Dyson Spheres And Ring Worlds... err... HOW?

HansMustermann

Penultimate Amazing
Joined
Mar 2, 2009
Messages
23,741
Well, to keep the question short and to the point: how would you keep a rigid structure centered on the star? Seems to me like basically the shell theorem works both ways, so there's not an awful lot holding such a mega-structure centered on the gravity well. In fact, it seems to me like nothing at all is keeping it centered.

So... err... how was that supposed to be solved?
 
Larry Niven got rather seriously caught out by his own fans on that one, who pointed out that the Ringworld system was gravitationally unstable, requiring him to retcon attitude control rockets in the later books. Presumably a Dyson sphere would need a similar system.

Dave
 
Wouldn't a dyson sphere, not having a magnetic field and ozone layer like Earth's, be a bit vulnerable to solar radiation?

Doesn't sound like a great idea to me.
 
Wouldn't a dyson sphere, not having a magnetic field and ozone layer like Earth's, be a bit vulnerable to solar radiation?

Doesn't sound like a great idea to me.

Yeah, I think the original idea of a Dyson sphere was to harvest all the energy from a sun, so I would imagine that the habitable zone would be around the equator - from simple physics considerations, it would have to be around the equator, because that's where the atmosphere goes if you spin it (you need to spin it to get any gravity on the inside; even artificial gravity wouldn't fix that one unless it violated the inverse square law) - and anything anywhere near the poles would just be energy collection machinery. That would limit the region where you have to provide solar radiation protection.

You'd also need some *********** enormous radiator fins on the outside, because waste heat management would be a nightmare of, quite literally, astronomical proportions.

Dave
 
Yeah, I think the original idea of a Dyson sphere was to harvest all the energy from a sun, so I would imagine that the habitable zone would be around the equator - from simple physics considerations, it would have to be around the equator, because that's where the atmosphere goes if you spin it (you need to spin it to get any gravity on the inside; even artificial gravity wouldn't fix that one unless it violated the inverse square law) - and anything anywhere near the poles would just be energy collection machinery. That would limit the region where you have to provide solar radiation protection.

You'd also need some *********** enormous radiator fins on the outside, because waste heat management would be a nightmare of, quite literally, astronomical proportions.

Dave

Well, they are the kind of tunnel-vision overkill "inventions" that you find in science-fiction and visions of the future. Like Asimov's mainframe computers in trans-Jupiter orbits (to harvest liquid nitrogen to keep superconductors cooled):

You focus on one current challenge and imagine future technology used specifically to solve the problem of the present, not realizing, or ignoring, that the same future technology will probably enable us to work around the challenge in a number of ways.

- So Earth had (some decades back) an exponential growth in energy requirement, while running out of fossil fuels? Well, use future tech to harvest enormous amounts of energy. ... In reality, we use the new tech to reduce energy consumption and convert to renewable sources (OK we're certainly not there, but it will come).

Hans
 
- So Earth had (some decades back) an exponential growth in energy requirement, while running out of fossil fuels? Well, use future tech to harvest enormous amounts of energy. ... In reality, we use the new tech to reduce energy consumption and convert to renewable sources (OK we're certainly not there, but it will come).

Remember, though, that "renewable sources" in this context basically means sunlight. That's precisely what a Dyson sphere is intended to optimize.

Dave
 
Remember, though, that "renewable sources" in this context basically means sunlight. That's precisely what a Dyson sphere is intended to optimize.

Dave

Yes, but of a scale quite out of proportion to our needs.

Hans
 
- So Earth had (some decades back) an exponential growth in energy requirement, while running out of fossil fuels? Well, use future tech to harvest enormous amounts of energy. ... In reality, we use the new tech to reduce energy consumption and convert to renewable sources (OK we're certainly not there, but it will come).

Hans

Energy efficiency would not do away with the need for such large scale energy use, as economic growth outstrips any efficiency improvement we care to come up with.

What will result in us not needing such mega structures is the lack of need for such economic growth.

As world population levels out (it will), and the rest of the world develops in line with the likes of the US (or some level above that), economic growth will level off. Why? Because when a population reaches a certain point economically, there is not the room for any more for people to buy more stuff. To put it another way - when people have enough "stuff", economic growth will stop. There is only so much stuff most people actually want.

Of course, innovations and fashions will bring about new products. Trends will see old and new products to become popular. But this idea that we will forever need more energy due to the eternal growth of the economy is misplaced.

Then there is the point that Dave said: We'll go to renewables. Only on Earth, as its easier. And we'll also go nuclear (fission/fusion).
If we continued with economic growth beyond what people want (unlikely), the main physical limit of economic growth is waste heat on Earth, as there is a limit to how much energy we can produce and still emit waste heat. There is a finite amount of space for solar/wind and energy efficiency is limited by the laws of physics (eg. you can only light a bulb with a minimum amount of watts per lumen), so this leaves fission/fusion for any economic growth. This would eventually produce enough waste heat to heat the Earth, thus limiting economic growth.

To put it another - the pill will do away with the need for any kind of Dyson swarm/shell/ringworld.
 
Last edited:
Well, to keep the question short and to the point: how would you keep a rigid structure centered on the star? Seems to me like basically the shell theorem works both ways, so there's not an awful lot holding such a mega-structure centered on the gravity well. In fact, it seems to me like nothing at all is keeping it centered.

So... err... how was that supposed to be solved?

That is indeed one of the big problems with rigid versions of Dyson Spheres and an even worse problem for Ringworlds as they are dynamically unstable so if they go off center gravity will pull them further off center rapidly resulting in their destruction.

The original solution for a Dyson Sphere was simple, don't be a rigid shell, but rather a very large swarm of independently orbiting solar collectors that capture the output of the Sun. A more recent proposed variation is a shell of statite used as collectors, as they are held up by radiation pressure rather than being in orbit and need not be a rigid, or even continuous, shell which avoids problems with shell theorem.

The solid shell version that is seen most often in science fiction is also by far the least plausible having massive engineering problems not simply collapsing while being built, require ridiculous amounts of materials, and not provide much in the way of habitable space. They would presumably have some kind of active station keeping system to deal with the instability.

And as I said above Ringworlds are the worst in terms of stability along with having the problems constructing them of a solid Dyson Sphere, and would also presumably have active station keeping
 
For any who are interested in this subject, Isaac Arthur has done a series of YouTube videos on this sort of thing. In particular, the Dyson Sphere and Ringworld videos should be relevant, and if you like those, he has a lot of videos worth watching.

(not doing them as embedded yt because they're 30+ minutes long and not really something I'd watch while reading a thread).
 
The original solution for a Dyson Sphere was simple, don't be a rigid shell, but rather a very large swarm of independently orbiting solar collectors that capture the output of the Sun.

The main problem with that, I think, is that it violates the Hairy Ball Theorem. You'd need multiple orbits crossing over one another, the usage of the collectors would be horribly inefficient because most of the time most of them would be occluded, and the orbital mechanics would be ridiculously unstable. There's a good reason why ring systems are planar.

A more recent proposed variation is a shell of statite used as collectors, as they are held up by radiation pressure rather than being in orbit and need not be a rigid, or even continuous, shell which avoids problems with shell theorem.

That could be feasible, yes (for extreme values of "feasible").

The solid shell version that is seen most often in science fiction is also by far the least plausible having massive engineering problems not simply collapsing while being built, require ridiculous amounts of materials, and not provide much in the way of habitable space.

True, the idea doesn't really stand up to close examination. They have the advantage of not being topologically impossible, though.

Dave
 
Well, to some extent, yes, SF generally tries to scale up existing technologies and problems, rather than predict whatever will happen in the future. I guess if the Romans ever wrote SF, it would involve how to fit even more rows of oars on a ship, rather than come up with a screw.

But I guess another problem with at least a certain kind of SF is that it feels a compulsion to awe. In a "go big or go home" way, so to speak. So you end up with thousand mile long ships, with power cores measured in TW, and, yes, remaking whole solar systems from scratch.

I guess nobody wants to read or watch the story of a realistic spaceship that has all the size, space and comfort of a WW2 submarine, including hot-bunking. Because cross section counts for drag when you go relativistic, length counts for tidal forces when you're anywhere near a gravity well, and MASS counts more than both combined when you have to carry your own fuel.
 
Well, to some extent, yes, SF generally tries to scale up existing technologies and problems, rather than predict whatever will happen in the future. I guess if the Romans ever wrote SF, it would involve how to fit even more rows of oars on a ship, rather than come up with a screw.

A Googolkonteres.
 
Well, orbiting stuff with big panels close to the sun also has the problem of the Poynting-Robertson effect, a.k.a., why dust orbiting a star spirals inwards instead of having a stable orbit.

The problem is that photons may not have mass, but have momentun. And while in the frame of the sun they radiate outwards along the local radius (but not really in practice), in the frame of something orbiting the sun, they come at an angle from the front. Kinda like how if you watch raindrops on a train window, they always seem to come from the front, even if there's no wind. So you collide into photons, and regardless of whether you bounce them towards the front or absorb them with your solar panels, you lose momentum. Not the same amount in both cases, but you lose momentum anyway.

And if you make something lightweight and with big panels, well, you have a problem. You'll need to use your thrusters all the time just so you don't spiral into the sun.

Now we come to the "not really" part about solar radiation. You may think, 'duh, of course the average of the radiation is emitted perpendicular to the surface', and you'd even be correct... for when it's emitted. But it has to go through the photosphere and corona, where it has the same braking effect on the matter there. But momentum conserves, so all the momentum that was stolen from the corona and dust by those photon collisions, is now present in the photons coming away from it. Light is actually coming at a very slight angle from the surface of the sun.

So now if you're close enough to be supported by just radiation pressure, effectively you get a slight acceleration sideways from those photons. Nothing unmanageable, but it's going to be playing silly buggers with your orbit anyway.
 
For any who are interested in this subject, Isaac Arthur has done a series of YouTube videos on this sort of thing. In particular, the Dyson Sphere and Ringworld videos should be relevant, and if you like those, he has a lot of videos worth watching.

(not doing them as embedded yt because they're 30+ minutes long and not really something I'd watch while reading a thread).

I just found his channel a couple of weeks ago and I'd highly recommend it. Good stuff and more in depth than the usual YouTube offerings.
 
Well, to keep the question short and to the point: how would you keep a rigid structure centered on the star? Seems to me like basically the shell theorem works both ways, so there's not an awful lot holding such a mega-structure centered on the gravity well. In fact, it seems to me like nothing at all is keeping it centered.

So... err... how was that supposed to be solved?

Technology indistinguishable from magic.
 
Just going back to keeping the dyson sphere/ringworld centred

Ringworld: It rotates doesn't it, so effectively, the whole ring is in orbit. It would have to rotate at an orbital speed commensurate with its distance from its sun. Also such a massive structure is going to be sunk into its own gravity well; a well which would extend all around its parent star.

Dyson sphere: Given what I said above, I can't see how this would work. If it rotated, the equator would need to be rotating at orbital speed/distance, but as you move away from the equator towards the poles, the radial velocity of any given point on the surface falls, and the centripetal force falls with it. The gravitational attraction of the star would tend to pull the sphere inwards, trying to distort it into an oblate spheroid shape.

There is also the problem of the solar wind, which varies in density, temperature and speed over time and over solar latitude and longitude. And therein lies another problem. If the Dyson Sphere is fully contained, how is the solar wind handled? These charged particles have to go somewhere. The Earth is protected by its magnetic field... how will the builders protect the inner surface of the Sphere? Also, the dynamic pressure of the solar wind might be less than 1 nanopascal, but that is enough to keep it doing speeds of over 500km/sec all the way out to the termination shock between the orbit of Neptune and the Oort cloud. If the Sphere is "sealed" the pressure inside would build up because the components of the solar wind (electrons, protons and alpha particles) have mass. Would most particles pass right through? Even if some do, not all of them will. Those that do not will impact and create pressure.
 
Ringworld: It rotates doesn't it, so effectively, the whole ring is in orbit. It would have to rotate at an orbital speed commensurate with its distance from its sun. Also such a massive structure is going to be sunk into its own gravity well; a well which would extend all around its parent star.

Doesn't matter that it's spinning, that keeps it from collapsing under it's own weight but it doesn't keep it centered. If it drifts off center, gravity doesn't correct for that, the increase in gravity on the side getting closer to the sun is exactly offset by there being more of the ring on the other side. Some sort of active stationkeeping is required.

A sphere has the same problem, and additionally has the problem of not being able to stay intact under it's own weight. As you mentioned spinning can keep it intact near its equator, but nothing can hold up the poles.
 
Last edited:
Yeah, I think the original idea of a Dyson sphere was to harvest all the energy from a sun, so I would imagine that the habitable zone would be around the equator - from simple physics considerations, it would have to be around the equator, because that's where the atmosphere goes if you spin it (you need to spin it to get any gravity on the inside; even artificial gravity wouldn't fix that one unless it violated the inverse square law)


It would have to be stationary, because if it was spinning it would be uninhabitable because of the thousand mile an hour winds at the equato...

Oh, sorry, wrong thread.
 
Yeah, I think the original idea of a Dyson sphere was to harvest all the energy from a sun, so I would imagine that the habitable zone would be around the equator - from simple physics considerations, it would have to be around the equator, because that's where the atmosphere goes if you spin it (you need to spin it to get any gravity on the inside; even artificial gravity wouldn't fix that one unless it violated the inverse square law)


Back in 1995, physicist and science fiction writer Robert Forward released an anthology called "Indistinguishable from Magic", which alternated scientific essays about advanced theoretical technologies and short stories featuring the technology. One pair dealt with using manufactured hyperdense collapsed matter to put enough mass into a small area to generate gravity without spinning. Something like sheets of collapsed matter under floors to create gravity pulling down against that floor. It's been so long since I've read the book, I don't remember if it seemed feasible.
 
Last edited:
If the Dyson Sphere is fully contained, how is the solar wind handled? These charged particles have to go somewhere. The Earth is protected by its magnetic field... how will the builders protect the inner surface of the Sphere?

Good question. If the Builders are smart enough to engineer a Dyson Sphere, but too stupid to provide for charged particle shielding on the inner surface, they're gonna have a bad time.
 
Also, the dynamic pressure of the solar wind might be less than 1 nanopascal, but that is enough to keep it doing speeds of over 500km/sec all the way out to the termination shock between the orbit of Neptune and the Oort cloud. If the Sphere is "sealed" the pressure inside would build up because the components of the solar wind (electrons, protons and alpha particles) have mass. Would most particles pass right through? Even if some do, not all of them will. Those that do not will impact and create pressure.

Old and Busted: Dyson Sphere

New Hotness: Dyson Rocket
 
Doesn't matter that it's spinning, that keeps it from collapsing under it's own weight but it doesn't keep it centered. If it drifts off center, gravity doesn't correct for that, the increase in gravity on the side getting closer to the sun is exactly offset by there being more of the ring on the other side. Some sort of active stationkeeping is required.

That was what I assumed when I mentioned the shell hypothesis. Turns out that... let's just say, it takes a big man to admit he was wrong, so... I'm going on a diet ;)

For the sphere, yes, i cancels out, you get zero pull in any direction.

But here's the thing. When you integrate in both cases for radius R and going along the circle with a dθ measured at the centre, for a sphere you get basically a mass contribution proportional with R*sin(θ)*dθ. For a ring you only have it proportional to dθ. The whole R*sin(θ) disappears from the expression you integrate.

It's not the same integral as for a spherical shell.

So, long story short, for a ring you don't get flat space over the whole space in the middle. You actually get an unstable system where getting closer to it on one side, actually results in a net pull harder on that side, making it basically collapse into the sun.

Essentially, if it helps understand the situation, it's 100% equivalent to the electric field inside a charged ring. Which actually is higher, the closer you get to the ring and thus the farther away you get from the centre (but still in the same plane as the ring.) It's not the same as the inside of a farraday cage.

I could try to transcribe the maths here, but I'd have to draw some pictures and put them in my album and so on. So I'll plead too lazy for a change.
 
Last edited:
For the sphere, yes, i cancels out, you get zero pull in any direction.

But here's the thing. When you integrate in both cases for radius R and going along the circle with a dθ measured at the centre, for a sphere you get basically a mass contribution proportional with R*sin(θ)*dθ. For a ring you only have it proportional to dθ. The whole R*sin(θ) disappears from the expression you integrate.

It's not the same integral as for a spherical shell.

So, long story short, for a ring you don't get flat space over the whole space in the middle. You actually get an unstable system where getting closer to it on one side, actually results in a net pull harder on that side, making it basically collapse into the sun.

You are correct sir, my mistake!
 
Yes, but of a scale quite out of proportion to our needs.
There is more to the human condition than mere survival. Say rather that our dreams and our desires will always scale in proportion to our available energy. And that however much energy we harvest, our dreams and desires will always spur us on to harvest more.

Someday, we'll be able to build such gargantuan collectors. And on that day, we'll already have in mind some use for all that energy. Hell, we'll even have people saying we need that much energy.
 
Maybe. But you'd assume that it would still be a matter of expecting a reasonable return on investment, within a foreseeable future.

1. A Dyson sphere is a bit of an all or nothing thing. You'd expect that putting another smaller collector around the sun, and then another, and so on, would promise quicker returns than waiting until we can build a gigantic sphere.

Basically it's the same reason why we'd rather build another nuclear power plant that is anywhere between 500MW (the smallest in the USA) and 4GW (the largest), rather than try to builld a TW one in one go.

RL solutions are more incremental than the spectacular show-boats to end all show-boats that SF thrives on.


2. Here's a more practical idea: OK, you build a sphere -- doesn't even matter if it's righid or several stations -- and collect the energy from the sun, and all. NOW WHAT?

So you have whole millions of square miles of solar panels collecting the light and turning it into electricity. NOW WHAT? How do you get that electricity from THERE to HERE?

Ah, you're gonna use some microwave emmiter or such, right? Sure, but that's transforming it BACK into photons to get that energy to Earth. But it was already in photon form to start with. Those were already carrying energy alone just fine. So, as Daffy Duck once said, "Was this trip really necessary?"

Better yet, they were originally in the perfect band to get through the air. Air becomes gradually less transparent the farther you get out of the visible spectrum.

Seems to me like you don't need some energy collector at all. What you need is just a frikken mirror, to guide those already existing photons to where you need the energy.
 
Is it worth pointing out that when Dyson originally proposed the idea of a Dyson Sphere he didn't propose a rigid structure, but:

He proposed a system of orbiting structures (which he referred to initially as a shell) designed to intercept and collect all energy produced by the Sun.
https://en.wikipedia.org/wiki/Dyson_sphere

Which, as far as I can see, is entirely physically possible. And could certainly be built in stages. In fact there's no need for an long term plan, just slowly taking advantage of available resources bit by bit.
 
If you're going the Dyson sphere route a Main Sequence G Type star wouldn't be my first choice.

A Dyson Sphere is going to be a looooong term investment for a civilization. A "Our Sun" type star is too large, too short lived, and too unstable. You've got a couple billion years before it either cooks or otherwise plays havoc with your civilization and only about 5 billion until it expands beyond the size of your sphere. By the time we get even begin to guess at the possibility of doing something on this scale we would only get a few billion years use out of it. That's like building an aircraft carrier to cross a stream.

Get yourself a nice Red Dwarf; very common so easy to find, much, much smaller in size (TRAPPIST-1 is only slightly larger, if much more massive, than Jupiter), some are much better behaved, and it is going to be one of the last energy producing items in the universe. And the downsides; that they tend to have only a tiny sliver of a habitable zone and the fact that any planet in that zone would probably be tidally locked, are a non-factor in this scenario.

Set yourself up a Dyson Sphere (or similar mega construct) around one of those babies and you're set on as long a time frame as civilization could really hope for or plan on. If your civilization is still around when the Red Dwarfs start dying off, you've had a good run by anyone's standards.
 
Last edited:
Seems to me like you don't need some energy collector at all. What you need is just a frikken mirror, to guide those already existing photons to where you need the energy.

I remember reading an article back in the 1970s about using a system of orbiting mirrors to direct the sun's rays down to a huge solar array on earth to generate gigawatts of electric power. It was the forerunner to this idea..

http://jxcrystals.com/publications/Self_Deploying_Mirrors_Solar_Power_from_Space_10-7.pdf

IIRC the artwork on page 4 is the same artwork in the original article I read, so it was probably the same one by this "Dr Ehric"

But we should be careful with space mirrors

https://www.theonion.com/astronomers-discover-planet-identical-to-earth-with-orb-1819577220
 
So you have whole millions of square miles of solar panels collecting the light and turning it into electricity. NOW WHAT? How do you get that electricity from THERE to HERE?

From there to here? You wouldn't do that. You'd live on the sphere. It's a gigantic space station with room for quadrillions of people to live.
 
It would have to be stationary, because if it was spinning it would be uninhabitable because of the thousand mile an hour winds at the equato...

Oh, sorry, wrong thread.

Well played sir. The bill for my new keyboard is in the mail.
 
Well, to some extent, yes, SF generally tries to scale up existing technologies and problems, rather than predict whatever will happen in the future. I guess if the Romans ever wrote SF, it would involve how to fit even more rows of oars on a ship, rather than come up with a screw.

But I guess another problem with at least a certain kind of SF is that it feels a compulsion to awe. In a "go big or go home" way, so to speak. So you end up with thousand mile long ships, with power cores measured in TW, and, yes, remaking whole solar systems from scratch.

I guess nobody wants to read or watch the story of a realistic spaceship that has all the size, space and comfort of a WW2 submarine, including hot-bunking. Because cross section counts for drag when you go relativistic, length counts for tidal forces when you're anywhere near a gravity well, and MASS counts more than both combined when you have to carry your own fuel.

An example is the way that science fiction from the '50's and '60's envisioned computers. There were many stories about computers with amazing capabilities -- self aware, intelligent, able to run the whole world, etc., but the computers were always envisioned as huge, expensive hardware installations, as mainframe computers in that era were. Very few stories foresaw anything like everybody having a computer on their desktop, and another one that they carried around in their pocket that also served as a telephone.
 
An example is the way that science fiction from the '50's and '60's envisioned computers. There were many stories about computers with amazing capabilities -- self aware, intelligent, able to run the whole world, etc., but the computers were always envisioned as huge, expensive hardware installations, as mainframe computers in that era were. Very few stories foresaw anything like everybody having a computer on their desktop, and another one that they carried around in their pocket that also served as a telephone.

Or a huge, massively capable computer that consisted of a network linking hundreds or thousands of individual, low-powered systems (i.e.-folding at home).
 
I won't be happy until we've built a discworld. :| ;)
 
Last edited:
Despite my job title at several companies, I never claimed to be an engineer (I preferred CAD Guy) and even I looked at both ideas and thought they were the dumbest Rule 10 I'd ever seen.
 
An example is the way that science fiction from the '50's and '60's envisioned computers. There were many stories about computers with amazing capabilities -- self aware, intelligent, able to run the whole world, etc., but the computers were always envisioned as huge, expensive hardware installations, as mainframe computers in that era were. Very few stories foresaw anything like everybody having a computer on their desktop, and another one that they carried around in their pocket that also served as a telephone.

Well, I'll mostly agree. But to be fair the problem is less that supercomputers still exist in that universe, and more like the last part of what you wrote: the absence of anything on the lighter side of things.

I mean, smartphones may be more known to everyone, but the mainframe isn't quite dead yet (BM is still making a mint selling them, only FSM knows why) and supercomputers are still room sized and not particularly dead either (Cray is still alive and well.)

Plus, the idea of putting just a dumb terminal on everyone's desk and making them connect to a big fat computer is one of those silly ideas that never die. You know, why have a computer on your desktop that can do all the compiling or CAD work you'll ever need, when you could pay ten times as much on a big fat supercomputer in the basement and on the kind of network that can haul the video for 1000 users in real time, and still end up slower than with the computer? It's practically a no brainer! (As in, people with a brain need not apply;))

Just when you think you've buried it at crossroads with a stake through its chest, someone drops a drop of blood on its tomb and it springs right up again. I'm very sure I've heard it AGAIN in the 21st century, not in the 60's.

So I'm not entirely fussed if some ship or institute or government in SF still has SOME big computers. It's more like the lack of anything below that that's a very failed prediction.
 

Back
Top Bottom