Hacker News Reader

Comment by simonh | Hacker News Reader

There's no guarantee it's a solvable problem. With the Manhattan Project we knew it was feasible and how to do it, all we needed was engineering.

With fusion we don't know how to do it. There are many, many ideas we could pursue, and they are being pursued, but we have no idea if any of them will actually work. It's quite possible none of them will, or that solving it might take enabling technology we haven't even thought of developing yet.

For context we have already spent several Manhattan Projects worth of fund on fusion research, in inflation adjusted terms.

View on HN

> There's no guarantee it's a solvable problem. With the Manhattan Project we knew it was feasible and how to do it, all we needed was engineering.

I mean, the Manhattan project wasn't about trying to figure out how to harness nuclear power, it was about how to deliver a nuclear explosion. getting some energy out of fission was never really a particularly hard problem once we understood what was going on (it's literally so easy that it's a problem), getting it to explode from a package you could load into a b-52 was the harder problem.

The difference isn't really about what we know is possible, it's that unlike fission, making a fusion bomb is easier than making fusion power, so the military funding dried up before it made much progress, like a person's interest in a porno once the fun part is done.

There's no guarantee it's a solvable problem.

And there's the more insidious variant of this: to solve it we need new physics. I say it's more insidious because, if we used all the money in current approaches, we would be diverting it from the research that could really solve it.

We need to understand the nuclei. The only way to do that is to create macro-scale working models of nuclei, to see what happens inside them. The «walking droplet» kind of experiment is a step in the right direction. These experiments are much, much cheaper than any other experiment in quantum or nuclear physics, but nobody wants to fund them.

We have certainty they will not be solved. Anyway, not to anywhere in sight of commercial competitiveness.

Fusion can never even come close to competing with fission, and fission is not today competitive. It gets less so every year.

You made 10+ comments along those lines, just for this article. How about providing at least some justification, rather than making absolutist statements?

DT fusion has inherently lousy volumetric power density. ITER's thermal power density, for example, is 400x worse than a commercial pressurized water reactor's primary reactor vessel. ARC's is 40x worse. Now couple that with those cores being far more complex than a fission core, and the difficulty of maintenance (they will be too activated for hands on maintenance.)

IMO, the only fusion company that has a real chance is Helion. Most of the breathless hype runs into generic showstopping issues like the ones above. If someone presents a new fusion concept, ask them about their power density, and how much complexity there is in that core.

Reply on news.ycombinator.com