Moving A Planet

[shynung]

This is a thought experiment that just crossed my mind. I'm considering submitting this to xkcd's What If? next.

Suppose someone built a rocket engine, and set it at the equator, nozzle pointing up. He set the firing sequence to fire the rocket at 0600 (so that the rocket nozzle faces Earth orbit's prograde), in an attempt to get closer to the sun.

How much of a rocket, in terms of both thrust and specific impulse, would this setup need in order to have any effect on the Earth's orbit at all?

Edited October 5, 2014 by shynung

[cantab]

A lot.

If you don't want to burn up a big chunk of the Earth's mass you're going to need specific impulse beyond any reasonable proposals. And a rocket engine/cluster the size of a small country.

http://qntm.org/moving

[MockKnizzle]

Well, plugging in the exhaust velocity of the SSME (4500m/s, a reasonable starting point) and a dV of 1m/s into the rocket equation, I get around 1.33x10^21 kg of propellant required. That's about 10% of the Moon. Where you will get that propellant, I have no idea.

[ZetaX]

You just might want to use that huge ball of stone and iron called "earth" below the engine

By the way, 4500m/s will not be enough. You need to overcome escape velocity to have an effect. Otherwise you just shovel earth in orbit around the earth (or it just comes back down).

Edited October 5, 2014 by ZetaX

[shynung]

I don't think 4500 m/s would be enough. The exhaust would be slowed down by the earth's own atmosphere, absorbing the kinetic energy of the the exhaust, like a sailboat with a fan blowing into the sail.

I'm guessing something with much higher exhaust velocity would be needed.

[Kryten]

I don't think 4500 m/s would be enough. The exhaust would be slowed down by the earth's own atmosphere, absorbing the kinetic energy of the the exhaust, like a sailboat with a fan blowing into the sail.

It'd have to do something, or conservation of momentum would be broken.

[N_las]

It'd have to do something, or conservation of momentum would be broken.

Yeah, it falls back down. So every bit of momentum the earth gained in the intended direction will be applied in the backwarts direction. The net effect is no movement.

As ZetaX said: The exhaust velocity has to be high enough to penetrate the atmosphere AND still be greater than escape velocity. And you can only use the resulting velocity (original exhaust velocity minus atmospheric effects minus escape velocity) as the "real exhaust velocity" in the rocket equation.

[shynung]

It'd have to do something, or conservation of momentum would be broken.

Well, it'd heat the atmosphere around it considerably, for one. A big gout of flame for another. It may also drive itself into the ground, if the mounts aren't sturdy enough.

[Tommygun]

I would think mounting the engine might be the biggest problem.

If you can imagine placing a 100kg moist block of clay on the floor and then try pushing it across the room with a toothpick.

No matter how deep or wide you make the mounts, the engine is going to sink into the Earth.

[Red Iron Crown]

I would think mounting the engine might be the biggest problem.

If you can imagine placing a 100kg moist block of clay on the floor and then try pushing it across the room with a toothpick.

No matter how deep or wide you make the mounts, the engine is going to sink into the Earth.

If you limit it to one engine, yes. Seems like a tremendous array of thousands of engines would sidestep that problem.

[Ralathon]

Why use engines? You could also detonate millions of nukes in the upper atmosphere. It would cause the upper crust to start vaporizing and get enough thermal energy to escape the earth thus providing thrust.

[shynung]

In that case, why use a nuke? Slap that engine left over to a big enough space rock, hurl it into a retrograde orbit around the sun, and set it into a head-on earth-impact trajectory. It'll be quite a smack, but at least the earth's perihelion should definitely be closer to the sun.

Huh. Didn't think of that one before.

Edited October 5, 2014 by shynung

[Kryten]

In that case, why use a nuke? Slap that engine left over to a big enough space rock, hurl it into a retrograde orbit around the sun, and set it into a head-on earth-impact trajectory. It'll be quite a smack, but at least the earth's perihelion should definitely be closer to the sun.

Given there aren't any large objects in retrograde orbits, you'd have to transfer pretty much all of that momentum yourself with the engine. Why bother when you could do it directly?

[shynung]

That was the original premise; slap the engine on the equator, and fire it skywards at dawn. However, there's an atmosphere in the way. Asteroids typically don't have an atmosphere, which would have eased the burden on the engine.

Of course, I never limited what kind of engine it is. Not much people are going to whine if I theoretically mount a nuclear pulse drive on a Chicxulub-sized space rock and fly it here.

[ZetaX]

Why use engines? You could also detonate millions of nukes in the upper atmosphere. It would cause the upper crust to start vaporizing and get enough thermal energy to escape the earth thus providing thrust.

I see no use in doing that in the upper atmosphere when you can also do it on ground, where you will transfer more heat.

And I am not convinded that this will actually cause thrust. Apart from it possibly being too weak, the gas might flow of in all direction instead of just one.

[TheSaint]

Read "A World Out of Time" by Larry Niven. I thought it was out of print, but apparently it was recently reprinted. His idea for moving the Earth out of the way of an expanding Sun was to rig up a giant fusion engine on Uranus. Anchor it to the core and use the atmosphere for fuel. Then you maneuver Uranus around so that its gravity alters the Earth's orbit the way you want it to go. It's a very long term project, it would take hundreds of years. And obviously it would take a level of technology far beyond what we have now. But the physics are there.

Edited October 5, 2014 by TheSaint
Actually in print

[ZedNova]

Why don't we start by moving a much smaller planet, like the moon.

We move that with several hundred nuclear pulse engines, then put it into a orbit that slowly pulls the earth where we want it.

[cantab]

If the atmosphere's in your way, get rid of it. As the source I gave mentioned, that's trivial compared to the main task. Alternatively, build the engines on an artificial mountain that puts them above most of the atmosphere.

And indeed the engine cluster will slowly sink, as its force bends the lithosphere down and pushes the asthenosphere out of the way. This may reach an equilibrium, just as mountains do, but if not then you will indeed need to keep infilling with rock to counteract the sinking. Maybe you could deliberately induce volcanic eruptions around the engine cluster so the magma will do the infilling for you.

[TheSaint]

Why don't we start by moving a much smaller planet, like the moon.

We move that with several hundred nuclear pulse engines, then put it into a orbit that slowly pulls the earth where we want it.

By the time we reach a level of technology that enables any of this, the Moon will almost certainly be heavily colonized. It won't be expendable, we'll actually want to bring it along. Plus it has the same problem that any other teresstrial planet will have: lack of reaction mass. Using a gas giant solves that problem handily.

[TheDarkStar]

Plus it has the same problem that any other teresstrial planet will have: lack of reaction mass. Using a gas giant solves that problem handily.

You can throw the planet's mass, at the very least. It still counts as moving the planet, since part of it (probably/hopefully the main part) ends up in one orbit and the other part (used as reaction mass) ends up in the other.

[cantab]

Indeed. If you're trying to move a planet with a reaction engine, then one thing you're not short of is reaction mass.

[Nuke]

i think i would use large fusion powered rail guns. and then sling iron rods into space at escape velocity. its going to consume a large amount of the earth's mass to move it even slightly.

ive always thought about moving icy objects about the size of pluto with nuclear thermal engines, where the whole of the object is to be used as propellant. the engines would need to be able to run on the various ices that compose the object. anything that is not ice gets filtered out and either utilized to support a population or loaded into railgun and ejected. one of these days im going to run the numbers and see if this is a viable method for interstellar generation ship.

[megatiger78]

warp drive? the thing with negative mass? that no one has built yet? ACCTUAL USE FOUND

[Tech Support]

interesting article, Scott Manley has an interesting video on this topic.

[metaphor]

We can do the math...

The Earth has a mass of 6*10^24 kg. Let's say you use nuclear pulse engines like the proposed Orion rocket, with an Isp of about 100,000s. In order to change the Earth's perihelion by 1%, that's a delta-v of about 300 m/s. Using an Isp of 100,000s, the mass of propellant (nuclear bombs) needed to accelerate the Earth through that delta-v is about 2*10^21 kg.

The total mass of uranium in the Earth's crust is about 10^17 kg. The amount of uranium in the entire solar system is about 10^21 kg (99.8% of that is in the Sun), of which only 0.7% is the kind that bombs are made of.

So there is not enough uranium in the solar system to make enough nuclear bombs using nuclear pulse propulsion to change Earth's perihelion by 1%. That option is off the table.

Let's say you use ion propulsion instead, using the gas Argon which is relatively abundant, it makes up 1% of the Earth's atmosphere (Xenon and Krypton are much rarer). The Isp of such an ion engine would be around 4000s, so the mass of argon needed is about 5*10^22 kg.

There is about 5*10^18 kg of Argon in the entire Earth, and about 6*10^24 kg of Argon in the entire solar system, 99.8% of which is in the Sun. Assuming we can't extract the argon inside the Sun, this option is also off the table.

So we head to ordinary chemical engines. Let's say hydrogen + oxygen combustion, 2 of the 3 most common elements in the universe, which can be electrolyzed from water. Assuming an Isp of 450s, the mass of water needed is about 5*10^23 kg.

The amount of water in all the oceans on Earth is about 1.4*10^21 kg, and the total amount of water in the Earth is about 10^22 kg. So there is not enough water on Earth. But the solar system is full of water. The planet Neptune is about 50% water, or about 5*10^25 kg. So if we could get 1% of Neptune's water we could use that.

Now, let's say we use hydrolox engines like the Space Shuttle main engine, and they are packed next to each other across the entire surface of the Earth and facing upwards (ignoring the atmosphere). That would be about 10^13, or 10 trillion engines firing at once (in the right direction as the Earth is rotating). They would have a combined thrust of 2*10^19 N. At that thrust, it would take 10^8 seconds, or 3 years of constant thrusting to make the Earth change its perihelion by 1%.

The energy for electrolyzing that water for the hydrolox engines would have to come from somewhere. If the electrolyzers are 50% efficient, that's about 4*10^27 Joules needed. That's about the same as the total energy Earth receives from the Sun over a period of 700 years.

So... plausible, but it would take a while.