Dyson spheres

[gbaji]

Surface gravity is GM/R² for idealized spherical masses, so the physical "size" of the object you're dealing with matters quite a bit; a solid object with the same mass as a hollow sphere will only have a similar surface gravity to that of a hollow sphere if both the solid object and the hollow sphere have similar radii. The surface gravity of the Sun is ~28g; a shell of negligible mass with a radius of ~3.7 million kilometers (~5.3 solar radii or ~0.025 AU) would have approximately Earth-normal surface gravity if centered on the Sun; a shell of negligible mass with a radius of 1 AU (~1.5e8 km) likewise centered on the Sun would have a surface gravity of about 0.0006g.

That's pretty much what I assumed. So to make the whole "folks living on the outside of the shell" concept work, you'd either have to have the sphere *really* close to the star, or somehow make the sphere itself out of some super dense material to make up for the rather large hollow space inside. Again, there are a host of other problems with a solid sphere concept, even before putting habitats on the outside, but was just checking on that part of the math as well.

While a full sphere/swarm is impractical, aspects of a swarm can be taken into consideration. For example, if a civilization did build a few hundred 1000 km radius space habitats it would be useful to both coat their exteriors in solar panels on whatever side faces the star and to have additional energy gathering units positioned throughout the system for both backup power, shielding (putting your habitat behind a big solar collector helps it you're worried about flares and stuff), and to power any sort of big interstellar project like a solar sail based starship or a giant ball of computronium full of digital people.

Space-based solar energy collection is even a potentially viable medium-term technology, especially if fusion doesn't work out.

Yup. That's where I was kinda going. This sort of solution scales up over time/construction too. You build cyliders or small orbitals where you want them, orbiting around the star. You can then put additional collectors (also orbiting around the star) as needed and beam power to the habitats. Keep doing this, and you'll eventually have a larger and larger swarm of such things. Some number of large habitation's and another large number of more or less automated power collection systems. Whether that ever reaches the coverage to qualify as a Dyson Sphere proper or not is really only a nitpick at that point.

Oh, and we can also use Newton's Third Law. Put a mass anywhere inside of a symmetric spherical shell, even someplace off-center, and the net force on it is zero. Turn that around, and you see that, if you have an off-center spherical shell around a star, then the net force on the shell from the star is also zero. A complete solid sphere is neither stable, like a marble in the bottom of a bowl, nor unstable, like a marble on top of an upside-down bowl. It's neutrally stable, like a marble sitting on a flat tabletop.

Yup. But just like a marble on a flat table, the slightest force will upset its balance and it'll roll in one direction or another. Which is why we'd have to assume some sort of station keeping methodology as well. Heck Niven more or less wrote Ringworld Engineers specifically to address this very issue (and others).

[Chronos]

A ring is actually unstable. If it drifts off slightly, then it'll start drifting faster and faster. With a sphere's neutral stability, if it starts off drifting very slowly, it'll continue to drift only very slowly. You'd want active stationkeeping for both, but the problem is much easier for a sphere than for a ring.

[LibraryOgre]

Probably requires incredible structure strength to keep it from collapsing into a more compact body, but I can't be certain without a more detailed description. It's more plausible than the Can Full of Sky, which had both vaguely Earth-like levels of (rotational pseudo-) gravity and noticeable atmospheric pressure at a much greater range of altitudes than I believe possible.

Schlock Mercenary has it's fair amount of unobtanium is post-transuranic alloys, not to mention active gravity manipulation and antimatter plants for energy.

A ring is actually unstable. If it drifts off slightly, then it'll start drifting faster and faster. With a sphere's neutral stability, if it starts off drifting very slowly, it'll continue to drift only very slowly. You'd want active stationkeeping for both, but the problem is much easier for a sphere than for a ring.

This was part of the impetus for later Ringworld books; engineering students at a science fiction con chanting that it was unstable, which led to Niven including a plot where some of the station-keeping engines were needing repair.

[Bohandas]

I just realized that my prior converting the planet into a giant building thing might work if we also greatly the planet's rotation to the point where the centrifuge effect mostly cancels out gravity in the upper levels

[Batcathat]

I just realized that my prior converting the planet into a giant building thing might work if we also greatly the planet's rotation to the point where the centrifuge effect mostly cancels out gravity in the upper levels

I don't know enough about physics to say for sure, but I have to imagine trying to balance two forces against each other like that has to be really bad for the stability.

[NichG]

I don't know enough about physics to say for sure, but I have to imagine trying to balance two forces against each other like that has to be really bad for the stability.

Stability isn't necessarily an issue, but tension can be. Materials have different compressive vs tensile strengths, so you want the right kind of equilibrium for what you're building with.

If it's concrete, compression is better. If it's carbon nanotubes, probably tension is better.

[halfeye]

I just realized that my prior converting the planet into a giant building thing might work if we also greatly the planet's rotation to the point where the centrifuge effect mostly cancels out gravity in the upper levels

That doesn't work for the poles, the Earth is slightly oblate because it rotates in 24 hours, Haumea rotates much faster, with the result that it looks like this:



https://en.wikipedia.org/wiki/Haumea

[Lord Torath]

This was part of the impetus for later Ringworld books; engineering students at a science fiction con chanting that it was unstable, which led to Niven including a plot where some of the station-keeping engines were needing repair.

So if your 'ring' was a section of a spherical shell (as opposed to a rectangular or circular cross-section), would it still be unstable? Or would it have the sphere's 'neutral' stability?

[Mechalich]

So if your 'ring' was a section of a spherical shell (as opposed to a rectangular or circular cross-section), would it still be unstable? Or would it have the sphere's 'neutral' stability?

I think you might be referring to a topopolis, which is like taking an impossibly large piece of spaghetti and wrapping it around a star, potentially multiple times. This is, as I understand it, both theoretically stable and also theoretically conceivable within the bounds of materials science if the individual loops are small enough that they don't tear themselves apart.

[Chronos]

You need the whole sphere to get the sphere's neutral stability.

[Maat Mons]

I had to think about it for awhile, but I can see how the ring would have stability issues. treating the plane of symmetry of the ring as the xy plane, if the center of mass of the star isn't in the xy plane, there will be a net force in the z direction. Treating the z axis as passing through the center point of the ring, if the center of mass of the star isn't on the z axis, that the force applied will result in torque. And if the ring is spinning to simulate gravity, the gyroscopic effect makes funny things happen that I haven't worked out.

[Lord Torath]

I think you might be referring to a topopolis, which is like taking an impossibly large piece of spaghetti and wrapping it around a star, potentially multiple times. This is, as I understand it, both theoretically stable and also theoretically conceivable within the bounds of materials science if the individual loops are small enough that they don't tear themselves apart.

Not what I was thinking of, no. I was imagining you take a 'dyson' sphere centered on the star, then take a giant blade and chop off the 'top' and 'bottom' sections leaving you with an 'equatorial ring' with the same curvature as the sphere.

I can definitely see how having a 'vertical' misalignment could cause all sorts of problems in that case, though. Thanks, Maat Mons! That helps a lot!

[halfeye]

Not what I was thinking of, no. I was imagining you take a 'dyson' sphere centered on the star, then take a giant blade and chop off the 'top' and 'bottom' sections leaving you with an 'equatorial ring' with the same curvature as the sphere.

I can definitely see how having a 'vertical' misalignment could cause all sorts of problems in that case, though. Thanks, Maat Mons! That helps a lot!

What actually would cause problems with the ringworld was that the ring can shift in the plane of the ring, and as one side of the ring gets closer to the star the gravity on it from the star increases. It doesn't matter whether the ring is rotating, if it is rotating the part nearest the star changes but there is still a part nearest the star and that part is more strongly attracted.

[Trafalgar]

So my question whenever something like a Dyson sphere comes up is "What is your civilization doing that requires that amount of energy?" If your civilization has the technology to build a Dyson sphere, it probably doesn't need to.

[Eldan]

In Alastair Reynold's House of Suns, they build dyson spheres to contain supernovae which threaten inhabited systems.

[NichG]

So my question whenever something like a Dyson sphere comes up is "What is your civilization doing that requires that amount of energy?" If your civilization has the technology to build a Dyson sphere, it probably doesn't need to.

Run Crysis 17 at 60fps...

[Anymage]

So my question whenever something like a Dyson sphere comes up is "What is your civilization doing that requires that amount of energy?" If your civilization has the technology to build a Dyson sphere, it probably doesn't need to.

What is the civilization doing with all that energy? Expanding, and powering projects on a scale that would boggle our minds. Look at energy dependency in our world and scale that up.

To flip the question around, if a civilization wouldn't need to tap into a huge amount of energy that the star is giving off anyways, where would they be getting the energy to power a vast futuristic civilization from?

[Tyndmyr]

You need the whole sphere to get the sphere's neutral stability.

It won't be stable, though. Even the slightest imbalance will have positive feedback, resulting in a sun/sphere collision.

Even granted the kind of insane material needed to make a physical sphere work somehow, a collision with the sun would pose notable difficulties for human habitation.

So my question whenever something like a Dyson sphere comes up is "What is your civilization doing that requires that amount of energy?" If your civilization has the technology to build a Dyson sphere, it probably doesn't need to.

That's actually a good point. It *might* make sense if your society is so densely populated that stars are a scarce resource, but that's...obviously very far from the present world. If any such society existed, it would be a large, growing black spot erasing stars. A fun sci fi concept, perhaps, but definitely not congruent with reality.

If you have that much energy, you can just...travel to a new star. Orbitals around a star is probably sufficient for any conceivable scale of society in practice. You just spread out, instead of building locally. Space has, after all, a lot of space. No reason not to use it.

[Mechalich]

That's actually a good point. It *might* make sense if your society is so densely populated that stars are a scarce resource, but that's...obviously very far from the present world. If any such society existed, it would be a large, growing black spot erasing stars. A fun sci fi concept, perhaps, but definitely not congruent with reality.

If you have that much energy, you can just...travel to a new star. Orbitals around a star is probably sufficient for any conceivable scale of society in practice. You just spread out, instead of building locally. Space has, after all, a lot of space. No reason not to use it.

The most likely producers of Dyson sphere(s) would be a hegemonizing swarm: a self-replicating machine civilization empowered by it's core programming to eternally convert matter into more and more copies of itself (AI-risk scenarios like the 'paper-clip maximizer' share a great deal with this concept), so the reason to build something that massive would come down to more, more more! Thankfully, this has not been observed, and we have used telescopes to go hunting for Dyson spheres/swarms.

[Jimorian]

It won't be stable, though. Even the slightest imbalance will have positive feedback, resulting in a sun/sphere collision.

No, what is meant by "neutrally stable" is that there is NO force exerted between a hollow sphere and an object located inside of it, even if the sun isn't centered inside that sphere.

That does mean that if the sun or the sphere start drifting because of some outside force, though, momentum would eventually carry them into a collision, but there is absolutely zero feedback to accelerate or decelerate the process.

[Trafalgar]

Thinking about this, what radius of a Dyson Sphere is possible in a Solar System like ours? Too close and the sphere melts. To far, and the sphere is too big to build. Someone said the goldilocks zone but 1AU would make a sphere that is far too large to build.

[Batcathat]

Thinking about this, what radius of a Dyson Sphere is possible in a Solar System like ours? Too close and the sphere melts. To far, and the sphere is too big to build. Someone said the goldilocks zone but 1AU would make a sphere that is far too large to build.

What is "too big to build" in this context? Any Dyson sphere presupposes a pretty crazy level of technology.

[Bohandas]

there needs to be enough material in the system for one thing

[halfeye]

there needs to be enough material in the system for one thing

Yeah, but you've got a star to play with.

[warty goblin]

Yeah, but you've got a star to play with.

Which is mostly hydrogen, helium and oxygen, which aren't good construction material. They're also inconveniently hot, and down a gravity well.

Sure in theory you can fuse that into heavier elements, but if you have the ability to do that solar panels are kinda a dumb investment. Just fuse Jupiter.

[Maat Mons]

I’m pretty sure you’d need a Dyson sphere around a Sun-like star to have a much larger radius than 1 AU. I mean, the Earth is 1 AU from the Sun, and if it didn’t rotate, the Sun-ward side would be pretty darn hot.

Ultimately, energy in must equal energy out. The energy in is the total output of the star. The energy out is the black body radiation of the Dyson sphere. Just figure how much surface area you need for a 70°F (approx. 20°C) object to emit infrared radiation at a fast enough rate to match the rate at which the star is outputting energy.

[Mechalich]

It's been speculated in some sources (notably Schlock Mercenary) that the ideal target for a Dyson Sphere/Matryoshka Brain (which aren't the same thing but both involve encapsulating a star) would be a white dwarf. This would still provide a great deal of energy at a much, much smaller radius. It's also possible that a high metallicity red dwarf might have a great ratio of potential materials to utilize compared to a yellow star like our Sun (though red dwarfs have certain stability issues that make this tricky).

Sure in theory you can fuse that into heavier elements, but if you have the ability to do that solar panels are kinda a dumb investment. Just fuse Jupiter.

You could conceivably do both. If you're trying to conquer the galaxy as a hegemonizing swarm (and if you're not a hegemonizing swarm you have no real reason to try), you'd use a dyson swarm to power immensely powerful lasers used both for interstellar communication and to push solar sails to accelerate your starships while at the same time using vast quantities of fusion fuel harvested from gas giants to slow said starships down upon reaching their destinations.

[Trafalgar]

I’m pretty sure you’d need a Dyson sphere around a Sun-like star to have a much larger radius than 1 AU. I mean, the Earth is 1 AU from the Sun, and if it didn’t rotate, the Sun-ward side would be pretty darn hot.

Ultimately, energy in must equal energy out. The energy in is the total output of the star. The energy out is the black body radiation of the Dyson sphere. Just figure how much surface area you need for a 70°F (approx. 20°C) object to emit infrared radiation at a fast enough rate to match the rate at which the star is outputting energy.

Well, let's use 1AU as a starting point. 1 AU equals 1.496 * 10¹¹ meters. A sphere with that radius would have a surface area of 2.81 * 10²³ meters². If the you made the sphere 1 meter thick, it would have an approximate volume of 2.81 * 10²³ meters³.

The volume of Jupiter is 1.43 * 10²⁴ meters³ so if Jupiter was entirely a usable material, you could build a Dyson sphere out of it. But Jupiter, along with Uranus, Neptune, and Saturn, is mostly gas and I think to build a Dyson sphere, you are going to have a very specific list of materials that are probably all solids. If you used every bit of these specific materials that could be from every other planet, moon, and asteroid in our system you would still not have enough. Which means going to other systems and transporting huge amounts of solid heavy building materials light years.

This goes back to my original point. If a civilization has the ability to build a Dyson Sphere, it probably doesn't need to.

[halfeye]

Well, let's use 1AU as a starting point. 1 AU equals 1.496 * 10¹¹ meters. A sphere with that radius would have a surface area of 2.81 * 10²³ meters². If the you made the sphere 1 meter thick, it would have an approximate volume of 2.81 * 10²³ meters³.

The volume of Jupiter is 1.43 * 10²⁴ meters³ so if Jupiter was entirely a usable material, you could build a Dyson sphere out of it. But Jupiter, along with Uranus, Neptune, and Saturn, is mostly gas and I think to build a Dyson sphere, you are going to have a very specific list of materials that are probably all solids. If you used every bit of these specific materials that could be from every other planet, moon, and asteroid in our system you would still not have enough. Which means going to other systems and transporting huge amounts of solid heavy building materials light years.

This goes back to my original point. If a civilization has the ability to build a Dyson Sphere, it probably doesn't need to.

If I recall correctly, over 98% of the matter in the solar system is in the sun (Oops, Wikipedia says 99.86% of the mass). Playing with the star is all you need to do to find enough matter. Supposing there is no FTL, then playing with the star is the way to go if you need a Dyson sphere. We can't do it now. As you say, probably nobody needs a Dyson sphere.

[Trafalgar]

If I recall correctly, over 98% of the matter in the solar system is in the sun (Oops, Wikipedia says 99.86% of the mass). Playing with the star is all you need to do to find enough matter. Supposing there is no FTL, then playing with the star is the way to go if you need a Dyson sphere. We can't do it now. As you say, probably nobody needs a Dyson sphere.

Of course, most of the sun is Hydrogen with some Helium which I don't think would be useful to build a Dyson Sphere with.