[Retsof]

So I was thinking about futuristic power sources, such as fusion.  Now, all the power sources we have, barring hydro, solar, and wind, basically work on steam.  Nuclear, coal, oil, all work by heating steam to turn generators.  So, would future power sources work the same, or are there better ways to get from raw energy to usable electricity?

[perihelion]

Depends on the source and the form the energy takes.  Most energy doesn't start out thermal.  In the cases you mention, the energy we're tapping into is either chemical (oxidation) or nuclear.  It is possible to convert some chemical energy directly into electrical energy, i.e. fuel cells.  There is ongoing research into making fuel cells work with something other than hydrogen.  Usually involves catalysis of one variant or another to try and get back to hydrogen for the final reaction.  I was friends with someone working on this maybe 10 years ago, but I don't know what if anything came from it.

But for the most part, chemical and nuclear energy is most easily captured as heat.  When you are dealing with thermal energy, converting to electrical energy directly is possible using the Seebeck Effect.  But there are major obstacles to doing this on the scale required to power a house, let alone a city.  While I do hold out hope that they will be surmounted in time, nothing yet attempted even approaches the robustness of a boiler / turbine loop.  One of the big advantages of these big inefficient heat engines is that you can kick the **** out of them and they keep on working.  Part of that is because it is an old technology that has gone through lots of refinement.  But another part is because there is a lot of "dumb iron" involved.  This is not delicate machinery, and it can therefore take a lot of abuse.  Currently known materials which would be suitable for Seebeck thermal-to-electric power conversion are not particularly strong.  Granted, you are taking the mechanical stress out of the system because it would have few-to-no moving parts, but thermally induced stresses would still be significant.

But hey, there's a fair amount of research in this area.  Someone may come up with an approach that scales up effectively.

[Nuke]

he3 fusion is supposed to be able to generate electricity directly. though id rather see money put into heavy ion fusion instead as we dont have a readily available he3 supply.

[Locutus of Borg]

Fusion would require somethng to be heated to 5000k

[Bobboau]

possibly, but the idea is that you would be able to tap the energy directly via magnetic flux. which would be cool to say the least.

[Nuke]

even then im sure the reactor would need to deal with waste heat, and what better to do with it to use it to make steam and generate more power. steam aint going anywhere, its gonna be used for hundreds if not thousands of years to come.

[Klaustrophobia]

Fusion would require somethng to be heated to 5000k

the ion "temperature" is in the millions actually.  or were you talking about the structures?

current magnetic-confinement fusion designs have a component of direct ion capture that is around 95% efficient.  the actual amount of energy extracted this way is still quite small however, and most of it is still a traditional rankine cycle.  it's hard to do direct conversion in the same quantities as thermal cycles.  there's also the fact that there's just no getting around energy ending up as heat.  might as well make the best use of it you can.

[Retsof]

Hehe, so all these futuristic space ships with fusion generators are steam-powered!