Hacker News Reader

Comment by dreamcompiler | Hacker News Reader

There's a huge amount of hype in fusion these days. Companies are getting big investments just by saying "we're over unity." None of their investors will recoup any money.

Getting above unity is important but it's still a very long way from systemic over unity of the entire lifecycle of the process that turns fusion into electricity on the grid. And that simply won't happen in my lifetime or most of your lifetimes.

What will likely happen during our lifetimes is we develop large-scale electricity storage mechanisms. Together with decentralized microgrids, storage will enable most of the world's electricity to be generated by renewables. The sun is a giant fusion engine, and it's the only fusion engine that will be practical for us during at least the next 50 years.

View on HN

Any green gas could be that storage. With solar power around $.01/kwh in some sunny places, converting that energy into storable gas/liquids can be profitable even at low efficiency rates. And we already have an existing network to transport, store and burn gases. Converting those from natural gas to other gases is likely easier than rebuilding everything.

From all that I read, Ammonia is a prime candidate. Especially because there's an existing large market for it, so producers don't have to bet on a new market to ramp up production. But there's no real reason why some power plants, ships and maybe even trucks couldn't be driven by Ammonia in the future.

I think investments are happening because recent progress in this space is indicating that this might in fact happen in the next decade or so. The cliche that fusion reactors are perpetually 30 years away, might have actually been correct up to about 20 years ago.

There are no guarantees here of course, but there are now quite a few companies acting like they are going to get their first test plants up and running in the next ten or so years. From there to commercialization is of course still a long and risky path. But it's very different once you can prove that the process works and more energy comes out then goes in.

As you say, fusion reactors will have to compete with absolutely dirt cheap solar, wind, and storage. As well as with future fission plants that may or may not become cheaper than the current state of the art (which is super expensive). Just getting things working is not going to be good enough. It will need to work cheaply.

The biggest problem to take any of these companies seriously, even the MIT spin-off, is that no one has ever attempted yet to extract a single watt of electricity from the fusion reaction. The part where the thermal energy is actually converted to electrical power is still entirely at the theoretical design level. While it is nowhere near as complex scientifically as actually containing the plasma, it will still be a huge engineering challenge.

When these companies come and claim that their better magnets will allow them to supply power to the grid 3 years from now when they haven't even arrived where JET is yet, so they haven't even started in the unknowns of actually running the reactor continously and extracting power from it, it is clear that they are gifting with their timeline. And if they are willing to lie about the timeline, I don't see why anyone believes them about the fundamental technology as well.

The "30 years away" thing assumed investment.

It's quite weird to think that seeing a potentially endless source of free energy we wouldnt be throwing a trillion dollars or so at it but thats humanity for you.

Scientists were similarly optimistic about space exploration after landing on moon shortly before NASA's budget dwindled.

Private Investments usually need returns within a 10-15 year time scale. Once practical fusion gets to that point, we should expect money to come pouring in which will help make it a reality.

We’re seeing something similar happening in self driving cars

My pet theory is that it's no coincidence the self-driving-car investment craze started the same time the baby-boom generation started to enter "grandpa can't drive himself anymore" territory.

In other words, I suspect it's not fueled by a technology trend, but by trying to capture a potential customer trend.

It seems like a stretch. Self driving car investments mostly coincide with AI getting to a good enough level that it seem feasible.

Helion are planning for systemic returns over unity quite soon really.

Reply on news.ycombinator.com