The Space Elevator: ‘Thought Experiment’, or Key to the Universe?
Editor’s Note: The following paper was first published in 1981 in Advances in Earth
Oriented Applied Space Technologies. It is reprinted here with the authors permission.
Visit the new Space Elevator Reference
PART 3
by Sir Arthur C. Clarke
THE MASS ANCHOR
In order to balance the weight of the lower portion, and to compensate for the reaction produced by ascending payloads,
the space elevator has to extend far beyond geostationary orbit. The upper portion may be regarded as the mass which keeps
the cosmic sling taut, as it whirls round the Earth every 24 h.
This mass could be provided by another tapered cable, extending out into space, but it has to be very much longer than
the lower portion to produce equilibrium. Indeed, calculations show that it must reach the enormous height of 144,000 km.
I do not think that a cosmic flail extending a third of the way to the moon will make the Earth a nice place to visit.
The alternative is to have a large mass anchored at a much lower altitude, not far above the geostationary orbit. The
closer it is, the larger the mass required; it might be many megatons, or even gigatons. Both Sheffield and I have
suggested that captured asteroids could be used for this purpose, and as many of them now appear to be largely
carbonaceous they could also supply much of the material of the elevator, the remaining debris providing the anchor.
CATASTROPHES
A structure extending right through the atmosphere and on into space for at least 50000 km would be a considerable
navigational hazard, both to aircraft and spacecraft. Very elaborate anti-collision measures would have to be taken and
all air traffic would have to be diverted from the equatorial danger zone. Probably the structure would be strong enough
to survive impacts at atmospheric velocities; cosmic speeds would be another matter.
The problem here is aggravated by the fact that, over a long enough period of time, all satellites with perigees below
geostationary altitude would eventually collide with the space elevator, as their orbits precess around the earth. So
before the elevator is built, there would have to be a thorough job of garbage collection, and thereafter all remaining
satellites would have to be closely watched. Whenever they approached too near the elevator, they would have to be nudged
into a safer orbit. The impulses required would be trivial, and need be applied only very infrequently.
Meteorites present a more difficult problem, since they would not be predictable. But the impact of a large one would be a
very rare occurrence indeed, and the elevator would have to be designed with enough redundancy to withstand any reasonable
danger. Thus if it was in the form of an open framework — like a boxgirder — a meteorite should be able to pass through
it in any direction without causing a structural failure.
But what if the elevator is severed?
Well, if the elevator is cut through at the Earth’s surface, it would de exactly
the opposite of a terrestrial building. It wouldn’t fall down — but would rise up into the sky! In theory, the loose end
might be secured and fastened down again; but that would be, to say the least, a tricky operation. It might even be easier
to build a new system….
If the break occurred at any altitude up to about 25000 km, the lower portion of the elevator would descend to Earth and
drape itself along the equator while the now unbalanced upper portion would rise to a higher orbit.
Hopefully, such major catastrophes can be avoided by good design; after all, it is very rare indeed for a modern bridge
to collapse. (Though it has happened!) Much more likely — indeed, inevitable — is that objects would accidentally fall
off the elevator. Their subsequent fate would depend upon their initial altitude.
The situation here is totally different from that encountered in orbital flight. If you step outside a Spacecraft, you
stay with it. But if you step off the elevator’ it’s rather like jumping out the window on the thousandth — or ten
thousandth — floor of a rather tall skyscraper. Even so, you might still be quite safe because you wouldn’t fall
vertically. You would share the structure’s horizontal velocity as it whirls round with the spinning Earth; in other
words, you would be injected into an elliptical orbit.
If your initial height is less than 23000 km, too bad. Your orbit will intersect the atmosphere in a few hours — or even
minutes – – and you’ll burn up on the other side of the planet. Above this critical altitude, you would be in a stable
orbit, skimming the atmosphere and coming back, after one revolution, to the place you started from. Of course, by then
the elevator would be somewhere else, but with luck your friends might be waiting for you with a net and some well-chosen
words of advice. Or if not on this revolution, on a subsequent one….
If you stepped off at the geostationary altitude itself, here, and only here, you would remain with the elevator, just as
in conventional orbital flight. At higher altitudes, you would be injected into orbits of increasing eccentricity, with
periods of one day and upwards.
That is, until you reached another critical altitude — 47000 km. At this point, you’d be slung off into space at more
than the local escape velocity, and would never return. You would become an independent planet of the Sun, and it might
not be possible, owing to budgetary considerations, to rescue you and bring you back to Earth.
The analogy with a sling is now complete. Payloads released anywhere above the 47000 km altitude would escape from the
Earth’s gravitational field, and by going to greater and greater altitudes any desired launch speed could be attained.
Pearson [9] has shown that all the planets can be reached by this technique, without the use of any other propulsion. The
energy comes, of course, from the rotation of the Earth.
BEYOND THE EARTH
The lower the gravitational field of a planet, and the quicker its speed of rotation, the easier it is to build a space
elevator. On a small asteroid the feat would be absurdly simple, and could even be achieved by a free-standing tower.
There would be no need for suspended cables made of exotic materials.
Pearson [9] has pointed out the advantages of lunar Space elevators — in this case, linking the Moon’s equator with the
well-known Lagrangian points in the line joining Earth and Moon. He calls them ‘anchored lunar satellites’, and they could
be constructed of materials already available. Working in conjunction with the earth-based elevator, they would permit
two-way traffic between Earth and Moon with almost zero use of rocket propellants.
The planet which seems ideally suited for the space elevator is Mars, with only one third of Earth’s gravity. What is
more, the outer satellite Deimos is only slightly above stationary orbit — in just the right position to provide a mass
anchor! Moreover, it appears to be largely carbonaceous, so could supply the required construction material.
But there is one big problem — about ten million million tons — in connection with the Mars elevator, and that’s the
inner moon, Phobos. Moving almost exactly in the equatorial plane, it would slice through the elevator at very frequent
intervals. Phobos is much too big to tow away, and blowing it up would only make matters worse. I refer you to
The Fountains of Paradise for one solution….
DYNAMIC SYSTEMS
A daring extension of the space elevator principle has been put forward by Hans Moravec of Stanford University [17]. He
imagines a ‘skyhook’ which is at a very low altitude, and is therefore not stationary with respect to the earth, but
orbiting around it.*
* Once again, Artsutanov got there first! (See Acknowledgements).
Consider a very elongated satellite in a two hour orbit, rotating like a propeller blade (remember them?) as it rolls
around the equator. The blades are just long enough to touch the earth, and if everything is properly synchronized, the
tips would always touch the same spots on the equator at regular intervals.
From the point of view of the earth it would be, as Dr. Robert Forward has put it, ‘a Jacob’s Ladder coming down out of
the sky, pausing for a moment, then lifting off again at 1.4 g‘. One could grab hold of the end, and get a free lift into
space — and of course come back the same way.
It’s a delightful concept, but the presence of the atmosphere, not to mention the fact that the equator isn’t a perfect
circle, and a few other practical details, make it rather unlikely. However, something similar may be possible in space,
because very large rotation systems might serve as ‘velocity banks’ an idea discussed by Pearson and Sheffield [9,15].
If you could hook on to the edge of a spinning disc — or an asteroid with a long extension from its equator — you could
let go again at the appropriate point and so obtain a major velocity change without using any propellant. However, we
would need such an enormous number of these ‘cosmic carousels’ scattered round the solar system that the idea is not
really practical, except perhaps for very special applications.
I’d like to conclude this section on ‘dynamic systems’ by mentioning, even more briefly, an idea that has just emerged
from Japan [19] — the ‘Space Escalator‘.
Imagine two satellites in circular orbits above the equator, one a few hundred kilometres above the other. Each carries a
launching mechanism and a catching mechanism, which could be something as simple as a hook and elastic cord. By means of
this mechanism, payloads could be transferred in either direction without the use of propellant. With a whole series of
satellites — about a hundred — you could hop, or leap-frog, all the way up to the stationary orbit. But it would be a
computational and operational nightmare, keeping track of all the constantly changing orbits, and launching and catching
payloads at the right time. I think I’ll stick to the elevator, rather than take the escalator.
POWER AND PROPULSION
The physics laboratories of British schools once boasted — and probably still do — an instructive device known as
Atwood’s Machine. I don’t know what it’s called elsewhere, and in any case Galileo was first with something very similar.
It’s an almost frictionless pulley over which runs a light cord, with equal weights suspended on either side. In this
state nothing happens, of course, but even a small additional weight on one side sets the system in motion, at a very low
acceleration.
This device may be regarded as the mechanical analogue of the Space Elevator. I don’t suggest for a moment that we would
actually use moving cables to lower and raise our payloads, but it demonstrates the basic principles involved. Such a
system is inherently conservative — if it’s properly balanced, it requires no energy to run it, except the very small
amount lost by friction. In principle, arbitrarly large masses can be raised or lowered through any distance. Unlike the
rocket, which wastes precisely 100% of its available energy on a round trip, Atwood’s machine wastes only a few percent.
And it’s a lot quieter.
In practice, the space elevator would almost certainly be electrically powered, and the energy generated by the returning
payloads during the braking and descent would be pumped back into the system — as happens with electric railroads in
mountainous country. But there is also another reason why electric propulsion would be mandatory.
Though inanimate payloads might be in no hurry to reach the geostationary orbit, 36000 km up, human passengers are easily
bored and have to be fed, entertained or at least tranquillised by alcohol and inflight movies. By the time the space
elevator is likely to be operating, no journey on Earth will last more than a couple of hours. I don’t think that the
average space tourist will tolerate a great deal of time in what will be little more than a glorified elevator cage,
though one with a magnificient view.
So we will require operating speeds of several thousand kilometres an hour, which can be provided only by some kind of
electric propulsion system with no mechanical contact — a linear motor, for example.
I am not competent to discuss the problems involved in switching huge amounts of electric power over distances a hundred
times greater than those encountered in terrestrial systems. Presumably superconductors will be available by the time the
elevator is built, but the weight penalty of the associated cooling systems may make them quite impracticable. It would be
marvellous, of course, if our superstrength material was also a superconductor — and at room temperature (or higher)! But
to expect not merely one but three miracles simultaneously is a little greedy.
Perhaps we can avoid enormously long transmission lines by using microwave or laser beams to get the power where it is
wanted. And if it ever proves possible to build small nuclear generators, then perhaps we can hang the power stations at
strategic points along the elevator.
However, this suggests an even more attractive possibility. There is no theoretical reason why small fusion — or even
fission — generators cannot be built. If they prove to be practicable, then we could forget electrical transmission
systems altogether and put the power plants in the vehicles. This would not be a retrograde step, because the weight of
the ‘fuel’ would be essentially zero.
The space elevator could even make possible a far more efficient chemically fuelled transport system. In this case, the
Earth-orbit structure would merely provide physical support — the railbed, as it were, for the equivalent of a
self-contained diesel locomotive, not a centrally-powered electric one. Unlike a rocket, the space-train would not have to
use much of its fuel merely to maintain altitude; it could do that simply by putting on the brakes. On the other hand, it
would be at a disadvantage over the rocket as it would have to lift some of its propellants all the way to the stationary
orbit. I have not calculated at what particular specific impulse the chemical elevator will be more efficient than the
chemical rocket.
SUBSIDIARY PROBLEMS
As is well known, satellites in the geostationary orbit will not normally stay above the same point on the equator, but
drift in longitude owing to the fact that the Earth’s gravitational field is not symmetrical. However, there are two
points of maximum stability — one in the Pacific over the Galapagos, and the other above the Maldive Islands, seven
hundred kilometres to the southwest of Sri Lanka. The latter point is the more stable; by an odd coincidence, it is
directly above the small island of Gan, which in the 1960s was one of the staging posts for the Blue Streak rocket when
it was being ferried from the United Kingdom to the Woomera launching site. If orbital stability is important, Gan —
abandoned by the Royal Air Force several years ago, to the great distress of its inhabitants, though not of the central
Maldivian government — may one day be the most important piece of real estate on Earth.
Other orbital perturbations — including ones in the north-south direction — are caused by the Sun and Moon. Probably all
of these are only important to free satellites, and will be insignificant in a structure which is tethered to the ground.
In any case, the upper section of the elevator could — and probably would — sway through an arc many thousands of
kilometres across without causing operational problems.
The effect of hurricanes on the lower portion of the structure has worried some writers; although high winds are rare on
the equator itself, they can occur, and if they did nothing else they would generate severe torsional vibrations which our
revered colleague Dr. von Karman studied in connection with the ill-fated Tacoma Narrows Bridge. So it might be worthwhile
siting the structure on a very high mountain to reduce aerodynamic loads; unfortunately, there aren’t any high mountains
near the stable points.
A RING AROUND THE WORLD
There are now scores of satellites in the geostationary orbit, and the problem of collision and interference — which not
long ago would have seemed an absurd fantasy — is already of practical importance. What is more, some equatorial
countries are attempting to establish jurisdiction over this large but still restricted narrow ring around our planet.
This has provoked the appalling pun, which perhaps fortunately cannot be translated from English, that there should be
another U.N. Committee — on the Useful Pieces of Outer Space.
In 1977, while working on the final chapters of The Foundains of Paradise, I had one of those sudden glimpsess of the
perfectly obvious out of which I have cunningly fashioned my reputation as a prophet. One way of preventing geostationary
satellites colliding or drifting around the equator would be to link them together with cables. As the forces involved
would be extremely small, for the most purposes nothing much stronger than a nylon fishing line would be adequate, and the
total mass needed to tie together all the satellites in the stationary orbit would be negligible.
But why stop there? The next step would be to build a continuous, habitable structure — a ‘Ring City’ — right around the
Earth. All the legions of geostationary satellites could be attached to it, and reached for servicing by an internal
circular railroad. And it could serve as a launch platform for almost all missions, manned or unmanned, into
deep space.
It would be reached, of course, by space elevators, which would take the form of several spokes linking the ring city
with the equator. The Earth would, in fact, now be the hub of a gigantic wheel, 85000 km in diameter. Passengers could
move up and down the spokes, or around the rim, just as freely as they now move around the surface of the Earth. The
distinction between Earth and space would be abolished, though the advantages of either could still be retained.
A Russian engineer, G. Polyakov had the same idea almost simultaneously, and published a paper with the title ‘A space
necklace about the Earth'[20]. However, as I might have guessed, we were both anticipated by Professor Buckminster Fuller.
To quote from the notes he wrote for the sleeve of my Fountains of Paradise recording (Caedmon TC 1606):
‘In 1951, I designed a free floating tensegrity ring-bridge to be installed way out from and around the Earth’s equator.
Within this halo bridge, the Earth could continue its spinning while the circular bridge would revolve at its own rate. I
foresaw Earthian traffic vertically ascending to the bridge, revolving and descending at preferred Earth loci.’
All that Bucky’s vision needs to make it reality is the space elevator.
And when will we have that? I wouldn’t like to hazard a guess, so I’ll adapt the reply that Arthur Kantrowitz gave, when
someone asked a similar question about his laser propulsion system.
The Space Elevator will be built about 50 years after everyone stops laughing.
Acknowledgements — My first thanks must go to the late A. V Cleaver, F.B.I.S., F.R.Ae.S., with whom I discussed
the subject of this paper for several years. It is a great sorrow to many, besides myself, that he never lived to see
the final outcome of our deliberations.
I would also like to thank Professor Harry O. Ruppe, Dr. Charles Sheffield, Dr. Robert Forward, Dr. Alan Bond, Frederick
C. Durant, and Jerome Pearson for much material and helpful correspondence.
Finally, I am especially grateful to Mr. Vladimir Lvovfor giving me biographical material on Yuri Artsutanov (Born 1929,
Leningrad) and for putting me in touch with him. Indeed, while this paper was in its final draft I was delighted to
receive a letter from Mr Artsutanov (dated April 1979) of such interest and importance that it demands quoting at
length:
‘It may be interesting for you to know how the idea of the space elevator (s.e.) originated. At the beginning of 1957 a
friend of mine, who like myself graduated at the Leningrad Technological Institute.. . told me about a material which
could hold its weight at the length of 400 km. I thought that at such a height the gravitating force is less and
consequently the length could be enlarged. Then it became interesting for me to calculate the strength of the material to
prolong the vertical rod made of it to infinity…. Immediately the thought came that this rod should have a changeable
section and it was easy to derive the equation . .. which showed that the rod could be done out of any material and its
mass did not become absurdly large….
‘At first I told some of my friends about this idea. Some months later the cosmic theme became very popular. In Summer
1960 I was in Moscow on business and visited the editor of Komsomolskayu Pravda with a proposal to publish my article
without any equations. .. to my mind, they could be derived by any student who understood the idea. A week later the
article was published under the title “Into space with the help of an electric locomotive. . .”;
‘In 1969 the magazine Knowledge is Force (Znanije-Sila) No. 7, p. 25, published my article developing the idea of the
s.e. It was proposed to sink the rods not from a synchronous satellite but from an ordinary one, for example 1000 km,
height. In this case the contact with Earth and the passing of denser layers of the atmosphere would take place at a
comparative low speed. The rods would be like spokes of a wheel rolling along the equator….Having attached itself to
the end of such a rod during a half-turn of this wheel the cosmic ship will gain the speed of 14 km/s. Similarly the
ship returning from the cosmic space will lose the speed and land during another half turn….’
It will be seen that this proposal is virtually the same as that put forward by Moravec 8 years later [17].
Until now, Yuri Artsutanov’s work has only been published in simplified form for the benefit of the lay public. Let us
hope that it will soon appear in its original version, so that his peers can fully appreciate the full genius of this
remarkable Leningrad engineer.
[First Published in Advances in Earth Oriented Applied Space Technologies. Vol. I, no. 1, 1981, pp. 39-48]
REFERENCES
[1] K. E. Tsiolkovski, Grezi o zemle i nene, p. 35.U.S.S.R. Academy of Sciences edition (1959).
[2] Personal communication, Edward J. Hujsak to Fred CDurant, January 6 (1978).
[3] Arthur C. Clarke, The world of the communications satellite, Astronautics, February (1964). Now in Voices From the
Sky. Harper & Row, N.Y. (1965); Gollancz London (1966).
[4] A. R. Collar and J. W. Flower, A (relatively) low altitude 24 hour satellite, J.B.I.S. 22, 442-457, 1969.
[5] John D. Isaacs, Allyn C. Vine, Hugh Bradner and George E. Bachus, Satellite elongation into a true “Sky-Hook”,
Science 151, 682-683 (1966).
[6] James H. Shea, Sky-Hook. Reply by Isaacs et al.,Science 152, 800 (1966).
[7] Vladimir Lvov, Sky-Hook: old idea, Science 158, 946-947 (1967).
[8] J. Pearson, The orbital tower; a spacecraft launcher using the Earth’s rotational energy. Acta Astronautica 2,
785-799 (1975).
[9] J. Pearson, Using the orbital tower to launch Earth escape payloads daily. AIAA Paper 7-123, 27th IAF Congress
(1976); Anchored lunar satellites for cislunar transportation and communication. J. Astronaut. Sci. XXVII,
No. 1, 39- 62 (1979). “Lunar Anchored Satellite Test”: AIAA/AAS Astrodynamics Conference, Palo Alto, 7-9 August, 1978.
[10] Arthur C. Clarke, The Fountains of Paradise. Harcourt Brace Jovanovich, N.Y.; Gollancz, London (1979).
[11] These include G. C. Weiffenbach, G. Colombo, E. M. Gaposchkin and M. D. Grossi, who arrived at the concept as a
result of their work on tethered satellites, and T. Logsdon and R. Africano [see T. Logsdon, The Rush to the Stars,
Franklin (1970)].
[12] Professor Harry O. Ruppe, Hyperfilament’s First Strand. 15 February (1979) (Personal Communication).
[13] For a classic example, see my Profiles of the Future Chapt. 1. Harper & Row, N.Y. (1973); Gollancz, London (1974).
[14] Willy Ley, Rockets, Missiles and Men In Space, Chap 5. Viking (1968).
[15] Charles Sheffield, How to Build a Beanstalk, in press.
[16] NSF Press Release PR 79-15, 2 March, (1979).
[17] Hans Moravec, A non-synchronous orbital skyhook. J. Astronaut. Sci. XXV, No. 4, 307-322 (1977).
[18] Charles Sheffield, The Web Between the Worlds. ACE (1979).
[l9] Space escalator, a quasi permanent engine in space. Tsuotomu Iwata, National Space Development Agency of Japan.
Application to the XXX IAC, 8 March (1979).
[20] G. Polyakov, A space ‘Necklace’ about the earth, (Kosmicheskoye ozherel’ye zemli). Teknika Molodezhi, No. 4
41-43 (1977). (NASA TM-75174).
