Bruce Bowdenscientium

Starship Designs
Sailing Ships

(These are mirrored and enhanced from the "Lunar Institute of Technology" Web site as a precaution in case that site should go down. Some material is a little dated, but I highly recommend visiting the Lunar Institute of Technology for some fascinating reading.)
- Bruce Bowden
Magnetic Sails
Relativistic Particle Beam Propulsion
Solar and Laser Driven Lightsails


Magnetic Sails

Contributed by Ges Seger

The magnetic sail (or Magsail) consists of a loop of superconducting cable several hundred km in diameter, which spends most of the spacecraft's flight stored on-board. To operate, the cable is deployed into space and a current passed through the loop. Once initiated, the current will be maintained indefinitely without further power input. The current loop's magnetic field guarantees it eventually reaches a circular shape, due to the stress it generates back on the hoop. Typical field stregths are ~10^-5 Tesla, so little structural strengthening is required.

The Magsail can be flown one of either two ways: in axial (hoop perpendicular to spacecraft velocity vector) configuration, or normal (hoop aligned along spacecraft's velocity vector) configuration. In either configuration, charged particles from a star's solar wind are deflected based on the magnetic field they encounter, which changes the momentum of the current loop. This creates drag, which will accelerate the spacecraft in the direction of the wind. In computer studies, Andrews and Zubrin (see reference) found that the normal configuration provided ~5.5 times the "reflection area" of the axial configuration. Depending on the current densities the superconducting material can handle without quenching (currently around 10^6 amps/cm^2), this yields at a distance of 1 AU from the Sun an acceleration of 0.0218 m/s^2. For an interstellar mission using a Magsail as a brake, a close (r<=0.6 AU) pass of the destination star would be neccessary.

Andrews and Zubrin go on to plug these numbers into performance calculations for fusion rockets. Without a Magsail, a one-way interstellar mission accelerating 1000 tons of payload with a D-He3 fusion drive to 0.1c, coasting, then decelerating to interplanetary velocity would require >32,000 tons of fuel. With a 1000-ton Magsail, propellant mass drops to >9500 tons since the fusion drive is no longer required to perform deceleration.

Andrews and Zubrin also examined performance of a laser lightsail without and with a Magsail. A Magsail gives a lightsail the capability of decelerating at its destination without resorting to beamed power from the Solar System, and actually decreases the amount of mass a lightsail probe would have to carry.


Dana Andrews & Robert Zubrin, "Magnetic Sails and Interstellar Travel," presented @ 39th IAF Congress in Bangalore, October 1988 (reprinted in JBIS vol 43 pp 265-272)


Relativistic Particle Beam Propulsion

(Scientium note: Please see the letter from Gerald D. Nordley linked from this sentence.)

Contributed by Ges Seger
Concept Introduction

Using a pellet stream of reaction mass shot out either ahead of or at an interstellar spaceship was first proposed by Singer in 1980. In the JBIS paper I am summarizing, Nordley takes this concept much further. He replaces the pellet stream with a particle beam, and uses an Andrews/Zubrin magsail to reflect the particle beam and propel the ship.


The Nordley Relativistic Particle Beam (RPB) drive requires a number of beam drivers "fixed" in the Solar System, which shoots a relativistic neutral particle beam at a magsail-equipped spacecraft. The spacecraft ionizes the incoming particle beam, which is them reflected by the magsail.

To minimize focusing distance of the beam, Nordley suggests performing acceleration to terminal velocity while still close to Sol. This requires accelerations approaching or even exceeding one earth gravity, which has the problem of requiring thousands of terawatts of power for massive (> 1000 tons) payloads.

A 1000-ton RPB-drive ship accelerating at 1 earth gravity requires a mass flow of 43 grams/sec from the beam drivers back home. If split among one million drivers, this results in 43 micrograms/sec/driver, with a required power input of 3 GW/driver. For those who think these are impossible numbers for engineering such a system, Nordley adds:

"The point is that a million beam drivers for an interstellar propulsion system is not unreasonable for a civilization that made ten million automobiles a year before robotics."
To get a feel for the energy required, he scales the acceleration back to 0.036 earth gravities and compares the a 1000-ton RPB-driven probe with a 1000-ton laser lightsail at the same acceleration. The RPB-driven probe requires 11 GW/ton to accelerate, as opposed to 65 GW/ton for the lightsail.

Nordley suggests a massive neutral particle, such as a heavy atom or molecule (C60 or C60F60 surrounding a heavy atom was suggested) for use in an RPB driver. Massive particles will not be disturbed by encounters with interstellar hydrogen atoms, but still be able to be manipulated by light pressure so as to collimate the beam downrange from the drivers.

The primary reflection scheme Nordley discusses is the magnetic sail concept developed by Andrews and Zubrin. He notes that while other reflector concepts (such as an Orion-style pusher plate and D-He3 pellets) are readily usable by this system, there would be losses due to heating of the reflector which would not translate to propulsion. Magnetic fields are conservative, require little if any additional energy (if using superconductors), and have little energy dissipation upon establishing the field.


Nordley investigated missions to Alpha Centauri using 1 to 3 earth gravities acceleration with different acceleration times (and thus different required beam energies, terminal velocities, etc.). For a 1000-ton ship, one gravity acceleration over 155 days results in a peak velocity of 0.71c, a trip time of 11 years (Earth reference), and require a 1263 terawatt beam focused over 0.091 light-years. The same ship accelerated at three gravities for 196 days (Earth reference) reaches a peak velocity of 0.99c, and requires a 8701 terawatt beam focused over 0.31 light-years. This mission, incidentally, required only 5.3 years Earth time (3 years ship time).


No interstellar propulsion scheme is perfect, and the Nordley RPB-drive is no exception. The main stumbling blocks of the drive are as follows:

Building the Beam Driver
Nordley assumed robotic space mining and manufacturing for driver construction. This allows the possibility of carrying self-replicating machines with you to the target system and "seeding" it. After a short time, the target system will be capable of accelerating/decelerating payloads.

Beam Driver Power
A 1000-ton spacecraft travelling at 0.5c has a kinetic energy of approximately 1.4*10^22 Joules (14 gigaterajoules, or GTJ). The actual value will be higher, depending on mass accelerated, driver efficiencies, and reflector efficiencies. Solar power collection, beamed to the drivers, was assumed.

Particle Stream Acceleration
Though this was not in the original Nordley article, I would suggest plasma particle (AKA plasma wake-field) acceleration be used for the beam drivers. Current accelerators are limited by magnetic stress in their electromagnets, which limits the final energy of the particle stream. Much more powerful fields can be generated in a plasma, which also allows for a shorter accelerator.

Reflecting the Particle Stream
Though a magsail is suggested for reflecting the particle beam, Nordley suggests that reflector engineering needs to be thought out further, and that different reflectors should be considered. He also suggests some thought be given to the optimal size and mass of the particles used.

Decelerating in the Target System
This is a big problem which receives a cursory discussion in Nordley's paper. His assumption for an initial mission to a target star is that a combination of magsail and lasers ionizing the star's solar wind would be used to affect deceleration. Considering the ship is approaching at relativistic speeds, this needs to be thought out much further. Assuming you make it and the target star posesses the required raw materials, beam drivers can be built in situ and a two-way propulsion link be established.


Nordley makes the following comments in conclusion:

(Scientium note: Gerald D. Nordley wrote a letter to site owner Bruce Bowden on August 14th of 2001 to ask that the following information be included.

"...shortly after submitting this paper I proposed that the particles could be made self-steering through microtechnology to elimnate the need for lasers stationed along the beam path to correct the course of the particles to the reflector, which would be on the order of a thousand AU's from the beam projectors at the end of the acceleration period. This idea is briefly documented in: Frisbee, R.H. (ed). JPL D-10673: NASA OACT Fourth Advanced Propulsion Workshop, Apr 5-7 1993, p.470. I will be addressing it at greater length at the IAF conference in Toulouse this fall.")



Solar and Laser Driven Lightsails

Contibuted by Alasdair Allan

Solar Sail Diagram


In the search for a replacement for based propulsion systems, several novel ideas have been considered. One such idea is the lightsail, which was first suggested by Tsiolkovsky and Tasander in 1924. Fundamentally, the concept behind the lightsail is to use a large reflective surface to provide propulsion for a spacecraft through the use of sunlight pressure for the motive force.

Physical Concepts

In consideration of the physics behind the sail a common misconception is that the propulsive force is provided by the solar wind. The solar wind is a stream of charged particles, mostly high velocity electrons and protons, emanating from the solar corona. It is a highly dynamic phenomenon, influenced by factors such as solar flares and X-ray bursts. The energy attatched to these particles however is several orders of magnitude smaller than the energy attatched to the flux of solar photons itself.

Hence the lightsail makes use of the solar photon flux for propulsion. While photons have no rest mass, they do of course have mass while in motion, and thus momentum. Upon collision with the reflective sail material the photon will be reflected and, in the process, apply a force to the sail. The size of the total force on the body of the craft will thus be proportional to the sail's area.

The applied force can be directed by tilting the sail with respect to the incoming photon flux, this will change the direction of the acceleration applied to the sail and thus allow changes in the sail's orbit.

Hence for an inward spiral, the pressure applies a breaking force to the sail allowing it to fall inward towards the Sun. this will place the sail into a tighter, faster, elliptical orbit around the Sun. Similarly, for an outward spiral, the photon pressure applies an accelerating force to the sail allowing it to move outwards away from the Sun. This will place the sail into a larger, slower, orbit around the Sun.

We can very roughly approximate the force applied to the sail, and hence determine the acceleration of the vehicle, by using the equation

F = m*a = A*L*r / 2*pi*R^2*c

Where L=3.9x10^26 Watts and is the solar luminosity, A is the total area of the sail, r is the reflectance of the sail, R is the distance of the sail from the Sun, c is the speed of light and m is the mass of the sail.

Current estimates [Friedman, 1988] give a sail density of between 5x10^-3 kgm^-3 and 8x10^-3 kgm^-3 for a working solar sail design. This implies a charateristic acceleration of between 1x10^-3 ms^-2 and 1.6x10^-3 ms^-2, if the solar constant is taken as a working figure for the power per square metre intercepted by the sail.

For orbital calculations involving the sail it must be borne in mind that the equations of motion of the lightsail are complicated due to the fact that it undergoes continuous acceleration. This is not the case for conventional ballistic spacecraft where any velocity changes can be assumed to occur instantaneously.

Considering a sail with a characteristic acceleration of 1x10^-3 ms^-2, the transfer time from Earth to Mars orbit would be approximatley 400 days. This is somewhat slower than for ballistic spacecraft. thus there is no performance advantage for low thrust vehicles over simple one-way trajectories. However for round trip insystem travel, where one way travel time is not considered a priority, the high payload characteristics of the lightsail make it an attractive alternative to the ballistic spacecraft for the efficient delivery of large payloads.

Design Considerations

A perfect reflector is not possible, some photons will always be absorbed by the reflecting material. However, it is obviously sensible to make use of the best material available. Silver would be an ideal reflector, except that it is expensive and undergoes rapid oxidisation. Aluminium is both cheap and does not oxidise readily, and reflectances of 85 to 88% are possible. Hence it is most likely that lightsails would be constructed from a thin film of aluminium backed by a plastic lining.

The possibility of wrinkling in the sail material must be considered. Not only would such wrinkling cause reduction of thrust on the sail, it would also possibly cause reflection of photons back onto the sail material itself, causing hot spots and perhaps burn throughs. This possibility introduces some important design considerations for the wires and edge members that hold the sail in position. This is one of the main reasons why the Aluminium reflective surface must be backed by a plastic material. It must be mounted so that it will be sturdy enough to survive folding and packing for launch and deployment. Another reason not to use pure aluminium is that the structure would be too heavy, a lighter smoother sail can be produced by spraying a very thin layer of Aluminium onto a plastic substrate.

Plastic film is now commerically available in thinknesses of approximatley 8 microns. This ultra thin platic is used as insulation in semiconductors, and other electronic devices, amoung the more common forms of this plastic are Mylar and Kapton. An 8 micron thick plastic sail will have a top acceleration of approximatley 0.8x10^-3 ms^-2, this is insufficient for interplanetary missions where an acceleration of around 1x10^-3 ms^-2 is the minimum desirable performance. For this we need, taking payload into account, a sail approximatley 2 microns thick. Since plastic sheets this thin can now be produced in limited quantities, it seems reasonable to consider a 2 micron thick sail technically possible with today's technology.

Indeed, while at MIT, Drexler developed a concept for space based amanufacturing of ultra-thin solar sails, whereby it would be possible to produce sails of approximatley 0.1 microns thickness by constructing the sails on a fibre web. Aluminium would be evaporated such that a thin film would form between the fibres giving an ultra-thin reflective coating. With such a sail accelerations of 8x10^-3 ms^-2 may be possible.

However, while restricted to ground based manufacturing techniques it will obviously not be possible to fabricate large areas of sail material in single continous sheets. Long strips may however be fabricated in managable widths, these being joined togerther to create sails.

In fabricating the sail material the problems of meteorite punctures must be considered. To prevent rips from spreading across the sail, shredding it, ripstops must be built into the sail material. These are producted by doubling the sail material, creating seams, or by reinforcing the material with tape.

Electrostatic charging of the sail must also be considered. Since the sail material is thin a potetial difference across the sail could cause arcing and ripping in the material. This problem can be dealt with by installing an electrical shorting mechanism which runs from the front of the sail to the back. This can be simply insterted at the joins of the stips of the sail material.

Rigging the Sail

The solar sail vehicle must be rigged with four main design criterion in mind, performance, rigidity, stabilisation and control.

There are three main classes of sail design, these are differentiated by the techniques used to introduce rigdity to the vehicle. A disk of sail material can be kept stable if the entire disk is spun, this is the first class of solar sail vehicle, the disk sail. A second class of sail is the heliogyro, which operates using the same principle as do helicopter blades, the blades of the heliogyro stay rigid and in the same plane due to rapid rotation. If, instead of spinning the sail, a support structure is attatched a third class of sail vehicle is produced. this third class is usually called the square sail, although the sail need not necessarily be square.

The first solar sail considered were disk sails. they are the simplest, in theory, since a large sail can be spun and requires no supporting structure at all. However, it is difficult to control a disc sail by the application of torques from either gas jets or vanes. In fact, the only practical way to control such a sail is to move the centre of mass relative to the centre of area. the offset will cause a torque to tilt the sail in the desired direction. However the mechanisms required to perform this offset would need to be quite eloborate, and manoeuvring would be slow.

The heliogyro is a variation on the disk sail which was invented in the mid 1970's by MacNeal and Hedgepath. The long blades of sail material would be constructed and spun like helicopter blades to keep the sail rigid. The chief advantage of this design would be easy controlability, unlike disk sails caontrol would be achieved by pitching the blades. Thus the heliogyro retains the advantage of a lack of supporting structure, but is nearly as controlabale as a three-axis stabilised square sail.

The three-axis stabilised, or square, sail is held rigid with a supporting structure of wires and booms. Control of the square sail is achieved by using solar pressure vanes. These vanes are small, independently controlled, sails mounted at the ends of the axes. The solar vanes allow force to be applied to give movement in roll, pitch or yaw. However, since the square sail has a supporting structure its performance is moderate relavtive to the spinning sail designs.

Sail Launch

Another design problem is provided by launch and deployment of the sail. If the sail must be launched from the surface of the Earth using a rocket careful thought must be given to the folding and packing of the sail during launch.

For the disk sail the packaging problem is almost insurmountable, and for that reason the disk sail has been dropped from consideration by most of the agencies currently concerned with sail design.

The heliogyro solves the problem presented by launching by rolling the sail blades up, and letting them be deployed by spinning the spacecraaft after orbital insertion.

Square sail deployment is perhaps the most analogous to the hoisting of a terrestrial sail. The masts and spars are deployed first, guy wires attatched to the ends of the spars then begin to unfurl the sails which would have been folded into a canister during launch. The wires then pull the sail straight out by its corners and hang it on the spars.

The Interplanetary Shuttle

The concept of the interplanetary shuttle was first developed in 1976, at JPL, by Wright during the NASA study of solar sail spacecraft for a prospective rendezvous with comet Halley. Wright envisioned the vehicle designed for that mission as being a reusable solar sail, capable of delivering spacecraft to various planets, asteriods, short period comets, or to solar orbit. The payload capabilities of the sail, a 12 blade heliogyro with a characteristic acceleration of 1x10^-2 ms^-2, are detailed below in table 1.

	Target Body     Flight Duration         Payload Capacity

	MERCURY         600 days                8300 kg

	VENUS           270 days                6800 kg           

	MARS            400 days                2300 kg           
			500 days                5100 kg           

	JUPITER         900 days                1500 kg           

Solar Sail Diagram

Laser Driven Lightsails

By using a large sail, and embarking upon a series of gravity-assist manoeuvers cumulating in a close fly-by of the Sun, we could conceivably give a solar sail vehicle a velocity of approximatley 2x10^5 ms^-1 on leaving the solar system. This woould allow the sail to travel to the nearest star in roughly 6500 years. Humanity, as a rule, is not usually that patient.

Alternativley, through the use of a large aperture high powered laser systems to provide photon pressure for our lightsail, higher velocities are obtainable.

The possibilty of laser powered lightsails may date back as far as 1966, when Marx produced a theoretical analysis of the special relativity behind such a vehicle. The analysis was later proved to have several major flaws, which nontheless did not alter radically the chances of success for such a laser driven interstellar craft.

A fly-through sail vehicle has been proposed by Robert Forward [Forward, 1984] whch could carry a 1000kg payloads to the Alpha-Centauri system in approximatley 40 years. The sail required is 3.6km in diameter, not that much larger than the JPL design for a Halley rendezvous sail, it would however be constructed of pure Aluminium only 16 nanometres thick. Unlike ordinary sails the pure Aluminium sail must be constructed in space. The sail would reach a maximum velocity slightly greater than one tenth the speed of light, with a characteristic acceleration of 0.036 gravities. Upon reaching its destiniation the vehicle would fly through the Alpha-Centauri system at 3.6x10^7 ms^-1. The main problem with the vehicle is the power needed for the laser system necessary to drive the sail, this being 10 Giga-Watts, powered over the lifetime of the vehicle.

If a sample-return mission is required Forward finds that a 7.2 Trillion Watt laser is required. the sail diameter also grows to 100 km, with our maximum velocity being one fifth the speed of light. As the sail vehicle approaches the target star the outer part of the sail is detatched. The inner part of the sail is turned around so that the non-reflective side faces the laser. The outer part moves ahead, and the laser light is reflected back onto the inner part, decelerating it.

Perhaps a more workable design for a laser driven lightsail is Forward's Starwisp. This is an ultra-light, 20 gram, wire mesh sail that would make use of a proposed solar power satellite, designed to beam microwaves down to Earth, for its launch.

Forward studies the case of a 10 Billion Watt solar power satellite. This would accelerate a 20 gram spacecraft at 115 gravities. The craft would reach a velocity of two tenths the speed of light in a week and would arrive at Alpha-Centauri in 21 years.

Proof of Concept

The World Space Foundation is currently constructing a proof of concept solar sail vehicle, the sail has a current launch date of late 1995 aboard an Ariane Launcher. The Foundation has opted for a square sail design, in its present form the sail is 55m on a side supported by a light X-frame. The sail material is 2.5 microns thick with a Kapton substrate, control is provided by four steering vanes.


The lightsail has aroused a great deal of interest from scientist and layman alike, possibly due to the elegance of the concpet. No matter what the motive for this interest however, the lightsail offers a practical alternative to rocket powered craft for interplanetary and perhaps interstellar exploration.


Starsailing: Solar Sails and Interstellar Travel
L. Friedman
John Wiley & Sons, Inc. [1988]

Sailing the Solar Seas: Part I
R. Pyle (ed)
Foundation Astronautics Notebook, No. 32

Sailing the Solar Seas: Part II
R. Pyle (ed)
Foundation Astronautics Notebook, No. 33

Interstellar Vehicle Propelled by Terrestrial Laser Beam
G. Marx
Nature pp. 22-23 [Jul 1966]

Interstellar Vehicle Propelled by Terrestrial Laser Beam
J.L. Redding
Nature pp. 588-589 [Feb 1967]

Was Marx Right? or How Efficient are Laser Driven Interstellar Spacecraft?
J.F.L. Simmons, C.R. McInnes
American Journal of Physics pp.205-207 [Mar 1993]

Roundtrip Interstellar Travel Using Laser-Pushed Lightsails
R.L. Forward
Journal of Spacecraft Vol. 21, No. 2 [1984]

Interstellar Migration and the Human Experience
B.R. Finney, E.M. Jones (eds)
University of California Press [1985]

Norton's 2000.0 Star Atlas and Reference Handbook
I. Ridpath (ed)
18th Edition, Longman Scientific & technical [1989]

This article is copyright Alasdair Allan (1993)

All rights reserved. No part of this article may be reproduced,
stored in a retrival system, or transmitted, in any form or by
any means, electronic, mechanical, photo-copying, recording or
otherwise, without the prior permission of the Author.



Starwisp Diagram

Starwisp is a concept by Dr. Forward

This design requires the use of a standard solar power satellite that converts solar energy to microwaves, and beams them down to Earth. This satellite focuses its microwaves on a Fresnel zone plate. As the beam passes through the Fresnel lens (which focuses the microwaves at a distance), it pushes a lightweight sail (see other, more standard solar sails for a more detailed explanation).

This sail is approximately one kilometer across and made of a fine wire mesh. It masses approximately sixteen grams. Each intersection in the mesh contains a microcircuit. The microcircuits together (approximately one hundred billion of them) would mass about four grams.

A 10-billion watt satellite would accelerate Starwisp at 115 G's. This would take the spacecraft to 20% light speed in one week, at which point the microwave power satellite is released to return to its proper functioning. Starwisp cruises the rest of the way. It would reach Alpha Centauri in twenty-one years.

Upon system-arrival, the destination system is flooded with microwaves from the power satellite. The wire mesh is designed to act as an antenna, which collects this power. This powered phase would last about thirty hours, and the spacecraft should pick up about ten watts of power.

The circuits measure the phase of the microwaves to determine the direction of the Earth. Each circuit has a diode which looks in a unique direction and color with a fairly diffuse reflector. So Starwisp is able to take live video of the destination system and transmit these images to the Earth.

Starships of the Mindscientium

Scientium Webmastering