Machines for a Lunar Economy

These sections focus on the machines that come to the fore once you’ve built the first permanent habs. To make it all work you have to scale up as quickly as possible so you have products you can sell, and lower your costs as much as possible by using local resources to the maximum. These machines achieve that.

Power Infrastructure

Unless you go back to plans from the 70s, reference to nuclear power has completely disappeared from proposals for the Moon. This is unfortunate, as nuclear power is constant day and night, small modular plants are extremely simple to deploy, and safety is quite easy. At the poles, sunlight is nearly constant on certain mountain peaks, making solar power a good alternative to nuclear. Anywhere else, to set up a settlement you need nuclear power. Nuclear is a really good idea on the Moon. Here, it is used while the initial colony at Lalande, Cernan's Promise, gets going.

Later on, once Cernan's Promise has developed the infrastructure, it can start extracting thorium from minerals in and around the crater and produce nuclear fuel on-site instead of importing it from Earth. But that wouldn't really be for power plants - it would be for nuclear rockets. Unless technology like Mach drives manages to produce high-thrust engines based on completely new science, nuclear engines will continue to be the best transport option between the surface of a celestial body and orbit. (Unless you've managed to build a space elevator... and not if the body has an atmosphere and someone lives there... there are a few wrinkles.)

Solar Thermal Power Plants - STePP units

The main power source once Lalande has a genuine town, and later for the city, isn't nuclear or solar panels. It's generators running on stored solar heat. Specifically, Stirling engines using a gaseous working fluid, heated by a mass of molten regolith, which in turn is heated by tubes filled with molten sodium that circulate through parabolic mirrors.

In an earlier version, the chemicals that boil off the lava were going to be collected for use elsewhere, and the plant was sized with extra capacity so that some lava could be drawn off and used to produce things like glass and construction materials. But the plant is much more efficient and easier to build if it runs at temperatures too low for that to be viable, so now a separate solar furnace will be designed for bulk molten regolith production.

Some of the components for STePP units are simple and can be produced on-site early in colony development - mirrors, the tanks, pipes. With 3d printing of parts out of sintered metal powder and casting of molten minerals, most of the radiators, pumps, and generators can also be made. Production of efficient working units made entirely locally seems feasible sooner than the same goal for solar panels or nuclear plants, especially if early units work at lower temperatures.

We have been working on a spreadsheet that roughly calculates how well this thing would run and the power it would produce. Very roughly, and it is being added to bit by bit. It does help in thinking it through and has been a good learning tool. It will be nice to have spreadsheets, scripts, and simulations for all of these models down the road.

This shows a single mirror unit, tank complex, Stirling engine, and radiator unit, for the sake of a clear explanation. Units placed in the colonies will have as many mirror, generator, and radiator units as the tank can support, and be scaled as large as simulations indicate would work well enough. That simulating remains to be done, but guesstimating from what we have now, this concept is a good option for units producing up to 10 MW of electricity, steadily day and night.

Nuclear Reactors

We'll come back to nuclear reactor designs in the Transportation section below, but there is a link in the sidebar to a very good overview of the matter, in power-point format. The current model Moonwards has for a nuclear plant on the lunar surface is quite crude. Expanding it into a good representation of a working reactor in space would be great for educational purposes. That is a goal for all the models of course, but as nuclear generators are so very misunderstood it is especially appealing to make that model. (Alas, i don't know how to interpret and adapt the drawings i've found of such things very well. When i work next work on it i'll seek consultations on the matter.)

On slide 35 is a great comparison: "Standing next to a NTP [Nuclear Thermal Propulsion] engine before launch for one year is less radiation than a diagnostic x-ray".

Extraction: Mining and Refining Elements and Compounds

The Moon has a marked tendency to either have lots of something, or none at all of something. Once we are able to really dig around there is reason to hope we will find some stuff there is no sign of on the surface. For the first years, it is prudent to assume what you have to work with is only what has been proven to be there.

That list is adequate for a lot of things, but some ingenuity will be required to turn it into all we need for survival. Every way we process ore must be reinvented to suit the Moon. The chemicals most used in batch reactions to refine out certain products - acids and bases - are entirely absent from the Moon. If water and other volatiles were extracted from the poles in sufficient quantity, some of those chemicals could be made. That will take time and the resulting resources will be precious for much longer. The more water and nitrogen and carbon chemicals you use in materials processing, the less you have for farming and living spaces. It will likely be a long time before volatile mining can satisfy all local demands. Chemical catalysts can be imported from Earth and carefully recycled, but a bit will be lost in each batch and more will need to be imported regularly. Ways to use plentiful local resources to reduce reliance on these chemicals are important.

Chemical makeup of lunar regolith.

HL = Highlands, LTB = Low-Ti Basalt, HTB = High-Ti Basalt.
From R. D. Waldron's chapter in Resources of Near-Earth Space
Things to remember about lunar chemistry Because water was not involved in the Moon's formation, and exists only at the poles, and only as captured vapors from meteor impacts, the absence of water in lunar minerals gives them somewhat different properties than Earth equivalents. One anticipated effect is that glass made of lunar materials will have much greater tensile strength due to the absence of the microscopic flaws created by water molecules in the matrix.

All lunar minerals are highly reduced, meaning they tend to have extra electrons. This also changes their chemical properties, but it won't be clear how until we can experiment on-site.
Pyroxene [?]
(Ca,Fe,Mg)2Si2O6 [?]
(mostly Ilmenite, FeTiO3)
HL [?]
(5 - 35) [?]
(42 - 60)
(42 - 60)
(0 - 35)
(0 - 36)
(0 - 10)
(45 - 95)
(17 - 33)
(15 - 33)
(0 - 5)
(1 - 11)
(10 - 34)
SiO2 [?] 51 - 55 41 - 54 44 - 54 38 - 40 34 - 38 29 - 39 44 - 48 44 - 48 47 - 53 ≤0.1 <1 <1
Al2O3 [?] 1 - 3 1 - 12 1 - 6 ≤0.1 0 0 32 - 36 32 - 35 29 - 35 1 - 65 ≤1.2 ≤2
TiO2 0.5 - 1.3 0.2 - 3 0.7 - 6 ≤0.1 0 0 ≤0.3 0 0 0.4 - 53 51 - 54 52 - 74
Cr2O3 0.3 - 0.7 ≤1.5 ≤0.7 ≤0.1 0.3 - 0.7 0.1 - 0.2 trace 0 0 0.4 - 4 0.2 - 0.8 0.4 - 2.2
FeO 8 - 24 13 - 45 8 - 46 13 - 27 21 - 47 25 - 29 0.2 - 0.3 0.4 - 3 0.3 - 1.4 12 - 36 44 - 47 15 - 46
MnO 0 ≤0.6 ≤0.7 0 0.1 - 0.4 0.2 - 0.3 0 0 0 0 0.3 - 0.5 <1
MgO 17 - 31 0.3 - 26 2 - 23 33 - 27 19 - 39 34 - 37 ≤0.2 0.1 - 1.2 ≤0.3 7.7 - 20 0.1 - 2.3 0.7 - 8.6
CaO 2 - 17 2 - 17 4 - 21 0.2 - 0.3 ≤0.3 0 19 - 20 17 - 19 14 - 19 ≤0.6 <1 <1
Na2O 0 ≤0.1 ≤0.2 0 0 0 0.2 - 0.6 0.4 - 1.3 0.7 - 2.7 0 0 0
K2O 0 0 0 0 0 0 ≤0.15 ≤0.3 ≤0.4 0 0 0

On Earth this mineral is different. It often includes sodium (Na) or zinc(Zn), and aluminum is so typical it is part of its formula.

The formula means that 2 atoms that could be calcium, iron, or magnesium combine with 2 atoms of silicon and 6 atoms of oxygen. The crystal lattice of the rock alternates between oxygen atoms and the others.

Figures in this row are the percent of volume taken up by the mineral above in rocks and soil of each type.

For instance, between 5 and 35 percent of the volume of highland rocks and soil are made of pyroxene. This covers a wide range of different rocks made of many different minerals that fall under the general category of pyroxenes. The usefulness of all these figures is to give a bit of an idea how much you can increase the content of desired elements in the rocks you gather by looking for ones mostly composed of certain minerals. Soil of preferred composition can also be collected by color or location.

Silicon dioxide, or silica - glass is mostly this. Pure, as a glass it is fused quartz, a stong, non-reactive glass that transmits most wavelengths - including ultraviolet light (for a tan, or a sunburn). It can withstand very high temperatures. It has an extremely low coefficient of thermal expansion, meaning it doesn't shrink when it gets cold or expand when it gets hot. So, glazing made of it won't develop leaks around the edges. Tubes made of it can run through heat exchangers and radiators, without cracking. At pressure of 1 atmosphere, it softens at 1665°C. In vacuum that temperature will be lower by several hundred K, but it would still be high enough to serve well in the applications mentioned. It can be drawn into fibers with high optical quality (fiberoptics).

Aluminum oxide, or alumina - in crystalline form, it is sapphire. It is very hard, and like silica very clear, transmitting light from 150 t0 5500 nm.

Then there is the question of how you get the stuff out of the ground in the first place. The powder on the surface that can simply be scooped up is good enough for many things, with the designs adapted for that shown in these sections, including the dust roasters, MIP stations, and the collectors skimming the vapors off the TESS reservoir. For purer minerals, at Lalande Crater there is a wealth of rocks exposed on the surface, with a variety of compositions. Some are chunks of the bedrock that was blasted sky high during the impact that created Lalande. There are enough such rocks to supply reactors with good feedstock for quite some time, you just have to map and catalogue what kinds there are and haul the good ones a few kilometers to the refineries. But Peary Crater doesn't seem to have many rocks, as it is a far older part of the Moon. In such places fine-grain regolith has long since buried larger rock chunks, leaving them many meters under the surface. And anyways, there are more immediate reasons why you will want to make big, deep, steep-walled holes on the Moon than simply for mining. That's where you are going to put your habitats.

Under the first 20 cm or so of powdery regolith (8 inches), the ground quickly becomes very hard packed. The individual particles of the soil look like shrapnel, because that is basically what they are - blast debris. Some of them are glass globules because they were from ground close enough to an impacting asteroid to be melted by it, and the molten spray of droplets cooled solid before they hit the ground, but it's largely shrapnel. That seems to be why the stuff clumps so well, it is all sort of gnarled together. So breaking it up isn't easy.

Low gravity is another issue. If you could teleport an excavator to the Moon and run it, what would happen when you tried to dig with it, is it would tip over on its side. When everything weighs one-sixth as much, the force of the arm pushing the bucket into the ground is enough to lift the body. Just like with refining, excavation has to be done very differently.

Refining to get metals, oxygen, and metal oxides

The alternative to using chemicals for refining processes is using energy. It takes a lot of it and comes with challenges of its own. This is the approach taken by the dust roaster shown in the next section. This one device can do a lot. It produces a range of pure metals, pure oxygen, slag that makes good refractory materials, and a vapor byproduct that can be applied as a hard coating and sealant on objects placed at the machine's outlet. Because any kind of regolith that is put into a dust roaster comes out as 100% useful products, it is reasonable to calculate it outcompetes batch methods based on chemicals despite using a lot of energy to do so. By filling its hopper with extra-hot lava melted in the MIP stations or from solar thermal power plants, its power needs can also be considerably reduced.

What the dust roaster can't provide is pure metal oxides. That is the unresolved issue that really hangs in the air. Building big and beautiful on the Moon takes a lot of glass, and making glass takes a very specific mix of metal oxides. There are a range of formulas for different kinds of glass, and they can stand a small amount of impurities. None of those formulas are close to the mineral composition of anything found on the Moon. Getting from what there is to what is needed requires either heating a suitable feedstock so hot for so long that most of it evaporates, or using an awuful lot of water and chemicals. The consistent, strong sun can easily do the heating with simple mirrors or lenses. The challenge is figuring out how to control that so specific substances can be collected, and the infrastructure can handle the heat.

Or, the process can be done in reverse - take pure metals and oxygen produced by the dust roaster, and combine them to produce metal oxides. That again requires a furnace, and mechanisms to properly mix the ingredients so they react completely, and then to shape them. It may be the most efficient method, if not in general than in certain cases. The process can be finely controlled which is a significant advantage, and heat is cheap as long as you get it by concentrating sunlight.

Olivine - that's (Mg,Fe)2SiO4 - can be purified out of molten regolith by spinning it in centrifuge. To do that, heat it to the temperature where everything in it except the olivine is fully melted - around 1200°C. On Earth, deposits of magma that sit at that temperature for a while often form a layer of olivine at the bottom because the crystals settle out. On the Moon, the gravity may be too low for that, at least in a timely fashion, so you help it along by spinning the melt fast enough that a force equivalent to several gravities push the heavy crystals to the outside of their vessel. Shall we say 10 g's of force? Right now, i favor spinning a flat, disk-shaped vessel to perform this process. The olivine would gather at the rim of the vessel, and the rest would stay towards the middle. Now you chill the vessel to around 1100°C so the whole thing hardens. Unclamp the top half of the vessel from the bottom half, pop out the hot stone, and cut the olivine part off. By lining the vessel with graphite, the stone will pop out easily. The vessel would be sort of like a flat burger, with the stone being the patty and the two halves of the mold being the buns.

The east wall of Lalande Crater is probably highlands material in parts. If you get to the bedrock in that area you might find dunite, or digging in the nearby hills might uncover that. Dunite is at least 90% olivine. Doing the above process with that feedstock would have much better yield - really, you would be removing 'impurities' of aluminium, calcium, and traces of other stuff.

Once you have pure olivine, you melt it and centrifuge it again. This time you are separating the two minerals that make up olivine: forsterite (which is Mg2SiO2 from fayalite (Mg2SiO2). The fayalite will melt first, when the stone is again above 1200°C. The magnesium silicate will remain crystalline, and as it is less dense, in the centrifuge it will gather in the middle of the disk while the iron silicate is concentrated at the edge. It could be from a third to a half of the mix. Cool the mould again so everything is solid, pop out the stone, and cut the iron silicate away from the magnesium silicate.

So, why do this... Well, it all depends on how well the fayalite reacts with hydrogen. If the fayalite is put in a fluidized bed reactor, melted once more, and hydrogen is bubbled through it, a reaction occurs in which oxygen preferentially bonds with hydrogen over iron. This needs a much higher temperature of at least 1700°C so that everything stays molten as the reaction proceeds. If the mix is all reacted this way, the yield is water, iron, and silica. That would be pretty great. However, if more than maybe 2% of the iron can't be removed, there is really no point. That is what is needed so that the silica can be turned into decently clear glass. Either this is a promising option to produce bulk glass, or it is a dead end.

The architecture in these colonies uses a great deal of glass. If you can't go outside, you have to have plenty of big windows. Thus there is a need to answer the question of how that glass is made in order for the colonies to be realistic. How to supply materials for everything else is sufficiently answered for our needs here. Glass is still an open question.

The model notes go through each stage of processing. Production was estimated in the AIAA paper to be 20 metric tons of silicon per year. If the efficiency for other metals is similar, that means about 10 tons of aluminum, 8 tons of iron, and a ton or so of titanium, considering the average proportion of metals in the regolith of the maria. Output of a desired metal can be increased by selecting better feedstock or pre-processing it with other methods before putting it in the dust roaster. The paper postulated power needs of about 200 kW. If processing targets one or more specific metals, power needs could go up or down.
This reactor uses the heat of the sunlight concentrated in the mirrors to melt the regolith loaded into the silo. Then hydrogen is bubbled through the mix. The reaction of hydrogen with the molten mix produces water and free metals. This model looks at processing ilmenite this way, but the process can be applied to many other mixtures. The hotter the contents can be made, the more can be done here. Pre-processing elsewhere can also make this unit more versatile. For instance, if fayalite can be sufficiently purified out of olivine, this reactor could be used to produce silica. The silica is then used to make glass, or panels of quartz. Even if the product is discolored by impurities, a bulk process like that would be invaluable.


About 3/5 of the velocity needed to get to the Moon has to be acquired just to get to orbit around Earth. So, unless you make that part cheap and reliable, getting to the Moon is always going to be extremely expensive. We are going to do something nifty about half way along the current timeline to achieve that - build a tether that would actually be practical in Earth orbit. Let us come back to that.

After you get to Earth orbit, getting safely to the Moon's surface involves a speed change of around 6 km/s, which is 21,600 km/h or 13,400 mph. The last part of that involves braking from a speed of close to 2 km/s (4500 mph) to a speed close enough to zero when you reach the lunar surface for a safe landing. Transport in space is a really big deal.

The plans for the transport system for Moonwards haven't been modeled yet, but they have been thought about. That contemplation is what first led to the policy of 'politics and budgets be damned, we're going to examine what we could do in space if we tried'. The question of transport is where it is really striking how much our capabilities would change if we made a serious commitment.

Nuclear Shuttles

These were mentioned above in the Power section. The kind of nuclear ship that could make a big difference, and wouldn't be very hard to produce as we already had a working prototype in the 60s, is a ship that would be used only in space, preferably beyond Earth orbit. Launching a ship like that presents absolutely no radiation risk. The fuel is not more radioactive than rocks until you start to burn it, and since it is only for use in space, you simply don't turn the engines on until you are safely away from Earth. If this was commonly understood half a century ago, we'd be living in a completely different world.

Nuclear ships could attain a specific impulse of 900 to 1000 seconds if we rebuilt that 1960s technology. That's a measure of propellant efficiency. Another way of saying it is that a nuclear engine could heat its fuel so hot it would be moving at 9000 to 10 000 m/s when it exited the rocket nozzles. That speed is what determines how hard the rocket is pushed in the opposite direction by that exhaust. The best chemical rockets have exhaust velocities of 4400 m/s in space, so a nuclear engine could speed up or slow down a ship by twice as much as the best chemical engines we have. The virtual colonies use them as shuttles from the lunar surface to lunar orbit. That involves about 4 km/s of delta V for a round trip. Cargo ships coming to lunar orbit can have larger payloads by handing off their cargo to the shuttle and staying in orbit. More information about how a nulclear rocket engine works can be found on the tech site.

Potential Future Engine Options

Ion egines have now been used to take probes to several asteroids. NASA's NEXT ion thruster is rated as having an exhaust velocity of nearly 40 000 m/s (Isp of 4190 s). They burn their fuel only a very little at a time, limited by electrical power. They are therefore slow to accelerate a ship enough to get it somewhere. That could make them good for getting a cargo ship from low Earth orbit to lunar orbit. The mass saved because less fuel is needed means a much bigger payload can be delivered. Not as much as the difference in Isp suggests - the mass of solar panels needed to power the drives eats a little into payload mass, and the slow, spiraling trajectories used by low-thrust motors takes a bit more fuel. But the payload is still several times more than chemical rockets, or even nuclear, could deliver. Ion drives are durable, too, so the cargo ship could make the run a bunch of times before it needed maintenance.

However, it would take about 6 months for the cargo ship to reach the Moon from Earth orbit, and that could be a damaging bottleneck for the overall project. The early missions are primarily to test prototype robots. Once that test data is in the next iteration of robots would be ready in much less than the 8 months or so it takes for the cargo ship to return. (The return trip is faster because it takes less delta V to break orbit, and the ship is much lighter.) Then the engineering teams would have to wait a further 8 months for the new versions to be tested, and again to send the following versions. Since there will be at least one crewed mission in support of that testing, which couldn't possibly use an ion-drive ship to get to the Moon, there will also necessarily be at least one large mission using conventional hydrolox (burning hydrogen and oxygen) upper stage engines. So, the cargo ship might as well have next-generation hydrolox engines. After the first tether orbiting the Moon is complete, there would be no further role for an ion-drive ship on the Earth-Moon run, either. A cargo ship with hydrolox engines will need maintenance in orbit earlier, but a space station in support of these missions is needed anyways, so that is no issue.

For launching from the Moon to deep space destinations like Mars or Venus, low-thrust ion drives could be a good choice. NEXT ion drives require a tenth of the fuel of hydrolox engines for the same delta v, a HiPEP ion drive about a fifteenth (that one is still in development though). Neumann drives would be even better if they are successfully developed. They look promising as long as no issues show up during on-orbit testing of a working model. They might be available in a decade or less. They could be fueled with magnesium sourced from the Moon, or aluminum collected from scrapped satellites and spent upper stages, or even carbon from the waste produced by a human crew. Other dark horses include Em drives, and Mach drives, but those need a lot more development even to be sure they are genuinely viable, and after that a lot more development to be working drives.

Pod ship from Low Earth Orbit to Low Lunar Orbit

The pod ship might as well have next-generation hydrolox engines, such as Aerojet Rocketdyne CECE engines, which are designed for up to 50 restarts. It will be so large that it would benefit from being assembled in orbit. That type of construction would allow the best overall design, as its shape would be less constrained by the size and shape of rocket fairings. Thus it could have an open, wine-rack sort of shape, with solar panel wings and rocket engines at one end. It may need to be modified after its first trip, as on that trip it takes the nuclear shuttle to the Moon, which will itself be large enough that it couldn't be placed in sectioned cargo bays. After that initial mission, payloads would go on the pod ship in pods designed to slip into bays, a standardized system like container shipping.

If the harvesting of water from the poles turns out to be as easy as some data indicates is possible, it could start bringing full loads of water back to the space station in short order, and be fueled from that source. If that task is instead difficult, fuelling from the Moon would only slowly ramp up.

This ship is unpiloted. Where autopilot programming is insufficient, remote piloting is done, but this will likely almost never be necessary. If Falcon 9 first stages can manage to land on a barge in rough seas on their own, a pod ship should be able to berth with a tether anchor station on its own.

When it is time for crew to go to the Moon, the pod ship can do that too. Cargo pods set up with all the needs of people would be loaded onto it. When transferred to the nuclear shuttle, not much needs change. The shuttle also pilots itself, the main thing would be to configure the pod for best radiation protection.

Vertical Skyhooks for the Win

The real peach is vertical skyhooks. Think of them as junior space elevators. They hang down towards the surface of the body they orbit from a large anchor mass, the same as space elevators do. They just aren't attached to its surface. The platform at the bottom of their cable (the foot) instead moves above the surface at some suitably low altitude. Ships launched from the surface need to catch up to the foot, and to brake to return to the surface. The advantage is the delta v required is much less than what is needed to get to orbit. For the first skyhook planned for the virtual colonies, it takes 1/5th of the normal delta v. That is the case if the foot hangs to 20 km above the surface and the anchoring asteroid orbits 5000 km up.

Managing a skyhook is far easier if you have a nice, heavy anchor mass so big that the center of gravity of the whole complex doesn't move very much as things come and go. That helps its orbit stay nice and circular at the right altitude with much less maintenance. One event such as a shuttle arriving at the foot or a ship being released from the tip can be compensated for enough simply by moving around counterweights, that you can afford to wait for arriving ships to correct the orbit by adding or subtracting momentum, and reduce fuel spent on station-keeping. In fact, you can even slowly alter the orbit of the complex by quickly moving a counterweight car up or down at the right point in orbit. Possibly that could handle station-keeping all by itself. However, skyhook complexes will always need engines - just to be sure there is always enough thrust around to maintain orbits properly. The 4/5ths of delta V saved this way can still be regarded as coming almost for free. What little thrust the skyhook's engines need to provide can happen over weeks or months by highly efficient low-thrust motors. HiPEP ion engines would be the best option here, or perhaps Neumann drives, both mentioned 2 sections up. An array of such engines are installed on the tips of the tethers that extend outwards from the anchors (the spaceward tethers). That is where their thrust has the greatest effect, once again reducing the fuel needed, proportional with the engines' distance from the center of the orbited body.

Zylon is the material most often considered for this sort of application. It is an existing product that is plenty strong enough to be the cable for any of the skyhooks around the Moon. The car that runs up and down the cable needs to be designed but that isn't a huge challenge. The best cable design, the best way to deploy it, matters of control and maintenance and such - these would all need to be worked out. They are pretty manageable engineering problems. If creation of such a skyhook was a priority, it is reasonable to expect we could have it working in a decade, for the kind of budget needed for development of a new rocket.

Asteroid Anchors

To get a good anchor for the tether, an asteroid retrieval mission along the lines of the Keck mission architecture would be needed. To get something worthwhile, the virtual colony version beefs it up by a factor of 10, meaning an asteroid massing about 7500 metric tons. A nice one full of carbon chemicals, pure metals, and hydrated minerals. That one goes into a polar orbit. With it, the small initial polar colony could go anywhere on the Moon every two weeks, and could lauch to Earth that frequently too.

How close the foot of the tether can safely be to the surface mostly depends on the size of the anchor mass. The larger it is relative to the shuttles, the less the tether foot dips down when a shuttle is docked to the foot platform. The mass of loaded cable cars is also something to consider. The calculations on these orbital perturbations remain to be done, so for now the foot height is set at 50 km. A healthy margin in this regard is definitely needed, but perhaps it can be lowered more. It makes a big difference to the amount of fuel needed to safely land.

One of the great things about combining asteroid retrieval with tethers, is that it makes mining the asteroid so much easier. Most of the delta v needed to get the products to a place where they can be used is looked after by the tether. The colony can send machinery to do the mining and run it from the surface. As material is removed from the asteroid for use in the colony, they can replace the mass with lunar material. In fact, they can replace it partly with modules and gear they've made for a space station. An asteroid mining business and a lunar colony would make great partners.

Once an asteroid has been placed in orbit this way, it is a lot easier to do the same thing again. The full design for the Moon is four tethers in staggered polar orbits and one in equatorial orbit. The polar tethers would make any point on the Moon accessible every 3.5 days, and the equatorial tether could launch to Earth, or anywhere else, twice a day. The equatorial tether would be much easier for any ship from Earth (or Mars, or the asteroid belt, or anywhere) to dock with than the polar tethers, requiring a plane-change of a few degrees at most and a much greater range of possible trajectories.

Skyhooks as Launchers, and Skyhook Control

Extend tethers outwards from the anchors, heading away from the lunar surface, and they become launchers. That's because the foot and the tip of the skyhook orbit the Moon at the speed of the asteroid at the center of gravity, and not at the natural orbital speed for an object at their altitude. The foot moves far slower than orbital speed, and the tip moves much faster. If the skyhook's spaceward cable is 15,000 km long, terminating at an altitude of 20,000 km, the outward velocity it imparts to a vessel released from there is 2.7 km/s. That's more than enough to return to Earth, or go to Mars, Vesta, Ceres, or Venus. It will get you most of the way to Jupiter.

If you manage to balance the mass that leaves the skyhooks with the mass that is received by them, you can do all this without any use of fuel. The momentum the skyhook complex loses when something leaves is balanced by the momentum gained when something arrives. In practice, a bunch of engines on the anchor masses are needed to keep orbits nice and circular and in the right spot.

The priority to bring asteroids that are as large as possible is primarily to create anchors that outweigh the objects that travel on the attached tethers so much that orbit control is not difficult. The ability to move the center of gravity of the whole skyhook quickly could still be advantageous to reducing fuel expenditure, and possibly be important to safety in extreme cases. This could be done by:

  • Reeling the spaceward and groundward tethers in and out.
  • Having semi-independent counter-weights that move on their own cables, or on arms, or on telescoping trusses
  • Having systems on the end platforms that reel them in or out, or telescope them up or down.

Oscillations and twirling of the cables is the other main control concern.

  • Making the platforms at the tip and foot of the tethers good and massive helps with this by increasing the tension on the cables.
  • The end platforms need thrusters that can be used to dampen oscillations, and to prevent or unwind twists in the cables.
  • The cable cars need moveable weights on rails that can be similarly used to dampen oscillations. It may be better to invest early on in infrastructure so that two cars can pass each other, to have the option of a car moving around simply to work on cable and orbit control.
  • However, with cables thousands of kilometers long, the main control strategy for waving and twirling is for cars to be able to adapt. There will always be some oscillation.

A Vertical skyhook for Earth

But wait - there's more! That skyhook in Earth orbit mentioned in the intro is the big brother of these skyhooks. Alright, it isn't a space elevator. It would still make a gigantic difference in the ease of getting to orbit. And in some ways it's better than a space elevator.

A vertical skyhook whose anchor mass orbits at an altitude of 10,000 km with a foot at 250 km has the foot moving relative to the surface at 4.5 km/s. The minimum speed needed to orbit Earth is 7.8 km/s. The speed needed to meet the foot of the skyhook is 43% less. The same force as before would be needed to overcome drag, and the rocket has to reach a higher altitude, so the overall net delta V savings to meet the skyhook foot would be more like a quarter. Since a rocket launching from Earth is at least 80% fuel, that quarter is a really big deal. Now making a reusable rocket is easy. There is so much leeway to beef up the structure of the rocket, the result would be a rocket that could be safely used hundreds of times with minimal maintenance and repair between flights, and the payload it could carry would still be several times more than what is possible today.

Then you extend the tether 5000 km spaceward and again you have a launcher that can get you to the Moon, Mars, Venus, or the asteroid belt. The only fuel you need is for course corrections and braking on arrival. Transport between the Moon and Earth becomes the acrobatics of tossing vessels to and fro between the skyhooks, timing it all as best as possible so the momentum economy stays as even as it can. If you had the material to do it, and extended the tether tip out to 20,000, it could send you out of the Earth-Moon system at 5.7 km/s. That's enough to be on your way to any of the outer planets, and gets you most of the way to Mercury. In fact, it would be important to release your vessel from the tether at the best altitude for your destination, otherwise you will need to use a lot of fuel to brake so you don't just fly right past it.

By developing skyhooks first for lunar orbit, where the demands are much lower and the consequence of total failure is limited to the loss of the equipment, a path to building the skyhook in Earth orbit is created. The technology is perfected and proven at the Moon before the venture is undertaken.

  • Zylon is still good enough to do the job, although it takes an awful lot more of it. The spaceward tether, going to the tip, is actually under the most strain - going beyond 15,000 km wouldn't be practical without better materials. At 15,000 km, the tether needs to be at least 10 times thicker where it attaches to the anchor than it is at the tip in order to safely bear the loads on it. For a decent safety margin, more like 15 times as thick.
  • The anchor asteroid needs to be as massive as possible. The figure sketched in right now is 200 metric kilotons. By the time this is done we are old hands at rustling asteroids, so we bring home one well worth the trouble, chock full of platinum-group metals. The size isn't for the sake of the billions of dollars of metals, though, it is for orbital stability. There will be plenty of engines on the anchor to do station-keeping, but the harder it is to nudge it around, the better. This complex will also get a counter-weight system that will allow the center of mass to be quickly shifted as much as is practical, maybe up to 50 or 100 km, and/or perhaps have a telescoping section on the end platforms of a similar length.
  • The anchor mass is just outside the inner Van Allen Belt, meaning the car travelling the tether has to traverse it to get there. To safely transport people it would need thick and heavy shielding.
  • Powering the car is much more demanding than on the Moon. A nuclear power plant like nuclear submarines use will be the solution used on Moonwards.
  • Earth orbit is crowded. The trajectories of everything up there would need to be much better tracked in order to anticipate collisions with the skyhook complex. It would need to be able to take steps to evade things, and to move things that are getting too close. Repair and maintenance on the cable would be a continuous task far more demanding than such jobs on a lunar tether.
  • It would be important, and logical, that the space station at the anchor host an extensive repair and maintenance service for satellites. It needs that just to protect itself from collisions with space junk.
  • An approach to satellite orbits based around the priorities of the skyhook complex would help a lot with collision issues. Three orbital altitudes would be strongly preferred: below 250 km, where the tether foot is, at the altitude of the anchor body (but only if the satellite will move in an orbit harmonious with the anchor), and above the tether tip. Since the skyhook complex can help a great deal with satellite deployment, and maintenance, there would be a strong incentive to choose orbits that pose no threat to it.

It's a much bigger project to build this thing than to do the same thing at the Moon. There sure is a huge payoff, though. The skyhook orbiting the Earth appears on the Moonwards timeline far enough down the road for it to be justified to assume stronger materials have been found by that time. This doesn't require the idealized carbon nanotube cables dreamed of in space elevator projects, but cables of less perfect carbon nanotubes or graphene would help so much they are being taken as existing - something that reduces the taper ratio to, say, 3 instead of 10.

There is lots of room to play with this idea. Down the road the anchor mass could be moved into a higher orbit, lowering the speed needed to reach its foot even further. If it orbits at 20,000 km the speed of its foot is 3.4 km/s relative to Earth's surface. If the spaceward tether extends another 10,000 km a vessel released from there would receive a speed boost of 2.6 km/s for interplanetary travel. The best balance of where to place the anchor mass would have to do with the power and time needed to travel along the cables, the maintenance needs of the complex, and the economics of using rocket engines rather than relying on skyhooks.

One thing that could be an important consideration is the ease of dropping payloads from the tether foot that then descend to the surface pretty passively. Every km/s shaved off the entry speed of a vessel reduces the demands on a heat shield a lot. At some velocity, it becomes viable to design drop vessels that use aerodynamic lift to reduce speed and maneuver. With rocket fuel sourced from space, it could become cheap to have rockets that leave the foot, accelerate retrograde down to a few hundred meters a second relative to Earth's surface, drop their load, and accelerate back to the tether foot again. That is about 8 km/s of delta V, but the rocket would be a lot lighter on the way back, which saves on fuel. And fuel from space could become cheap. Then you would need drop vessels that can reduce their speed and direct themselves enough to deploy parachutes and land safely in a reasonably well defined area, probably at sea. That could be worth it, if each one is carrying a ton of platinum. At any rate, the point is this creates a different paradigm.

Advantages such a structure has over a space elevator:

  • The climb to the anchor is less than a third of the distance to an anchor in geosynchronous orbit. As cars on the cable can only go so fast, the time savings in the trip is helpful. Beaming the power the car needs for the climb from the anchor is also viable over a larger fraction of its journey, allowing the onboard nuclear reactor to be smaller and safer.
  • The cross-section of the cable is much less than that of a space elevator because the loads it must bear are much lower. Not only could several such tethers be made with the amount of material needed for one space elevator, if there is a grave accident and the cable is cut, the risk it represents is far less. Cable designs are usually based on ribbons with an open weave. By the time an object with that shape and mass had passed through Earth's atmosphere, it may either have mostly disintegrated, or have slowed to a minimal speed and drift to a soft landing.
  • The security needed to protect an installation that starts 250 km above the Earth is much less than what is needed to protect something that reaches the ground
  • There is also no need to deal with Earth's atmosphere, whereas a space elevator has to solve the problems of lightening strikes and the effects of weather on the cable.
  • The launch boost the spaceward tether can provide is more, over a shorter distance, than a space elevator can provide, because it is deeper in Earth's gravitational well.
  • Because satellites can be shifted over time to orbits that don't interfere with the skyhook, the demands of avoidance collision will eventually be minimal. A space elevator would always cut through all orbital elevations below geosynchronous.