5.Task Report : Intermediate Term Architecture for a 

RadiationShielded Habitat

The purpose of this NIAC project is to look at concepts for building large-scale habitats in orbit. While that is clearly several decades in the future, it is necessary to lay out a possible path to get there from the present. At the outset, we point out that our interest is more in laying out a path to build massive habitats using automatic, low-recurring-cost techniques, than to get into the Moon vs. Mars debate, and hence our choices are not necessarily optimal; it is enough for our purposes to show what it takes to make them feasible. The work described in this and the following chapters was in fact motivated by a question from a NASA engineer in the audience at the First Space Resource Utilization Roundtable in 1999 how will you build one of those massive colony shells using your automatic techniques? We considered techniques that would work in vacuum and of course, quasi-steady magnetic fields are the most obvious choice there. In this chapter we discuss the intermediate-term architecture, which involves considerations of education, economics and public support, in addition to engineering. To chart this path, we present a technique to build one of the grandest projects envisioned in the 1970's; the first large habitat beyond Earth. In 1975, Gerard O'Neill presented his concepts for the first habitats, to be built at the L-5 point of the Earth-Moon system. Features of the O'Neill habitat [1] concepts are summarized below:

 

5.1  O'Neill / NASA 1970s Model for Habitat

· Economic opportunities as motivator. The precise industries foreseen as the leaders of this enterprise may not today be the prime movers, but the basic concept that economics - rather than exploration or national / military motivations - would drive the construction of the habitat, remains valid today.

· Moon as first source for extraterrestrial resources. The Moon was seen as the logical first place for the extraction of resources such as oxygen and metals. This choice remains valid today, since there is far more quantitative knowledge about the composition of lunar soil, and the availability of various resources, than there is about any other heavenly body. O'Neill pointed out that much of the material processing might actually be done in the habitat itself, where varying amounts of artificial gravity would be available.

Today this choice triggers considerable argument, which is driven by the perceived need to make either/or choices between the Moon, the Near-Earth Objects, missions to Mars, and missions beyond Mars. However, we present this as the first step in the systematic development of infrastructure on the scale needed to enable the realization of all these dreams. In that context, O'Neill's choice of the Moon as the best-understood destination is still valid.

· L-5 as the logical location for the settlement. The argument for a habitat in Space, rather than on the Moon, was that the economic reasons for the habitat involved access to other locations such as GEO, to service other satellites and Space Stations, and provide access to Mars and beyond. The access cost would be greatly reduced by locating the habitat in orbit rather than in a gravity well.

Again, this is a choice that triggers much debate. Concepts for developing radiation-sheltered sub-surface habitats have been developed (e.g., Boston [29]) and offer attractive options for quick and relatively inexpensive development of sophisticated long-term habitats with controlled atmospheres. These will be attractive, and indeed essential, for resource-development operations on planetary and lunar surfaces. However, the argument about access to multiple points in Space holds today. The major advantage of a sub-surface habitat is that there is no need to transport huge amounts of material into orbit to construct a radiation shield and until now this was a clincher. With the concept that we present below, this is no longer a clincher. With our concept, the facilities developed to build the radiation shield in orbit will remain as permanent facilities for lunar transportation and manufacturing, and reduce the marginal cost for shield construction by several orders of magnitude. The advantages of variable-gravity facilities for manufacturing are strong additional arguments for a Space habitat.

· Bernal sphere + toroidal agriculture stations on either side, with near 1-G at the equator. Studies existing at the time were cited for evidence that long-duration exposure to low gravity would harm the health of humans. This has since been corroborated by experience with the Mir station. In addition, studies were cited to the effect that over 3% of humans would find an angular velocity of more than 1 revolution per minute to be disorienting. To achieve a gravity level of 1-G, with less than 1RPM, the habitat would require a radius on the order of 1 kilometer. Various shapes of habitat were considered in a NASA-ASEE study in 1975-77 [2] and toroidal habitats were selected. Examination of the reasons for picking the toroid shows that a cylindrical shape would have been better, but was considered unnecessary because it would offer far more than the required surface area to support 10,000 inhabitants. Thus the toroid chosen must be viewed as a minimum-length cylinder.

· Shell made of aluminum and glass (to admit sunlight); support structure made of aluminum ribs and/or steel cable

This was based on then-prevalent construction techniques for lightweight, mobile structures. Human labor was assumed, in order to not make assumptions about the availability of robotic machines. The aluminum and glass were assumed to be shipped from the Moon initially, and from orbital manufacturing facilities for subsequent construction projects. We avoid the need for such detailed construction, and present a system amenable to automatic construction.

Our construction sequence, summarized below, involves deploying several rings of metal cable, attached to solar-heated gas thrusters which will provide a small rotation rate to provide some tension and retain a circular shape. Rectangular metal containers (boxcars) of size 20m x 2m x 2m will be launched from the Moon to arrive at this site with a very low apogee velocity. They will be maneuvered into place using hybrid shepherd spacecraft, and welded together using robotic welding arms. The final stage of the maneuvering, which is the actual construction stage, will be performed by interaction between electromagnets on the Shepherds, and currents in an electric grid held by a construction Spider, a robotic platform. A ring of these containers will provide the nucleus of the structure. A central beam structure will be built along the axis as the nucleus of the zero-g manufacturing facility. A mobile construction spider and a tether system from a central metal structure will be used to reduce the repetitive work in capturing and positioning loads. Additional cable rings will be attached to the first ring of boxcars, and used to support the next set of boxcars, and so on. The angular rotation rate will be maintained, and gradually increased, by gas thrusters. After the first ring of boxcars is completed, following boxcars will arrive filled with lunar regolith to form the radiation shield. As the 2km-long cylinder is completed, the ends will be sealed using a combination of water-filled inflatable bags, support structure, solar collectors and transportation gateways. It is not considered necessary to enclose (other than for radiation shielding) and pressurize any part of the cylinder except those regions intended for human or other live occupation (e.g., agriculture sections). The agriculture sections may be in concentric cylinders at much lower g-levels closer to the axis, with filtered sunlight directed in at appropriate angles using thin-film mirrors.

· Projected earth-LEO launch costs of $110/lb. Costing was attempted, assuming this launch cost to Low Earth Orbit using the as-yet untested Space Shuttle, with boost to the lunar L-5 requiring the same amount of energy again. While this number proved to be an underestimate by two orders of magnitude, the implication now is to rule out the use of Earth-launch for any recurring-cost or mass-manufactured items that can conceivably be manufactured on the Moon or in orbit. It is still necessary to Earth-launch such items as control equipment, the spacecraft needed to shepherd loads to the construction site, hydrogen, nitrogen, and the robotic arms for manufacturing on the Moon and at the habitat site.

· Lunar-basedmass driver. The bulk of the material for the radiation shield, which was to be built of lunar regolith, was to be launched as baseball-sized lumps of regolith, accelerated at 30-G over a 10km track, and at a rate of about 10 per second. A catcher system positioned near L-2 would receive these ballistic payloads and take them to the construction site. We depart from this launcher, which was optimized for the single project of building a habitat, and instead argue for a versatile launcher system, which will form the nucleus of the future translunar surface transport system. Our basic payload unit is a 160,000 kg boxcar filled with regolith, launched by a carriage which keeps the electromagnetic components of the launcher on the Moon for re-use.

· Radiation shielding dominated the mass of the settlement. Ref.[2] envisaged a stationary radiation shield around a revolving toroid, with some means for maintaining a suitable gap between the shield and the moving structure. This was seen to greatly reduce the strength demands on the spokes and other structure of the habitat. In our concept, we use metal cables to provide the initial scaffolding, followed by a welded-box rib and longitudinal beams for the shell strength.
 
 
 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

 
 

5.2  Present Model of a Habitat

 
We adopted a strategy where by the habitat project would itself serve to bootstrap an entire Space-based economy. Rather than optimize everything for the most efficient construction of the habitat, we looked at how to set up the many other industries in a synergistic economy. Our approach also assumes that human presence at the construction site is not necessary until the radiation shield is complete. Recurring costs are minimized, and thus the project can be spread out over a longer period. Since the construction is automatic (with at most telepresence supervision and control from Earth) we can afford to consider building a cylinder 2km long, with the entire radiation shield gradually accelerated to 1 RPM by the time the shield is complete. Major differences with the 1975 approach are summarized in Table 5-1.
Table 5.1: Important Differences Between 1975 and Present Models for Space-City Construction
#
1975 models
Present model using Tailored Force Fields (TFF)
1
$110/ lb  Earth- LEO 
$1,300 to $14,000 per lb to LEO
2
Human labor on-site 
Robotic with Earth-based telepresence supervision 
3
Geometry: Toroid with non-rotating shield.
Cylinder with flat or hemispherical end-caps forradiation shielding.
4
Construction at L-5
Shell construction at L-2 followed by slow moveto L-5
5
Lunar H2-powered mass driver.Baseball-size loads.  30g; 10km run [2,30]
Lunar-equatorial Solar-power fields, 20 launchers; round-the clock launches; fuel is lunar-generated electricity. Railcar-sized loads. 8-g, 40km track.
6
Entire interior pressurized.
10 to 30 meters at rim pressurized, using membranewith 30-meter bubbles to provide micro-climates. 
7
Machinery required to make panels  etc. 
Non-contact shape formation with solar-heated powder sintering & furnaces. Robotic assembly of payloads. 

 
 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

 

5.3 Choice of Construction Location

 

The Earth-Moon L-2 is chosen over L-5 as the construction location to minimize the lunar launch energy. The final move from L-2 to L-5 (if needed) could be done by gradual orbit transfer over a period of months, with solar-heated gas thrusters providing the energy. The choice between Earth-Moon L-2 and L-1 is somewhat arbitrary strong arguments could be made for locating the station at L-1. Either location is convenient for telepresence operation from earth satellites in GEO or at L-4/L-5 can serve as convenient observation platforms. Launches to L-2 would usually occur from the visible side of the Moon, which will be more convenient for initial operations. The resource extraction facilities on the Moon, including the lunar solar-power fields and power-beaming plants, are more likely to be located on the visible side, so that the first launchers will be built on this side as well. Hence the choice of L-2.

These issues have been studied in the past in detail. Heppenheimer and co-workers [31-33] evaluated over 48,000 test trajectories to obtain all achromatic trajectories from the moon to any of the earth-moon libration points L-1, L-2, L-4, L-5, as candidate sites for a mass-catcher to receive material launched from the moon. They found ten such achromatics, and gave their characteristics and a photographic atlas of their launch sites. The best transport mode found was to launch from Mare Tranquillitatis to L-2, with the mass-catcher maneuvering near L-2; acceleration of the mass-catcher due to momentum transfer from the mass-stream was considered. Three propulsion modes were considered: ion-electric, Rotary Pellet Launcher, and Advanced Space Engine. A reference catcher design was proposed. A critical launch longitude was shown along the lunar equator (33.1 deg E) for a certain class of trajectories to L-2, to minimize the dispersion such that a miss distance of lunar materials at L-2 of 50 m would result from launch velocity errors of approximately 10 cm/sec along the track, 1 mm/sec vertically, and 1 cm/sec laterally [31]. The acceleration requirement for the mass catcher to follow lunar librations was 1.5 E-4 m/s2, plus modest additional station keeping with respect to the local force field. In the O'Neill system, the manufacturing facility was located within a DV of 10 to 30 m/s from L-2.

Optimal launchers for the O'Neill system had the constraint that they were powered by gas or nuclear energy, either option requiring large earth-shipped mass to operate. Their model imagined most of the regolith processing to occur inside the manufacturing facility part of the habitat so there was little incentive to develop other infrastructure on the moon. These considerations drove their optimum solution to be one where a stream of mass, sized to approximate baseballs, would be accelerated at 30-G's by a 10km accelerator track [33]. It made sense there to build a complex mass-catcher and transport the mass to the manufacturing facility.

In our model, we make use of other people's motivations to build infrastructure on the moon, and design our habitat project to enable all such projects to become realities. Thus in our system, there can be multiple launchers, distributed around the moon's equator, and accordingly, multiple solar-power fields, power plants, mines and metal processing sites. Thus our launchers are part of a permanent lunar export infrastructure, with dual-use as the nucleus of a lunar surface transport system. We also need metal box beam structures shipped to the construction site, to form the outer shell of the cylinder. These metal structures, accordingly are built as containers regolith launches greatly reduced, it makes sense to have a few Shepherd spacecraft performing the triple functions of (a) mass-catcher, (b) transporter to the construction site, and (c) maneuvering the loads into position as the assembly of the structure. Interlocking appendages on the containers enable much of the structural loads to be carried before the boxes are welded together by robot arms.

 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

5.4  Construction Sequence

Cable Grid Deployment and Construction Spider.

The first 4 boxcars will be launched in quick succession, empty except for eight cables - four for the first set of cables, and the rest for the second set. A hub beam structure and a "construction spider" equipped with an autonomous power supply, an electromagnetic wire grid, four robot legs with grapplers, and a robot welding arm, will be brought from Earth and positioned by the "shepherd" craft. The first four rings of 12.5mm dia.cable segments, 1km in radius, spaced 4 meters apart, will be connected by longitudinal cables. Thrusters will start the cable rings in rotation. The tension is kept low until first is complete. The dynamics of cable grid deployment in orbit require further analysis.

Figure 5.1 Deployment of initial cable loops for the cylinder construction project Regolith Transport and Positioning

  
 
 


 
 
 Figure 5.2 Boxcar launched off the lunar surface in an orbit with apogee at L-2
 
 
 
 
 

Figure5.3 Rendezvous with Shepherd craft
 
 
 
 
 
 
 

Lunar regolith 2 m. deep is brought up in iron/steel railroad boxcars. Each boxcar is met by a Shepherd craft. Each regolith-filled boxcar is brought by a hybrid gas/e-mag shepherd craft, and guided towards the grid. A winched tether attached to the rotating grid captures each arriving load-train. Axial momentum is transferred to radial and tangential momentum, bringing the load to the periphery at 1kmph, into the space between the outer grid and an active, powered electromagnetic Spider construction grid. Electromagnetic interaction between the loads, the construction grid, and the shepherds, moves the loads into position against the outer grid. The shepherds leave the grid. Robots attached to the construction grid complete the attachment of the boxcars. Figure 5.5 shows a load arriving, attached to an electromagnetic Shepherd, aimed by the tether into the space between the Spider and cables. The currents in the grid held by the Spider interact with electromagnets on the Shepherd, making the final adjustments and maneuver the boxcar into place.

Figure 5.4 Load is captured by a tether at the entrance to the cylinder grid, transferring axial momentum to tangential and radial momentum

 

The Shepherd detaches from the boxcar and pushes off from the cylinder site, to return to lunar orbit for the next rendezvous. The Spider is a robot, which can attach its legs either to the cables or to already-positioned boxcars. A set of welding arms on the Spider fixes the boxcars to each other. Note from Fig. 5.6 that the boxcars are built with interlocking wall geometry, so that the tensile load on the cylinder rim is transmitted.

 

Figure 5.5. Regolith-laden boxcars being delivered by Shepherd. Shepherds maneuver boxcars into place using e-mag fields.

 

 
 

The ends of the cylinder are sealed in any of various ways. Once the shell structure is completed, the radiation shielding for the ends can be accomplished using flexible bags containing water and/or hydrogen or more boxcars containing regolith. However, in our costing in Chapter 6, we assumed that the same regolith/ boxcar system was used to seal up the entire side faces, with radial cables for initial support. There is no need to pressurize and shield regions other than those to be inhabited or cultivated. To create atmospheres, inflatable balloon structures are adequate. Sunlight will be directed in through the end walls, with appropriate filters. Figure 5.7 is an artist conception of the completed habitat, with directional solar arrays shown attached.
 

Figure 5.6 Assembled boxcars


 
 
 


 
 
 

Figure 5.7 Artist's conception of the completed cylindrical habitat.
 
 
 
 
 
 
 
 
 
 
 
 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

 
 
 
 

5.5 Summary of Construction Parameters

Table 5.2: Construction Parameters of 1km cylinder radiation shield:
 
 
· Radius= 1km; 0.945 rpm  for 1g
· Length= 2km
·Shield  Depth 2m

·Grid  current = 35 amps

· Loops of cable; Wire dia =12.5mm 

·Solar  Panel area  to power grid = 350 m 2

·Boxcar  dimensions: 2m x2m x 20m
·Regolith  Mass/ load: 160,000 kg. 
·10  launchers operational at any time (20 total around lunar equator)


·
Shepherd  unit current: 5 amps

·Time  to build:  10 yrs.


 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

 

5.6  Summary of Construction Sequence


 
Lunar Solar-Power Fields made by robotic rovers around the equator

Lunar metal extraction; cable manufacture using robotic plants

Lunar launcher construction

First cable-set deployment and spin-up

First ring of loads forms framework for subsequent cables and loads

Solar collectors, thrusters; hub with tethers and Construction Spiders

Oxygen/propellant extraction from regolith for thrusters

Cylinder completion; end caps sealed with regolith and water bags

O2/N2 bubbles for habitation near rim; micro-g axial facilities

Human habitation commences  

 
 
Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References

 

5.7 Issues requiring further detailing

Shepherd craft - propulsion

Current concepts (e.g. Figure 5.8) for Orbit Transfer Vehicles visualize solar-heated gas-powered vehicles, which will perform transfer missions between Low Earth Orbit and Geosynchronous Earth Orbit. This is appropriate for the shepherds; however, it is not possible to have a large solar collector attached to such craft because the shape must enable it to move in the narrow space between the Spider grid and the cables. Focused sunlight beamed from the Moon or the cylinders constitute one option for heating the gas. The gas supply must be replenished either from supplies on the cylinder, or (more likely) from a pressurized cylinder carried with some of the boxcar loads. This will leave a large number of such cylinders to be disposed at the cylinder or used for storage of gas extracted from the regolith at the cylinder. One option is to use the robot arms on the Spider to refill these cylinders from ISRU units, and attach them to the thrusters needed to replenish the angular momentum of the cylinder.

Figure 5.8 STUS concept. Credits: NASA Marshall Research Center

Shepherd craft electromagnetic force.

 

A set of 3 electromagnets arranged in a delta on each end of the shepherd can serve to provide enough maneuvering forces and moments during interaction with the field due to the grid held by the Spider. Super conducting magnets may be an option for this use, with developments in the technology. Power for these magnets could come from beamed power from the cylinder, charging storage units on the shepherd. Coincidentally, when electromagnetic Shepherds were first being proposed to NIAC (Feb.2001) for this application, LaPointe [34] was also proposing Shepherd electromagnetic craft for formation flying.

Optimal orbits and launch sequences.

As discussed above, there has been considerable work done on determining optimal trajectories and launcher locations, but with the O'Neill system architecture in mind. These should be adapted to an architecture, which considers solar-power fields and metal mining/extraction sites on the Moon. Other issues related to this architecture are discussed in the next section where the cost of this project is considered.

 


Abstract 
Intro
Theory
Near-term: Acoustic
Mid-Term: L2 Habitat
Space Economy 
Far-Term: Radio-Wave Construction
Comments
Issues
Conclusions
Acknowledgements
References