots will probably end up as dedicated specialists who are getting paid, limiting you to one paying customer per mission instead of 3 (or 5 or 6). Since a Dragonrider is designed to carry seven people into orbit, it would make sense to be billing at least five of them, even if not all of them land on the moon. “Gold level passengers get to stretch their legs. Economy passengers get to go round and round the moon.”

      1. But Dennis Tito spend eight days at his destination, not eight hours.

        You can’t get to the Moon and back in 8 hours. The round trip will be at least a week.

        it’s that it wasn’t until the J missions that we started really exploring.

        This is the big problem with the space movement. Everyone wants to redefine “exploration” to exclude everything except his own pet project. “Suborbital isn’t exploration.” “LEO isn’t exploration.” “Asteroids aren’t exploration.” “Non-Red planets aren’t exploration.” “Exploration without science is merely tourism.”

        Just because it isn’t the type of exploration you want to see doesn’t mean it isn’t exploration.

        So the question isn’t whether you conduct one mission with four people versus fourteen missions with two people (four customers versus 28), it’s whether you’ll sell more one-week EXOTIC vacation packages packages than eight-hour exotic vacation packages. I think everyone in the travel industry knows the answer to that one.

        I am not planning to sell any vacations at all. I am not Alan Stern.

        You’re making an assumption about the elasticity of the demand curve with respect to mission duration. “Everyone in the travel industry” would ask to see the marketing research to support that assumption.

      2. The round-trip will take a week, but that’s like counting the time spent riding across the continent on a Grayhound bus to get to Yuma as vacation, but without anything to see in between the Earth and the moon. Most people would of course go for that, because it’s a ride “in space, to the moon!” But once they’ve arrived in the middle of nowhere, having a small hotel room, a jeep rental, and three or four days to kick around in the desert is a huge plus. Otherwise it’s like getting to Yuma and having to stayi within walking distance of the bus station for six or so hours, which isn’t a vacation, it’s what you do when you missed your connection in Yuma.

        From a customer perspective, the rover may be more important than the lander, because the lander will probably land in a nice, smooth, boring area (just like early Apollo missions), whereas the rover gets you to big mountains and big craters where the views are spectacular. The ideal would be a lunar RV with a tiny ascent module, so you don’t even have to retrace your path back to a lander, and can just drive cross-country for days. Then you didn’t just arrive and walk around the bus station for a few hours, you roved, collecting samples and video across a big swath of terrain, naming countless new features after your in-laws, pets, and sponsors. That would be an expensive architecture and probably isn’t in the offing, but adding some sort of vehicle and extending the mission duration is pretty easy.

          1. Certainly true, but economics demands something beyond flags and footprints for it to be sustainable.

            As Trent has observed, sending four to mars one way is a suicide mission.

            It’s not so bad sending a few tourist on a round trip to the moon. But for how many would that be worth $750 million? What’s a moon rock souvenir worth these days?

          2. Ken,

            Personally I think they have only one “commercial” market in mind, NASA. I suspect you will now see a lot talk about NASA doing a lunar COTS…

  4. I wish them the best but agree with Rand and George, more re-usability and a more robust set-up.

    The problem with this is it looks too much on the edge, a lot of effort when only a little more input would make the trip far more rewarding.

  5. That lander would make a cool point to point transport for the surface, operating from a base to areas of interest, like a little Robinson, but the idea of traveling thousands of km through space in it, feels .. a bit daunting.

  6. I would have liked to see a trade between LLO and EML-1 or EML-2, though. It looks to me like they settled on LLO early on.

    Agreed. There are slow routes to EML-1 that only take 3.2 km/s of delta-v, vs 4 km/s for LLO, which would help offset the inefficiency of using storable propellant as well as reducing the impact of having a small EDS.

    1. Also if they are planning multiple missions wouldn’t there be benefit to a Bigelow Aerospace habitat at the EM L1 that a reusable lander could be based at using ISRU for refueling it when on the surface? Add a Bigelow Aerospace surface habitat and you have a base where instead of a short stay you could have a months long rotation, time to really do something useful beside take a few photos of flags and footsteps.

      Really, rather than a railroad, which is sustainable and reusable infrastructure, this seems more like just a few quick trips tossing everything away, a cheapo version of Apollo and about as sustainable. Almost like a NASA mission approach 🙂

      1. It seems to me that a Bigelow hab at EML-1/2 might even draw some super-wealthy private clients, especially if they get to stay for a couple of weeks.

    1. Not if they pull it off, which is admittedly very uncertain. But if they do, it will mean the end of the Shuttle political industrial complex. I doubt NASA would ever be permitted to develop their own hardware again and a massive cut to their budget would seem likely.

  7. They look high to me (a hundred million for training?)

    They list crew and training as a “non-recurring cost” (strangely), so I’m not sure what that includes. More than one crew, obviously.

    I think a week aboard ISS would be valuable (if not required training) for lunar missions. So, I can easily see how they might get to $100 million.

    1. They list crew and training as a “non-recurring cost” (strangely)

      Right. So it makes no sense.

      I wish that Alan had asked me to be one of the “peers” who “reviewed” the paper.

  8. Even the $7 to $8 billion estimated development cost of Golden Spike may be over inflated. Keep in mind these were NASA guys using NASA costing models. But SpaceX has shown that following a more commercial approach can reduce development costs by 1/5th to 1/10th that of the usual NASA approach. Then we could have a commercial return to the Moon plan for a development cost of a few hundred million dollars!

    Bob Clark


    FRIDAY, JUNE 1, 2012
    On the lasting importance of the SpaceX accomplishment.

    1. Bob, you have no basis for suggesting Elon a liar: But I wonder if the real reason is…

      Your conclusion also has some problems: Manned spaceflight becomes routine world-wide

      Launch costs coming down would stimulate more companies to take advantage, but that isn’t really major.

      The thing that is major is having motivation (money! profits!) to send out large masses BEO every year because a city somewhere BEO has needs or more and more people want to go to that city.

      Profits could exist now at today’s costs with just some minor developments (sending more at a time for a marginal cost.)

  9. EML-1 and EML-2 are attractive but if trying to stage the whole thing from Earth without in-situ propellant, the extra delta-V to and from the lunar surface compared to a lower Lunar orbit loses in the trade.

    One of the interesting things I learned studying this was that the mythos about lunar orbits not being stable is somewhat exaggerated. There are lunar orbits which are reasonably stable (mostly polar or near-polar) so if one is willing to live with those, one can stage hardware in Lunar orbit for long term.

    Of course when one moves past Lunar exploration to Lunar development, the economics of what you’re exporting become more interesting, and in-situ derived propellant features prominently in those economics. That makes it advantagous to move your staging hub to one of the Lagrange points, burning a little more (relatively more affordable) Lunar propellant and saving a little (relatively more expensive) Earth-derived propellant.

    Also, the propellant is more valuable to other users the further away from gravity wells it is, which again favors a Lagrange point.

    But the savings for earlier expeditions of staging in a lunar “frozen orbit” were significant.

    1. EML-1 and EML-2 are attractive but if trying to stage the whole thing from Earth without in-situ propellant, the extra delta-V to and from the lunar surface compared to a lower Lunar orbit loses in the trade.

      That’s not true. L1/L2 comes out ahead if you use three-body trajectories to preposition the lander and its propellant to L1/L2. It takes a hundred days to get there, but since the lander would be unmanned that’s not a problem.

      But more importantly, if you have only a small EDS, or lack cryogenic refueling in LEO, or both, then a stage about the size of Centaur or DCSS is about the largest you can launch without an HLV and that gives an upper bound on the size of individual pieces you can move to L1/L2. In that case the lower delta-v to L1/L2 could easily be decisive.

    2. Actually NASA has identified four stable low (under 100 km) lunar orbits at inclinations: 27º, 50º, 76º, and 86º”.

      A Bigelow Aerospace station in the 86º one would be a good tourist destination as well as a site of studying the Moon. However all orbits would be useful for the hand off of fuel launched by mass drivers or similar fixed installation to solar electric tugs (ion or Plasma) that could cheaply deliver them to where the customers want them. Remember, as with oil its the cost not the speed of delivery that matters with fuel.

      In the long term, beyond Golden Spike, I see good transportation architecture being the Dragon or equivalent taking passengers to the Bigelow station, then having multiple reusable landers that use ISRU fuel ferrying them down to a surface installation at one of the poles.

      BTW the GRAIL mission should refine the measurements of the lunar gravity field to allow better prediction for other low orbits making it possible for solar electric thrusters to be used for station keeping.

  10. BTW does anyone have good costs figures for a Saturn V mission?

    I seem to recall they ran about $500 million a flight in terms of 1970’s dollars, which would probably be about $5 billion today. So in terms of cost it would seem this is only about a third of what the Saturn V missions went for.

  11. But the savings for earlier expeditions of staging in a lunar “frozen orbit” were significant.

    There is a stable frozen orbit at 150 km apogee over the north pole and 50 km over the south. That is a very good staging orbit.

    Looks like it is time to update my architecture as this one is really set up for a sortie type operation, more of a flags and footprints for rich people….

  12. Table 2 of the technical presentation by Stern is wrong. The GTO mass delivered is only 12,000 kg, not 20,000. Thus the TLI delivered mass is apt to be less as well.

  13. These guys may get to the Moon, but they aren’t going to help humanity get into space; not on purpose anyway. They’re a bunch of ex-NASA guys and politicians, and this is a NASA-like and political program. They’re just selling flags and footprints to second-rate powers. I don’t see any thought given to really investing in an infrastructure that permanently lowers costs over the long run. This is just low-rent Apollo.

    It’s Bigelow and Musk (and maybe Bezos or some other sub-orbital outfit) that will make the investments in sustainable space.

    1. Well, launching a lot of mass would help, especially if they used propellant transfer. Simply making the lander refuelable would help a lot, and that’s easy enough with storable propellant. Having a small cryogenic stage that can loiter in LEO for a couple of days and is capable of being launched on several commercially available launchers would also help. Such a stage might be relatively small, but if you offload most of the spacecraft propellant, use L1/L2 rather than LLO as a staging point and refuel at L1/L2 rather than in LEO then it would be large enough.

      So I disagree we need any additional infrastructure. All we need is to make the lander refuelable. Then, given enough demand, commercial infrastructure will follow, most notably small RLVs.

      1. Martijn,

        The key problem I see is they are making the same mistake NASA did with Project Apollo. In order to reduce up front costs and time to get humans to the Moon they are trading off sustainability for one shot missions using LOR.

        It appears they haven’t learned the lessons of history which is ironic since they are using the Transcontinental Railroad, the very symbol of a sustainable cost reducing infrastructure as their symbol.

        1. I don’t think they’re trading off sustainability at all. LOR is a good idea, and Lagrange rendez-vous is probably even better. They are focusing on the lander, which is wise since we already have launch vehicles and capsules. The only thing that is missing is propellant transfer, and that could be solved easily by making the lander refuelable or even just fuelable once.

          Cost-reducing infrastructure needs a certain level of demand and an exploration program could provide that demand. In an open market this will lead to private investment in infrastructure. To the degree a certain technology or a certain piece of infrastructure contributes to reducing cost, the market will use it. To the degree it doesn’t, the market won’t and shouldn’t. Premature investment in infrastructure is like carrying excess inventory, it’s inefficient.

          1. Martijn,

            A reusable lander would be good if it comes out of this venture. But everything seems geared towards a single use throw away lander which is probably its key weakness.

            But other than a lander there is really nothing her that Space Adventures has been trying to sell for a decade, a trip around the Moon, for an order of magnitude less which is why it is difficult to see it being any more successful given it has such a higher cost barrier to over come.

            As for research/science/resource exploitation on the Moon, at this stage rovers would be able to accomplish a lot more for much less. That is why I believe the next humans returning to the Moon will be individuals to support/repair/partner with rovers, not Apollo style missions.

  14. I’m not confident that they’ll get those nations lining up for the prestige of having a national walking on the Moon. You have to look at the amount of money involved, and how long the prestige will last. But most importantly you need to consider just how prestigious the people of such nations buying the ticket will see such an event, whether they’ll see it as a good return on their tax dollar.

    They won’t because: “space isn’t important”.

    1. I looked at the Wikipedia list of astronauts, and there are astronauts from 30 nations that lack the ability to launch humans into space (not counting the USA, and not counting 8 former Soviet states) that have flown on other nations’ spacecraft. Clearly, people have no qualms when it comes to hitching rides into space. NASA does it after all….

      1. Warren,

        Yes, but its done as part of foreign policy, as a demonstration of the good will between the nations involved.

        Also in most cases the launching state is picking up the tab. NASA’s paying the Russians for flights is the exception for nations using another one’s launch system, not the norm except of course for private space tourists. That is the one weakness of the sovereign space market.

        Its one thing selling hardware to other nations, its another to just sell rides.

        1. There is almost always a quid pro quo. E.g., ESA or JAXA astronauts may or may not get “free” rides (I don’t know the details), but on the other hand, it is also the case that their respective governments have spent billions helping out with the construction of the ISS.

          Also, WRT to NASA being the exception that pays for rides on other people’s systems, that bodes well for Golden Spike. After all, it is rather hard to argue that NASA should NOT become a GSC “customer” at these prices. If NASA contracted out for 2 flights a year for 3 or 4 years, GSC would be in the black.

          1. Warren,

            If you drill down in the data you will find most of those astronauts flew on Soyuz or the Shuttle before the ISS so there was little “quid pro quo”. It was pure geopolitics with Russia and the USA giving free rides for “good will” purposes.

            [[[If NASA contracted out for 2 flights a year for 3 or 4 years, GSC would be in the black.]]]

            That I suspect is their real target market, NASA, and the real reason they were formed. You will probably see them spending their time pushing for a lunar COTS rather then focusing on any extent to locate true commercial markets.

  15. @Thomas Matula:

    Even an expendable lander that is able to accept propellant in orbit and can therefore be launched dry would be very useful. It could create a competitive market for propellant launches, which in turn could accelerate privately developed RLVs. Once we have those the future of commercial spaceflight is assured. By contrast a reusable lander without that market and without those RLVs would be useless. The only crucial thing that’s missing is the propellant market, and that could easily be established by allowing the lander to be launched dry.

  16. About the price of an SLS. I really don’t see the savings…by Grabthar’s hammer.

    Some more interesting blurbs from the net:

    “Even with this “head start” approach, the company will still need
    significant amount of money to develop this system: Stern said they
    estimate the cost to be $7–8 billion”

    “But John Pike, a longtime expert on space policy who heads, said he was “deeply skeptical” about Golden Spike’s
    business plan. “If you could do it this cheap, somebody would have
    already done it,” he told me.

    “Landing pod looks like-Cobra Flight Pods from 1980’s G.I. Joe cartoons”

    “Except that the development cost is far more than $1.5 billion. Stern was clear that it would require multiple billions to get to the first mission. So your first 50 kilograms might cost $8 billion, and your second 50 kilograms might cost only $1.5 billion, or $9.5 billion for 100 kilograms. Hard to see how that is a bargain.”

    “And why does anybody believe these numbers? I listened carefully and never heard anybody state who did their cost estimates. Were they done by an independent assessment team that has a reputation for producing accurate cost estimates?”

    This is why heavy lift matters. Had Apollo–or Constellation–been allowed to continue, you could have had real infrastructure to allow for a true moonbase with sizable rovers and an ability to DO WORK. One way or another, BEO is going to cost billions–so you might as well spend a bit more, preserve infrastructure here, and have real capabilities out there.

    Apollo allowed for a pretty good haul of moon rocks that were selected in situ and by hand no less. The LEM, already pushing it mass constraint wise, was a tank compared to this contraption. A thimble too much of regolith and you’re not coming back. The LEM allowed a lander which covered more ground faster than any robotic rover before or since.

    GS is basically asking for its astronauts to dance barefoot atop a razor blade over the Pit… that’s the margin you are talking about here.

    What they are selling as their plan’s biggest strength is actually a weakness. No new LV capability that will allow simpler, more robust missions farther afield like SLS for comparable amounts of money.

    Therefore GS’s plan is more “Flags and Footprints” than Apollo itself was–because that’s all the blasted thing will hold. In retrospect–Apollo allowed more real science than GS affords.

    Were I a very wealthy investor, I would launch a Bigelow module to ISS, try to inherit that for a song–and put the other 6 billion into MCT and Skylon development.

    A couple of things from the nasaspaceflight site

    “I was telling people that “tumblr” site was full of nonsense.”
    “GS gets a FAIL on mass communications for something so EPIC.”

    Big “admiration” for NASA noted in closing remarks.

    Jeff Foust tweet: “Stern: this would not be possible without our current forward-looking space policy.”

    “NASA is tackling the very important and difficult challenges of human deep space exploration by developing systems including the Orion spacecraft and Space Launch System. The capabilities being developed by Golden Spike compliment NASA’s deep space exploration program.” said MacCallum.

    Maybe he was being diplomatic. Other responses:

    “But they really aren’t taking advantage of the Falcon Heavy.”
    “I have a hard time believing a Centaur will ever be integrated with a SpaceX vehicle”

    “There’s a bunch of others errors in this paper relating to the Dragon. It seems they’ve gotten no better information out of SpaceX than us amateurs. I think Golden Spike needs to go pay SpaceX for some mission analysis. That’s the only way they’ll get real numbers on the Falcon Heavy / Dragon capability.”

    I think we need to let MSFC take the lead on BEO, and let Musk and others deal with LEO only for now.

Comments are closed.