Imagine a small spacecraft carrying an inspection party has landed in the middle of the vastness of a powersat. From the observation ports, a few meters above the flat, level expanse of silicon cells, the powersat extends in all directions to the horizon, like a featureless, trackless stretch of desert.

Now the spacecraft begins to rise, to move away from the powersat like a grasshopper flying up from a beach. (The relative dimensions are rather similar.) For quite a while, there is little new or different to see. The world outside continues to be sharply bisected by what to all appearances is a plane extending to infinity. A while longer, a further elevation, and the plane is seen to have edges. It does not fill the universe, but its extent still cannot be grasped.

Only much later, when the distance has increased to tens of miles, can the true dimensions be seen. It is a flat rectangular slab, with circular disks at each end—and the whole of it is larger than Manhattan Island. It spans the area of 20,000 football fields. It has the weight of an aircraft carrier, yet could enclose the flight decks of 6,000 such carriers.

Is this the work of a super-civilization of the year 3000? The artifact of extraterrestrial visitors, commanding sciences and technologies of which we can only dream? No, it probably is the work of people who are alive today, and the year could be as early as 1995. And if a Manhattan-size spacecraft is hard to conceive, then consider an everyday item familiar to all, the production of which far outstrips the dimensions of powersats: the daily newspaper.

The Los Angeles Times, which I read every day, illustrates this point most concretely. The masthead carries the legend: “Circulation, 1,034,329; 152 pages.” That is, between Monday and Thursday of each week, the Times’ printing plants turn out enough newsprint, printed neatly on both sides, to more than cover the total 114 square kilometers of a complete power satellite.

The proposals of astronautics may boggle the mind; but its history shows that time and again, even the most startling concepts have swiftly become commonplace, even prosaic. The Boeing 747 is a case in point. Another is the communications satellite, which only fifteen years ago was grist for the science fiction of Arthur C. Clarke. Twenty years ago an artist’s rendering showing the space shuttle maneuvering near the space telescope could only have been wild fantasy; today such a picture is nothing more than an advertisement for Lockheed Missiles and Space Company. So, too, will it be with the powersat. Vast though it is, it lends itself to rapid building by construction techniques that will in time appear so straightforward as to be, indeed, prosaic. Sooner than we think, a new powersat may be no more remarkable than a new day’s edition of the Times.

The initial goal of work on large space structures will of course be nothing so elaborate as a power satellite; it will rather be the building of large communications platforms of the type discussed in Chapter 3. But the people who are studying such platforms are well aware how the techniques needed to build them would serve as well on a larger scale to build powersats. These techniques, even in their simplest applications, are quite novel and represent a considerable advance over what has been done to date.

It is no new thing to build a satellite larger in dimensions than the payload compartment that
carries it to orbit. There has long been ample need for creativity in designing spacecraft that unfold solar panels, deploy large antennas, or extend long struts once in orbit. In these areas examples of ingenious concepts abound. The FRUSA (Flexible Roll-Up Solar Array), for one, packages a spacecraft’s power supply into something very much like a roll of linoleum. The Flex-Rib Reflector, used on the ATS-6 satellite, is a parabolic antenna whose form is shaped by forty-eight flexible ribs, such as those of an umbrella. The ribs are furled or wound round a central hub to form a lightweight, compact disc. Once in orbit, the ribs release and unfurl, standing out from the hub to form an accurately shaped antenna thirty feet in diameter. The Astromast is a long beam, which extends out to thirty times its packaged length.

In the era of the space shuttle, these approaches will appear as expedients, as ways of getting around the small payload bays and lack of on-orbit human assemblers. But within a few years, it will become routine to have work crews fashioning spacecraft structures on-orbit, working with a very basic and familiar structural element: the beam.

Even in the early 1970s the earliest studies of space structures showed that they would be built on-orbit with frameworks of beams. At first it was believed that the beams would be fabricated on the ground and carried to orbit by the shuttle. But such beams then would necessarily be limited to the sixty-foot length of the shuttle’s cargo bay and in addition, would have to be narrow and compact, somewhat like aluminum pipes to take advantage of the shuttle’s 65,000-pound payload capacity.

What the designers really wanted, though, would be beams of any length, up to hundreds of meters if need be. But the beams could not be narrow or they would bend easily. Instead, they had to be of a light, open design. Such beams could be produced by work crews in orbit, but an early Grumman Aerospace Corporation study of powersat construction projected an orbiting crew of 258, just to fabricate the beams. With this, it became clear that the beams would in no way be built on Earth. Instead, a new and daring approach was in order: What the shuttle (or its advanced successors) would carry to orbit would be rolls of aluminum, structural parts— and a new type of machine known as a beam-builder. This machine, operating in orbit, would then form and shape the aluminum into beams of the proper type, in what would amount to an orbiting construction plant.

So it was that NASA’s Marshall Space Flight Center awarded a $635,000 contract to Grumman to build a “Space Fabrication Demonstration System”; that is, a beam-builder. The first such device was completed and delivered to Marshall in 1978. On May 4, 1978, it produced its first beam in ground test.

The beams it fabricates are both lightweight and strong. A one-hundred-foot length weighs only 85 pounds, yet will support a load of 1,260 pounds. The beams are triangular in cross section and a meter deep. (The depth of a triangular beam is the distance from one corner to the opposite side.) They are made up of long strips of angle aluminum supported by cross-braces. The long aluminum edge members are formed from rolls of sheet aluminum; the machine pulls out aluminum strips from the rolls and forms them into the proper angled shape. The cross-braces are made beforehand and packaged in magazines, which fit to the side of the beam builder. They are withdrawn automatically, somewhat like giant staples, and the machine automatically welds them to the edge members. With one supply of rolls of sheet aluminum and of full magazines of cross-braces, the machine can turn out a thousand feet of beam in as little as two hours.

The beam-builder will not long remain on Earth. As early as 1983 it may be modified for flight, placed in the shuttle’s cargo bay, and put through its paces at an altitude of several hundred miles. This shuttle-beam builder system would do more than merely fabricate beams in orbit, though. The shuttle carries, as standard equipment, a long manipulator arm that is controlled by a crew member, the payload handling specialist, who works at the rear of the shuttle flight deck. This manipulator arm can act as a crane, wielding beams and partially assembled structures. A proposed 1984 flight test may demonstrate the full power of this system, known as the Space Construction Automated Fabrication Experiment (SCAFE), which has been studied by General Dynamics under contract to NASA’s
Johnson Space Center.

In SCAFE, the beam-builder and its raw materials would be carried to orbit along with a jig or assembly frame mounted to the shuttle. Once in orbit, the manipulator arm would serve to deploy the stowed equipment. Then the beam-builder, moving to successive positions along the jig, would automatically make four triangular beams, each 200 meters long and held parallel to each other by the jig. The beam-builder would then move to the outboard end of the jig and fabricate the first of nine short cross-beams, 10.6 meters long, spanning the four long beams. To attach these cross-beams at the proper sites, like rungs along a ladder, the jig would shift the partially completed beam assembly along its length, stopping every 25 meters to permit the beam-builder to fabricate another cross-beam.
Jig-mounted welders would join beam to beam where they cross. In this fashion a complete platform would be built, a grid of four long beams braced by nine cross beams.

A number of important experiments will then be possible. With the platform still firmly attached to the shuttle, the shuttle can fire the maneuvering engines and accelerate. This procedure will place a load on the structure, tend to bend it, and test its strength, particularly the strength of the welds made by the beam-builder. When the shuttle passes in and out of Earth’s shadow, the structure will be subjected to sudden outside temperature changes and their effects can be measured. There can be tests with the manipulator arm, using it to release and recapture the platform, retrieving it as it floats freely in space. In addition, there can be tests of the performance of astronauts, who can attempt to accomplish such matters as attaching instruments and electronic equipment to the platform.

The shuttle, however, was never designed as a space construction center. Even with the aid of a beam-builder and of its manipulator arm, it can construct only modest structures. The shuttle has only limited power (seven kilowatts) from onboard sources; the manipulator arm cannot reach very far. To extend the capabilities of the shuttle and to permit more ambitious tests and construction work there will be need for a space construction platform. It will be a large orbiting facility operated from a shuttle docked to one side of it. It will serve to greatly increase the ease of building truly large structures.

As proposed by Tom Hagler and his associates in a study at Grumman in 1976, the platform will feature a 360-foot-long boom, which can rotate to any angle and will be outfitted with tracks along. which manipulator arms and equipment carriers can travel. The boom will also carry a small cab resembling the cab of the familiar cherry-picker used by telephone linemen, with room for one or more crew members. The boom will serve in building the platform, which will be 236 feet long by 105 feet wide, in orbit 210 miles up. A large opening in the platform structure, 105 feet long by 79 feet wide, will serve for developing and demonstrating procedures for mounting solar arrays, thin-film mirror surfaces, wire mesh, and other components that will span wide areas of finished large satellites. To provide power, thirteen solar-cell arrays of the type used in the Solar Electric Propulsion Stage (SEPS) project will produce 250 kilowatts. Total platform weight will be 42 tons.

Three flights of the shuttle will serve to construct the platform. The first flight will deploy as a single unit a core module, which includes the control equipment. To this core module will be added a 124 short section of the rotating boom, as well as one section of the solar array. The second flight will construct the inner area of the platform (105 feet square), the remainder of the long boom, and the rest of the solar array. The third flight will complete the platform structure and install the power distribution system, which will provide electricity not only for construction work, but for the shuttle itself. By relieving loads on the shuttle’s own power system, the platform will greatly increase the time that a crew can stay on-orbit.

The resulting platform will provide a powerful and flexible means for assembling large communications satellites and other space systems. The first such system, to be constructed using the platform, may be a 330-foot-diameter radiometer, or microwave antenna, used to measure the natural radio waves generated by Earth’s surface. It will detect the content of moisture in soil as an aid to crop forecasting. To build it, crew members working from the cab on the platform boom will install
structural members as well as the reflector mesh surface. A crew of seven could build the radiometer in eight weeks.

Another early candidate for assembly at the platform would be a multibeam communications antenna of two-hundred foot diameter. It would provide 256 fixed beams aimed at specific ground stations, and 16 movable or scanning beams. The antenna would be built from 226 hexagonal elements, fitted together like floor tiles. Again, crews working from the boom cab would assemble these elements.

The most exciting uses of the platform, of course, would involve experiments on the construction of power satellites. Like the other space structures, powersats would be built of beams. But whereas ordinary beams would be one meter deep, fabricated with standard beam-builders, the beams for powersats would be twenty meters deep. Rather than being built with edge members and cross-braces cut and formed from sheet metal, the twenty-meter beams would be formed from assembled sections of one-meter beam. They would thus resemble the massive structural members of great suspension bridges that are built from lengths of standard steel beams, which alone suffice to build an ordinary highway bridge.

To build sections of such great beams, the platform would mount a construction frame carrying six standard one-meter beam-builders. The first test specimen would be 246 meters long and, when completed, would be suspended beneath the platform, then shaken for vibration tests.

The platform will support other tests of critical questions in powersat construction. A key problem area involves installation of electrical conductors. In a full-size powersat such conductors will stretch for miles, carrying high voltage and current. It will be possible to test means for fabricating and installing such power conductors simultaneously with the construction of their supporting twenty-meter beams.

Another problem area involves installing large panels or blankets of solar cells. A promising approach is to produce such solar panels as roll-up arrays, attached at their edges to a crossbeam of a twenty-meter main beam. As the main beam extends in length during construction, the solar array unrolls and can be fastened at the sides. Still another area for study involves formation of tightly focused microwave beams for power transmission to Earth. To check this out, the platform’s boom can mount an array of klystrons or microwave generators. A small satellite will operate with a beam-mapping system at a distance of several tens of miles. As the boom swings from side to side, the satellite will study the characteristics of the transmitted microwave beam.

Moreover, while beams (both of microwaves and of aluminum) are the key to construction of power satellites, there are other ways to build large space structures. When the structure has a simple shape, as many will, there is the opportunity to use the Space Spider, invented by J. D. Johnston of NASA-Marshall. This novel construction technique lays a braced aluminum structure built from fifteen-foot-diameter rolls of aluminum, around the path created on its last pass. The simplest such structures are spiral-shaped and have the pattern of a lawn mower cutting a spiral path to cover an entire back yard. The Space Spider can advance at twenty feet per minute, thus constructing an eight-hundred-foot diameter structure in only seventy-two hours.

In the more distant future, the Space Spider may be the best way to build a true space colony. The best-studied colony design, the Stanford torus, features a living area shaped like a bicycle inner tube, 400 feet wide by a mile in diameter, rotating within a “tire” 6 feet thick and filled with lunar sand, as a radiation shield. Six spokes, 50 feet in diameter, lead from the colony rim to the hub: a 425-foot-diameter sphere. All these structural shapes must be strong and airtight, like the hull of an airliner; and none lend themselves in a simple way to being built from light, open latticeworks of beams. The Space Spider may fabricate their forms directly, in an elegant design resembling the
preformed fiberglass hulls of catamarans.

Such projects may be far away, but beam-builders and the shuttle-supported construction platform lie just around the corner (or the next presidential election, if you will). They will take us to the very verge of being able to build power satellites. They will not take us all the way, for the shuttle will never serve the immense logistical needs of a true powersat. But once these systems have been built and proven, once their experiments have succeeded and the world has grown accustomed to the communications platforms and other large structures they will build—then the needed confidence will be in hand for the next step.

That next step will be to go forward and build the vast orbiting station needed for building powersats. As long as space construction is constrained by the limitations of the shuttle, it will have the character of a machine shop in someone’s garage: Its equipment will call for only a few people, for only occasional use. In preparing for the powersat, the need will be for a space factory. It will have many items of equipment, at diverse work stations, most of which will see nearly constant use. The number of people in space will leap from a few to several hundred; their stay-times will go from weeks to months. For all this, the jobs and activities in this factory will first be tested in the shuttle era, in the orbiting construction base.