# Using a solar mirror to deflect a star

Gold Member
I just finished Stephen Baxter's Manifold: Space. Great writer. Vast worlds he builds.

In it, he has a project to deflect a star (a neutron star, to be exact) using a colossal mirror-sail placed on one side of the star. The idea is that the solar radiation bounces off the mirror and is reflected back.

The sticky point here is that the mirror is stable and ostensibly bound to the star by gravity.

There's really not much more to it, as he describes it.

That seems wrong. Sure, the mirror-sail could remain bound by gravity while being pushed away by radiation, but I can't see how that could set up a net movement in the whole system.

Are my instincts wrong?

Related Science Fiction and Fantasy Media News on Phys.org
hutchphd
I will not make a value judgement of your instincts but they certainly match mine. External force required.

Edit: I take it all back.
The system will work fine. It is the old "can you blow on your own sail" question which indeed works sort of. It is like a thrust reverser on a jet engine: all the expelled radiation is in one direction so it produces thrust. Its why you put the front cap on your solid fuel rocket.

Last edited:
russ_watters
BWV
Hopefully no Romans in spaceships in this one, I really hated Ultima

Gold Member
It is the old "can you blow on your own sail" question which indeed works sort of. It is like a thrust reverser on a jet engine: all the expelled radiation is in one direction so it produces thrust. Its why you put the front cap on your solid fuel rocket.
Yes, at first glance.

But the sail is only held by gravity, so enough solar pressure on it and it will just float away. If the thrust provided by the neutron star is enough to push a whole star around, then surely it's plenty enough to eject the solar sail instead.

(I mean, that would work once - the sail would simply act as ejected reaction mass. But I'm not sure you can suspend the sail stable-like and get the continuous effect. Like trying to get magnets to chase each other around perpetually.)

It just seems like a 'you can't have your cake and eat it too' kind of situation.

To analogize:
If I suspended two magnets vertically and let them drop in freefall, could I get the bottom one to push against the top one, slowing them both down? Of course not. (I think that needs refining to be a valid analogy, but you see where I'm going?)

Last edited:
Yes, at first glance.

But the sail is only held by gravity, so enough solar pressure on it and it will just float away. If the thrust provided by the neutron star is enough to push a whole star around, then surely it's plenty enough to eject the solar sail instead. (That means the sail would simply act as reaction mass, and only work once. I'm not sure you can suspend the sail stable-like and get the continuous effect.)

It just seems like a 'you can't have your cake and eat it too' kind of situation.

To analogize:
If I suspended two magnets vertically and let them drop in freefall could I get the bottom one to push against the top one, slowing them both down? Of course not. (I think that needs refining to be a valid analogy, but you see where I'm going?)
Sounds like a class A Stellar Engine (Shkadov thruster)
https://en.m.wikipedia.org/wiki/Stellar_engine

stefan r and DaveC426913
russ_watters
Mentor
Yes, at first glance.

But the sail is only held by gravity, so enough solar pressure on it and it will just float away. If the thrust provided by the neutron star is enough to push a whole star around, then surely it's plenty enough to eject the solar sail instead.
A sail on a ship is fixed in place by the structure of the ship. This idea would balances gravitational force against the force of the radiation. So it can be stable/perpetual, but the thrust available is a function of the mass of the sail.

hutchphd
Well in fact the maximum thrust possible from the sail is exactly the weight of the sail! Of course this is a neutron star so there is potentially a very large gravitational field so the practical limit is likely the net efflux from the star, not the gravity..

russ_watters
I just finished Stephen Baxter's Manifold: Space. Great writer. Vast worlds he builds.

In it, he has a project to deflect a star (a neutron star, to be exact) using a colossal mirror-sail placed on one side of the star. The idea is that the solar radiation bounces off the mirror and is reflected back.

The sticky point here is that the mirror is stable and ostensibly bound to the star by gravity.

There's really not much more to it, as he describes it.

That seems wrong. Sure, the mirror-sail could remain bound by gravity while being pushed away by radiation, but I can't see how that could set up a net movement in the whole system.

Are my instincts wrong?
Nothing can be wrong in fiction

BillTre
I just finished Stephen Baxter's Manifold: Space. Great writer. Vast worlds he builds.

In it, he has a project to deflect a star (a neutron star, to be exact) using a colossal mirror-sail placed on one side of the star. The idea is that the solar radiation bounces off the mirror and is reflected back.

The sticky point here is that the mirror is stable and ostensibly bound to the star by gravity.

There's really not much more to it, as he describes it.

That seems wrong. Sure, the mirror-sail could remain bound by gravity while being pushed away by radiation, but I can't see how that could set up a net movement in the whole system.

Are my instincts wrong?
Yes, your instincts are wrong. The star is pulled by the sail's gravity as much as the sail is "anchored" by the star's gravity. Action=Reaction. The net momentum gain is from photons leaving the system asymmetrically, thanks to the huge mirror.

Gold Member
Yes, your instincts are wrong. The star is pulled by the sail's gravity as much as the sail is "anchored" by the star's gravity. Action=Reaction. The net momentum gain is from photons leaving the system asymmetrically, thanks to the huge mirror.
It makes more sense now others have pointed out the maximum thrust that can be imparted is simply the mass of the sail.

hutchphd
Good. You almost sucked me in. As I think a little more about it, I believe the max thrust will be one half the local neutron star weight of the sail of the at the altitude where it is stable. The factor of ##\frac 1 2 ## comes from it reflecting the radiation and so it feels twice the impulse of the net thrust

I believe the max thrust will be one half the local neutron star weight of the sail of the at the altitude where it is stable.
The maximum thrust per weight ratio is almost 1. However, a lot of energy would be lost if the sail operates at this ratio because its solid angle needs to be almost zero. Increasing the solid angle increases the total thrust but the thrust per weight ratio would be decreased.

The factor of ##\frac 1 2 ## comes from it reflecting the radiation and so it feels twice the impulse of the net thrust
That's something different but the maximum ratio between the weight and the force acting on the sail due to the reflection of the radiation is also almost 1.

mfb
hutchphd
Yes I see I was overcounting. I revert to my previous statement.....thanks for the correction. The thrust will be approximately the weight.

member 656954
Hopefully no Romans in spaceships in this one, I really hated Ultima
YES!! What a hot mess that novel was. Two-star review on Amazon from me, such a disappointment after Proxima.

In it, he has a project to deflect a star (a neutron star, to be exact) using a colossal mirror-sail placed on one side of the star.
@DaveC426913, I've not read Manifold: Space, over what time period is the neutron star deflected? If the thrust is so merely the mass of the mirror, that seems like it would take quite a while to move it anywhere.

hutchphd
The good news is that it is the weight of the mirror in the gravity of a neutron star. The bad news is it has to accelerate a neuron star!!. I wonder what the momentum emission from a neutron star looks like......probably not that much....I think that is the real problem with the idea. Hard number to find a generic value.

I wonder what the momentum emission from a neutron star looks like......
For the case that the full emission of a hemisphere is reflected (which is quite hard for x-rays) it would be

##F = \frac{L}{{4 \cdot c}}##

The crab pulsar has a luminosity of 0.9 L and a mass of 1.4 M. That would result in a thrust of about 3·1017 N and an acceleration of 10-13 m/s² (or 3.3 m/s per million years).

hmmm27, hutchphd and member 656954
hutchphd
Very nice thank you. Not a neck-snapper!!

stefan r
For the case that the full emission of a hemisphere is reflected (which is quite hard for x-rays) it would be

##F = \frac{L}{{4 \cdot c}}##

The crab pulsar has a luminosity of 0.9 L and a mass of 1.4 M. That would result in a thrust of about 3·1017 N and an acceleration of 10-13 m/s² (or 3.3 m/s per million years).
Absorption gives you half the thrust. You can radiate the energy in one direction and recover the other half and for full thrust.

People usually assume that a Shkadov thruster is also a statite. That it is a stationary single mirror which does not orbit. Instead you could do a Dyson swarm. Mirrors with orbits in a planes that intersect the direction of travel knife edge on the trailing side, angle 45 after 90 degrees and are a full mirror 180. On a perpendicular orbit the angle could be 45 around the full trip.

In order to be stationary the mirror has to be an extremely thin sheet. With a swarm setup the satellites can be greenhouses or computers running mind simulations. The thin reflective sheet is the most efficient if you are minimizing material and maximizing thrust. The other way you get to full utilize the energy for something useful and the thrust becomes an added bonus.

If we all installed our air conditioners on the north side of our houses we would eventually push the solar system away from Polaris by more than a Planck length. If we installed our AC on the west side of our houses Earth would rotate faster and our orbit would rise.

People usually assume that a Shkadov thruster is also a statite. That it is a stationary single mirror which does not orbit. Instead you could do a Dyson swarm. Mirrors with orbits in a planes that intersect the direction of travel knife edge on the trailing side, angle 45 after 90 degrees and are a full mirror 180. On a perpendicular orbit the angle could be 45 around the full trip.
I was assuming something between statites and satellites. The mirrors could orbit around an axis that is parallel to the acceleration and form a parabolic mirror.