There are a lot of answers on Worldbuilding about how to destroy planets - just as an example - and they seem to be the method of choice for many people intent on destroying the world. My objection to this is that this is extremely difficult; moving a massive, orbiting planet isn't like picking up a marble and throwing it.

What is a feasible method to move a planet (gas giant or Earth-like; I'm interested in both) from an orbit around a star to anywhere else in the galaxy? Stellar engines exist, but they require . . . well, stars. I only want to move the planet, not its star.

I've considered things like creating large scale rockets - really big ones - and attaching them to the planet, but something tells me that this isn't realistic. In fact, using any sort of conventional propulsion on such a scale doesn't seem feasible.

So, how can I move a planet? Please, try to use some science here, although keep in mind that I'm asking from the perspective of a Kardashev Type II civilization. I don't have any timescale in mind; I'll go with what works.

  • 14
    $\begingroup$ Considering you're on Astronomy, I'd have thought you would have seen this... $\endgroup$
    – Frostfyre
    Commented Jun 20, 2016 at 19:43
  • 4
    $\begingroup$ Would it be okay to move it bit by bit and reassemble it at destination? Don't you say I'm silly, you're the one who wants to move a planet. $\endgroup$
    – PatJ
    Commented Jun 20, 2016 at 19:51
  • 2
    $\begingroup$ Go activate the Face on Mars or the Butt on Mercury. $\endgroup$
    – Devsman
    Commented Jun 20, 2016 at 20:35
  • 44
    $\begingroup$ "A sufficiently long lever and a Fulcrum to place it upon" $\endgroup$
    – Aron
    Commented Jun 21, 2016 at 3:50
  • 6
    $\begingroup$ "Simple! Change the gravitational constant of the universe." (Locally, I guess...) $\endgroup$ Commented Jun 22, 2016 at 12:30

17 Answers 17


The essence of moving a planet is just the same as altering the trajectory of a space probe - just on a larger scale.

When a probe (say New Horizons) flies by Jupiter, it steals some of Jupiter's orbital velocity. It is also possible to have the probe be on the other side of Jupiter and donate some of its velocity to the planet to slow the probe down.

With a sufficiently massive planetoid, one could have the planetoid donate energy. As described in gravity assist by the planetary society there are a number of options. In particular, the one in option A, B, C, and D:

gravity assist

In these cases the less massive object goes in front of the planet and slows down after the encounter. That slowing down transfers momentum to the planet and speeds the planet up. The faster the orbit, the further it moves from the sun.

With appropriate timing of repeated encounters it would be possible to use an asteroid or similar object to transfer momentum from Jupiter (it has 1000x more than the Earth (reference)).

It should be noted that this isn't something that can be done in a day, or year, or even century - but rather over the course of millennia. That said, it also doesn't require any fancy physics or technology. Just the right math and a lot of patience.

As an aside, look at Nasa Trajectory Search (example query) to get an idea of how often such flyby opportunities occur with minimum additional orbital mechanics. Granted, this is intended for space probe missions as opposed to a flyby round trip - but it gives you an idea that for each Jupiter flyby its a 6 year round trip.

  • 9
    $\begingroup$ So you still need the energy to be supplied by some means, and you hqve to move the other object. So how do you move that and how's it different from the original question? $\endgroup$
    – JDługosz
    Commented Jun 21, 2016 at 0:45
  • 1
    $\begingroup$ @JDługosz it doesn't take significant amounts of energy (when compared to the overall expenditures) to grab say, Ceres, Vesta and Pallas and put them on the proper flyby orbit for Earth / Jupiter. The less massive the body transferring the momentum it actually gets easier (less delta V needed to do adjustments) - it just takes longer as you don't transfer as much with each pass. But with sufficient time you can transfer momentum from the gas giants to an inner planet. io9.gizmodo.com/5923828/… suggests 1M passes needed. $\endgroup$
    – user21914
    Commented Jun 21, 2016 at 1:05
  • 8
    $\begingroup$ You have to supply all the energy. You just spread it out over more passes. $\endgroup$
    – JDługosz
    Commented Jun 21, 2016 at 1:10
  • 7
    $\begingroup$ You supply the energy to adjust the orbit of a minor celestial body. You are transferring the energy for moving the major celestial body from another one. Momentum is conserved. You are not spending millions of kg m^2/sec to move the planet but rather picking up a a small amount in a gravitational assist from Jupiter and then donating it to the Earth. And then doing it again. As part of this process, Jupiter will slowly orbit closer to the sun, though it has lots of angular momentum to spare. $\endgroup$
    – user21914
    Commented Jun 21, 2016 at 1:27
  • 3
    $\begingroup$ Oh, ok, you are moving 2 planets in opposite directions, and just need to be a middleman. I missed that, as your animation focuses on just single slingshots. You burried the lede :) if that's the main idea, with a brief mention in the middle paragraph. $\endgroup$
    – JDługosz
    Commented Jun 21, 2016 at 2:16

Diplomacy: Befriend a level 3 or 4 civilization. If you get a level 3 civilization have them find a level 4. The level 4 people open a wormhole and bing, bang, boom you're there.

Your only real hope is to open a wormhole directly in the planet's orbit and connect it to the destination somehow. Anything else would take insanely long.

Any towing beyond moving 1 orbit to another around the same star would likely kill the inhabitants, because as soon as you leave the habitable zone of the local star your planet will freeze. You would need fast transport to avoid your planet freezing, even if your planet has a molten core that will only last so long in the cold of space. Even if you do make it, your core is frozen and will take 100's of years to reheat if not more.

Additionally, if the planet has a lot of water, massive earth quakes will occur as the water freezes in its, potentially, long journey to another solar system.

You would have to harvest the energy the local sun to open the wormhole as the current theories say it requires vast amounts of energy.

Otherwise a series of star trek like transports and relay stations to get it to the destination.

A tractor beam, but it would require insane energy and the ship pulling even more. Then you would also have to tractor a part of the sun so your planet didn't freeze to death on the way. Assuming you could harness the hydrogen/helium directly energy might not be a problem. The problem is the bigger the chunk you need the more mass, the bigger it needs to be.

  • $\begingroup$ fraction of energy used in towing will be enough to heat and illuminate earth, not a problem here with freezing. Very like that diplomacy part. $\endgroup$
    – MolbOrg
    Commented Jun 24, 2016 at 19:15
  • 1
    $\begingroup$ Undetected particle are undetected for reason week interaction with matter or small numbers of them. Small changes may be, but if we stay scientists, I'll say go for it, core will not freeze. Although sun magnetosphere interaction with core etc. But I do not expect something super exciting, which is significant and could not be fixed. But yes more preparation have to be done, not a problem. $\endgroup$
    – MolbOrg
    Commented Jun 24, 2016 at 20:12
  • 1
    $\begingroup$ @DevilApple227 oil is not energy source. it is a mix of fancy chemicals barely different in energy from other chemicals. Be a big boy, like star, thermonuclear and we start talking. We talking about moving a planet, chemical energy in entire solar system is not enough to even make scratch on that task. I meant literary, in solar system there is not enough chemical energy to start that task by directly applying that energy to that task, instead of to invest it in to approaches for extraction energy from the star or a like. It is so because of gravity wells of most planets in our system. $\endgroup$
    – MolbOrg
    Commented Sep 1, 2016 at 22:08
  • 2
    $\begingroup$ The question was about moving a planet. Having its inhabitants survive wasn't a requirement. $\endgroup$ Commented Dec 25, 2016 at 13:17
  • 1
    $\begingroup$ Nitpick: Having a molten planet core lasts for cosmological time frames (the earth's core's heat is partly a leftover from back when). Planets are that big. But of course you need impractical amounts of energy to use the earth as " night storage heating", so heat is just one aspect of the answer "you can't because planets are too big". $\endgroup$ Commented Jun 16, 2017 at 16:46

In World out of Time, Larry Niven explains it well.

book cover

He had a motor in the atmosphere of one of the ice giants (I forget which) that shot the planet's atmosphere out to cause reaction. That planet was guided to pass the planet to be moved, nudging Earth with its slingshot effect. This pretty much destroyed the ice giant, but carefully moved Earth, intact.

Note that the planets did move around, substantially, even switching order, ejecting some, dropping others into the sun, smashing some together.

So why can't "nature" end up putting them where you wanted all along?

My novel answer is chaos. Nudge small rocks using a small amount of energy. That influences larger rocks to nudge their orbits. The larger rocks influence still larger rocks, etc.

You cause a dynamic instability, and through continued application of small changes, make it settle down the way you intended.


TL;DR Build a planet big space ship, move stuff with gravity of that ship, using star energy for propulsion solar sail like. Or by pressing on the planet.

As reasons to move planet are not given, so are not defined approaches to do so. So I will pick few of possible solutions. Also it have to be noted, CII may have better ways to deal with that particular subject, if let's say eLISA will lead to deeper understanding of gravity and CII may be able to manipulate that gravity which will make moving planets piece of cake, and alter all I have wrote.

So my suggestion is rather like how our today civilization with CII energy capability's may deal with that. And I ensure you, gap by energy isn't such big deal actually(It's way much closer as usually people thinks), but difference how to use it may be like steam arithmometer vs Top500. I feel like I can trow stones to Kardashev scale all nights.

Today knowledge, Sun energy

Sun energy is a lot, but not too much actually

  • Power 3.828×1026 W

Few numbers to represent it as kinetic energy, velocity mass, 1k=1000, 1kk=1'000'000 etc, relativistic kinetic energy $E_\text{k} = m \gamma c^2 - m c^2 = \frac{m c^2}{\sqrt{1 - v^2/c^2}} - m c^2$

  • 1c, just energy mass conversion: 4.3kk tonne
  • 0.99c, 0.7kk t
  • 0.95c, 1.93kk t
  • 0.90c, 3.29kk t
  • 0.80c, 6.38kk t
  • 0.50c, 27.5kk t
  • 0.30c, 88.1kk t
  • 0.10c, 833kk t

Not proud to write numbers, mostly for my personal reference, but as u may see from 0.9 and above u transport more energy than mass, and mass is just carrier for that energy.

Even if we may send everything we have at the moment, in 5-10 minutes with resulting speed 0.1c, but compared to the planet, everything we have isn't so much.

  • 1km/s, 765e+18kg

There comes difference, as result what we wish to accomplish, for which purpose we do it, etc. We are not equal interested in all 100+ elements we know, and from civilization stand point of view, importance may differ from value, we can't eat gold, but we happy to eat carbon based stuff, and as technology develops, no one can guarantee that technology valuable properties of gold will be so much important as it is now - it's subject of changes, but until we stay carbon based live, carbon will be important.

Also, carbon-based technology, wonderful (atm) strength of CNT may be very important for future tech, specially for moving planet's projects.

Option 1, take what you need, planets disassemble

When it comes to specific elements, then definitely there is a reason to mess with whole planet, not necessary do that but if civilization do not have tech to fuse elements easily(this is rather knowledge challenge then energy challenge) it may have sense. But we have force, me not thinks, me dissemble planet, hugh hugh, rrr - besides it's fun, why not.

Disassembly may be done in different ways evaporating by focusing light-energy on surface of planet (some one suggested moving planet that way, man think again ISP will not help here, just imagine what it means for a planet, just magma ball, not a planet)

It may be more gently dissemble, which is more energy efficient and more control over stuff, less mess and less after work.

But evaporating is the easy way to estimate max energy we need for the process.
Disassemble Venus will take: mass_kg*E(escape velocity, 1kg)/Power(sun, 1sec)/Seconds_in_year

(4.867*10^24 * 10360^2/2)/(3.828*10^26)/(365*24*3600) == 0.022 years or 8 days

This is rough estimation, which isn't counting escape velocity changes because of planet mass loss, but it also not counts efficiency of process, which is less then 100% because loosing energy by heated plasma trough electromagnetic waves emission. But overall I'm ok with that number.

Same for Jupiter

(1.8986*10^27 * 59500^2/2)/(3.828*10^26)/(24*3600) = 101613 days or 278 years.

I'm practically ok with that number, but I think 2000 min of my time worth to improve efficiency of process at least for 1% be more efficient, even humanity may think one year or two about that situation, before to begin some movements in that direction.

Sure, we have to recuperate our energy, probably we do not need that hydrogen cloud flying around, if we need hydrogen we may scoop it in any place in universe. So probably we will make 3 piles for hydrogen, one pile for He, and small moons with elements, one moon for each.

As we already sorted the Venus, so I expect 0.2-0.5% by mass carbon stuff with 100GPa strength to operate, which will be useful to deal with decomposition of gas giants. ref 1, 2

I'll take optimistic number 0.2% - it means 1e22 kg CNT, not bad it's already 1/7 of moon mass, so we may already start to move earth, but thinking spares time in that planet moving business. So we will use that material to cook Jupiter first.

3 Pile H and 1 Pile He, each is 1/4 of Jupiter mass, each will have radius something like 44-50k km and escape velocity 38 km/s and having Jupiter in 4 such piles - when recuperating energy, will save us 40000 days. Not bad not bad. (you may wish to play with escape velocities here

I could be satisfied, not each day you may save 100+ years of work for entry civilization, sure they could figure that for them selfs, but, u know ...

I'm not satisfied with that 60000 days of decomposition, not only because it's long in time, but because of slow start of such process, before we may start to recuperate reasonable amount of energy from placing mass in a pile, first pile will be ready in something like 10k days , and at that time we will recuperate much less then 40%, and we actually do not get something useful from that mass, it's still just big pile which we actually do not need.

We need max 1% or less of that gas giant, so max number of days which looks good is 1000 days or less, for first giant, for CI tech, but bad part is most interesting stuff is in core, and to access core we have scoop out most of GG.

Gas Giant or reason to move planets

  • Man said - before selling something unuseful, u have to buy something unuseful.

And we already have that unuseful, deal of the life, 99% of Gas Giant we do not need, and we may exchange it for what we need in star

As a body orbiting around star, planet, or gas giant already have all energy we need for such exchange.
But we as just CI civilization with a big hammer, we may need energy from the star as a catalyst for that process and compensate our losses etc. Efficiency is one limiting factor, so 90% efficiency means 10 times faster disassembly.

So exchange GG mass, for something useful from star is one of the reasons to move or change something in planet orbit.

Note about Venus scrap, snake elephant

probably most important part of that answer

I expect, and I have reasons for that(humanity lazy and smart is at least one of them), Venus scraping has been done more gently, and we have products as result, not a big cloud of materials.

After successful scraping Venus we have 1e22kg CNT, and I have to explain what I consider as my knowledge about what that actually means, and probably that is the only reason, why I write that answer.

You probably have seen that Tesla snake funny elephant manipulator, and if you search youtube for more, you will find more, maybe not best search keyword but still, elephant manipulator

And as one interesting, and most importantly, simple design, I will point that video: Robotic arm inspired by the Elephant trunk, time 4:08

What is good about the design its simple, device made of same or similar parts - simple to produce, efficient to scale its production etc.

There are other interesting use of properties of CNT this, this

this here Carbon Nanotube Muscle #2, the material itself is not important, important is the ability to make strings from it and their conductivity and strength.

Both of that principles combined(and they are not all possibilities, we already know), allows us to make 100GPa strong, very flexible manipulators.

And as CNT are very thin by their nature, such manipulators may be very thin too, and strong, and they may form more ticker manipulators.

So imagine that tesla snake but made from at least 500 times stronger material(I bet that tesla snake is weaker then 200MPa, which is the strength of ordinary steel cable), and definitely more flexible.

So imagine one unit same thickness as tesla snake but longer, 100-200 meters long, each equipment with some processor, some swarm algorithms, some sensors over surface: light pressure temperature etc - everything we may need for that unit is made from one material (maybe with some little additions of other materials, not as parts but as additive to change some properties of CNT's in desired way - but mostly 99.9% it's just carbon). And it is assembled from thin actuators.

So that one unit, which we may control by programs, with strength like space lift cable, may change shape, bend as we need, react as we need, be thick as we need (from microns to how much you have), works from 0K to up 2300K temperature, is very precise in form making-changing, is dynamic in form-shape, stiffness.

If you understand that moment, you will never wonder about howto make big constructions in space, huge ships, big thermonuclear reactors, your cubic worlds, many things considered like handwavium stuff, may be done from or with that.
If you go deeper you will not wounder about speed, anything under 1c is not a problem for you. It's not nanobots trough, it's better, stronger, it will pass any reality-check, it's real.

There are downsides too, you will begin to wonder how things may break at all, why they do not change shape, why they just make just one thing all the time, why you cant just upgrade thing like phone today, oh wait why I have to buy new phone why not just take small piece from that big chunk which plays jet at the moment and convert it to phone, who needs space suites, why people think gauss gun of any kind is wunderwaffe, why someone have to resupply something, why someone can't gather another 10kkk people and fly to some star on vacation, or make honey moon in center of our galaxy and return back 100k years later, why all will live on planets instead of comfortable space habitats made from smart material, why someone thinks pressure 1000bar is too big, why slicing 100-1000km asteroid in dust is too hard.
Planet stuff is most annoying from them all.

True limitation will be energy, and law of physics, there will stay things you can't do, taking core just away from gas giant is probably one of them, also taking core from earth size planets one of them too(but you may take stuff from 2000-3000 km deep), moon size object core will be not a problem, moon can be mined just as it is. Slicing earth size planets is not a problem, by removing upper layers - layer by layer.

Tool, Space swiss knife

Main valuable resource, from decomposition of Venus, is 1e22kg of active meta-material, actually it is our tool, which have to help us exchange 99% of mass Jupiter to stuff from star to make even bigger tool.

Tool consists of parts with different sizes and ticklishness, typical muscle let's say 1km long, square 10x10cm (I'm lazy mess with Pi, or any complex form), density 1 t/m3, strength 50 GPa, they may stick together with good seal and slide like linear motor, be reprogrammed to other form with accuracy 0.1mkm
They may store energy 10MJ/kg at least, as mechanical energy like spring, and release it like capacitor(fast if needed, mechanically or by generating electricity), with approximately 0 storage discharge.
They may store and convert electricity to kinetic energy and back.
They may conduct electricity, they may regulate temperature like peltier modules probably close to theoretical value.

I assume 100% efficiency, but even if it is 50% this is not a problem, but I expect it to be 90% and above, like high power electric motors efficiency.

1e22kg it will be 1e19 of TMU (typical muscle unit), it is also 1e19 km long cable 10x10cm, which is 66,666,666,666.7 a.u. long cable, or 11.5x11.5km cable with 5 a.u. length.

and all that mass orbiting on orbit of Venus with rest of Venus scrap, which is 99.8% of previous Venus by mass, which may be used as reactive mass for that tool, with wide range of ISP values actually, this linear motor sliding of TMU is quite handy.

Current form is probably ring-toroid(venus like orbit or close to that), to keep tool less dense, and to avoid need to wait 8days of star work to unfold it(with all that 99.8 not so much useful stuff) to something useful. But tool alone may be moon size(which is 7 times more we have atm) at least, and unfold pretty fast, something up to mars sizes is ok for dense and compact storage(everything with less then 3.3km/s escape velocity, which is around limit of static energy storage capability, is ok for tool, but it could be much bigger then that with other types of folding it). We could exchange venus scrap first, but we do not have to, and we rather will have metal elements(everything above He), then loose them, because they may be used for transmutations by neutron capture(forgot process name, something like Nuclear transmutation), it's specially useful if you have star as neutron source and ability to efficiently expose material to it, some isotopes of ordinary materials like Fe as example, are more valuable then other isotopes of same material, also it can be used as passive protection layer preventing degrade our Carbon based material, specially if we wish to dip some parts of our tool in to star inner, and bunch of other reasons).

With 10MJ/kg storage capacity we may store 260 sec of star energy, not bad, but it may store way much more than that (as kinetic energy).

Because tool consists of elements, which may slide against each other(let say 1m/s, not top speed but reasonable speed of sliding), flex on command, we may separate them inside in 2 rings, two sets of MTU.

Energy stored in Venus motion is 90 days of Sun work.

Gravitational potential is: $U = -G \frac{m_1 M_2}{r}\ + K$

Difference of potential energy between Venus orbit and Earth orbit, will be:

1.98855*10^30 * 6.68408*10^-11 / (108*10^9) - 1.98855*10^30 * 6.68408*10^-11 / (150*10^9) = 344597744.4 J/kg

To move Venus to a different orbit, Sun has to work with 100% efficiency :

  • to Earth orbit, 1 a.u. - 51 days
  • to Jupiter orbit, 5 a.u. - 155 days
  • to fly away - 181 days

To move Jupiter:

  • to Saturn, 10 a.u. - 4693 days
    with proper tool we could form binary system of them, and refine at least one body pretty fast, saving some years. Or refine them both in 3th body. But we have to have tool for moving GG first, but with such tool we could refine them in place.
  • fly away, 9779 days
  • to Venus orbit, get energy 60847 days of sun work, although we can't do it now, but that's interesting number, ~150y of star energy, possible number for moving hot GG to more distant orbits.

Sun energy work days to change venus orbit

Just as notes:

  • We can change inclination, by splitting ring again in orbit plane, if we have to
    Venus inclination is 3.39 deg, Jupiter 1.3 deg, Saturn 2.49 deg
    and because 99.8 percent of Venus is just scrap, which we use as we need without much care about, we may make small moon perpendicular to ecliptic - I notice that just for ease of understanding, we do not have to loose any reactive mass in that case, we need just energy, and compared to other task it's rather small. But yes, we have to respect momentum and impulse conservation.

  • Friction between rings or any other energy loss isn't big issue, surface area of tool is pretty big, and if it stacked such way as just disc, it may dissipate 100% of sun energy at temperature 900K, which is 627 °C. And this is without other 99.8 mass venus available to use as parts of heat dissipation system.
    Actual friction and energy loss is on level of good air bearing or better(which they actually may be, but this isn't only option). For those who isn't familiar with air bearings u may have to look at this and this as examples

  • As we have 2 rings rotating in opposite directions, in earth orbit it will be 60km/s difference (30 in one direction, 30 in another direction), as we set TMU slide speed to 1m/s (to be suitable for different approaches and implementations) it means 60000 layers separation between two main rings, as TMU is 10x10cm it means that area is 6000m wide intermediate ring

  • I took air bearing principle because that way implementation does not depend on the internal structure of TMU and that way it is easier to refer to today's technologies, but this isn't only option.

  • layers could be thinner actually nothing stops us from using 1mm tick layers, or 0.1mm tick layers, this is more question how strong we wish them to be, and how much sliding force we wish to have.

  • There is no centrifugal stress from rotating rings, they orbiting, but just close together, so zero force for them. There is no difference (practically) in which rotation direction to orbit, just in case.

  • only part which is not orbiting properly are separation layers, but forces are small, 6km wide separation layer (if assume it not rotates at all) on venus orbit will press on inner ring with pressure 6800 Pa, on earth orbit 3500 Pa, so actual pressure between layers will be less then 1Pa typically.

  • with 1,5,10,20 a.u. ring radius, we still may make enormous amount of layers, if with TMU(10x10cm 1km long) we have 66'666'666'666 a.u. cable to play with. As TMU consists from less smaller strands, we may split it in smaller units or build bigger units from them - so it's just typical unit we operate at the moment.

  • gravitational influence from other bodies, may be compensated by playing with layers, and counterweight strands. Also this is one of the ways to tune star system, and affect orbits of all bodies in system at once.
    it may be a way to convert potentially unstable (for billion years) system to stable one. Way to move planets actually. But long way, not efficient.

  • I do not talk about micrometeorites, asteroids etc - you may guess, not a problem (just collect them , omnomnom)

  • We may have an elliptic ring, changing orbital velocity along orbit is not a problem with sliding strands. We may convert circular ring to elliptic, at least several ways to do so. One by splitting rings.

Rings arranged something like that, black are rings: Jupiter Sun exchange, ring configuration

GG scrap, lift setup

  • To scrape Jupiter for 10years or less - we have to have mass transfer something around 60'185'185'185'185'185'185 kg/sec or 60kkkkk ton/sec

  • Elliptical orbit of ring on Jupiter to Venus orbit, have period something around 5 years, so first year or two we will be kinda limited to Sun power, which allows us to lift 2.16e+17 kg/sec or 216kkkk ton/sec

  • to scrap Jupiter in 10 years, we have to lift 6e+19 kg/sec

  • Orbit velocity at Jupiter orbit is 13 km/s, orbit period is 11.9y

  • would be GG on earth orbit, it would make operation easier

  • Sphere from TMU, one layer tick (10cm), approximately 318km radius under 1 bar pressure Hydrogen+, will be 12'170'840'439'815'458 kg of Hydrogen mix, or 1.2e16 kg. TMU mass will be 1% of Hydrogen mass. I will refer that sphere as Spoon Unit (SU)

  • 10y scrape plan means approximately 500 Spoon Units per second to lift

  • some gravitational effects are omitted because they can be compensated, and it's not only one way to do the job.

  • originally I wished another approach to describe, but this looks simple to explain.

  • challenge is big, and the tool is too small, so 10y plan have to be smarter then I describe.

Plan is simple, we will make a balloon from Jupiter. 1e22 kg TMU is enough to cover entry Jupiter with 23.5 km tick layer, at his 1bar level, it will squeeze Jupiter up to 2347 bar pressure inside that balloon, just by gravity force. With using 99.8 percent of Venus aka dead scrap this pressure could be higher up to 500 times and probably more. We do not apply force by tool, it's just gravity of Jupiter, and our limit is structural strength of TMU, which is around 50GPa or 500'000 bar.

I'm ok with 1bar pressure near TMU shell, so we need 10m tick layer or 100 layers of TMU, this will be a not perfect sphere, but we ok with that because of the flexibility of our shell and mobility of our TMU units so we may dynamically compensate what we have to compensate.

For that shell, we have to allocate approximately 1/2300 of our tool, by mass.

The shell may be formed in different ways, hm that's stupid but like that, I wished guys did that better than that, but... it illustrates.

After we formed shell over-around Jupiter (it will not fall it's just replacement layer for what was there before(part of atmosphere)) we have 1 bar pressure inside one side shell, and 10m over on other side of shell, we have vacuum. We do not apply force, we just chill on hydrogen couch.

On vacuum side, over that side we may wish to form 2 rings, same 3 layer structure, plane of this set have to be same as plane of main ring, and they will stiffer our balloon, pre-stretch let say equator region, to allow us to lift SU units to vacuum side (E=mgh style). and to accelerate SU units to orbital velocity, same principle as Launch loop but instead single rotor there is 2 rings and intermediate layer.

1 SU with Hydrogen, near hull, will weight 1.2e16*23.12=2.8e17 N, and to be able to withstand that force cable should be approximately square 2.5x2.5 km (everything could be done inside the hull itself, by forming same structures or proper equivalent of them, it can be done in different ways)

We open hole in the shell, pressure blows our SU up, like glass blowing process, the bubble moves away and we begin blow next - continuous process.

3layer ring system which accelerates bubbles one ring in one direction, another in opposite direction.

layers have to be pretty big, to be strong enough to to be able accelerate pretty massive SU, but we may do that with smaller SU's if we have to.

from 1% of 1e22kg active mass with resulting volume of 1e17m3 we may make 15x15km ring with radius 75000km, we also may wish to add all(or significant part of it) venus scrap to act like rotor in launch loop, we need just inertia mass to distribute stress over ring.

Accelerating SU at 1m/s2 is pretty reasonable value. So at max productivity there will be 500*59000 SU units, the mass of active material used for that will be 35% of our tool.

  • 35% of TMU's for accelerate process
  • 5% for Acceleration rings and reinforcement rings
  • 50% of scrap for rotors for acceleration rings, and for rotors of reinforcement of shell.
  • 0.05% for shell around Jupiter

I reconsidered approach slightly, will be below, but I leave this part as it is, as possible use case

After acceleration to orbiting velocity's, we attach SU to ring on close to Jupiter orbit. There will be different proportion in both directions, because of 13km/s Jupiter orbital speed, and we may wish to keep the momentum of the ring.

Refine Jupiter mix, problem

Intensive disassembly of Jupiter is actually challenge, there is set of problems: tool is too small, some processes still needs years to accomplish like cooling SU's, separate mix in to components, transfer closer to sun. Although some problems may be solved, I wish more general overview of the process, without going deep details of possible solutions.

To imagine what intensity of process is that 10y plan, we have numbers to tell us that. To make something near 10y plan, we really have to blast Jupiter, just nonstop blasting. Sending 500SU/sec, with content mass 1.2e16kg each, means each 1m2 of 75000km radius sphere, should have flow with velocity 970 m/s at 1 bar pressure. SUvolume*500/Jupiter_surface(75000km radius) = (320000^3*4/3*3.14*500 / (75000000^2*4*3.14)).

As we have regions where SU are forming it means compression(just adding more mass over shell top in that region) will make pressure and density more and thus flow speed less.

Probably we have to use entry equator region to fill SU, and this region will be our acceleration ring, which we will divide to SU's later. So it kinda shell flows to equator and accelerates perpendicular to flow, so on equator we have most energy it needs - everything is preloaded with venus scrap. Pretty much is happening there in that process, but I ran out of space to describe.

  • we do not have to stress entry jupiter, it's enough just to bend equatorial part to desired pressure (any pressing we do by just placing more weight in that place. That ring starts somewhere 300km deep under surface, Scale height 27 km, pressure 70000bar)

jupiter acceleration rings

Growing Active Material is critical for the whole process. Each SU content contains 0.3% of CH4 by volume or 2.4% by mass, and SU itself weights 1% of that content. So when we manage to separate CH4 from that mix, we may make 2.4 new SU for each SU we send down to venus orbit.

Extraction should be done directly from Atmosphere of Jupiter by shell AM. We have to extract building material until we close cycle, and SU will begin to return back for reuse. After that it can be done anywhere underway. That way we may have a continuous growing flow of SU up our max needs.

1st priority to grow tool first. Separating CH4 may be done in many ways but over all just usual gas separation, but on large scale with AM. Good about that, more we get, faster we get next portions and we have more than enough AM for such task.


I'll not describe mass exchange in details because the topic is bigger then I have already written.

  • local magnetic fields in sun spots are 0.3Tesla(probably not extreme case but above average field which is 1/10000 T, twice of earths), we do 2T with current approaches, and we definitely can do more with stronger materials.
  • there are some suggestion for probes and support-stabilizing strucures for thermonuclear reactors which are inside that reactor surrounding by plasma, they are protected by their own magnetic field. Same way we may protect parts of our tool inside sun.
  • as we transport mass from Jupiter, we have disproportion of momentum in tool(because of jupiter's momentum), mass exchange with sun is way to compensate and move that proportion to desired equilibrium.
  • Sun scooping may be done with same 3layer ring system
  • Because surface area grows like x^2 and volume like x^3, bigger part is, longer it may stay in hot area. Despite passive materials, with moving strands inside part, we may do much better and faster heat distribution in part, also we have plenty of Jupiter scrap to use too, as protection gaseous layers, if we have to.
  • Probably we cant dip to deep. Density of suns photosphere is 0.0002kg/m3 and that's not bad actually, pressure is also not very high Sun.

  • Energy flow in photosphere is 68MJ/sec/m2, but with AM we may have entry ring inside photosphere, and separate heavy isotopes just directly there. (sure with some cooling setup outside sun, connected to that ring)


I hope, at least partially, I have answered OP's Q, even if I skipped some details, because of A size limitations.

The whole concept heavily uses Centrifugal force direct or indirect(like orbiting bodies them selfs) aka inertia - so you have to understand and be familiar with Orbital ring, Launch loop, Space fountain, Orbit

Playing KSP help to understand some basic principles of orbital mechanics, fly safe. There are other games with some realistic orbital mech, and that's good.

C. Clarke is genius, Gerard K. O'Neill is great.
Special tanks for Google and Internet, without your help guys, writing would be impossible.

xkcd I Do Not Laugh Anymore, Ever, Thank You Very Much C.C.

  • 2
    $\begingroup$ Gratz on longest answer on WB.SE! $\endgroup$
    – Anoplexian
    Commented Dec 20, 2016 at 18:39
  • 1
    $\begingroup$ @Anoplexian woow, nice)). Was forced to trim it, it is about max length to post on SE, less then 50chars left. I guess I trimmed it about 25% to be able to post it. $\endgroup$
    – MolbOrg
    Commented Dec 20, 2016 at 18:58

Stellar Laser Propulsion

The problem with strapping any sort of engine, such as a chemical rocket or an ion engine, to the planet is coming up with enough energy to actually move a planet. Luckily, planets tend to be found close to sufficiently large sources of energy to move them: stars.

The trick is to figure out how to use the energy of a star to move a planet. To do this, we'll take inspiration from a fairly novel form of propulsion engineers have been playing with on Earth: laser propulsion, which typically works by focusing high-powered lasers on a reaction mass. The lasers rapidly heat the reaction mass, which vaporizes and produces thrust. In this sort of system, there is no energy stored on board the vehicle, either in the form of chemical potential energy, like with chemical rockets, or in nuclear energy, which is generally used to provide the electric power that ion drives use.

What we'll do is construct a Dyson web comprised of satellites with solar panels and lasers attached to them. These satellites will convert all of all of the electromagnetic radiation given off by our star into radiation we can direct towards the planet. All of these lasers will be focused using a lens or mirror array onto a single point on our planet, at which point the planet's surface will begin to vaporize from the heat, emitting a jet of high energy particles which will act as a thruster, pushing the planet towards wherever you want to go. Energy-wise, it takes the sun about a hundred days to produce an amount of energy equivalent to the kinetic energy of the Earth, so it will take a few years to accelerate the Earth to escape velocity. At that point, it will caroom across the galaxy towards its destination, where you'll have another satellite array waiting to slow it down at the end of its journey, assuming you've aimed everything right.

The planet may require some cooling once it reaches its destination.

How much mass will we lose? The Earth travels at around 30 km/s, and its escape velocity from the sun is 42 km/s at a distance of 1 AU. This gives us a $\Delta v$ of about 12 km/s. Ablative laser propulsion has a specific impulse of about 1000s, which we can plug into our equation for our fuel fraction: $e^{\Delta v/-v_e}=m_f/m_0$ to get a final mass fraction of about 30%, meaning that we will ablate away 70% of our target planet's mass getting it up to speed.

That being said, the more we can focus the light of the sun onto a small area, the higher the effective exhaust temperature will be, resulting in a higher specific impulse for our thruster and a smaller mass fraction. If we can get the specific impulse up to 10000s, our final mass fraction would be 89%, meaning we'd only need to ablate away 11% of the planet's mass to get it out of the solar system. The value I used in my initial calculation is based on the best that human society has achieved until now using ground-based laser systems, so it's entirely possible that our Type II civilization could achieve $I_{sp}$ values an order of magnitude higher.

  • 14
    $\begingroup$ Randall Munroe has gone into the math of why this would be a catastrophically bad idea: what-if.xkcd.com/141 $\endgroup$ Commented Jun 20, 2016 at 20:36
  • 3
    $\begingroup$ Randall Munroe also goes into why the whole mirror setup is going to be a lot less effective than you might hope. Sure, it'll work way better with sunlight than with moonlight, but you're still only going to be able to dedicate a tiny fraction of the sun's power output toward propulsion. $\endgroup$ Commented Jun 20, 2016 at 22:08
  • 6
    $\begingroup$ Push a second planet you don't care about so it pulls the first one. $\endgroup$ Commented Jun 20, 2016 at 23:09
  • 2
    $\begingroup$ Note that "laser bees" is a project funded in part by grants from the Planetary Foundation, to do exactly that to asteroids. And I help fund the Planetary Society, so... I feel like this is something I'm workin on. :) $\endgroup$
    – JDługosz
    Commented Jun 21, 2016 at 0:54
  • 1
    $\begingroup$ @user2357112 Randall Munroe also goes into why user21914's idea is not going to work what-if.xkcd.com/146. It almost makes me think that we should cross post this question in its entirety to What-If. $\endgroup$
    – Aron
    Commented Jun 21, 2016 at 3:58

Build an Alcubierre drive-system around the planet, assuming you can build them that big, and then transport the planet to your preferred destination. Essentially an Alcubierre drive moves a volume of space, and if that volume of space just happens to contain a planet so much the better.

There will be a few technical problems as the planet will retain its state of motion inside that volume within the Alcubierre vessel, but a civilisation this advanced should be able to cope.

Recent thinking on Alcubierre drives suggests they work best below lightspeed, so the planet most likely will relocated at sublight velocity.

Gravity tractors using asteroids have been proposed to move the orbits of planets in the solar system. This technique might work for the interstellar transfer of planets, but it would be hellishly slow. But there is better alternative for a gravity tractor. Namely, neutron stars.

Observations have been made of runaway neutron stars. Let's assume our advanced civilisation can construct sufficiently large wormholes, say, with a diameter around 100 kilometres. Using a series of wormholes it might be possible to shepherd the neutron star close enough to the planet to act as a gravity tractor.

This would involve multiple close passes of the neutron star to the planet. The neutron star would have to emerge from the mouth of the wormhole near the planet and move away from it in the direction of its destination. The planet will accelerate due to the neutron star's gravitation.

On approach to its destination, the neutron star will go through repeated manoeuvres to decelerate the planet. This means at the halfway point in its galactic journey.

This arrangement will require a reasonable number of wormhole mouth pairs attached to space vehicles to get into the correct positions to enable the right passes of the neutron star relative to the planet.

The motion of the neutron star will need to be reoriented so it is moving in the vector towards its destination. Since neutron stars are extremely robust objects extreme measures can be taken to do this. Dropping extremely large amounts of hydrogen leads to thermonuclear explosions on the surfaces of neutron stars and this might be used to steer the runaway neutron star to line it up correctly.

My third method is purely hypothetical. Use a Herman Bondi-style of diametric drive to move the planet.

This will involve placing a large positive mass and a large negative mass on either side of the planet. The masses in question will be relatively large compared to the planet's mass. I will assume that the two masses will be drawn out of the quantum vacuum and since they involve equal masses of positive and negative mass-energy the nett mass-energy budget will be zero.

In this configuration, both masses will accelerate away in the direction of the positive mass. The theory behind this can be found at https://en.wikipedia.org/wiki/Negative_mass and that of the diametric drive at https://en.wikipedia.org/wiki/Breakthrough_Propulsion_Physics_Program#Diametrical

To slow down the planet and its diametric drive the configuration of the two masses is reversed and the system will undergo deceleration.

It is nice to notice that the two masses don't need to be equal for this drive to work. This would be the fastest method of the three I have suggested for transporting planets across the galaxy. A diametric drive planet transport system could reach near-lightspeed.

  • $\begingroup$ I like this answer. But the problem is that we cant build an Alcubierre Drive yet and even if we could we couldn't power it because we'd need massive amounts of energy, AND that energy needs to be negative energy! So to build a planet sized one would be... well you get the idea. However, its very clever. So with a little handwavium... why not? However, I don't like the sub-light speed issue. I feel like that creates planet wide problems during the travel that would be worse than the solution. $\endgroup$
    – Len
    Commented Apr 11, 2018 at 16:31
  • $\begingroup$ @Len I am entirely familiar with the energy problems. For the purposes of the answer I have assumed they were solved. The task of moving a planet was for extremely advanced civilizations. The concepts were based on current science, theoretical concepts but not wholly disproved. Sublight travel problems? True, but they're solvable. Three solutions to one question too, what a waste. I appreciate your comment. $\endgroup$
    – a4android
    Commented Apr 12, 2018 at 6:03
  • $\begingroup$ well let me ask... what sort of physical effects do you think would be happening to Earth if it were "moved" at near light speed? $\endgroup$
    – Len
    Commented Apr 16, 2018 at 18:44
  • $\begingroup$ @Len A planet moving in Alcubierre bubble, even at near lighspeed, it will cool down in interstellar space. It won't undergo time dilation, so transport a planet 100 light years at near lightspeed it will take 100 years. The population will need to move into bunkers to survive the trip. If Alcubierre drives were FTL, especially if it were very, very, very fast, then cooling might not be a problem. $\endgroup$
    – a4android
    Commented Apr 18, 2018 at 8:31
  • 1
    $\begingroup$ @Len Rebuilding a planetary biosphere should be in scope for a civilization capable of moving planets. A gigantic wormhole would be the way to go. getting the orbit right at the destination might be tricky. There must be a way of doing it right. Probably, all a matter of orbital mechanics. $\endgroup$
    – a4android
    Commented Apr 19, 2018 at 12:39

For something like this, it'd likely be more feasible to manipulate spacetime to get your planet where you want it to go, rather than moving the planet itself. If the planet in question has inhabitants, this becomes even more of an attractive solution, as you'd have to protect life on the planet while it's being moved. Folding space effectively eliminates the hazard of a long voyage through the deep black.

  • 1
    $\begingroup$ How feasable is it to manipulate spacetime? How do the numbers actually compare? $\endgroup$
    – JDługosz
    Commented Jun 22, 2016 at 12:12
  • 1
    $\begingroup$ @DJlugosz It really depends on how advanced the technology is. For instance, currently on Earth, some physicists do believe it to be possible, though they have no idea on how to do such a thing. Essentially you'd have to be capable of opening a stable wormhole larger than the diameter of the planet you're moving, long enough for the planet to pass through completely. You'd then have to target the end of it to be at the exactly correct destination vector, so as to not disrupt the orbit. $\endgroup$
    – X0r
    Commented Jun 22, 2016 at 16:43
  • 3
    $\begingroup$ I don't think you made a case for "it'd likely be more feasible to manipulate spacetime". $\endgroup$
    – JDługosz
    Commented Jun 22, 2016 at 21:12
  • $\begingroup$ I downvoted this for the same reasons that @JDługosz mentioned. Unfortunately, the ideas you're talking about are really speculative. $\endgroup$
    – HDE 226868
    Commented Jun 23, 2016 at 22:02

Use dynamic compression members - basically a stream of high velocity pellets transferring momentum (preferrably from the Sun). Description and calculation is in the article Paul Birch: How to move a Planet. In: J. Brit. interplan. Soc., 46, 314 (1993), available online at http://www.orionsarm.com/fm_store/MoveAPlanet.pdf

Summary: by transferring momentum from the rotation of the Sun, it is possible to move e.g. Venus to Mars orbit in decades, given enough pellets, quite within established science (if not technology). Of course, moving a planet somewhere else in the Galaxy is considerably more difficult and time and energy demanding.


If you're not concerned with killing everything on the planet, the idea that pops into my head is to throw things at it. Instead of turning the rockets of a spaceship on at the apoapsis and periapsis, find a rock that's big enough to have an effect and make it collide with the planet at the apoapsis/periapsis. The issue with this idea is that anything big enough to matter would also break the planet so badly I don't think it would count.

Second idea: aim for a near miss. Throw the biggest thing you can attach rockets to at the planet, but miss by a very small amount. Essentially the same as a gravity assist on a vastly different scale. This could work if applied strategically (mostly at the apoapsis and periapsis) over many of the planet's years. Your civilization could place a number of super heavy spaceships into the same orbit as the planet to give a gravity assist regularly. With this setup, acceleration caused by rockets could be very gradual because the ships have most of the year to adjust their orbits before they are near the planet again. Over a long period of time, you shape the orbit of the planet into a very irregular ellipse pointed so that when it finally reaches the tipping point and becomes a parabola, it will be headed toward your intended destination.

  • $\begingroup$ So you'd need to send a massive enough "thing" at the planet to move it. I'm thinking in the order of the moon. So how do you move this thing? Just rockets? $\endgroup$ Commented Jun 20, 2016 at 20:24
  • 1
    $\begingroup$ One slightly too close and... Your near misses better miss. $\endgroup$ Commented Jun 20, 2016 at 20:56
  • 1
    $\begingroup$ "If you're not concerned with killing everything on the planet" Considering that the OP is trying to move a planet without also bringing it's host star along, I'm going to assume that there's no concern about killing everything on the planet. Once the planet gets to wherever it is going, it's going to be a large frozen rock. $\endgroup$
    – Ellesedil
    Commented Jun 21, 2016 at 0:08
  • $\begingroup$ The "thing" that'll move the earth out of orbit will probably shatter the planet into pieces. It's like trying to make a basketball move with another ball. You either need a equal or bigger ball, or the smaller ball will need to move much faster. However, when it gets small enough and fast enough, it'll act like a bullet and penetrate it. $\endgroup$
    – Nelson
    Commented Jun 21, 2016 at 6:09

Hyper-density Black holes

Considering the fact that the weight of Earth is over a billion billion tons. Yes, I said Billion billion. There is no realistic way that it can be pulled. Let's look at gravity then. Using a super collider-like device, small hyper-dense black holes can be used to use gravity to pull the planet.


  • It is simple and easy to do.
  • It can be done with modern technology (theoretically)


  • The black holes may attract unwanted celesial bodies.
  • It will be extremely difficult (if not impossible) to do this with planets with moons or rings.
  • Black holes can be very unstable
  • It will take a long, long time.
  • $\begingroup$ Why not use a comet or moon of the same mass? $\endgroup$ Commented Jun 20, 2016 at 20:49
  • $\begingroup$ @DonaldHobson wht moves the moon or comet? $\endgroup$
    – TrEs-2b
    Commented Jun 20, 2016 at 20:53
  • 3
    $\begingroup$ A black hole of a given radius always has the same 'density', defined as the ratio of mass to the volume enclosed by the event horizon. More mass will result in a bigger black hole. The black hole will also be affected by the gravity of the Earth, so even if you create one big enough for its gravity to pull the Earth, they'll orbit around a common barycenter, rather than the black hole pulling the Earth across the galaxy. $\endgroup$
    – ckersch
    Commented Jun 20, 2016 at 21:42
  • $\begingroup$ @ckersch ah, physics is not my specilty $\endgroup$
    – TrEs-2b
    Commented Jun 20, 2016 at 21:43
  • 4
    $\begingroup$ That begs the question of how you pull the black hole. $\endgroup$
    – JDługosz
    Commented Jun 21, 2016 at 0:56

Build Shkadov Thrusters to move the planet. That’s right, Stellar Engines work on planets too.

You build mirrors in space that are held within the gravity well of the planet. The mirrors reflect the energy escaping from the planet. My calculations based on lower estimates of reflected solar energy comes up with an acceleration of 1.748 * 10 ^-8 m/s or .55 m/year. This is a slow acceleration, but it is constant. You could also increase the acceleration rate with the addition of ground based radars. Also, heat radiating away from the planet adds to the acceleration, but I could find no accurate measurements for earth’s radiant heat.

Bonus 1: Usable with current tech.

Bonus 2: You can still use nuclear weapons if you want.

Calculations: 174 pettawatts solar energy in * 30% reflected from the atmosphere = 52.2 pettawatts. 52.2 pettawatts = 5.22 * 10^16 watts. Mas of earth = 5.972 * 10^24 kg.

Watts/kg = m/s (then we double that since mirrors provide double the thrust [don’t ask me why the reference video said so])

1.748 * 10 ^-8 m/s


You don't. At most you scoop off the atmosphere and biosphere, soil, some oceans... One rocky core is much like another but it contains most of the mass. If sending the planet interstellar expect million year transit times and a very cold planet. (this is if you want more than a ball of molten rock.)



Since you are not in a hurry, you can increase, or decrease the attractive force of the star on the planet. Changing gravity is though, but creating ions that generate a very similar force is easy.

I will now assume you want to get your planet closer to the star, it is easy to do the opposite and have the opposite effect.

The plan is to turn the star into a big positive electric charge dump and the planet a negative one. Why not the opposite? Solar winds! Stars eject part of their mass at space all the time, so you don't want to trow anything light at them.

So, you build a satellite that launches a stream of very heavy ionized nucleus at the star. That same solar wind will try to get some undesirable electron into your stream and foil your world moving plans. To avoid that we will probably need a chain of satellites with which to shield the stream by clever use of magnetic fields (made easier because you know from which direction the particle wind comes).

Still the same annoying solar wind will cause us more trouble. By moving particles from the star to the planet we can get a dreaded electric current between them even if the whole stream is perfectly protected.

The solution? Making our ion stream massive. The stream should move more change than the solar wind can counter. The increase in total change inside the star might influence the amount of wind expelled, but it will go in all directions and the extra amount that arrives on the planet should be small.

In short: place an array of satellites between planet and star guiding and protecting a massive stream of super-heavy futuristic nucleus into the star.

After thought: Differently from the atmospheric engine in other answer, this stream is not massive enough to move the planet by action and reaction in any appreciable manner.

If the solar wind is really powerful, then the charge buildup will be canceled and the stream will actually pull the planet away from the star very slowly.


Antimatter-based propulsion. When antimatter comes in contact with matter, both annihilate, transforming all of their mass into ridiculous amounts of energy (mostly gamma radiation). The testing of a similar engine is described by Stanisław Lem. Build an antimatter engine outside of the atmosphere, anchored firmly to many spots on the planet so that it doesn't dislocate continents or anything like that. It will be shielded from the side of the planet. It will propel the planet by pushing the shield, and by Newton's third law, like a next gen jet. The energy absorbed by the shield can be then used to heat up the planet, to compensate for moving away from the planet's sun.


  1. Simple linear propulsion.
  2. No need for external objects.


  1. Antimatter has to be somehow produced and stored.
  2. Uses up matter and antimatter.
  3. Shoots a beam of death into space.
  • $\begingroup$ So where do you get the energy to make the antimatter? Why not use that energy directly for the engine? $\endgroup$
    – JDługosz
    Commented Jul 16, 2016 at 4:36
  • $\begingroup$ Because you can store antimatter, using it similarly to a battery. Produce antimatter using solar energy, atomics, whatever floats your boat, store it and then use it for the engine. $\endgroup$
    – Deuxis
    Commented Jul 16, 2016 at 20:13
  • $\begingroup$ You can use the energy incrementally to change the momentum of the planet, too. So why store it? I also don't follow: if the rocket is anchored to the planet, why do you push on the shield? And doesn't tge planet's rotation make it difficult to point the rocket in the right direction? By “anchored” you mean to the crust? How much force could that take without wrecking the world? Why does the shield heat up? $\endgroup$
    – JDługosz
    Commented Jul 16, 2016 at 22:50
  • $\begingroup$ It's like a capacitor, allows you to accumulate stored energy at your own pace. That's firstly. Secondly, you also need a way to use the energy, and that engine provides that. It pushes the shield, because all the gamma radiation spreads in all directions, and upon hitting the shield dumps some of the energy into it, but of course most of the pushing comes from newton's third law. Finally yes, the planet's rotation does make it hard, but you can do alternating engines or one that rotates in the other direction. And about force, use any amount you want. You control the antimatter usage. $\endgroup$
    – Deuxis
    Commented Jul 16, 2016 at 23:15
  • $\begingroup$ The shield heats up because it gets hit by a lot of energy via gamma rays, and the most common byproduct of absorbing energy is heat. And about force... I don't know how much, but without using relative transport like wormholes that problem exists anyway, no matter which method you use. You might as well reinforce the planet if that is a problem. $\endgroup$
    – Deuxis
    Commented Jul 16, 2016 at 23:17

The basic principle of moving an object by thrust is the same no matter the size of the object. Just the amount of force you need goes up. With enough engine power, you could potentially move a planet.

"The Wandering Earth" is a short story by Liu Cixin that has this exact topic and reads at the believable level, i.e. no fancy super-tech, wormholes, unobtanium.

In the story, gigantic rocket motors are installed on the planets surface, and fed with entire mountains (so it appears they are some kind of fusion drives or matter-energy converters). It takes decades for Earth to leave orbit, and a century or two to exit the solar system. There is actually a "braking period" where the engines first stop the rotation of Earth, so that they can then provide forward thrust.

  • $\begingroup$ The rocket equation would still apply & doesn't depend on how fast you accelerate, just what your ultimate change in velocity is. If you only use a tiny fraction of the Earth's entire mass (implied by only using a small fraction of the Earth's crust like mountains, keeping in mind that the entire crust is less than 1% of Earth's total mass), your fuel mass is going to be tiny compared to your payload mass, so the change in v will be small even if exhaust velocity is light speed. $\endgroup$
    – Hypnosifl
    Commented Jun 1, 2019 at 22:23
  • $\begingroup$ Let's be generous and say you convert 1% of the mass of the Earth's entire crust to fuel (which would be a lot considering the crust ranges from 5-70 km depth, converting all the mountains on the planet wouldn't get you close) and that the whole crust is 1% of the Earth's mass, in which case 0.0001 of Earth's mass is turned into fuel and expelled & after burning that fuel the Earth is 0.9999 of its previous mass. If the exhaust goes at the speed of light, c, then the shift in velocity is c * ln(1/0.9999) = 0.0001c = about 30 km/s, while escape velocity from Sun at Earth's distance is 42 km/s. $\endgroup$
    – Hypnosifl
    Commented Jun 1, 2019 at 22:37

lets keep this simple, a rogue planet, flung from its own star in a supernova, passes through your planetary system. the gravity of the planet is strong enough and it passes in just the right place to alter the course of your planet outside the gravitational influence of its own star. what now? well... it simply floats away i guess... goodbye planet.. we'll miss you :(...

now i see you mean to move two planets, one basic idea is an eruption of the star, in the hope that both planets are far enough away that they dont disintegrate. another idea is a phenomenon called a "white hole" on a small scale, its never been witnessed but is possible. it occurs when so much energy is forced upon some amount of matter that the matter performs a quantum warp. we know this is possible becaus in computers we have made the barriers between pathways for electrons so small that the electrons have been witnessed building energy and warping to the other side of the so called barriers. this isnt exactly a white hole but it proves the idea of a quantum warp. some scientists believe that black wholes do generate enough energy that when something falls into one it builds enough force to warp somewhere else, you could do a few things with this, you could have a white hole appear naturally and blast the planets away. or if your society is technilogically advanced enough you could compile nuclear energy from fusion in such a high amount and blast your planet into a warp tunnel. if your a halo fan you could think of it as sending your planet through slipspace.


If matter can attract, anti-matter can repel!

Have the engineers build a anti-matter dispensing engine in the direction opposite to the path the planet needs to take. Pardon my pseudo-science, but as we all know (or all will know in the future) the power by which anti-matter repels is exponentially higher than that of gravity's attractive force.

This means that the anti-matter generator would have to produce a fraction of what a matter generator would have to produce to pull a planet.

  • 3
    $\begingroup$ Experiments so far show that antimatter has the same gravitational behavior as "normal" matter, so this wouldn't do anything. $\endgroup$
    – HDE 226868
    Commented Oct 6, 2017 at 0:34
  • $\begingroup$ @HDE226868, so to be clear by the same gravitational behavior you mean that it pulls/attracts like normal matter? $\endgroup$
    – Len
    Commented Apr 11, 2018 at 16:41
  • $\begingroup$ @Len Yes, that's it. $\endgroup$
    – HDE 226868
    Commented Apr 11, 2018 at 16:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .