Will the Helion Fusion Reactor ACTUALLY work? - Nuclear Engineer Reacts to Real Engineering

[RealEngineering]

Appreciate you handle these things. Not a huge fan of reaction videos, but you actually provide value and add additional context. But most importantly to me, you aren't an asshole about it.

[javaman]

Gigawatts of power; enormous capacitors; huge magnets. Some audiophile is going to make a kickass stereo system out of this.

[thamiordragonheart]

I like Helion as an aerospace engineer, but not because I think they have an above-average chance of practical fusion. The specific technology they're developing was originally a pulsed plasmoid electrodeless Lorentz force thruster in Dave Kirtly's lab at the University of Washington. even if we never get fusion out of it, I bet we'll get a pretty good electric thruster out of it that can use any gas a propellant, which would be great for exploring the solar system.

[MlokKarel]

I'm glad you made this reaction. I had just about the same problems with the concept you apparently have too - but I wouldn't be able to articulate them the way you did. 👍

[perryallan]

I'm a mechanical engineer who has worked most of my life in Power Plants (Fossil then Nuclear). One of my nuclear plant specialties was heat exchanger engineer so I understand a lot about heat transfer and keeping things cool (or hot as the case may be).

An interesting aside is that between my fossil and nuclear carrear I worked for a company that was involved in building test reactors (or technology demonstrators) for the UW_Madison Physics Department. I helped build the Helically Symmetric eXperiment (HSX) stellarator which has been proposed as an alternate possible fusion technology to the tokamak. At the time the Physicists involved told me that one of the problems with the tokamak was that in theory it could not be cooled if run at full power continuously. The HSX twisty curvy shape can be cooled. Germany subsequently built a larger version of the HSX design which solved some of the problems identified at UW Madison. But the research was more advanced for the tokamak so it got the major funding as a fusion reactor; when there are actually a number of real advantages of the HSX design for fusion. I suggest that you research the HSX device and the more modern German version, and how it relates to possible use for a fusion reactor.

Back to Helion. The biggest challenge that I see is that they are going to have to cool the compression and reaction chamber once they progress to several "shots" per second (or even multiple for minute on a continuing basis). This will keep the rest of the machine relatively cool as well. However, it adds effective thickness to the pressure vessel which means that the magnetic coils for both the compression and for recovering energy will be large in diameter which I believes reduces their efficiency. Material selection to prevent radiation embrittlement will be critical (and not my area of expertise).

But, the Helicon approach may actually work and produce more electricity than it takes to run the machine after more test reactors. There are huge advantages if they can remove the very inefficient steam cycle not to mention the large physical plant that it requires.

I see the potential; but I suspect they are at least 3 generations, or more, of test machines away from being able to demonstrate both continuous operation without overheating (even if only few shots per minute) and that they can effectively recover the excess energy directly as electricity and show net positive power output. Keep in mind that I believe EBR-1 could only light I believe 4 light-bulbs as the 1st nuclear generated electricity in 1951.

I believe you misunderstood about their comment on size and the next plant that is supposed to come online this year. That's just the next test reactor to see if they have solved some problems and to figure out what they have to solve to be able to increase temperature and future reactor output. Also, because these test reactors are very small compared to the tokamak test reactor. They can build them much cheaper and build multiple generations of them faster and cheaper than you can build a single tokamak test reactor at the size of the current one being built. So they should be able to progress through all the challenges faster and cheaper.

As a comparison they have been building the ITER since 2013 for 11 years now... Perhaps it will startup in 2025. More likely 2027 or later.

Helion as a company also started in 2013 using key people from the Inductive Plasmoid Accelerator (IPA) experiments (which is where the 1st 3 generations machines for this technology were built). They then built the next 3 generation of test machines (4th, 5th, and 6th - Trenta - which you saw in the video you critiqued which demonstrated 100 million degrees in 2021 in 9 years. All for a fraction of the money of ITER. The development cycle is much faster with small machines and much cheaper.

I am quite confident that they will have the 7th generation machine online before ITER, and they might even be well into building their 8th generation machine by the time ITER starts up (or even have started up the 8th generation machine).

They are already aware of the neutron radiation problem as their control room is behind a concrete wall. I am sure that they are aware of the requirements for added shielding as they increase in size. Also note that no one is present in the reactor room when they run the machine.

You are correct in that the base theory of operation is not new, nor the idea that they would collide dual shots from opposite ends. But, based on my review of several other projects trying different approaches... Helion appears to be well in the lead of doing it successfully from the other contenders.

I'm fairly convinced that its these small privately funded fusion research efforts that are more likely lead to net energy positive fusion generator than the massive very long lead time government funded project like ITER.

In the end how well Helion works is unknown. But they sure are moving a lot faster than the government funded programs (and there is no guarantee that the government funded programs such as ITER will work either).

Have a great day,

[kstricl]

I decided to scroll through the Helion Twitter while watching this. So From what I see, they are apparently on pace for starting Polaris in 2024. They are also putting the reactor into a room with 2.5ft thick borated concrete walls, stepping up capacitor manufacturing, creating testing procedures and equipment and analyzing the plasma output from their test reactors. Hoping they get to net zero by the end of the year or at least running. I suspect there are more gains to be had by capturing the neutrons with a breeder shell, but let's see if they get Polaris functioning first.

[hebijirik]

As an electrical engineer I find it exciting that there could potentially be a reactor that skips the whole steam engine part. I can see it possible to have the capacitors discharge to create the fusion pulse and be recharged back by it to a higher state of charge than they started with. In essence you would have to charge the thing from the grid to kick start it and once it is pulsing you need to steadily discharge the surplus back to the grid. If it can work even at relatively small sizes (lower tens of MW for example) it might even make a good variable source for the grid if the startup prodecure is just minutes long. Even if it is not that fast it still seems like it would have a faster controlability than a big fission reactor so it could have its place in the grid mix.
As you pointed out the efficiency is a big question mark and all the inefficiencies are coming out of there as kinetic energy of neutrons and as various electromagnetic radiation. That needs to be shielded, the shielding will be heated up by it and if that is enough heat it might be worth it to re-introduce the steam engine part back in to capture some 30% of that waste heat. Basically making it a hybrid engine, part electric part heat.
And I agree this is at a science experiment level now and therefore far from a powerplant someone can buy from them and connect to the grid. What I find optimistic about them is that the scale they run their experiments at for now lets them learn the lessons they wanted and implement them into the next evolution in a cycle of just few years. It is not taking decades and trillions in international funding right now to see if this direct-to-electricity method can be done or if it turns out the negatives outweigh the positives. It might get to that stage quickly if they find out they need to build it as big as a tokamak to make it workable. But from what I see it might be viable to keep it order of magnitude smaller even with all the future shielding and such. If that is the case then the speed of iterations to get it to commercial state can be much faster than other fusion projects.

[mariemccann]

Confirms what we already knew, "fusion reactors are less than ten years away, just as they have been for the past 70 years".

[teardowndan]

Getting energy out of it works the same way as putting it in: the coils used to accelerate and compress plasma into the reaction area and confine it there are used as generators after the reaction, much like a free-piston linear generator. As for what sort of output you'd get from that, field output will likely go through regen-braking type circuitry into the pulse capacitors used for the initial compression.

[XavierBetoN]

10:30
Not a nuclear engineer, but an electrical engineer here.
If the conductors don't melt, any conversion from electrical to electrical energy is far more efficient than say converting from heat to kinetic to electric. You'll have to just convert the waveforms, and PWM Inverter can work above .95 efficiency, while the other conversions fall back far behind it. Hence, It will be usable in the grid. But can't have any ideas about how to isolate the conductors. If what they imply is true, my guess is that they use magnetic fields to keep the superheated atoms and particles away from the conductors.

[bencoman]

Regarding the "10 million degrees" being too low, that seems an error by the narrator. Helion FAQ "What are Helion's technical achievements?" says "Helion's 6th prototype reached bulk ion temperatures greater than 9 keV, equivalent to 100 million degrees Celsius"

[andyhu]

10:15 I can answer those questions: the plasma never touches the conductors. This is the basis of all confinement of fusion systems. The question makes me doubtful about the video maker's knowledge about fusion. On the other hand, the electricity generated would be pulsing, which is a legitimate concern. I suggest we borrow a page from those old car engines and put multiple of these reaction chambers together. When one is generating electricity, one other is doing compression. Most electricity will go to the compressing coils, with the excess entering the grid. The waveform will still be spikes, but much more manageable. An added benefit is that the massive cap array doesn't have to be pulsed hard every cycle, which will probably reduce wear. They still need to be charged and used to start the process, pretty much like how a car engine is started.

[kizuro]

there is a video response by Improbable Matter, titled "The problems with Helion Energy - a response to Real Engineering" talk a little bit about some of your points

[nicksantos]

A person drinking too much deuterium would be a funny scenario for one of those medical drama shows

[marcelgranetzny]

I don't know where to begin here but this video is in parts really painful to watch for someone who has a degree in plasma physics and nuclear engineering with focus on nuclear fusion. And I'm saying that as someone who is doubtful about Helion myself. The problem seems to be that Tyler's background is in fission not fusion. Those two fields actually have very little overlap in how to build a practical reactor. Fusion is 90% plasma physics. Fission on the other hand doesn't involve any plasmas, so nuclear fission engineers don't get trained in it As a consequence Tyler seems to not really understand plasmas and how they are confined. He says that the sun is fusing deuterium and tritium which is not true. He constantly seems to forget that the plasma is in a vacuum and isolated against the vacuum vessel by magnetic fields. He confuses plasma pressure with magnetic pressure. I could go on The end result is that he is not able to properly evaluate the claims being made. I'm not siding with Real Engineering or Helion but Tyler clearly doesn't have the necessary background to evaluate their claims. It would be equally difficult for someone trained in nuclear fusion to evaluate a new nuclear fission concept.

[HavocStylesJoe]

3:16 No, no, no, no, no, this sucker's electrical, but I need a nuclear reaction to generate the 1.21 #gigawatts of electricity I need.

[MrTommispilot]

Further more, the purpose among other things of ITER is to TEST different materials at the wall and different configurations to investigate the yield of tritium. The plan is not a dogma to use only beryllium.

[ChakatStormCloud]

My opinion of Helion's approach was that most of their challenges were relatively simple, they have more challenges, but they've dodged some of the biggest ones that tokomaks struggle with. Meaning they should have a more straight forward path in terms of developing towards a design that's ready for regulatory testing.

In a tokomak, the walls of the fusion chamber have to handle: Neutron impacts, heat/energy collection, fuel injection/breeding, and likely more, meaning improving one of those factors will likely lead to complications with others. in Helion's case the injection happens separately, and the energy collection is electromagnetic instead of thermal, which should make the design easier to handle. Also, pulsed magnetics avoids the issue of super conducting electromagnets and the frequent issue of quenching.

In truth it remains to be seen who will end up having the easiest or best path to viable fusion power, but I'd personally say that Helion and laser inertial confinement are the two most likely to see success. Tokomaks have just... they take decades to build just 1 and have only gotten bigger and bigger with little progress. They just take way to long to ITERate (sorry I had to)

[robertheinrich]

the thing, as far as I understand it, is: they once developed ion engines, and plasma engines. essentially eject material out one end at really high speed (therefore high specific impulse).
in fusion, particles that can fuse need to collide at really high speeds, usually achieved with high temperatures. with higher temperature, particles travel faster, but all the particles go in different directions. here, they want to send them all into the same direction. you can assign a fast particle with a high temperature. you could in theory assign a temperature to a car on a highway (at highway-speeds, cars are actually quite cold).
so they fire two of these "engines" at each other and in the middle, the fusable atoms collide. for real fusion, they need to build bigger and have higher speeds. but for testing the tech it's a good method.
funnily enough, there is another british startup that wants to create fusion by firing the fusable material with a gun. and it is driven by gunpowder. they actually have fusion, but it is far from Q = 1

regarding beryllium: world production: 250t anually. so not exactly a lot. but again, there is not much need for it.

[Channel_]

If the man wants to spend his money on something useful, lets hope he can make this work. Better than owning an island with lots of dodgy visitors.