All Astronautical Evolution posts in 2020:

Stellar Engines (August)

Voyage to the Large Magellanic Cloud (July)

Why the Human Exploration of Space? (May)

Artificial Gravity for the Journey to Mars and Return (April)

Cruising in Space (March)

All Astronautical Evolution posts in 2019:

The Destiny of Civilisations – Fire, Iron and Gold (November)

The Destiny of Civilisations – A Problem for SETI (November)

The Holy Grail of Space (October)

Return to the Moon, 50 Years On (August)

The Case for Interstellar Flight (June)

SpaceX Dragon 2 Success (April)

Killing the Doomsday Fallacy (Feb.)

All Astronautical Evolution posts in 2018:

How Far Can We Take the Copernican Principle? (Dec.)

Dawkins and the McGraths: a Biologist versus two Theologians (Nov.)

The Atheism Question (Oct.)

The Religion Question (Sept.)

I, Starship (June)

Back to 2017:

Scenario Block Diagram Analysis of the Galactic Evolution of Life (Nov.)

Comments by Alex Tolley (Oct.)

Elon Musk’s “Great Martian” (Oct.)

Elon Musk’s Mars Plans: Highlights from His Second Iteration (Sept.)

What is a Supercivilisation? (Aug.)

Quantifying the Assumptions Behind the METI Debate (July)

Five Principles of a Sustainable Manned Mars Programme (June)

Pale Red Dot: Mars comes to Oxford (May)

Back to 2016:

Elon Musk and Mars: Looking for a Snowball Effect (Oct.)

New in 2020:

Download science fiction stories here

AE posts:

2022: What’s to do on Mars?…

2021: New space company Planetopolis…

2020: Cruising in Space…

2019: The Doomsday Fallacy, SpaceX successes…

2018: I, Starship, atheism versus religion, the Copernican principle…

2017: Mars, Supercivilisations, METI…

2016: Stragegic goal for manned spaceflight…

2015: The Pluto Controversy, Mars, SETI…

2014: Skylon, the Great Space Debate, exponential growth, the Fermi “paradox”…

2013: Manned spaceflight, sustainability, the Singularity, Voyager 1, philosophy, ET…

2012: Bulgakov vs. Clarke, starships, the Doomsday Argument…

2011: Manned spaceflight, evolution, worldships, battle for the future…

2010: Views on progress, the Great Sociology Dust-Up…

Chronological index

Subject index

General essays:

Index to essaysincluding:

Talk presented to students at the International Space University, May 2016

Basic concepts of Astronautical Evolution

Options for Growth and Sustainability

Mars on the Interstellar Roadmap (2015)

The Great Sociology Debate (2011)

Building Selenopolis (2008)


Issue 156, 12 August 2020 – 51st Apollo Anniversary Year

Site home Chronological index Subject index About AE

Read this post on Wordpress if you wish to leave a comment

Stellar Engines

Can one propel an entire solar system?

Part of the Small Magellanic Cloud

In the previous post we considered intergalactic travel using the planetary system of a rogue star which happened to be going the right way. Here we ask: can we make our own intergalactic star and planetary system?

The Shkadov thruster is the original star-moving engine, the idea dating from 1987. It consists of a giant mirror hovering over the Sun (or another star) – rather than creating an imbalance in radiation pressure, as has been claimed, the real physical process is that it uses the radiation pressure from the Sun to balance the Sun’s gravitational attraction, keeping it in place without being in orbit, and it is the mirror’s own gravitational pull on the Sun which drags the Sun in the desired direction.

Since the acceleration is only enough to impart a speed of something like 20 metres/second to the Sun per million years, the practical value of such a scheme is questionable.

The Svoronos Star Tug

Now Alexander A. Svoronos, a graduate of Yale School of Medicine, has a paper scheduled for publication in Acta Astronautica (Nov. 2020, p.306-12) entitled “The Star Tug: An active stellar engine capable of accelerating a star to relativistic velocities”. Sounds good. But will it work?

His method is to place, not a mirror, but a rocket vehicle – the Star Tug – in suspension above the Sun and close to its surface (or another star, but he uses data for the Sun in all his calculations). A separate piece of infrastructure hovering over the Sun sucks up hydrogen and helium, presumably magnetically, and channels these gases to the Tug. The Tug feeds them into a reaction chamber where the hydrogen undergoes nuclear fusion (the hot CNO cycle), creating an exhaust stream which is directed back past the Sun at greater than solar escape velocity. The rocket thrust balances the Tug’s weight, maintaining its distance from the Sun and allowing its own gravity to drag the Sun behind it.

Svoronov starts with the assumption that the entire output of radiant solar energy (all 3.85 × 1026 watts of it) is harvested by a Dyson sphere and employed to lift gas from the Sun to the Tug. He finds an exhaust velocity for the Tug’s engines of 0.1c.

He then discusses four variants of the design (together with some intermediate cases):

Svoronos’s poster child is the most optimistic case in which a Tug close to the Sun operates at 100% efficiency: in this case he arrives at an acceleration imparted to the Sun of 2.3 × 10–6 m/s2, equivalent to 73 m/s per year. This acceleration begins to fall off after the first thousand years as the Sun’s mass is used up and its luminosity reduced. In order to reach a speed of 1% of the speed of light (3,000 km/s), about 200,000 years of continuous thrusting are required; 10% of light speed (30,000 km/s) is reached after several tens of millions of years of thrusting (as shown in Figure 3 of the paper). Whether these results justify the claim that the Star Tug can usefully accelerate the Sun to relativistic velocities is a moot question.

Clearly, the more realistic assumptions of less than perfect use of energy and standing a little way away from the Sun’s surface reduce the performance of the Star Tug by orders of magnitude.

If, for example, one assigns 10% efficiencies to the uses of fusion energy to create rocket thrust and of solar power to do the work lifting gases from the Sun to the Tug, and if one places the Tug at two solar radii from the Sun’s centre and angles its rockets at 45° to the Tug-Sun line, then an acceleration for the Sun of 1.353 × 10–9 m/s2 is obtained, thus 0.043 m/s per year, or 43 km/s over a period of one million years of thrusting. Given that this requires the Tug’s engines to run at a combined power level of 47 times the power output of the Sun itself, and at a power density of half a gigawatt per kilogram of the Tug’s material structure, I wonder whether even this modest performance is possible.

The paper has an interesting level of detail in some respects and is certainly worth reading for its mind-expanding ideas, but I noticed a couple of obvious errors (which should have been picked up at the review stage). I would also say that the placement of the Tug at only 10,000 km above the Sun’s surface is not a practical possibility. Although Svoronos does not state the dry mass of this Tug, it is easy to calculate that it must be 1.6 × 1022 kg (= 0.22 lunar masses). This is not likely to be enough for a rigid structure that has to stretch across 1.4 million km in order to fire with two engines past opposite limbs of the Sun.

Otherwise, I have only three major reservations about this system.

Problems with the Svoronos Star Tug

The globular cluster M13 in Hercules

Certainly, his rocket-propelled Tug is very much more effective than the Shkadov thruster with which this discussion began. Yet the acceleration of the Sun in the desired direction is still achingly slow on a human scale, and the technologies needed to do so – if they are feasible at all – represent capabilities far in advance of those required for interstellar travel using more conventional worldship designs. Therefore I query whether the attempt to move one’s star together with its planetary system accomplishes anything that cannot be attained much more easily and cheaply by other methods?

For example, Svoronos suggests that civilisations more advanced than us might wish to control their star’s motion in order to avoid cataclysmic cosmic events, or to facilitate interstellar and intergalactic colonisation. But there are stars everywhere one looks in the galaxy, and no special reason for preserving the star one is currently resident at. Probably all stars have orbiting matter which can be used to construct habitations, and, due to the constraints on biological evolution, very few of them can have indigenous industrial species. Any organisation in that advanced civilisation which uses vastly smaller worldships to make the necessary journeys would settle suitable unoccupied systems at a much earlier date than those who stayed behind in order to travel with the entire original system.

My second reservation has already been alluded to above. Svoronos’s most effective Tug (10,000 km from the surface of the Sun; all efficiencies at 100%) has a power level of 6.4 × 1031 W, equivalent to 166,000 times the solar luminosity. If we assume that half of the Tug’s dry mass is devoted to propulsion (the other half to structural stiffening), the engine has to handle a power density of 8 GW/kg, or, in the more familiar format used in present-day space engineering, 1.25 × 10–7 kg/kW (the present-day capability is in the region of 50 kg/kW). All one can say is: good luck with that!

Can One Handle the Waste Radiation?

My third question about the Svoronos Star Tug is strongly illustrated even by his most modest proposal: the version in which the Tug hovers at 0.4 AU from the Sun, and has energy efficiencies of 10% in its three major processes.

Again, he does not state the mass, but it works out at 5.22 × 1021 kg, or one fourteenth of a lunar mass. Since it does not have to stretch halfway around the Sun, this version of the star tug can be much more compact, and its entire mass devoted to engines and associated machinery. The propulsive power is 9.2 × 1026 W, or 2.4 times the power of the Sun itself, requiring a more modest power handling capacity of 0.176 MW per kg of engine mass. Note that the requirement for the Daedalus first stage was much higher, at 24 MW/kg, so it appears that we are well into the realm of feasibility here. But the acceleration imparted to the Sun is down to 1.0 × 10–10 m/s2, producing a velocity change of only 3 km/s after a million years of thrusting.

One question which the author skates lightly over is the meaning of 10% efficiency for the fusion engine. He states that this is the efficiency of conversion of gamma rays produced by fusion reactions into thrust, but any sort of combustion, chemical or nuclear, also has an issue in that the fuel is never completely burned, and Svoronos implicitly assumes 100% burnup of his nuclear fuel. If we divide that 10% efficiency equally between gamma ray cleanup and fuel burnup, then, of the theoretical total energy that can be extracted from burning a kilogram of fuel, 68.4% is represented by fuel which does not burn before being ejected into the exhaust stream (still in the form of hydrogen), 21.6% is in the form of gamma rays, and 10% is in the form of the kinetic energy of hydrogen and helium in the exhaust. This generous interpretation of Svoronos’s statement means that, whatever the propulsive power of the engine may be, more than double that power goes into uncontrolled gamma rays.

Even this most modest version of the Tug, then, floods the Solar System with at least five solar luminosity’s worth of gamma rays. Their effect on the habitability of Earth would certainly be severe…

Returning to the design on which the claim of being able to accelerate the Sun to relativistic velocities is based, namely the Star Tug poised 10,000 km above the Sun’s surface and enjoying 100% energy efficiency: we recall that the propulsive power is 166,000 times the solar luminosity. The ambitious target of 100% efficiency in the fusion engine is now an absolute necessity: leakage of only one part in 166,000 of this energy flow would match the power of the Sun itself. Again, on any halfway realistic design assumptions, we have a problem protecting Earth from the intense flux of gamma rays which the engine will produce.

Surrounding the Star Tug with shielding massive enough to absorb that flux does not help much, as it must then radiate away the same amount of power in the form of waste heat. The only way to protect Earth is really to move it into the outer Solar System, but I suppose a civilisation capable of moving a star and with the patience to wait periods on the order of a million years to see results would not have difficulty moving one small planet.

Towards the end of this paper, Svoronos suggests that people, or at least the biological members of the civilisation which built the Star Tug, might prefer to live off-planet, in megastructures (microstructures in comparison with the Star Tug and with the machine that feeds fuel to it!) like O’Neill cylinders. But this of course removes the entire rationale for building a Star Tug in the first place: if living in a space colony is acceptable, then so is living in a worldship. A vast fleet of worldships could be built and flown for a tiny fraction of the effort that would go into constructing even the most modest of Svoronov’s Star Tugs.

The bottom line is this: the power of sunlight by itself is not commensurate with the power needed to accelerate the Sun and planets at a useful rate. Svoronos achieves his results – or apparent results – by adding more power: the nuclear fusion of the hydrogen lifted off the surface of the Sun. That gets him into the realm of power levels orders of magnitude greater than the radiant power of the Sun itself; levels which are able to do the job. But now his power is so high that the slightest deviation from 100% efficiency will be enough to render Earth uninhabitable, making the operation worthless. The game is simply not worth the candle.

An Alternative to the Star Tug

About 150 globular clusters surround the Milky Way

So let us design a worldship for interstellar or intergalactic travel. Let us take Svoronov’s smallest Star Tug, thus with a mass of 5 × 1021 kg (five million trillion tonnes, or one fourteenth of the mass of the Moon). We assign half of this mass to payload, thus the habitable volume occupied by the passengers, and half to the engine, power generators and propellant tanks.

This version of the Star Tug has an exhaust velocity of 9500 km/s. If we give it a mass ratio of 4.0 then its cruising speed will be about 2% of the speed of light. It could make the crossing from our Solar System to the Large Magellanic Cloud, a distance of about 163,000 light years, in a little over eight million years. Over the same period of time, the same version of the Star Tug would manage to accelerate the entire Solar System by a speed of only about 0.01% of the speed of light; the Svoronos Star Tug at 10,000 km from the Sun but with realistic energy efficiencies of 10% would manage 0.2% of the speed of light.

If we assume that the worldship is fuelled by the same supply system as proposed for the Star Tug, the time taken to load it with 15 × 1021 kg of fuel works out at about 24 years. If we allow the total engine burn time, to accelerate the worldship away from the Solar System and decelerate it at a star in the LMC, to be 240 years, then the power level drops by a factor of ten, to 9 × 1025 W = 0.234 sunpower.

The power that goes into waste radiation is double this, if our assumptions above are correct. We would not want to launch this vehicle from anywhere close to Earth, so clearly it needs be moved to the outer Solar System before switching on its main engine. A disk 10 metres thick with a diameter of 2000 km and the density of rock would have a mass of about 1017 kg, which represents 0.00004 of the mass we have allowed for the passenger accommodation, so there’s no problem shielding them from the waste radiation.

Such worldships could basically be launched as fast as they could be built and fuelled. Each would have accommodation for over one trillion passengers, even allowing a very generous one million tonnes per person.

Of course in reality one would not build a single vehicle as large as that. The mass budget for ships and propellant would be spread over a fleet of smaller units in say the tens of millions up to billion-tonne size range (see my discussion of worldship fleets in JBIS). Propulsion would use the easier deuterium/tritium or deuterium/helium-3 reactions, perhaps with lithium deuteride fuel. Deuterium and helium-3 can be mined from the atmospheres of the giant planets (a lot cheaper than trying to lift matter from the surface of the Sun!) and lithium from small rocky worlds. Lithium is needed to breed tritium by fission, as the latter is radioactive and so cannot be stored for the timescales needed. See a paper by Robert Freeland of the Project Icarus Study Group describing a fission-fusion engine using lithium deuteride fuel.

Despite all the verbal handwringing about the difficulties of rocket propulsion and the tyranny of the rocket equation which one so often hears, nuclear fusion is eminently capable of powering massive interstellar vehicles on voyages of a few centuries to the nearest stars, as we discussed earlier. With mature technologies and convoy tactics, the journeys could be extended to thousands, and – who knows? – perhaps even millions of years, given a large enough convoy, allowing access to globular clusters and the nearest galaxies beyond the Milky Way.

While the subject of star tugs and stellar engines is a fascinating intellectual pursuit, I would not advise SETI astronomers to spend much time on making any special searches for them.

To make comments, please visit this post on Wordpress.

Site home Chronological index About AE