What if we made a really really huge one? I'm talking 15 × 10^28 kg worth of hydrogen. At that size it would start and maintain itself. We could place it far enough away that we could harness its output safely here on earth with solar cells!

The problem is, something that massive would affect the motion of the Earth itself.

I think it was a joke hinitng at the sun, but it seems to be @ 1 order of magnitude off, so I'm just confused now.

I was going for the smallest amount of hydrogen that would form a star. I suppose the joke would be improved if I used a number measuring the Sun's initial hydrogen content. I suppose that's why I'm not a comedian :P

And we won't have to solve the problem of plasma confinement!

It would bombard the Earth with dangerous levels of radiation.

I suppose it could work, we would just need to be careful about atmospheric composition to manage radiative cooling, and block some harmful em frequencies. But I don't foresee any problems with that.

The radioactive waste is quite unlike fission. Fusion reactors will generate low-level radioactive waste (from parts of the machines being irradiated), but fission generates high-level waste which must be sealed of for thousands of years (from fission products).

I support your plan, and present a proposal for phase two: control and efficiency improvements.

The density of the plasma determines the rate of fusion, and is itself determined partly by temperature. If we were to reflect a significant fraction of the emitted radiation back, then the plasma and its surrounding gas should expand, reducing pressure on the fusing core. That should in turn reduce the rate of fusion, lowering fuel consumption.

In theory, we could even diffuse the entire hydrogen supply into a warm cloud that is not dense enough to fuse at all, and store this until we need more heat. As the system cools, it will contract, eventually reigniting itself.

Even if we fail and die, it will still reignite itself, which clearly demonstrates the inherent safety of my proposal.

> And just the other day we had a link here where a retired physicist was saying fusion is not likely to work, and if it does it will produce radioactive waste much like fission.

He didn't really say it won't work, he mostly just said it wouldn't be as clean as people think it will be. But you are right, it does put news like this one into perspective.

Link to the HN discussion from two days ago: https://news.ycombinator.com/item?id=14202488

> We could make it safe by having it operate in vacuum at a safe distance from the earth.

And if the thing happens to be rotated even by a milli-arc, for example due to asteroid strike or a simple software error, the receiving area will be fried by microwave. Or hackers might want to introduce a deliberate rotation to use it as an orbital based frying weapon. No thanks, that's way too risky.

Your parent comment is a joke about the sun as a fusion reactor, in case you couldn't tell.

The actual failure case to be worried about is that the reactor blows up towards the end of its life cycle.

He's talking about the sun. We already have a giant fusion reactor that we're not using.

"Not using"?! Literally all of the energy we consume on earth, except the radioactive material used in our nuclear plants, is a byproduct of said fusion reactor :)

And it's reactors all the way down, mate! That radioactive material was created by an ancestor of our current fusion reactor.

Geothermal isn't :)

This criticism is usually brought up with this kind of energy transfer (solar beamed down from orbit etc.). But would it not be rather simple to build in some kind of feedback mechanism, where the ground station beams back at the orbital unit as long as it is receiving an incoming beam?

It would act as a switch, where, as soon as the incoming beam stops, the feedback stops and the orbital transmitter would stop too, e.g. if the energy beam went out of position.

Don't forget about the speed-of-light delay...