Hacker News new | past | comments | ask | show | jobs | submit login

As someone who has spent a few years on this problem: what do you think are the the fundamental physical reasons why controlled fusion is unlikely to be a practical commercial power source?



Unlike designing a fusion reactor, or going to the Moon, here nature is fighting you every step of the way. There are a host of instabilities working against confining a burning plasma. It’s not that it’s impossible to create a practical reactor; it’s that mitigating all of these natural processes will lead to a device of immense complexity. The key point is this: even the most optimistic fusion fans say we might have grid power from fusion, if all goes really well, in another 30 years or so. Look at the recent progress in photovoltaics; the only remaining obstacles are in energy storage, which also has seen great progress in the last few years. Is there any serious doubt that in 30 years we could have a solar power economy if we really wanted it? Why, in that case, would we choose fusion instead, even if it worked? It would be more expensive, more prone to failure, more fragiley centralized, and it comes with some hazards: low-level radioactive waste, tritium, and more. So I’m not saying that it’s impossible. It’s not a perpetual motion scheme. However, further expenditure aimed at fusion energy is simply pointless.


Helion is projecting a small amount of net electrical energy out of their next reactor, which they project to come online in 2024. And that reactor is designed primarily for producing He3, which they will use in their follow-up reactors designed for power generation.

So any fusion fans who still say we won't have grid energy from fusion for another 30 years are either misinformed or not very optimistic.


Some brilliant investor-speak from Helion's FAQ:

"A handful of organizations have performed bulk fusion, where a large volume of particles reaches temperatures high enough for fusion to occur on a large scale. [...] none of the organizations that have managed to do bulk fusion have done it in a practical way that can be used to make electricity."


Seems pretty straightforward and accurate to me. Helion is already re-capturing most of the energy from each pulse. I'm not aware of a single other approach that has demonstrated direct conversion from plasma energy to electric energy.


Hint: The United States first demonstrated bulk fusion in 1952.


What are your thought? You have actual expertise to contribute. No need to test the GP.


I see no physical limitations. There are engineering hurdles primarily addressed by materials. Making HTS coils cheaply is a big deal and a WIP problem. Test reactors such as ITER and SPARC need to be built as platforms to experiment with first wall and blanket designs. Making nuclear facilities is expensive. I personally do not see any showstoppers here, just expensive work that needs to be done.

Will fusion ever be economical? I don't have a crystal ball, but properly charging for the economic cost of carbon emissions is necessary before answering that question. I also can't predict how much cheaper reactors would be than test machines and how long they would run for. The science needs to be funded to answer these questions. When society decides these answers are worth 5x F-35s then we'll have our answers.


Thanks! One quibble:

> The science needs to be funded to answer these questions. When society decides these answers are worth 5x F-35s then we'll have our answers.

Isn't society funding ITER for $22 billion, and much else? The American and UK people are investing in some of the companies, per the OP.


The 5x F-35 is hyperbole. ITER is currently being funded and does account for about 80% of the current global fusion research budget. Fair enough, its results will be valuable. US pulling of funding throughout the Reagan era slowed progress by a few decades. We won't make up that lost time no matter how much money we spend. For now all the eggs are in the basket and we wait for the results.

MIT's HTS coil winding is the space to watch now imo. If it isn't all smoke and mirrors then we could see real test reactors for less money than ITER in our lifetime.


Having less money might have increased progress. I tend to agree with my old boss, David Montgomery, that the eagerness to build big machines distracted the community from investigating the basic plasma physics and hydrodynamics that we needed to understand first.


It's not an eagerness to build big machines because they are fun, but out of necessity to push the experimental regime forward.

Given constrained budgets theory has continued to thrive. Cutting edge turbulence models are being made by theoretical plasma physicists. This is a problem no other engineer wants to touch with a 10 foot pole yet has incredibly far-reaching implications for many engineering fields.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: