Compact nuclear fusion reactor is 'very likely to work,' studies suggest • Hacker News Board

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#261

Post by grandinj » Wed, Sep 30, 2020, 8:10 AM UTC

physicsguy wrote: post #259

So, I'm a physicist, and I went to a number of talks from people involved with the JET fusion reactor over the years, though fusion is not my area. And my understanding is not that it's difficult to make plasma, or even build a reactor in particular that is the main problem (though instabilities can be problematic), but it's that the internal structure degrades very rapidly and becomes highly radioactive, because you…

This++++++. Nuclear reactors are primarily bug generators for chemists and metalurgists. The things that need solving mostly boil down to dealing with the insane chemistry and metallurgical problems that arise when you have a wide variety of incredibly nasty compounds arising that do not normally occur, and consequently require out-of-the-ordinary solutions.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#262

Post by ClumsyPilot » Wed, Sep 30, 2020, 8:27 AM UTC

Maursault wrote: post #220

ClumsyPilot wrote: post #173

Earlier quoted context omitted.

Thats not accurate at all, Uranium itself is much cheaper than coal / any fuel per unit energy. Some contries have permanent, final storage for nuclear waste, it does not cost any more once created. Furthermore, not only was solar power borderline non existent when those reactors were built, it is also still intermittent. You are just glossing over the biggest challenge of energy - balancing the powe grid. No-one nee…

I'll respond to your straw man. It isn't about the energy within the thing, the exorbitant cost of Uranium is in mining and refining. No one said anything about coal, but it doesn't need refined, and is cheaply mined... but you've also made a false equivalency, because coal isn't the only competition to nuclear. The sun just shines, wind just blows, water just flows, and the geothermals just produce heat, without any…

Two questions:

You keep repeating this idea thay uranium is expensive, by what metric? Where do you get the idea that its expensive?

If nuclear is so expencive, why does France have cheapest elecrticity in EU? And Denmark /Germany have invested in renewables have expensive electricity.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#263

Post by nmca » Wed, Sep 30, 2020, 8:28 AM UTC

jagiammona wrote: post #195

sien wrote: post #148

Earlier quoted context omitted.

The Fusion Triple Product, which is the key determinant of whether there is a self-sustaining reaction, increased faster than Moore's law up until about 2005. Then all the money went into ITER. https://physicsworld.com/wp-content/uploads/2004/01/pwhoa4_0... From the article at : https://physicsworld.com/a/controlled-fusion-the-next-step/ more information about the triple product and the Lawson Criterion. https://en.w…

Here's a great article showing progress improving the triple product until now. https://www.fusionenergybase.com/article/measuring-progress-...

Well now I really want to see the projected SPARC reactor on that plot. Does anyone have the necessary familiarity to add it?

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#264

Post by OJFord » Wed, Sep 30, 2020, 8:42 AM UTC

varjag wrote: post #260

physicsguy wrote: post #259

So, I'm a physicist, and I went to a number of talks from people involved with the JET fusion reactor over the years, though fusion is not my area. And my understanding is not that it's difficult to make plasma, or even build a reactor in particular that is the main problem (though instabilities can be problematic), but it's that the internal structure degrades very rapidly and becomes highly radioactive, because you…

That's a luxury problem at the moment. A fusion reactor that can sustain positive output operation for 5 years is still a dream.

Sort of, but only if you exclude the amortised cost of fabrication from your calculation - it's part of the same goal, unlike say 'they're hard to make, we can't make them fast enough to put a dent in energy production'.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#265

Post by uCantCauseUCant » Wed, Sep 30, 2020, 8:43 AM UTC

physicsguy wrote: post #259

So, I'm a physicist, and I went to a number of talks from people involved with the JET fusion reactor over the years, though fusion is not my area. And my understanding is not that it's difficult to make plasma, or even build a reactor in particular that is the main problem (though instabilities can be problematic), but it's that the internal structure degrades very rapidly and becomes highly radioactive, because you…

It needs a paradigm shift, away from a fixed structure, towards a structure that repairs part of its "exposed" skin constantly.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#266

Post by pm90 » Wed, Sep 30, 2020, 9:02 AM UTC

physicsguy wrote: post #259

So, I'm a physicist, and I went to a number of talks from people involved with the JET fusion reactor over the years, though fusion is not my area. And my understanding is not that it's difficult to make plasma, or even build a reactor in particular that is the main problem (though instabilities can be problematic), but it's that the internal structure degrades very rapidly and becomes highly radioactive, because you…

Not disagreeing with you, but it might still be ok. I don't believe early fission reactors were cost effective either. If we build a few with heavy subsidies and create an economy around them, we may get more commercial players involved, and a lot more investment dollars chasing the problem.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#267

Post by pjc50 » Wed, Sep 30, 2020, 9:14 AM UTC

samatman wrote: post #167

Tossrock wrote: post #150

I wish these people the best, and I really hope they get a working fusion plant soon. That said, I can't resist sharing Admiral Rickover on academic reactors vs practical reactors: Important decisions about the future development of atomic power must frequently be made by people who do not necessarily have an intimate knowledge of the technical aspects of reactors. These people are, nonetheless, interested in what a…

It's a good contribution to the conversation, keeping in mind that Adm. Rickover wrote these words in 1953. Anyone could have said similar things about computers in 1953, and been just as correct. Are nuclear reactors computers? Of course not, and neither have practical reactors kept pace in development with practical computers. But neither is it inevitable that steady progress cannot grind down the latter set of cha…

Similar things can be said about computers: the Mill CPU is simple, small, cheap, fast etc and of course can't actually be bought right now.

Vaporware is a real problem in every discipline.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#268

Post by yholio » Wed, Sep 30, 2020, 9:25 AM UTC

dnadler wrote: post #140

yholio wrote: post #134

About time to consider alternative paths to fusion. ITER and similar projects are abject failures from non-scientific perspectives, they fail to improve on the economic weaknesses of fusion (radiological waste, massive capital costs, scarcity of fuel, proliferation risk), and only deliver on issues that have become irelevant for modern fision, like the risk of a meltdown. There is zero economic potential for any ITER…

I wasn't aware of any radioactive waste from these fusion processes. I was also under the impression that fusion fuel is quite abundant (Deuterium and Lithium as an example discussed in one of the videos linked in the comments).

There are no fission waste products but the reactor deals with massive neutron fluxes that activate all internal components, coolants and the internal fuel fluxes that need to be processed in order to keep the reaction going. The Deutherium-Lithium route is not direct, you need to breed Tritium, a hazardous radioactive gas. There is poor neutron economy and you need neutron multiplicators that also become activated.

While I agree the quantity of waste is low, that doesn't really matter for the general public. We had for decades the technology to put fission waste in deep geological storage, what held it up were political concerns - the same for fast breeders that could burn the waste. So I cannot for the life of me understand how NIMBY-ism, the major cost driver of fission plants (via political challenges, court actions, schedule slips etc.), is allayed, when the plant will regularly ship out tons of hazardous materials through the communities they serve.

As for the fuel cost issue, you cannot use natural Lithium due to low cross section, but blankets made up of tons of enriched Lithium that has a large Li6 content, that are continually circulated and need to be topped off as H3 is bred. No estimate of cost for this feed-stock exists, but it is likely more expensive than natural uranium that can be burnt in a heavy water reactor, for example. (that reactor design has it's own issues with heavy water inventory costs, but this material is most probably easier to produce than enriched Li6, and is not consumed as a fuel)

When you draw the line, the best prospects of ITER derived fusion plants (not existing experiments, mind you, but the theoretical future, practical designs) is at best comparable to existing fusion plants. Why should we waste money on them, if they cannot improve on the current state of the art?

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#269

Post by Mvandenbergh » Wed, Sep 30, 2020, 9:38 AM UTC

Tossrock wrote: post #150

I wish these people the best, and I really hope they get a working fusion plant soon. That said, I can't resist sharing Admiral Rickover on academic reactors vs practical reactors: Important decisions about the future development of atomic power must frequently be made by people who do not necessarily have an intimate knowledge of the technical aspects of reactors. These people are, nonetheless, interested in what a…

Yes, that's very important but it is also important to remember that this was in a particular context and that the "pick a design and just build it" approach only works when you have a design (however imperfect) that is going to work.

First, the context. Rickover needed to get a reactor designed, series built, and operated right away. In that context, he was absolutely right.

Second, the PWR design works and it works very well for submarines. I don't know that there has been a better design concept for submarines (the USSR built a few liquid metal cooled reactors which caused them lots of problems). When you have a basic design that does work, then going ahead and building it is the right thing to do.

We can see the consequences of what that approach does to earlier stage research programmes as well because some of Rickover's people were put in charge of Argonne's reactor research programme which they did not understand and where they caused an immense amount of trouble by trying to duplicate this approach in a context where the basic approach had not been definitively decided. This lead, for instance, to spending an enormous amount of money and time pursuing an oxide fuel fast reactor concept despite being warned that there were fundamental problems with that design and that spending more time up front on metal fuel design work would be a better idea. They didn't want to hear it - oxide fuel works (which it does - in thermal PWR and BWR reactors) - so go with that and just build it.

Rickover was right of course about the principle of understanding technological predictions in the context of what we would now call Technological Readiness Levels and optimism bias but in understanding this we also have to keep in mind the other side of the coin which is not to entirely commit to premature optimisation.

Re: Compact nuclear fusion reactor is 'very likely to work,' studies suggest

#270

Post by Mvandenbergh » Wed, Sep 30, 2020, 9:41 AM UTC

physicsguy wrote: post #259

So, I'm a physicist, and I went to a number of talks from people involved with the JET fusion reactor over the years, though fusion is not my area. And my understanding is not that it's difficult to make plasma, or even build a reactor in particular that is the main problem (though instabilities can be problematic), but it's that the internal structure degrades very rapidly and becomes highly radioactive, because you…

One of the advantages with the MIT superconducting magnet design is that it allows the vacuum vessel and neutron shielding to be lifted out and replaced wholesale (n.b. this doesn't have to be every 5 years but it does have be done at some point during the reactor lifetime). The ITER design would require essentially disassembling the whole reactor to do that. In both cases, the neutron absorbing first wall will be replaced every 5 years or so which is indeed a substantial challenge requiring robotic operations.