Starship Moon Base Design Principles

In previous blogs I’ve talked about how Lunar Starship can save the Artemis program to build a sustainable Lunar base. I’ve discussed how a Lunar Starship can deliver roughly 210 T to the Lunar surface if traveling one way. And I’ve talked a bit about some of the challenges of managing electricity generation during the Lunar night.

One missing piece of the puzzle is speculation about how a Lunar base assembled from Starships might actually look. What would it be like to live there? This is a fun question, and this blog is intended to discuss some aspects of the problem. It’s also a less serious question as the particulars of interior design are dependent, in large part, upon individual taste. Unlike, for example, hard numbers about Lunar transfer orbits and mass allotments, there is considerable latitude here for speculation and flights of fancy. The Lunar Starship represents approximately 2000 cubic meters of pressurizable volume, which is more than 4x the size of my house.

Starship to scale with a model of my house. There’s room for all kinds of stuff.

My intention with this post is to discuss some existing ideas and talk about some of the more obvious design considerations and constraints. Let’s not take this too seriously!

SpaceX's Starship interior concept by Erik Corshammar (ErcX) & smallstars - docked to ISS
Concept by Erik Corshammer (Erc X) and smallstars (

I’m not the first person to develop concepts in this area! Here’s a post at HumanMars with a list of speculative designs, and Google Image Search turns up a variety of other concepts. Many of these look pretty cool and draw on older concept art developed by SpaceX. In particular, they often feature capsule-like sleeping areas, reclining couches, lounge areas, at least one deck devoted to treadmills, and a large central tunnel with ladders for moving between floors.

All Decks of SpaceX 100-passenger Starship design by Ace & Michel Lamontagne
Concept by Rick Kiessig and Michel Lamontagne (

When I think about Starship Lunar Base interior design, I think about how the space will be used, what its mission might be, and what mechanical constraints exist on the design.

Concept by Joseph Lantz (

In terms of space utilization, a well designed base should support the daily flow of activities from sleep and self care, to eating, recreation, work, and maintenance. People will need private space and public space, and some common areas will be noisier than others at different times of day. If the base, or part of it, is operating on shifts, how will that work?

Concept by Paul King (

Is the design so mass and volume constrained that we need to draw on submarine design language and have one hundred sweaty people sharing two showers (one of which is full of food) and hot bunking in shifts? Or do we turn to the design of Antarctic research stations, which must support teams of variable size performing multiple independent and extended missions in a deeply hostile environment?

The mission of the Lunar base at the Lunar south pole is to support science and technology objectives. Over time, the base will grow but it is also necessary that each Starship be capable of operating as an independent and self-sufficient surface space station. Long term, larger facilities may be build directly on the surface but with 2000 cubic meters per lander, and 210 T of cargo to play with, I don’t think that there’s a strong forcing function to start digging tunnels through Lunar rock. This would be different if the largest possible lander was another HLS contender or a tiny Skylab variant, like Jamestown Base in the first season of “For All Mankind”.

Finally, let’s consider mechanical constraints. The Starship is nominally a stainless steel tube about 50 meters long and 9 meters in diameter. The top half is payload volume, and the bottom half is divided between two tanks for oxygen and methane fuel. These fuel tanks can also support pressurized habitable volume after landing, but would arrive as little more than empty caverns with, perhaps, airlocks and mesh flooring built in.

The Starship’s load paths are through the skin. Therefore loads on the structure, such as heavy cargo and human passengers (if present during launch) should be mounted closer to the edge. Starship can support something like 15 decks from top to bottom, but mass and geometry constraints preclude any of these (except the prop tank domes) being 1 bar-capable pressure bulkheads. So keep the floors thin, and put heavy stuff near the walls.

Stairs Lighthouse Spiral Staircase - Free photo on Pixabay

This structural constraint is similar to silos or lighthouses. In both cases, access throughout the structure is effected not by a central shaft with incandescently hazardous unfenced edges and ladders, but gently curving spiral staircases running around the perimeter. In the case of Starship, the stairs could include runner tracks and rack/pinion for moving heavier cargo, and could taper towards the top to reflect reduced traffic at the extremities of the structure. Outside the main airlock just forward of the fuel dome, a spiral access ramp could be deployed on the exterior of the vehicle with fold-out slats. Similar fold-out or keyed structures could support external loading with sandbags containing Lunar regolith for thermal, radiation, and micro meteoroid protection.

A spiral stair access strategy also raises the design possibility of using split levels to take advantage of relative heights. For example, a level devoted to being a dormitory needs less head room than a basketball court, and there’s no rule stating that each sub-system or separate function needs exactly one (1.0) deck worth of floor space. Here’s another approach to helical and redundant stairs for Starship.

So, how does internal space usage break down? What is the optimal population per Lunar Starship base?

Let us consider the floor plan of the Amundsen-Scott South Pole Station, which can house up to 150 people. During busy summers with even more people, the overflow are housed in tents! My wife Dr Christine Moran spent most of 2016 there so I’ve some second hand familiarity with how it works.

The image below shows a plan view of the main part of the station. There are various satellite buildings, storage, telescopes, runways, ruins, crashed planes, and tunnels not shown here. But it presents a good summary of the core functions of the station. The triple structure to the bottom right (“the arches”) is devoted to logistics. There is a garage, power plant, cargo storage, and space for 600,000 gallons of jet fuel.

The quadruple structure to the upper left is the elevated station, and exists on two floors, with a plan given in the next image.

Elevated Station Design for the South Pole Redevelopment Project at  Amundsen-Scott South Pole Station – Ferraro Choi

The elevated station, which is the newest part, is given in rough plan view below. Much more detailed design documents are available for the interested reader. The structure consists of two core structures end to end, with four wings on the north side. Three of the four contain single bedrooms roughly 2.5 m x 3.3 m, while the fourth contains a gym and basketball court, which seems extravagant but going outside requires procedures to avoid becoming an ice block.

The core contains offices, laboratories, computer facilities, medical facilities, dining, kitchen, a sauna, post office, store, greenhouse, laundry, and recreation specific rooms devoted to quiet reading, arts and crafts, music, TV, and games.

Home Sweet Amundsen-Scott Station – Mad Adventures

In terms of overall area, roughly half the station is logistics, and half is habitation. Of the habitation, half is private quarters and half is common space, divided up by about a dozen core functions. While I don’t think it’s remotely necessary for the Lunar Starship base to have underutilized rooms for very specific functions, I think there is enough space that it doesn’t have to be run like a WW2 submarine.

In total, the south pole station has about 5000 square meters of floor space. A single Lunar Starship with 15 decks each 3 meters tall has about 900 square meters of floor space. Operating on the Moon presents logistic challenges exceeding even the south pole, so I think it’s reasonable for a single Starship to accommodate between 16 and 20 people. Given that the Lunar Starship mass constraint is less stringent than the volume constraint, it’s not impossible to imagine a stretched Starship providing more living space at the cost of some of that 210 T of cargo.

There is room for seven decks above the methane dome, and nominally eight below, though some of them may have higher ceilings to accommodate bulkier functions.

Cupola with low poly person for scale.

The top of the Starship should have some kind of oculus or cupola. The uppermost deck is quite small, and makes a natural place for a secluded reading room enjoying natural light and the best views. The next deck down is also smaller and can serve as storage for infrequently required items or matter that can be moved through pipes, such as water and gases.

Oblique cutaway of speculative Starship interior

I generally think there should be a gradient of noise and activity across the Starship, with the sleeping decks positioned a good distance from the machine shop. So the next two decks can be sleeping decks, which are also a natural place to include water storage for additional radiation protection. 16 crew members can be housed in 16 cabins split across two decks. Each cabin is similar in scale to a college dorm room with a bed, desk, chair, porthole, closet, and perhaps a sink. Each sleeping deck also has a small central common lounge area and common bathroom facilities. As the decks are accessed via a helical stair mounted to the inside of the pressure hull, the sleeping deck floor plans are clocked relative to one another.

Multiple rooms per deck, plus bathrooms and small common space.

The next deck down handles the transition between introverted self care and work/life functions, so is the natural place for a “mess”, a large open space used to serve meals, with an adjacent galley, bathroom, food storage, and a variety of wall-mounted cabinets to facilitate the space’s usage for other forms of recreation.

Multipurpose galley/mess/rec room.

The next deck down is devoted to work functions, and includes office space, a laboratory, a small medical office. This deck can be naturally split into adjacent split levels depending on ceiling height requirements. Remember that in lunar gravity ceilings may have to be a bit higher, and steps can be a bit higher too.

Lab and office space.

The lowest deck with the habitation part of the base interfaces with the methane tank dome, the primary external airlock, and gravity-fed logistics functions. For example, sumps for gray and black water storage are housed at this level before treatment and pumping back to header tanks beneath the oculus. Being adjacent to the external airlock, this deck is also a natural place to store cargo and equipment, including EVA suits, machine tools, and gym equipment.

In this concept, the main cargo deck airlock is extended beyond the structure to meet a spiral access ramp.

The lower part of the Starship, that was formerly devoted to storing liquid oxygen and methane fuel, provides enough room for eight decks with three meters of head room.

Below the thrust puck and oxygen tank lower dome is a space between the skirt and the ground. This space provides additional room for cargo and is also immediately adjacent to the ground. I think this is the natural place to store rovers prior to landing, and to locate a large airlock that can interface with vehicles on the ground. While the airlock 28 meters off the ground can serve well enough for pedestrian access (given an elevator or ramp), driving a fully loaded dump truck up a spiral ramp seems tough. So the space inside the skirt is cleared of raptors and becomes an unpressurized garage.

Quick sketch shows complete profile with mesh decks in former fuel/oxidizer tanks, and garage space beneath skirt at ground level.

The inside of the tanks is initially empty. Mesh floors and stairways are compatible with cryogenic fluid flow, as are pre-installed raceways for power and water, hatches, and airlocks. But any desired equipment will have to be moved into and around the space. Therefore the top of each dome should have mounting points for a capable bridge crane and the mesh floors should be locally removable to facilitate access to any point. Pre-loaded equipment and cargo can be stashed in the cargo deck above then, on arrival, lowered through a hatch into pre-built mounting points for easy installation.

The logistics part of the station is responsible for keeping 16 astronauts alive indefinitely, and closing the environment well enough to sustain life support. This requires provisioning of adequate power, atmospheric regulation, water recycling, communications, controls, and thermal control including refrigeration. It also requires the ability to store or fabricate spare parts, support maintenance cycles of all kinds of equipment, and operate space suits and rovers.

What is the mission of the base? As specified, a Lunar Starship base could keep a dozen or so astronauts alive on the surface in comfort for extended periods but what are they there to do? The whole point of using Starship as a lunar base is to jump start surface operational capacity at a sustainable cost. If a Starship base can support a crew of 20 people for 6 months between crew changes, it is conceivable that the whole program could operate below $150,000/person/day, which compares extremely favorably with the ISS program ($10m/person/day), although a LEO space station using similar hardware would be an order of magnitude cheaper if only because the delta-V requirements are lower.

What fraction of the crew is devoted to serving logistics functions and what fraction are research scientists? The south pole overwinter crew of 50 is split roughly 90% logistics to 10% scientists, although if required many more scientists could be housed with little marginal increase in cost. If this pattern repeated on the Moon then each crew might only have two or three rock-obsessed geologists to be gently humored by 15 or so technicians keeping them alive and taking the rovers out for a spin.

Of course, there is no reason to build only one Lunar Starship base, and a sustainable program would aim to deploy them at a cadence of perhaps four per year. Each would be customized to support its specific mission and in response to lessons learned along the way. Combined, they represent a real opportunity to build a series of self-sufficient huts all around the rim of, say, Shackleton Crater. If the landing legs can articulate then they can be walked into closer proximity after landing and joined with pressurized tubes.

Perhaps, down the track, one can be deployed containing only a huge power system and vertically oriented tunnel boring machine, with which it can start excavating the second generation lunar base, complete with enormous underground caverns and Starship-derived towers.

60 thoughts on “Starship Moon Base Design Principles

  1. “Starship’s load paths are through the skin.”

    I’m surprised. I imagined Starship as having just enough strength in the tank walls to hold in the pressure, and then having the payload bay made of the same metal for ease of construction and rapid prototyping. I imagined the compressive load being borne by the pressure in the tank, to the point where a full payload and no fuel would mean that the skin crumples like aluminum foil. So I assumed that any internal structure would have to be load-bearing.


      1. I was going to comment on the lack of pressure in the repurposed tank section may cause it to collapse but I forgot about the moon’s 1/6G…whoops!

        I think though, SS will be cheap enough that sending another SS will make better use of engineers rather than repurposing the tank section. I mean it seems like a dead end (ie. Old space) use of engineers. Why not use those engineers for something that will have increasing return over time?

        I also think that SS will be so cheap compared to the rest of Artemis hubris that repurposing is basically being economical on the cheapest part of the system….

        For a completely silly idea I’m curious if “bunker bombs” can be made big enough for human habitation? Just crash one down and then enjoy the 10’s of meters of radiation shielding on top.

        Some great ideas. Loving your blogs!


    1. Starship’s main tanks have internal stiffeners and anti-slosh baffles. Those tanks are not unreinforced stainless steel balloons.


  2. I’m an optimist so I dont think that Lunar Starship is going to be designed to hold a crew of more then half a dozen people before we move on to something better.

    As I see it, inflatable habitats are just way too good to be ignored. Unlike using them for space stations they dont need to be a rigid that will stay in place relative to the rest of the space station. They don’t need airlocks that double as structural elements and rigid supports capable of transferring the force from orbital reboosting rocket burns. There’s not need to go wet workshopping the fuel tanks for another thousand cubic meters when you can just add another habitat with twenty times that volume which could be connected to other such habitats.


  3. For the nose tip, I would put a rotor on it, with masses (e.g. bags of regolith, or disused raptor engines) on cables; and spin it up for kinetic energy storage. At the ends of crossed booms, at rest the masses would hang down 50 m to the ground, and would lift up and out as it is spun up. The long cables mean radial accceleration is limited even at very high rotation speed, so it can store a great deal more energy for its structural mass than is possible for the tiny flywheels typically designed for use where outdoor air resistance, and worries about safety, would be constraints.

    Liked by 1 person

      1. Certainly you wouldn’t want to make sudden changes to the torque applied, or periodic changes close to the natural frequency of the system, so you also want supercapacitors to smooth out the load curve and smart algorithms watching it. There are people who are good at that. (Not who coded attitude control on the Mars ‘copter!)

        It would better be mounted on a supply ship parked on the other side of a hill than on the habitation module.


      2. Probably any large flywheel rotor would need to be mounted on a separate cargo ship, not on a habitat module.

        Assuming two 1500kg Raptor engines on the ends of 50m cables, swinging at 2380 m/s tangential speed, lunar escape velocity..

        Energy stored (2 * 1500kg * 2380^2) / 2 = 8.5 GJ = 2.36 MWh

        Available average draw 2.36 MHh / 2 weeks = 7 kW

        Circumference is 50m * 2*pi = 314m

        Rotation rate = 2380 / 314 = 7.58 RPS, 455 RPM, 47.6 rad/s

        Centripetal acceleration = 50 * 47.6^2 = 113288 m/s^2 = 115 G.

        Cable tension = 1500* 50 * 47.6^2 = 170 MN

        Carbon fiber strength = ~ 150 kN/mm^2, derated to 100 kN/mm^2

        Cable cross section = sqrt(170 MN / 100 kN/mm^2) = 41 mm square

        Volume of cable 100 * .041^2 = 0.17 m^3

        Each pair of Raptor-mass counterweights spun up to escape velocity adds 7 kW average capacity. Most alterations scale results linearly. Probably bags of regolith will prove more practical as kinetic mass than Raptor engines, which might even be worth exporting with returning personnel.

        It will be essential, in any case, to spray a thick coat of insulating foam over the whole exterior of the vehicle to control heat loss at night, and with a reflective surface to control heating in daylight. New solar panels will need to be deployed outside the foam, or on the ground. If it is to be applied after landing, the nose cupola will be taken up with equipment for that. Foam might be more conveniently applied while in orbit, instead, before landing.

        Strontium-isotope radiothermal generators are looking pretty appealing, given the high freight capacity available. Raw heat output of strontium-90 is 460 W/kg. A half-ton of strontium-90 would provide plenty of heat and light through the lunar night. Means to dissipate excess heat would be needed during the lunar day, and while in transit.


      3. I would *like* to be able to sell you 500 kg of strontium-90 (probably stirred with lanthanum, titanium, and some graphene). But it is mostly governments that have strontium-90. It is a waste product from nuke plants, so Hanford probably has lots, and France. Chernobyl and Fukushima vaporized a great deal to broadcast to the world, where it unfortunately displaces biological calcium.

        The Soviets distributed easily a half ton out to lonely lighthouses in the ’70s and ’80s, most since gathered up again. With a 29-year half-life, it must be half stable zirconium-90 now.


      4. 100 kg of strontium-90 would produce 46 kW of heat, continuously, which seems like it ought to be enough for one ship throught the night. You need to keep it from overheating on the way to the moon. If you keep the rocket nozzles pointed at the sun, the whole rest of the hull is radiating to space at 2.7 K, and you only need to dissipate some 140 W per square meter of hull around the crew compartment. So maybe it’s not hard to keep it cool on the way, if you just circulate enough air through.

        Fortunately strontium-90 produces practically no gamma rays or neutrons, so needs minimal shielding.


  4. This is fascinating to think about, but I wonder about the part at the end where you talk about linking them up to each other as part of larger base complexes…this seems like a brilliant initial base architecture but after a while I’d think people would get pretty frustrated by the vertical layout of everything and want to move to a more conventional horizontal base layout.

    Then again, maybe a vertical base is actually perfect for moon gravity – fun AND bone-loss preventing exercise every time you change decks!


  5. How do we make sure it doesn’t ever tip over? Maybe it isn’t necessary, but I was thinking you could use some kind of bulldozer-rover to pile up some regolith around the base, for added stability.


  6. If you wanted to have a more even distribution of mass on the floors, couldn’t you just run cables from the top to the bottom of the non-tank decks section for extra support? Starship is designed to be lifted from the nose on the ground on Earth.


  7. A housing solution has to serve wide needs with longevity.
    The one achieved through extendable habitats has the advantage of extendable, multipurpose, and longevity.


  8. Would it be useful, or even possible, to develop a Starship variant that lies horizontally on the surface, on its belly, rather than vertically? This would have the benefit of placing the airlocks right next to the ground, rather than 28 metres above it.
    I imagine that, even in low lunar gravity, falling from a gangway or lift at Starship-height would pose a significant risk to astronauts and their spacesuits. The outer hull of the base would also be easier to access for repair and maintenance.
    A horizontal Starship base would need to have a new form of landing system designed, different to the HLS system planned for Artemis, but once on the lunar surface it could be moved around with a reduced risk of toppling over due to its low centre of mass. You could even give it a set of motorised wheels, similar to the crawlers used in Boca Chica, which would make it much easier to link Starship modules together to create an extended lunar base.
    A Starship version with a tilting nose, like a Super Guppy aircraft, might even be useful for moving large and bulky cargo items to the Moon and Mars. Once landed it could simply be rolled off onto the surface.
    Are there practical or risk constraints to this idea that I’m missing?


      1. Also, based on my experience trying to fill up a bilge storage area in a large sailboat, I believe the vertical cylinder may result in more easily-usable interior volume.


  9. I think Elon has already shot down the horizontal Starship idea.

    Casey, is your thinking on subsurface tunnels evolving? You seemed pretty negative in earlier posts.


    1. Nobody seems to be leaving any interior space for thermal insulation. With external temperatures alternating between 200 degrees above and 200 degrees below ambient on a one-week-on / three-week-off cycle, structural metal fatigue from thermal cycling is by itself a serious worry, even neglecting protection of crew from searing and cryo-cold surfaces.

      My interpretation is that external, foamed insulation will be needed. It is probably most easily applied in earth or lunar orbit, but could also be applied, given some cleverness, using equipment deployed from the nose after landing. The alternative of applying insulation before launch remains, but then it must survive supersonic airflow.

      Doubling ramps for emergency exit concerns seem poorly justified. In any conceivable emergency, the interior will remain much safer than outside, unless there is some other interior to escape to, and the bottleneck would be at the space-suit foyer and air lock, both of which involve minutes-long cycles. If ramps are doubled, it has to be to aid day-to-day traffic in both directions. With two ramps, the ramps need only be wide enough for one person, so need take no more interior space in total than a single ramp.

      No one seems to make provision for sealing off one part of the vehicle from another, in the event of a hull breach, e.g. as by a micrometeorite.


      1. With vehicle surface area exceeding 1300 m^2, covered with bags of regolith 10 cm thick at, what, 1000 kg/m^3? That is 130 tons of regolith for the crew to shovel into bags and hoist up the sides, before they get down to whatever they came to the moon to do. Shoveling 130 tons of regolith would be very … let us say team-building. Then, how well does the bag material handle thermal cycling?


      2. I was theorizing they would use remotely operated bulldozers and dump trucks to cover the SS with regolith that was compacted and zapped with microwaves to stabilize it. Along with a spiral ramp up to the level of the airlock deck. The extra height of a few SS along the Shackleton crater would help provide a better locale for solar panels. The regolith would be the radiation barrier, thermal mass, and micrometeoroid barrier.


  10. Getting into and out of spacesuits, and cycling airlocks, could get prohibitively time-consuming, at times to the point of fatality.

    Better, instead, to keep most of each spacesuit permanently exposed to vacuum, with a hatch on its back to drop in through and close and seal, and a dome that can be closed over that to hold back habitat pressure when the suit is in use. Then, to come back in, you back up to the dome, clamp a seal around your hatch, then open the dome and your hatch. The only part of the system cycling between vacuum and pressure is the gap between hatch and dome, which may have negligible volume, and so need no pumping.


  11. The most difficult part of living on the lunar surface is survival during the ~14.5-day lunar night (350 hours). So, the first Starship to land on the lunar surface should be the central electric power station for the base.
    Something like 200kW of solar panels and four Tesla Megapacks (3MWh, 23t each) should suffice for starters.
    Arriving Starships plug into this power station. No assembly required for the power station. A single unit. Plug and play.

    Liked by 1 person

      1. Why do that? Those Megapack battery units take up a large amount of space in the cargo bay that could be better used for other types of payload (e.g. liquid nitrogen to dilute the LOX to make a gas mixture that people can breathe without adverse effects indefinitely, water, food).


    1. 4 x 3 MWh amounts to about 36 kW averaged over the two weeks.

      2 MW of solar panels is some 6000 panels x 15 kg = 90 t, maybe less because they don’t need to stand up to aerodynamic stress like terrestrial panels, and because insolation is more intense; but they need to tolerate 400 K temperature excursions. Batteries weigh in at 4 x 23 t = 92 ton, so we are into almost 180 tons. Add in labor to distribute 6000 panels out over the regolith: “no assembly required”, really?

      180 tons is quite a large load, considering you can have the same generating capacity from 0.1- 0.2 ton of strontium-90 plus a few tons for power extraction and cooling apparatus.

      But if we are looking for ways to use up our freight allotment, solar + battery would help there.


      1. Oops. We don’t need 2 MW of panels, but 200 kw, or only 600, for 9 tons. Labor to set up 600 panels seems tolerable.


    2. NASA can land as many Starship central power units it desires since the operating cost of a Starship mission to the lunar surface is the cost of one lunar Starship and five tanker Starship launches. At $10M per launch, the operating cost is $60M, the cost of a single Falcon 9 launch.

      The cost of a Tesla Megapack is ~$5M. So $20M for four Megapacks and a few million for the 200kW solar panel array.

      Assuming six Starship landings per year at the lunar base, two of these could be Starship central power units while the other four would land more hardware and consumables to supply a population of 20-30 humans.

      Liked by 1 person

  12. I was just wondering whether an ablation shield can be made entirely from lunar materials. Apparently zinc peroxide (ZnO2) can exist. Perhaps it could be combined with suitable other materials (alumina maybe?) to make a composite that produces gas as it overheats and crumbles away, i.e. a material that could be the basis of an ablation shield.

    However, a search for “zinc peroxide ablation shield” returns nothing relevant.


  13. Building solar arrays on the rim peaks of Shackleton Crater on the lunar south pole looks actually practical.

    The principle is that, at any time, one or more peaks of the crater rim are in sunlight. So, a solar array mounted vertically would rotate into sunlight on a monthly basis. Mounting them on a vertical axis to track the sun when it is in view would be maximally efficient.

    A 66 km superconducting cable, running around the inner rim, perpetually in shadow so naturally chilled, would deliver power to fixed points, probably one or more rim stations, and a telescope on the central peak. Shackleton Crater might be the best place in the solar system for an infrared telescope. (There is some range of distance, thus red-shift, where it functions as an x-ray telescope.) It only ever points one direction, but there is a practically unlimited amount of stuff to look at if you are looking far enough away.

    At 100 g per meter of superconducting cable, that is under 10 tons of cable. Thus, a single Starship could deliver all the cable needed, plus the towers, and the first allotment of solar panels. Stringing the cable would be a big project done manually, but hopping robots might be built to do the job. Or, the cabling might be shot ballistically from the rims of spinning drums. Or, it could be reeled off on a single flight around the rim and then lowered into shadow.


      1. Furukawa Electric Group’s subsidiary, SuperPower Inc., would like to sell you it.

        Interestingly, though: the interior temperature of Shackleton crater is right at the threshold of superconduction for RE-BaCuO tape, so it might need a circulating LN2 pipe bonded to it to keep it below the line, and to increase critical current capacity. That adds complication and expense.

        A conventional copper conductor ought to suffice, at least at first, with some sacrifice of awesomeness. Resistive losses over the up to 33 km range could be made up with a slightly larger generating capacity.


      2. The higher the voltage, the smaller the loss. But at some voltage you risk arcing to the ground, so would then need to separate it from ground with insulator stacks, which would be another complication. So, probably you would limit voltage to what the insulation cladding can withstand, and tolerate any loss. It is common to distribute power to people’s houses at 18 kVAC, where 33 km is not a prohibitively long run.

        You need an alloy of copper with good conductivity that doesn’t get brittle below 100 K; or use fine-gauge multi-strand braided cable. Many things you have to worry about on terrestrial distribution systems are not a problem, including safety, weather, and corrosion.


      3. Where placed in permanent shadow, temperature excursions the cable must experience would be minimal.

        So, cabling only needs protection where it gets variable sun exposure, e.g. on the way up to a crater-rim peak solar array. It seems like shallow burial with a reflector laid on should suffice — or possibly even just the reflector stood off a few inches.

        It is amazing how many critical details come up when designing for exotic environments. It would be surprising not to miss important ones. We need a small-scale pilot project, to start, with a power cable whose reliability the mission does not depend on.

        A base at a crater-rim peak that gets sun 80% of the time, with a battery pack for the 20% time remaining, would be much better off than one well-illuminated less than 50% of the time. They could experiment with placing extra collectors at other peaks to increase coverage, at incrementally increasing distances as power needs grow, without need to increase battery storage capacity.

        Cable laid on level ground would, at the pole, get only crepuscular illumination, so a trench only an inch or two deep would keep it in perpetual shade.

        It will be necessary to identify some level ground on the crater rim to land on. There might not be very much of that, but they don’t need much.


      4. Pure copper turns out to get better — more conductive, stronger — as it gets colder, down to well below 100K. And kapton is advertised as good down to 4 K. So at least the cable materials won’t pose a problem.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s