# Is there an upper limit on the amount of light that a Dyson Sphere can attenuate?

• I
I'm not sure this a strictly Astronomy question; perhaps it should go in Aerospace Engineering. I have always thought that Dyson's work was more of an Astronomy topic, although admittedly, from the POV of observing some other system's sphere.

In any case, I was thinking about future intelligent life on Earth having to deal with the Sun going into its red giant phase. It would seem that a Dyson Sphere could reflect the radiation away and keep Earth nice & cool. I suppose that alternatively that there could be some type of sphere put around the Earth itself instead, as the surface area would be much lower (no reason to attenuate the Sun's radiation to everyone else!) If the latter were done, it seems that based simply on the heat-retardation technology of the Webb telescope, only a foot or so of material in the Earth's crust would be needed to construct the actual shield, which could be doable by a civilization completely harnessing solar energy and robot construction workers.

EDIT: I've just read about a conceptual technology called star lifting. I suppose that future civilization could simply blow most of the red giant Sun away? Of course, this would come back to bite them once the Sun becomes a white dwarf. Or perhaps after blowing away most of the Sun, it becomes a red dwarf, thus lasting trillions of years?

Last edited:

In such a far future there is no reason to keep Earth alive other than sentimentality and whoever wants to preserve the Cradle of Humankind wouldn't disfigure it with mega structures. Star lifting makes more sense.

If we have the ability to make a dyson sphere, we'll have been long already had the ability to get at least some part of humanity away from the solar system, even if it doesn't actually involve live humans traveling through space.

sophiecentaur
Gold Member
2020 Award
The first question that comes to my mind about a Dyson Sphere is 'what could you build it from?'. Presumably it would need to be a sphere with radius about 1AU (temperature considerations). Available materials in the present solar system would limit the thickness of the sphere (less than 1mm probably as you can ignore all the gases of the gas giant planets). They'd need to go elsewhere for materials - nearby planetary systems?

Wouldn't any civilisation that could contemplate a Dyson sphere already have fusion technology totally sorted out? So why would they need 'Solar' Energy?

Eric Bretschneider
Presumably it would need to be a sphere with radius about 1AU (temperature considerations).

1 AU is quite small for a Dyson Sphere. The bigger the better.

Available materials in the present solar system would limit the thickness of the sphere (less than 1mm probably as you can ignore all the gases of the gas giant planets).

I would estimate around 1 mm for a radius of 5 AU (not considering transmutation).

Wouldn't any civilisation that could contemplate a Dyson sphere already have fusion technology totally sorted out? So why would they need 'Solar' Energy?

It is impossible for us to guess the motivations of such a civilisation.

sophiecentaur
Gold Member
2020 Award
1 AU is quite small for a Dyson Sphere. The bigger the better.

I would estimate around 1 mm for a radius of 5 AU (not considering transmutation).

It is impossible for us to guess the motivations of such a civilisation.
I guess my 'off the top of the head' guess of 1AU wasn't't too bad. I have read abut the Dyson sphere but it doesn't really float my boat. Guessing motivations would be more a matter of guessing the motivations of the human proposers, I would have thought. I don't know why bigger size would be better or worse. Any size would be impressive for another civilisation to stumble across, a billion years after it was built. Clearly the builders would have been near the end of their civilisation if things like the DS were all they had to occupy themselves with. It's the sort of activity that I sometimes consider for the front garden. Folly would be a good word for it. Human Follys are / were built on a more modest scale.
We could do Angels on a Pinhead next.

hutchphd
Guessing motivations would be more a matter of guessing the motivations of the human proposers, I would have thought.

Even understanding humans can be challanging. Why were some ancient civilizations (e.g. Egyptians or Mayas) so obsessed with pyramids? They invested an insane amount of resources to build monuments with no obvious purpose. We wouldn't beleave it if they wouldn't be there. Archaeologist explains it with religion (as everything we don't understand). That would just mean that civilisations sometimes act irrational and therefore are unpredictable. Maybe there have been practical reasons (e.g. impressing inferiors or potential enemies) or maybe it's part of human behaviour (do we need Burj Khalifa?). Who knows? If we can't explain what our ancestors did, how can we predict what our successors will do?

I don't know why bigger size would be better or worse.

That refers to the use as solar power plant. Increasing the size increases the maximum efficiency because the outer surface gets colder (and therefore releases more entropy with the waste heat). Low temperatures also make it easier to use superconductors.

Any size would be impressive for another civilisation to stumble across, a billion years after it was built.

Dyson Spheres are instable and need active control. I don't think they would survive for a billion years without somebody or something that understands the physics behind it.

sophiecentaur
Gold Member
2020 Award
I have to be a bit skeptical about the DS. For a start, any civilisation that could actually build one would need to invest so much Energy that they would already have immense resources. I reckon the whole idea sounds a lot like a human thought-experiment with loads of post hoc justification. It's so big a scheme as to be non-falsifiable, at least for a few millennia.

As for the motivation for building huge, whacky structures in history, it was all based on religion (all the ones I'm aware of) and the rationality of modern, tech based societies means they are less susceptible to religious influence / control. There are other available methods for controlling the population so religion is less and less likely to work for 'the ruling classes'.

To be fair, the DS is probably more believable than ftl because the limits are practical rather than fundamental.

I suppose a Dyson Swarm is more realistic than a sphere, but they're both incredibly unrealistic.

sophiecentaur
Gold Member
2020 Award
I suppose a Dyson Swarm is more realistic than a sphere, but they're both incredibly unrealistic.
They certainly are. Collecting the energy is only part way there. You'd need to 'beam it' down to home planet in such a way as not to fry the population, yet efficiently enough not to waste most of what the DS produced in the first place. We have enough problems now with the climate getting out of hand so adding to the Sun's supply of Energy would be yet another consideration.
But (I hear you all cry) we could sort all that out. Oh yeah? Who would be in charge of doing that?
Sorry for raining on parades but the DS almost deserves to be re-directed to a more fictional forum. We have enough trouble with the Space Elevator, when that is only just unrealisable.

They certainly are. Collecting the energy is only part way there. You'd need to 'beam it' down to home planet in such a way as not to fry the population, yet efficiently enough not to waste most of what the DS produced in the first place. We have enough problems now with the climate getting out of hand so adding to the Sun's supply of Energy would be yet another consideration.
But (I hear you all cry) we could sort all that out. Oh yeah? Who would be in charge of doing that?
Sorry for raining on parades but the DS almost deserves to be re-directed to a more fictional forum. We have enough trouble with the Space Elevator, when that is only just unrealisable.

In my opinion, the most realistic way some far, far future ultra technolohically adcanced civilization would get their power would be fusion. Stripping the gas giants if they had to. Sounds unrealistic, but much more realistic than a dyson sphere. If they were totally efficient, theres enough raw materials too sustain them for a very long time.

If it turns out fusion cant be mastered, I doubt any civilization would ever be able to exist that long honestly.

You'd need to 'beam it' down to home planet

Why does the energy need to be used on a planet? Why would you have a planet in the same system at all? That is just waste of material.

stefan r
sophiecentaur
Gold Member
2020 Award
Why does the energy need to be used on a planet? Why would you have a planet in the same system at all? That is just waste of material.
OK. But it has to be used somewhere for making or doing something. That energy needs to be converted and 'beamed' efficiently and beam dispersion over large distances would nullify any advantage that the DS would give you. The aperture (several AU) would be a help with this but then, at the other end of the link, the energy would need to be re-directed to somewhere where it could actually be used. I guess the idea is Manufacturing. But then the timescales get very difficult if the distance scales are stellar.
My feeling is that the idea is to add more and more orders of magnitude to the model that people just stop being able to give it serious thought. We're into Drake equation figures. Just what sort of civilisation would be that far-seeing to invest so much into the fortunes of individuals, living millennia in the future?
Lets face it, humans are now doing their best to spend their own kids' inheritances and are more than happy to accept that the next generations will have no chance of buying their own homes until they are into their forties and fifties. So, not only are we discussing a technology that's unbelievably advanced, compared with ours but also a totally alien attitude to the future of their species. That's the real SciFi part of the idea, imo.

OK. But it has to be used somewhere for making or doing something.

You can use it on the Dyson Sphere itself (e.g. to power a computer that covers the entire surface).

That energy needs to be converted and 'beamed' efficiently and beam dispersion over large distances would nullify any advantage that the DS would give you.

If you beam it at all why not to stations orbiting the sphere close to the surface?

I guess the idea is Manufacturing.

The idea is to harvest energy. That's something we can deal with. It makes no sense to speculate what this energy might be used for in a far future. A caveman could deal with fire but nobody would expect him to come up with the idea of turning the heat into electricity and use it for Bitcoin mining.

Just what sort of civilisation would be that far-seeing to invest so much into the fortunes of individuals, living millennia in the future?

Nobody knows. We don't even know if it would be a civilisation. All we know is physics. We can discuss if and how a Dyson Sphere could be build in general but not by whom and why.

sophiecentaur
Gold Member
2020 Award
Nobody knows. We don't even know if it would be a civilisation. All we know is physics. We can discuss if and how a Dyson Sphere could be build in general but not by whom and why.
Your answers are all fair enough and pretty consistent. It's just a matter of personal taste as to how attractive a person finds this sort of speculation. I find it far too much extrapolation to take seriously but I realise it grabs many people.

I was thinking along the lines of putting a sphere around the Earth, so as to deflect energy when the Sun starts getting hot. Even if Earth gets swallowed up by the Sun when it is at its peak red giant stage, by then the temperature of the outer planets would get warm enough to boil away most of the atmosphere, and Earthlinkgs could move there.

Think about the practality of constructing a DS in the first place the unconnected parts would have to be in orbit otherwise they would just fall into the star. This means you would first have to construct a ring in the orbital plane. How would you construct the rest of it as anything out of the orbital plane would tend to be pulled in by the imense gravity of the star unless it was made strong enough to withstand these forces. I agree the whole concept is a thought experiment.

Think about the practality of constructing a DS in the first place the unconnected parts would have to be in orbit otherwise they would just fall into the star.

That applies to a Dyson Shell but not to a Dyson Bubble.

swampwiz
Once the "bubble" is completed are you suggesting that the radiation pressure would be enough to keep it inflated against gravity.

Yes, that is possible at least in theory. Building the Dyson Bubble could start with a Dyson Swarm and than with increasing radiation pressure gradually turning the satellites into statites.

stefan r
In my opinion, the most realistic way some far, far future ultra technolohically adcanced civilization would get their power would be fusion. Stripping the gas giants if they had to. Sounds unrealistic, but much more realistic than a dyson sphere. If they were totally efficient, theres enough raw materials too sustain them for a very long time.

If it turns out fusion cant be mastered, I doubt any civilization would ever be able to exist that long honestly.

Fusion power plants still use the Carnot cycle. In order for the Carnot cycle to work you need a heat sink. There will be radiators. The radiator surfaces give off infra red. That star has a Dyson swarm.

The Sun is a fusion power supply. Why would you not use it?

Lithium and deuterium stocks exists but they are limited. As energy supplies they would be useful in the Kuiper belt, Oort cloud, and in interstellar colonies. Deuterium has uses in transmutation and lithium has numerous technology applications. Accessing deuterium reserves in Jupiter, Saturn, and/or the Sun would require the energy surpluses that become available with Dyson swarms. If you do not have one then taking apart planets will be difficult.

OK. But it has to be used somewhere for making or doing something. That energy needs to be converted and 'beamed' efficiently and beam dispersion over large distances would nullify any advantage that the DS would give you. ...

This is not standard for descriptions of Dyson Swarms people want to build in the Solar System. Check out Gerard O'Neill island III cylinder. It is not likely to look exactly like that design. Technology like LED lighting make the windows and Sun centered mirror panels unnecessary. The habitats in a Dyson swarm would occupy the habitable zone. Most energy collectors would deliver to a very nearby customer.

...
My feeling is that the idea is to add more and more orders of magnitude to the model that people just stop being able to give it serious thought. We're into Drake equation figures. Just what sort of civilisation would be that far-seeing to invest so much into the fortunes of individuals, living millennia in the future?
Lets face it, humans are now doing their best to spend their own kids' inheritances and are more than happy to accept that the next generations will have no chance of buying their own homes until they are into their forties and fifties. So, not only are we discussing a technology that's unbelievably advanced, compared with ours but also a totally alien attitude to the future of their species. That's the real SciFi part of the idea, imo.

The idea that economy and energy consumption will grow exponentially is religiously clung to by economists and politicians. Perhaps that wing of the university may be insane while you are perfectly sane. They are in control of civilization though so do not tell them too bluntly. Just suggesting they look for a way to stop exponential growth will get you labeled an anti-natalist neoMalthusian or communist.

The math is straight forward. If energy consumption grows at 3% annual the doubling time is under 24 years. In 240 years a factor of 1000x. Within in a millennium it needs to be 10^12 higher energy consumption. We can plug in other growth rates. At 1% annual the doubling time is 70 years.

Energy return on investment for silicon solar panels is currently estimated at well under 2 years. The solar power available on Luna's surface is much higher than on Earth's surface because there is no atmosphere or clouds and very slow dust accumulation. Getting to the point where you have a lunar surface industry that can make 1 functioning PV panel is a huge hurdle. Lets not minimize the effort that project will take. However, once (if) that hurdle is cleared there is not anything obvious preventing production from expanding exponentially. If half of production is diverted to other things the Lunar semiconductor industry could still be doubling itself annually while sprawling across the far side. After the first square kilometer solar farm is in place it takes only a few decades to cover Luna. Kids in school today could see it built before they retire even if there is no life extension.

A hundred terrawatt power supply on Luna can power a substantial mass driver. We could eliminate all extraction activity on Earth's surface. Lunar calcium could be used to make Portland cement. That would build houses and also work as a carbon sink.

You and I do not have a reasonable motive for wanting more than a 100 terawatt power supply. But the replicating system of automated robots that did this on Luna will be made of known off the shelf technology. Why not send some missions to the belt and to Mercury? Why stop the replication process after Mercury's arctic is utilized? Is there a reason we would want to insist on radiating visible light out the Sun's north pole instead of infra-red?

I think it is nice to consider civilizations might put a stop to expanding energy consumption. Humanity could increase energy supply by a factor of 1 million and the Sun's infra-red excess would still be lower than the current zodiacal light. That is enough power to easily boil the oceans. Interstellar travel might be possible using much less energy. This limited growth in energy allows for the possibility that aliens could be anywhere or almost everywhere (except here) in the Milky Way. Is possible that civilizations will put a stop to most asteroid collisions and will accelerate cleaning up the dust. The colonized stars could have a lower infra-red excess. Sexy aliens living all around us is definitely more fun than the drudgery work of expansion the economics department wants to force us to do.

For astronomy the important thing is measuring the infra-red excesses. Alpha Centauri has 10 to 100 zodis. Vega like stars have orders of magnitude more infra-red.

swampwiz