Hacker News new | past | comments | ask | show | jobs | submit login
Nuclear fusion plan at former Aberthaw Power Station (bbc.co.uk)
55 points by dane-pgp on April 25, 2021 | hide | past | favorite | 28 comments



> The Vale of Glamorgan council is bidding for the site to be used to house new technology to produce "limitless" energy ... However, the plant would not open until 2040.

I'm glad that fusion is now less than 20 years away.


Well, it is all contingent on someone getting high temperature superconductors to work on an industrial reliability level. If they manage, we will have power producing fusion (though probably not economically viable).

It is a big if, though.

Fusion has always been 20 years away, as it was always contingent on some science panning out, but the universe has not been in our favour so far. It is 20 years away, with some decent but <50% probability.


If you listen to the omega-tau episodes on fusion and ITER the 20 years thing is discussed quite a bit, it’s a question of money, not time. The technology is known to be possible but the experiments that need to be made to confirm how to do the engineering to build a working and reliable plant are big and very expensive so the constraint is getting governments to spend £100bn (or whatever) that it will cost to do this work. E.g ITER could be finished already and they could be starting on a prototype plant for energy generation if they didn’t have to struggle to justify funding for every little thing to all the various governments funding the project.


For ITER money has little to do with it. The problem is mostly politics.

For example, take the Vacuum vessel. Currently there is only one section on-site. The other sections are in various stages of completion in different countries around the world when they are due to be on-site already. Instead of one contractor manufacturing each of the nearly identical sections they are each being made in separate member countries for political reasons.

The governments involved don't have a problem with spending the money, but they want a return on that investment so they look for that return in jobs and contracts for their engineering firms.

The international collaboration isn't working. If ITER was only an EU project, it could probably be complete already.


The reasons you give look quite OK to me. As a international, huge project I can understand government who want to participate and make sure they get to understand the knowledge in the project by having firsthand experience.

I'd say it's more a question on how we evaluate the fact an international organization is working or not. Is it "delivering the project's goal" or "making sure all participants get the most out of the project".


That's an interesting take on it. I expect if you posed that question to the DG he would tell you the answer is "both". He is simultaneously a scientist/engineer and a politician.


ITER, much like the SLS, is an industrial development program. Countries will use it to develop their own capabilities.


Exactly. You put it better than I did.


Check out the papers from the MIT SPARC folks. Getting the magnets working is not the bottleneck at all. REBCO and similar commercially available products are physically robust. Seriously, two grad students set a new field record doing nothing more complicated than spooling the stuff into a spiral.

In terms of engineering, the diverter, and a wall that can last years are far, far, more difficult than the magnets.


> Well, it is all contingent on someone getting high temperature superconductors to work on an industrial reliability level.

Is that really the only path-critical bottleneck that needs to be solved? Has everything else been figured out?


Far from it. That old power station will sit idle for a long, long time... unless they build a solar farm around it.

Seriously, there will never, ever be one solitary erg of commercially competitive energy from any process dependent on heating by neutron irradiation. Every cent going into work on those is, in effect, stolen from work on viable technologies that can have any impact on global climate disruption—mainly solar, wind, geothermal, and storage; and digs fusion deeper into an already tens-of-$billions-deep hole it can never climb out of.

The only reasons spending on neutronic fusion continues are secondary—employing hot-neutron physicists so enough are ready when weapons work comes up, hi-temp superconductor industrialization, plasma fluid dynamics praxis, hot-neutron-tolerant metallurgy.

At the end of it all they will shrug and say, "We guessed wrong", and not even apologize; instead: "... but we have this other great idea!"


No it's more complicated. High temperature superconductors are neither strictly necessary nor sufficient on their own, but I'm sure they'd help quite a bit.


Isn't REBCO tape the thing that is going to save fusion?


Yes, but turning the tape in a supermagnet that is reliable and doesn't burn itself to pieces on every quench event is non-trivial.


heh, it's been less than 20 years away for the last 75 years.


I guess I have to get the chart now.

https://i.imgur.com/3vYLQmm.png

The real world is more complicated than an old chart but it's definitely worth keeping in mind.

And supercomputers have been very helpful recently to design balanced tokamaks!


No, I think the joke is that it’s always 30 or 50 years away, can’t remember which.

20 would be a bit more optimism


As long as it’s actual production of limitless power in 30 - half the time these things are r&d stuff. 20 years to limitless power in production is great.


That's more than the traditional "15 years" for technology that never happens though. Don't know what to make of this ;)


Later in the article it's clearer that it's talking about a prototype/research thing.


> The Vale of Glamorgan council

That sounds like something straight out of Tolkein!


Or a book your username comes from? :)


Headline is a bit misleading.

> Council officials are hoping a prototype nuclear fusion reactor could be built at a former power plant.

They've bid to be the site for Spherical Tokamak for Energy Production (STEP). The selection hasn't bee made yet, even if the project goes ahead.


and I'm pretty sure we've had a HN story for a whole bunch of the other applications for STEP sites.


Isn't this putting the cart about 10 light years in front of the horse?


No, only 30 or so.


I wonder who's spouse, in-law or sibling will get most of that £222m.


"The devolved nations are getting antsy. Quick! Tease them with bridges and tunnels and infrastructure projects"




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: