Hacker News Reader

Comment by sp332 | Hacker News Reader

How is energy extracted from the reaction? Once you get a bunch of energetic particles flying around, how to do you get from there to electricity?

View on HN

I don’t know if stellarators have a different plan, but in a production tokamak, it’s a combination of:

- deuterium-tritium fusion, which produces lots of high-energy neutrons

- the neutrons help to keep the plasma hot, but being neutral, they escape the magnetic bottle; there is a lithium blanket around the reactor to catch them

- the neutrons that hit the lithium produce tritium for the reactor, and steam for the turbines

Wasn't there a fusion project where the end product was a charged particle so energy could be extracted from currents generated by changes in the magnetic field as they slowed down?

Here's one such promising project: https://lppfusion.com/technology/fusion-energy-generator/

"Dense Plasma Focus device with aneutronic hydrogen-boron (pB11) fuel"

Same way it is pretty much everywhere, including in nuclear power plants. Heat water and use the steam to run a turbine.

I have always wondered, is heat->steam->turbine->generator efficient? In my mind it just doesn't seem like it would be efficient at all. Surely there must be other ways to convert heat to electricity (Thermoelectric effect?), but why do we still use steam?

I’ve long wondered this too. Intuitively, it just seems like the losses from all those energy transfers would be unacceptably great.

The thermal efficiency of turbines can approach 90%, which is far and away better than any other method we have. Turning energy into heat is usually not a problem.

And in a lot of cases extra heat can be put to use beyond electricity production. Steam heat and cooling is a thing for large metropolitan areas, and it can also be used for desalination water. A few Soviet nuclear plants were designed to provide both electricity and water, actually.

Peltiers and other thermoelectric generation is really really inefficient compared to basically any other energy capture method. They're about 5-15% max that we know how to make currently. Basically they only get used for things like RTGs where you need a zero maintenance very long lasting electric source; in the US that's mostly been spacecraft but in the USSR they were used for extremely remote lighthouses around the Arctic Circle where solar was impractical and resupply (and maintenance) of a diesel generator was prohibitively expensive.

Current thermoelectric tech has very low efficiencies. It's an active area of research where any small improvement could be significant. It would be an ideal way to produce power more or less directly from thermal neutron bombardment, IF a method / material can be found that can make direct thermoelectric conversion efficient enough to compete with a water or molten salt heat exchanger loop.

Any engine is limited to the carnot cycle; the themoelectric effect still requires a cool side and heat gets conducted along the two leads to equalize the temperature. That's where the power comes from, the temperature differential across the thermocouple wires, and you consume the temperature differential when you generate electricity.

Different types of engines have different details of where that energy gets lost, but ultimately they are devices that if left to run indefinitely would equilibrate in some way and the way that they idle tells you a fair bit about where the energy flows are.

Turbines are pretty impressively efficient, but something can always be better. They often utilize the low grade waste heat in steam heating though, so a lot of times the efficiency is extremely good when you include free heating or process heat (in a plant they might use that steam to heat another piece of equipment).

> ...why do we still use steam?

It hits the sweet spot on fiscal, engineering, safety, and science fronts. Energy extraction is very challenging, we're likely hundreds of years away from aneutronic fusion, for example. The research area you are asking about is direct energy conversion, of which aneutronic fusion is one small branch (though within it, there are many scientific and engineering branches to explore).

I think I may have said this before, but a statement like "we're likely hundreds of years away from aneutronic fusion" seems to me necessarily meaningless.

We're only a bit over a hundred years from developing quantum mechanics and relativity, how can we possibly say anything about hundreds of years from now?

If something is well enough understood to accurately predict the timeline, we could do it much sooner.

If we have no idea how to do something, then "hundreds of years" means nothing, except maybe "not proven impossible yet".

I think it basically heats up the walls of the chamber, from which you can use the heat to make electricity in the usual way.

Reply on news.ycombinator.com