The Dyson sphere temperature.

[Connor MacLeod]

I think he's assuming that if it's not "natural", then the applicable laws of physics must be totally different, so it must require a constant input of energy. Pretty common reasoning, actually. My falling-out with Robert Brown was over this exact issue, as it related to landspeeders in SW.

I think I remember that. I can also remember you trying to repeatedly drive that point home that such devices don't neccesarily need a constant input of energy to function (especially since it begs the question of where all that energy goes)

Same with deflector shields as I recall, actually, since shields also are commonly assumed to require a massive, constant input of energy to function (despite the thermodynamic implications.)

[Wyrm]

We don't just need to keep it inside, we need to keep it on the inner surface of the sphere. I am merely asking by what measures can you achieve sufficient gravity that no air will be sucked into the star from any spot of the sphere. The sphere is not a planet, its inner surface has no natural gravity.

Spinning the sphere. The things stick to the inner surface of the sphere by centrifugal forces. (Or, more accurately, to a relatively thin belt about the equator, but that accounts for millions of Earths.) For an internal gravity of 9.8 m/s², a 200 million kilometer Dyson sphere requires a rotation rate of ~0.13 revolutions a day (once every week and a bit).

Please note that no power is needed to maintain this gravity — the sphere is in a nearly frictionless environment so there is no loss of rotational energy.* (The Dyson sphere will not be filled with gas. Atmosphere sticks to the belt for the same reason you would: the artificial gravity. Stars are noisy things — if space were filled with breathable atmosphere, we would be instantly deafened by the noise from the sun.)

* In general, power input is required if there is power loss elsewhere, or if the system is measurably accumulating energy. CoE

[Junghalli]

Spinning the sphere. The things stick to the inner surface of the sphere by centrifugal forces. (Or, more accurately, to a relatively thin belt about the equator, but that accounts for millions of Earths.)

Seems like rather a waste of material to build a whole sphere if you're only going to use a thin slice at the equator though. If that's your plan it would seem much more efficient to build a ringworld instead. Unless perhaps the uninhabitable regions are giant solar collectors tapping the central star's energy for whatever such a civilization might need huge amounts of energy for.

[Wyrm]

Spinning the sphere. The things stick to the inner surface of the sphere by centrifugal forces. (Or, more accurately, to a relatively thin belt about the equator, but that accounts for millions of Earths.)

Seems like rather a waste of material to build a whole sphere if you're only going to use a thin slice at the equator though. If that's your plan it would seem much more efficient to build a ringworld instead. Unless perhaps the uninhabitable regions are giant solar collectors tapping the central star's energy for whatever such a civilization might need huge amounts of energy for.

The point of a Dyson sphere is to catch all the radiation from the star, which will upgrade your civillization to Type II. The extra elbow-room is just gravy.