≡ Menu

Pondering the ‘Dyson Slingshot’

Let’s start the week by talking about gravitational assists, where a spacecraft uses a massive body to gain velocity. Voyager at Jupiter is the classic example, because it so richly illustrates the ability to alter course and accelerate without propellant. Michael Minovitch was working on this kind of maneuver at UCLA as far back as the early 1960s, but it was considered even before this, as in a 1925 paper from Friedrich Zander. It took Voyager to put gravity assists into the public consciousness because the idea enabled the exploration of the outer planets.

Can we use this kind of maneuver to help us gain the velocity we need to make an interstellar crossing? Let’s consider how it works: We’re borrowing energy from a massive object when we do a gravity assist. From the perspective of the Voyager team, their spacecraft got something for ‘free’ at Jupiter, in the sense that no additional propellant was needed. What’s really happening is that the spacecraft gained energy at the expense of the planet. Jupiter being what it is, the change in its own status was invisible, but it lent enough energy to Voyager to prove enabling.

According to David Kipping (Columbia University), the maximum speed increase equals twice the velocity of the planet we’re using for the maneuver, and when you look at Jupiter’s orbital speed around the Sun (around 13.1 kilometers per second), you can see that we’re only talking about a fraction of what it would take to get us to interstellar speeds. But the principle is enticing, because traveling with little or no propellant is a longstanding goal, one that drives research into solar sails and their fast cousins, beamed lightsails. And it has been much on Kipping’s mind.

For gravitational assists from planets are only one aspect of the question, there being other kinds of astrophysical objects that can help us out. Depending on their orbital configuration, some of these are moving fast indeed. In the early 1960s, Freeman Dyson went to work on the physics of gravitational assists around binary white dwarf stars — he would ultimately go on to consider the case of neutron star binaries (back when neutron stars were still purely theoretical). Such concepts obviously imply an interstellar civilization capable of reaching the objects in the first place. But once there, the energies to be exploited would be spectacular.

While I want to begin with Dyson’s ideas, I’ll move tomorrow to Kipping’s latest paper, which addresses the question in a novel way. Kipping, well known for his work in the Hunt for Exomoons with Kepler project, has been pondering Dyson’s notions but also applying them to what would seem, on the surface of things, to be an entirely different proposition: Beamed propulsion. How he combines the two may surprise you as much as it did me, as we’ll see in coming days.

Image: An artist’s conception of two orbiting white dwarf stars. Credit: Tod Strohmayer (GSFC), CXC, NASA, Illustration: Dana Berry (CXC).

Nature of the Question

If we talk about manipulating astrophysical objects, a natural objection arises: Why should we study things that are impossible for our species today? After all, we can get to Jupiter, but getting to the nearest white dwarf, much less a white dwarf binary, is beyond us.

But big ideas can be productive. Consider Daedalus, conceived in the 1970s as the first serious design for a starship. The idea was to demonstrate that a spacecraft could be designed using known physics that could make a journey to another star. The massive two-stage Daedalus (54,000 tonnes) seems impossible today and doubtless will never be built. Was it worth studying?

The answer is yes, because once you’ve established that something is not impossible, you can go to work on ways to engineer a result that may differ hugely from the original. Breakthrough Starshot is built around the idea of using lasers to propel a different kind of spacecraft, not of 54,000 tonnes but of 1 gram, carried by a small lightsail, and designed to be sent not as a one-off mission but as a series of probes driven by the same laser installation.

Once again we’re stretching our thinking, but here the technologies to do such a thing may (or may not, depending on what Breakthrough Starshot’s analyses come up with) be no more than a few decades away. The current Breakthrough effort is all about finding out what is feasible.

Again we’re designing something before we’re sure we can do it. The challenges are obviously immense. Consider: To go interstellar with cruise times of several decades, we need to ramp up velocity, and that takes enormous amounts of energy. Kipping calculates that 2 trillion joules — the output of a nuclear power plant running continuously for 20 days — would be needed to send the Breakthrough Starshot ‘chip’ payload to Proxima Centauri. And that’s just for one ‘shot’, not for the multiple chips envisioned in what might be considered a ‘swarm’ of probes.

Working with Massive Objects

Are there other ways to generate such energies? Freeman Dyson’s extraordinary white dwarf binary gravitational assist appears in “Gravitational Machines,” a short paper that ran in a book A.G.W. Cameron edited called Interstellar Communication (New York, 1963). Conventional gravity assists aren’t sufficient because to be effective, a gravitational ‘machine’ would have to be built on an astronomical scale. Fortunately, the universe has done that for us. So we should be thinking about the principles involved, and what they imply:

…if our species continues to expand its population and its technology at an exponential rate, there may come a time in the remote future when engineering on an astronomical scale will be both feasible and necessary. Second, if we are searching for signs of technologically advanced life already existing elsewhere in the universe, it is useful to consider what kinds of observable phenomena a really advanced technology might be capable of producing.

Dyson’s considers the question in terms of binary stars, specifically white dwarfs, but goes on to address even denser concentrations of matter in neutron stars. Now we’re talking about a kind of gravitational assist that has serious interstellar potential. A spacecraft could be sent into a neutron star binary system for a close pass around one of the stars, to be ejected from the system at high velocity. If 3,000 kilometers per second appears possible with a white dwarf binary, fully 81,000 kilometers per second could occur — 0.27 c — with a neutron star binary.

Hence the ‘Dyson slingshot.’ (As an aside, I’ve always wondered what it must be like to have a name so famous in your field that everything from ‘Dyson spheres’ to ‘Dyson dots’ are named after you. The range of Dyson’s thinking on these matters certainly justifies the practice!).

The slingshot isn’t particularly effective with stars of solar class, where what you gain from a gravitational assist is still outweighed by the possibility of using stellar photons for propulsion. But as Dyson shows, once you get into white dwarf range and then extend the idea down to neutron stars, you’re ramping up the gravitational energy available to the spacecraft while at the same time reducing stellar luminosity. An advanced civilization, in ways Dyson explores, might tighten the orbital distance until the binary’s orbital period reached a scant 100 seconds.

Now a gravity assist has serious punch. In other words, there is the potential here for a civilization to manipulate astrophysical objects to achieve a kind of galactic network, where binary neutron stars offer transportation hubs for propelling spacecraft to relativistic speeds. As you would imagine, this plays to Dyson’s longstanding interest in searching for technological artifacts, and we’ll be talking about that possibility as we get into David Kipping’s new paper.

For Kipping will take Dyson several steps further, by looking not at neutron stars but black hole binaries and coming up with an entirely novel way of exploiting their energies, one in which a beam of light, rather than the spacecraft itself, gets the gravitational assist and passes those energies back to the vehicle. Kipping calls his idea the ‘Halo Drive,’ and we’ll begin our discussion of it, and a novel insight that inspired it, tomorrow.

The Dyson paper is “Gravitational Machines,” in A.G.W. Cameron, ed., Interstellar Communication, New York: Benjamin Press, 1963, Chapter 12. The Kipping paper is “The Halo Drive: Fuel-free Relativistic Propulsion of Large Masses via Recycled Boomerang Photons,” accepted at the Journal of the British Interplanetary Society (preprint). For those who want to get a head start, Dr. Kipping has also prepared a video on the Halo Drive that is available here.


Comments on this entry are closed.

  • Alex Tolley March 4, 2019, 13:37

    In the spirit of reducing size to manageable technologies, what about mini/micro-sized black hole binaries? Could these be used to slingshot microscopic objects? A binary micro black hole of a Planck mass could have an “orbiting” BM at just a few Planck lengths with a velocity close to light speed. This could be scaled up to the sort of mini BH as featured as part of the propulsion of the Sirius in Clarke’s Imperial Earth.

    While I do not propose that we create such BHs, perhaps finding such objects might be the means to harness such binaries for propelling microscopic objects on interstellar journeys. Such microscopic packages may be the best way to reliably transmit large quantities of information across interstellar distances rather than the bandwidth limiting use of EM systems over such distances. As targeting and capture would be “difficult”, the package could transmit its data as it neared its target.

    It might also be useful for spraying out microbes for directed panspermia, limiting the transit time to high fractions of c and using massive redundancy to ensure success.

    • Charley March 4, 2019, 17:15

      Alex Tolley, while you said you don’t want to have any black holes being created, what you think of the idea that a ‘tiny’ black hole, such as could be formed in a collider, which would exist for just a fraction of the second before it would evaporate and use that tiny black hole as a point source for acceleration?

      • Alex Tolley March 4, 2019, 18:28

        My main reason for not wanting to create BHs is the energy cost. Finding them if they exist, and then harnessing them seems like a better way to go. Ideally a pair of mini BHs would be the sort of scale I was thinking of.

        • Charley March 4, 2019, 20:37

          Carrying this idea just a little bit further with you, putting aside the energy cost, wouldn’t a relatively small binary pair of black holes be a problem in terms of their stability? It’s the Goldilocks issue all over again; too small and they won’t last very long, too big and they might rip you apart as you attempt to use them. By the time you find a given pair that is just right, they’ve already evaporated.
          That’s why I was suggesting that making them upon demand gives you a window in which you can utilize them to your advantage before they evaporate. Your point is well taken on the energy cost and I don’t even know if it’s in the capabilities of the present time to create such a entity. But I assume that as knowledge advances that there will be times in which it may be tried.

          • Alex Tolley March 4, 2019, 23:53

            A mini BH of around 1E8 MT would take centuries to evaporate. As mass increases, the evaporation time increases as the cube of its mass. To put 1E8 MT in perspective, that is smaller than asteroid Bennu that OSIRIS-REX is to visit for a sample return, and about 2x the mass of Ryugu that Hayabusa 2 is currently visiting.

            If we were to find such BHs, the interesting thing is how would we move them and then get binaries to dance? Such BH binaries should have stable orbits around larger bodies, even the Earth. [ If they can slingshot tiny particles at fractions of c, they might make formidable anti-satellite weapons. ]

            • Brett Bellmore March 5, 2019, 18:30

              The problem with mini-black holes is that they don’t have any more gravitation than non-black holes of the same mass. You can just get closer to them.

              And the smaller the black hole gets, the worse the tidal effects get. The reason tiny black holes evaporate, after all, is basically the tidal forces “ripping” space apart, liberating virtual particles into the real world by swallowing half the pairs.

              I suspect tiny black holes would be better used to enable efficient mass energy conversion for powering a star ship.

              • Alex Tolley March 6, 2019, 1:35

                mini BHs allow very fast orbits simply because all that gravitation is compressed to allow the radius to be very small. You cannot slingshot anything with a pair of asteroids gently orbiting each other.

                But I agree that simple dropping matter into such a BH is probably a better way to extract energy for a drive, although it would be less efficient than an anti-matter rocket for propelling a payload if only due to its dead weight mass.

  • Bruce D. Mayfield March 4, 2019, 14:38

    Look forward to reading this keenly. Two obvious issues with these extreme gravitational slingshots: Surviving the radiation environments and the extreme tidal stresses.

    • Jens March 4, 2019, 22:04

      I take it the tidal stress would be the easiest one to quantify?

      • Andrei March 5, 2019, 10:21

        Many years ago I did read a story where science fiction ship used a binary neutron star system as a way station to boost speed when flying between an start inhabited by humans, to investigate a mysterious radio signal they picked up for another star.
        On arrival the single pilot finds an long abandoned base falling to pieces where the self repairing machines had lost part of their programming so the radio dish had gotten stuck in some kind of emergency mode sending a strong signal trying to call for help.
        It were a bit of a horror story as the pilot for a while got the idea that the animals found on this moon, which slowly were loosing the atmosphere with terraforming no longer maintained – were descendants of the original pioneers, but they turned out to be no more than animals.
        The story took a twist when he continued to explore the system and the planet in the habitable zone – but anyway here I learned the slingshot with the neutron stars where the ramscoop ship gained nearly a third of the speed of light was based on Dyson’s idea.
        But even when reading it I did wonder of such an increase in speed were possible but it now seem correct. But the question remain about tidal forces, how the ramscoop ship would not be torn apart.

  • Ron S. March 4, 2019, 14:40

    First a general observation (that I’ve made before). The ability to send a spacecraft to one of these naturally occurring (or altered) systems precludes the need to do so. That is, you must already have effective inter-stellar propulsion to get there, so it is not needed.

    The second is that you can only effectively travel to another system with similar parameters so that it is possible to decelerate. Otherwise it is only useful for flybys. If you do have another means for deceleration you also have another means for acceleration, again rendering use of the system superfluous.

    I skimmed Kipping’s paper and mentally noted a few problems. Towards the end he addresses all of them, how successfully I don’t yet know. When I have time I’ll read the paper more thoroughly.

    • Jens March 4, 2019, 21:37

      My thought on your first point: The performance requirements would be different using a ‘propelled’ starship to/from a binary system. These bodies will slingshoot anything, including very heavy vessels. The difference in requirement can be useful.
      Second: Yes. We have before us something like a ‘hub – and – spoke’ story here. Getting to/from a ‘slingshoot system’ is one story, and the transfer between them is another. Again, I am thinking that an ‘interbinarystellar’ vessel can be designed with totally different mass constraints from an ‘ordinary’ starship.

      • Ron S. March 5, 2019, 10:43

        This is really just begging the question. Here you have effective inter-stellar propulsion to get to and from these systems, a distance of 10s or 100s of light years, and these systems add value? I suppose one could contrive a justification but I doubt there could be one.

        • Jens March 5, 2019, 11:47

          I see your reasoning.

        • Brett Bellmore March 6, 2019, 13:47

          Basically only valuable if you happen to have the (mis?)fortune of being located quite close to such a pair. It seems very unlikely it would otherwise be worth going out of your way by even tens of lightyears.

  • Robin Datta March 4, 2019, 14:44

    Yes indeed, The Halo Drive!

    • Alex Tolley March 4, 2019, 16:24

      Thanks for posting this link. Very helpful in explaining the concept.

  • galacsi March 4, 2019, 17:25

    I suppose the tidal effects would be monstrous

    • Jens March 4, 2019, 21:58

      It would be possible to put some numbers on that. But what do we know about radiation in regions close to black holes?

    • Ron S. March 5, 2019, 10:30

      For stellar mass and larger BH the tidal effect is only “monstrous” well within the event horizon. However even the small tidal effect can greatly impact the performance of the proposed Halo Drive as Kipping shows in his paper.

  • David March 4, 2019, 18:07

    Thought of the Breakthrough energy requirement. I always assumed that a nuclear plant would have to be built to power the laser array.

  • Andrew Palfreyman March 4, 2019, 18:24

    From a quick calculation from Kipping’s own data, the average distance between BH binaries in our galaxy should be on order 50 LY. Although daunting, it’s not impossible to imagine ourselves bootstrapping to that level to take advantage of a “gravity web”, just as our current intent is to
    set up an interplanetary “beamer web”. Baby steps.

    A huge potential consequence from Halo Drive research would be a successful confirmation that the spin-down rate of a neutron or BH binary significantly exceeded the natural decay rate. Indeed, were we to observe this in several cases, our SETI antennae should be primed at full alert.

    • Michael March 4, 2019, 18:37

      50 ly seems too close for comfort, once formed they migrate to the centre of the galaxy farely rapidly.

      • Bruce D. Mayfield March 6, 2019, 13:33

        Err, no. The laws of gravity work on all massive bodies in exactly the same way, so BHs would orbit throughout the galaxy just like stars do.
        See Newton’s laws of motion.

        • Michael March 7, 2019, 19:26

          The migration of heavy objects to the centre of the galaxy is know as dynamic friction, I believe it was predicted by Morris. BH’s are on average heavier than other stars so tend to push lighter stars outwards. There is predicted to be tens of thousands of BH’s at the centre of the galaxy.

  • Alex Tolley March 4, 2019, 18:37

    While I haven’t read the paper yet, the video posted by Robin is very intriguing. The clever use of recycling the laser light reminds me of those old ideas of spaceships recapturing their exhausts, only in this case there is no violation of the laws of physics.

    What is not shown in teh video is how exquisite the precision of aiming the laser beam would have to be. It must skim just above the event horizon, all the while the ship is moving relative to the BH which in turn is in an orbital dance with its partner. Creating a gravity lens telescope would be child’s play in comparison to this technique.

    I am also intrigued to the estimate of the number of binary BHs in the galaxy. The large number would make for an interesting transit network. Unless they are near their destinations, travel to and from these transit points is going to have to be a much lower sub-light velocities. I have to wonder whether Bob Forward sized light sails would be a lot easier to create and push ships to fractions of c and then decelerate them (which can be done initially with the source beam).

    • Charley March 4, 2019, 20:40

      I too have an read the paper as of yet, but I read the abstract and it is extremely interesting. Essentially pass a light beam past and orbiting pair and it will increase in energy and use that as a beam to focus upon your ship to increase its speed. That’s as far as I’ve gotten but it seems like on the surface, it’s a new idea.

    • Jens March 4, 2019, 22:01

      Is there such a thing as a maximum distance between the black hole and the craft tapping its energy using a laser beam?

      • Jens March 4, 2019, 22:08

        Are we also looking at a requirement for a black hole with reasonably clean surroundings, with little ‘stuff’ in its vicinity that could obstruct a laser beam?

  • Michael Spencer March 4, 2019, 19:27

    “the maximum speed increase equals twice the velocity of the planet we’re using for the maneuver”

    For an interstellar voyage, then, perhaps to Proxima: a hypothetical spaceship could gain, from Sol, twice the relative velocity between Sol and Proxima?

    • Andrew Palfreyman March 5, 2019, 4:05

      The mass of Sol is many orders down on a neutron star or BH. Its gravitational focal length for a source at infinity is >=550 AU. Thus it’s not so easy getting a halo drive to work with such an insubstantial object as Sol.

      • Andrew Palfreyman March 5, 2019, 4:10

        CORRECTION: Replace “The mass of Sol” with “The surface gravity of Sol”

  • Andrew Palfreyman March 4, 2019, 19:39

    The advantage of Halo Drive over Beamer Drive is that there is realistically no mass limitation on the spacecraft, and so even a Jupiter mass craft travelling at a sizeable fraction of c may be envisaged. How’s that for outrageous? :)

  • Andrew Palfreyman March 4, 2019, 19:45

    Lest we not forget, Einstein’s equations contain some juicy goodies. Reading on the Halo Drive reminded me of this nice little paper from back in 2006

    • Project Studio March 10, 2019, 10:43

      Still untested at LHC.

  • Alex Tolley March 4, 2019, 23:21

    One obvious downside appears to be that the spacecraft must carry enough energy to create the beam. While the energy is supposedly more than recovered I see 2 issues:
    1. The mass of the energy supply, e.g. batteries, is going to be huge, making acceleration from the beam extremely small. This seems to me to lose all the advantages of not carrying an energy supply for light craft.
    2. The conversion losses between each beam recovery, energy storage, and beam emission may be much greater than the hoped for energy gains. I do not see any accounting for this in the paper. The only issue the paper raises is whether the medium around the BH may be less than perfectly transparent to the beam and therefore dissipate some of the beam’s energy.

    • Ron S. March 5, 2019, 10:38

      If I understand Kipping he’s saying that part of the return beam is used to power the laser, thus only requiring enough energy to get the process started. A motor vehicle with a dead battery can run fine, if you can get it started by other means. But that won’t be easy (as you note) even if it can be made to work at all.

      • Alex Tolley March 5, 2019, 14:16

        To use your analogy, the motor is powered by gasoline, the battery just to get the ignition started. But in the Kipping model, the laser provides the initial kick, then the return beam provides another kick and due to the higher energy of the blue shift, provides another kick and power to fire the laser beam again. However, unless the conversion efficiency is near 100% to fire the laser again, then the motor runs down. Today, lasers are about 3% efficient? Energy conversion using PVs are maximally around 67% (I don’t know what that figure is for laser light).

        I also think there is a bit of hand waving about catching the return beam. Kipping suggests a low power guide beam to control the laser direction, but the time delay between aiming and receiving is potentially large.

        • Ron S. March 5, 2019, 15:37

          I wasn’t arguing that the Halo Drive can work. Your points are valid. I was merely attempting to illuminate this one aspect of Kipping’s proposal.

  • Hal March 5, 2019, 9:59

    The next best thing to FINDING an ancient stellar transportation network (ala Stargate), if speculating about how it could be built in the first place…
    Next time you see a semi truck moving a steel cargo container, pause and consider that it originated thousands of miles away. But remember that people in the 1820s were moving similar sizes & weights of goods hundreds of miles using a canal network and mule-drawn boats.
    It’s ok to start slow and speed up incrementally.
    Perhaps future Earth-craft will use gravity assist from Earth, moon, Jupiter, Sun out the 8.6 light years towards Sirius B to explore the nearby stars.
    (And setup a cool Einstein-Ring telescope to figure out where to go next)
    And then take aim from Sirius B towards the neutron star Calvera, 600 light years away, to explore the rest.

  • Gary Wilson March 5, 2019, 11:56

    This is all wonderful speculation and I encourage people to continue it, but let’s also continue getting mass beyond near Earth orbit. I know you have all been waiting for the announcement that Canada has joined the U.S. in the Lunar Gateway program. We will contribute a specialized Canadarm for use at the lunar space station and a couple of billion dollars of total government investment including a program called LEAP (Lunar Exploration Acceleration Program). The money will be spread over 24 years so don’t get too excited but at least it is a small commitment and a partner for the U.S. :)

  • ljk March 6, 2019, 11:11

    If you want something named Dyson that could not only push light sail vessels to relativistic speeds but also serve as a huge habitat and an interstellar weapon if the need arises, then we should build a Nicoll-Dyson beam:




  • Michael T March 7, 2019, 4:19

    Seeing as white dwarf / neutron star / black hole binaries can propel spacecraft to relativistic velocities, they could also propel a very lucky NATURAL object to such velocities. So, could there be relativistic particles, meteors, asteroids, or even planets?

    Even (to take it to extremes) relativistic Jupiter-sized (to use Mr Palfreyman’s description) Jupiters? I assume the chances would go down rapidly with increasing mass owing to the orbital precision or number of passes needed.

    • Bruce D. Mayfield March 9, 2019, 20:23

      Also, we do see particles with relativistic velocities; cosmic rays.

  • Bruce D. Mayfield March 9, 2019, 20:16

    Michael T., Indeed there are natural objects that have been given gravitational slingshots, only not to relativistic velocities. So called hyper-velocity stars have been found exceeding galactic escape velocity. Those and other runaway stars are thought to possibly be the survivors of binaries in which the less fortunate member was consumed by a BH while the other was shot away. Rouge planets are assumed to get flung out of newly forming star systems quite commonly. Even ‘Omuamua, our recent interstellar visitor is thought my many to have been sent our way by a natural slingshot past a gas giant in its home system.

    Another natural but hair raising way to achieve high velocity is to be close to a Supernova when it explodes. Not recommend.