Hacker News Reader

Comment by _0w8t | Hacker News Reader

I am very skeptical about Helion, https://www.youtube.com/watch?v=3vUPhsFoniw

FRCs have been around a long time but Helion isn't just making an FRC, they're making two, slamming them into each other, and then compressing with a magnetic field.

The neutrons from D-D fusion are much lower energy than D-T fusion, and don't cause the same damage to materials. They're similar to fission neutrons, and modern fission reactors last sixty years without refurbishment.

According to Helion and other sources I've seen, their hybrid reaction would only release 6% of its energy as neutron radiation, compared to 80% from D-T. They could also build reactors dedicated to D-D and hence He3 production, designing their power plants to mostly burn D-He3. That would reduce power plant neutrons to less than 1% of power output.

Regarding net power and scaling, I'm not competent to talk about it but here's the CEO of Helion going into some detail on their view: https://www.youtube.com/watch?v=G1vyMcqiVtA

With Helion pulse design peak neuron flux luminosity will still be much higher than that was observed in nuclear reactors. Compared with tokamaks their magnets will be directly exposed to it.

Plus their idea to convert fusion energy directly into electricity depends on many unknowns to assert that it will work with reasonable efficiency.

So even if their approach may work to produce positive energy output in few years, a practical device will still be far away.

Edit: Helion itself admits that net positive in electricity with D-3He is tricky even with their design at that presentation.

So most likely their first prototype will be D+D with all the problems associated with high-energy neutrons.

Bear in mind that Helion's magnets will not be superconductors. Last I saw, they were looking at an aluminum alloy. Exposure to D-D neutrons isn't likely to be a structural problem at least, since you don't get the helium buildup in materials that you get with D-T.

The presentation says they think they can get a 5X gain with D-3He, and they think that's sufficient given the efficiencies of their heating and extraction.

The new video by the same guy https://youtube.com/live/wK5v1Nw98xs from 32 min describes how Helion will get a lot of D+T neutrons from the secondary reactions despite all their efforts. Even without that the neutron flux from D+D reaction by Helion own admission will be enough to do a lot of damage within days of operation based on experience with slower fission neutrons.

Helion as already demonstrated recovering 95% of the energy from non-fusable plasmas and Q << 1 plasmas. Thousands of times.

It's perhaps worth noting that the high speed, high power electronics required to implement their approach are fairly new. I don't know how old the concept of FRC merging is, but if it existed in the 70's they wouldn't have had the technology to build it.

Edit: and to be clear, I'm talking about the dynamic control circuits, not the capacitors. That's old technology.

The new video by the same guy who created the original criticism https://youtube.com/live/wK5v1Nw98xs starting from 2:03:00 explains that 95% recovery efficiency is not the whole story.

Ion escape might be a significant consideration, but given the plasma density and magnetic field strength, I'm guessing it wouldn't degrade the efficiency too much. Presumably they've done the calculations. I have not.

As for the input heating loss, presumably that's included in the 95% recovery figure. If it's not, then it would definitely be good to know what the actual figure is including input heating losses.

The nice thing about colliding two FRCs is that they form a new FRC with all their kinetic energy converted to heat, and it's very efficient to accelerate an FRC.

The theoretical problem with fusion energy isn't neutron activation; if you could generate electricity you could easily amortize the cost of disposal of any induced radioactivity over the lifetime of the plan. But no known projected fusion reactor is anywhere close to even being financial viable with todays technology. Iter won't demonstrate actual D-T fusion until 2035 and it is too small to project what the economics of a takomak power plant would look like.

To make fusion viable you have to generate a substantial surplus of power that can be converted into usable energy. Nothing in that video touches on that subject at all. The most he can say is that it is possible D-He3 can reach scientific breakeven, which of course, is no where close to net power generation.

It is strange to see claims about Q > 10 using D-T fusion being possible and then immediately pivot to talking about being excited about Q > 1 using D-He3. If you can build a fusion reactor that could produce Q > 10 why would you bother pursuing Q > 1? The 'benefits' of direct electrical generation from D-He3 are not relevant if you can't even produce net positive energy in the first place. Electrical generation is a solved problem; if you can produce net energy you can produce heat that can power a turbine.

If the most you can do is talk about how Q > 1 is possible it seems unlikely you are going to start producing electricity any time soon.

Heat to electricity conversion ratio with a realistic Tokamak design will have 30% efficiency at best, accounting for other losses will require Q at least 5. But even that may be too low so to play safe the latest Tokamaks target at least 10 or more.

But Helion hopes that with their design where the theoretical efficiency of electricity generation is 100%, they do not need such high Q allowing one to be excited about Q > 1. Their models predicts Q around 5.

The problem is the model uncertainty. If the real Q will be just 2, I very much doubt that they can limit the total losses below 50%.

But even if they can get net gain in power per pulse, they are too optimistic about neutron flux and its consequences. Again, model uncertainty is big.

So overall there is very little room for unexpected in their design so I am very skeptical based just on that alone.

That's a pretty pathetic summary of the Helion technology. The youtuber seems to have watched the Real Engineering video but not actually done any research on the other information that Helion has published in the last 12 months. A lot of the criticisms are due to Real Engineering video missing details. The youtuber is unprofessionally rude as well.

If you look at the extended video he released after, there's a lot of heavily motivated reasoning and straw manning, a few pieces of slight of hand to dismiss holes in his reasoning.

I was pretty sceptical about achievable power density going in and the gish gallop actually pushed me into believing helion a bit more if someone who allegedly is an expert needs to be this duplicitous to criticize it.

While I do agree that the extended video went into hand waving arguments on few occasions, that is not because Helion may know something that general public do not, but rather because nobody knows! Helion own admission is that they use 2-d code to model the whole thing. While they validated that model with the current design, but one cannot predict how the real physics will deviate from the model at fusion condition. That adds a lot of uncertainty and, as such, skepticism as the reality can be much worse than even the hand-waving arguments from the video.

For example, Helion’s patent https://patents.google.com/patent/US11469003B2/en hints how critical is suppression of the secondary D+T reactions and removal of T to their design, but patent itself does not describe how they are going to archive that. The only claim is that there would be no time for secondary D+T in their design. But that is based on those imprecise models at best. Nobody knows if that is true for sure. Plus the patent does not tell at all how they are going to remove mechanically all T after the pulse as keeping even 1% of T can be problematic.

In addition the extended video cites Helion own admission that the neutron flux from D+D alone will be like 1e18 neutrons/day based on their models. That type of flux damages a lot of materials within days with slower fission neutrons.

While their aluminum coils may handle that better especially if partially shielded, nobody knows for sure again as modeling effects of radiation damage is hard.

Reply on news.ycombinator.com