← Back to context

Comment by Gud

4 days ago

https://news.mit.edu/2024/commonwealth-fusion-systems-unveil...

So according to the industry leaders, we will have the first 400MW plant within 10 years.

22 comments

Gud

Reply
andriesm  4 days ago

How can they build something commercial/grid-scale when not a single research-level reactor truly generates net energy out, and none can do it anywhere near continously enough to be of any practical use?

This news is either based on misleading the public, or I am about to be updated with where Fusion is?

  • Gud  4 days ago

    I think you are about to be updated.

    Watch this clip with Prof. Whyte from 8 years ago. It’s the team behind CFS(then still at MIT). Highly interesting. He will explain exactly what they will do(now doing), how they will do it, and why they will do it they way they are doing it.

    Please note that they are pretty much on target since. I have been following CFS closely.

    Essentially the breakthrough has been the ability to manufacture more powerful magnets. CFS makes the most powerful magnets in the world.

    That was always the main issue, how to contain the plasma.

    https://m.youtube.com/watch?v=KkpqA8yG9T4&pp=ygUrYnJlYWt0aHJ...

    • pfdietz  4 days ago

      Now ask: what is the power density of their proposed reactor?

      The predecessor, the ARC reactor in the 2014 paper, had a power density of 0.5 MW/m^3.

      In comparison, a commercial PWR, which you can buy today, has a power density of 20 MW/m^3.

      So, even with all the HTSC hype, their machine is still a factor of 40 worse than existing proven fission technology.

      How is such a bloated, much more complex machine even going to compete with fission, never mind the alternatives that have sidelined new fission construction? The capex will be completely out of bounds.

  • cladopa  4 days ago

    Because they are more or less certain (because of their magnets) they will get the necessary temperatures for ignition, giving more energy that it consumes.

    If they get ignitions, all the other problems will be solved very fast, because there will be an enormous rush from investors wanting to invest billions.

    • nradov  4 days ago

      I hope they succeed, but throwing more money at problems like this doesn't necessarily accelerate progress. There is still basic research needed. Building large, high-precision devices can't be rushed.

      3 replies →

epistasis  4 days ago

What makes them industry leaders? Do they have a prototype? Can they get Q>1, much less >5 or similar for what will be needed to break even on all the rest of the inefficiencies?

If they don't have a prototype, and are going straight to plans for a 400MW "commercial" plant, why should we believe this is possible? What evidence is there that these plans for a massive breakthrough ten years from now will work out?

This looks, walks, and talks like a ploy to get in on AI energy demand hype. It may not be, but it has all those features, and not many other features.

  • gpm  4 days ago

    > What makes them industry leaders?

    They have a plausible relatively well understood path to fusion, have credibility with their background (coming out of fusion research at MIT), and have raised something like 2 billion dollars in funding.

    > Can they get Q>1, much less >5 or similar for what will be needed to break even on all the rest of the inefficiencies

    They think so

    > and are going straight to plans for a 400MW "commercial" plant

    They aren't. They're currently developing "SPARC", a Q>1 demonstration plant targeting 2027. The 400 MW commercial plant, ARC, is a follow on design targeting 2030s.

    > This looks, walks, and talks like a ploy to get in on AI energy demand hype

    They predate the AI boom by a lot. The project started in 2018. They had a $1.8 billion dollar funding round in 2021.

    The basic concept is "hey look, someone figured out how to build better superconductors. What if we took what ITER is trying to do, but used modern super conductors to make it smaller and actually achievable". I'm not saying I think they're certain to succeed, but I don't think they're a scam and I think it's very reasonable to include them amongst the group of "industry leaders"

    • orwin  4 days ago

      The reason Iter is so big is not to achieve controlled fusion, but to be able to capture the fusion energy output. I'm not doubting their ability to reach the self-sustaining plasma, I'm doubtful about their ability to capture energy from it though.

      1 reply →

bufferoverflow  4 days ago

I looked up this Commonwealth Fusion Systems company.

"The company plans to produce its first plasma in 2026 and net fusion energy shortly after."

Looks like your argument is build on just promises, not backed by any tech developments.

  • Gud  4 days ago

    No, they have developed the world’s most powerful magnets based on superconducting tape which is essential for containing the plasma.

  • thinkcontext  4 days ago

    Their key tech development is the manufacture of high temperature superconducting magnets. This is the key to their reactor design.