Hacker News Reader

Comment by mywittyname | Hacker News Reader

Nuclear power can be fast to build. It would need to be constructed in volume, with immutable designs. Right now, America builds so few nuclear plants that they are essentially all one-off designs. Mass production is key here.

Vis-à-vis costs, the government can poof into existence trillions of dollars with no ill effects. Cost is an easily overcome obstacle.

If a mandate came down to produce 1000 new nuclear reactors by 2030, I think it could be achieved. The country just needs the political will to make it happen.

View on HN

> Nuclear power can be fast to build. It would need to be constructed in volume, with immutable designs.

Exactly. You need look nowhere further than the french nuclear timeline for that to be clear: the country grew from 4.5GWe capacity to 49.5GWe between 1977 and 1987. 900MWe class reactors took 5~6 years to build, and the 1300MWe 7 to 8. And when you've worked out the kink, this can be parallelised massively (as long as you have sites to put them on), for about 20 years the country had a dozen reactors being built concurrently, the slowdown in construction times really started as the number of plants being built decreased: https://upload.wikimedia.org/wikipedia/commons/2/24/Chrono-p....

The design got more complex and there were teething issues with the N4, but Chooz took 16 years to enter service where the P3s took 8 years at most.

Between this, and their track record with rail transport, it seems like the French are really good at building stuff. We should learn whatever we can.

Were, things have gotten way less great since the 80s, it’s mostly coasted.

Too bad really: nuclear power was a way to gain independence (from the US and the oil) but then it kinda fell by the wayside as the country largely went with oil anyway (though the grid is both powerful and rather clean owing to the high ratio of nukes).

Would have been interesting for the country to ride the contrarian gallic spirit and decide to go all in on electricity and renewable way back then.

Maybe we’d have working large-scale SMR too.

Also include Airbus. Indeed an impressive bunch!

I had the privilege of watching a bunch of (6 or 7) French engineers make an impromptu visit to the ticket desk (of the airline I worked for) at Charles de Gaulle airport and work their magic, running comms and power with no fuss, just teamwork and that brilliant "why not now?" European attitude.

Meanwhile back at Heathrow that would have taken months of permits and planning and dealing with useless subcontractors.

State monopolies FTW.

In 2020, nuclear energy accounted for 70.6 percent of France's total energy production what is stunning, if you take into account its one of the G7 countries:

https://www.statista.com/statistics/270367/share-of-nuclear-...

It gets 70% of its electricity from nuclear, but electricity is only a minority of the energy we use, so France still gets most of its energy from fossil fuels:

https://ourworldindata.org/grapher/fossil-fuels-share-energy...

Thanks for the correction. I meant to say 70% of electricity not 70% of the energy use. The link was clear about it, but my post not.

Nuclear power could have been fast to build if we had kept building it. We did not, and most of the people who knew how to do it are now retired or dead. Because nuclear power had no future for decades, few young people chose to make a career in it. The workforce needed for designing, building, and operating nuclear power at scale no longer exists and cannot be trained quickly enough. That alone would take 10+ years, which we do not have.

And thus, we will spend the next 10 years playing around with building piddling amounts of renewable power generation and storage instead of actually having a solution in 20 years.

The best time to plant a tree was 20 years ago. The second best time is now.

> the government can poof into existence trillions of dollars with no ill effects

This is because the federal reserve's printed money goes basically into treasuries and stocks.

The rich save basically all of their money. If they decided to spend it on real-world goods, there would be huge inflation.

But it's probably all nil. The cost of building enough nuclear power plants with economy of scale is probably near the same as building enough solar and wind farms.

Except that you miss the part where solar and wind need normal power plants to provide power when they aren't running.

Balancing systems do exist - I used to swim in Smith Mountain Lake[1] as a kid and that entire lake is a gigantic battery. During low demand periods they pump water into a reservoir to then run the turbines during peak demand hours. None of these systems are perfectly efficient - but we've got a lot of options in our tool belts for ways to balance peak demand that don't require more continuous production capacity.

1. https://en.wikipedia.org/wiki/Smith_Mountain_Dam#Hydroelectr...

Yes, pumped storage does exist, but to provide a nation sized battery using pumped storage, you would need to use two of the Great Lakes and pump one of them up by 200 feet.

The power needs of the occasional cold wind still winter week are fairly large.

Fun fact: Nixon actually wanted 1000 new nuclear reactors by 2000.

Beyond the fact that electricity-generation is once again only part of our emissions issue, there's about ~450 reactors worldwide and the uranium ore reserves give us around ~120 years of operation on current 3rd (/3+) gen reactors (short of having fast neutron reactors basically). Having 1000 in the US (or rather 900 more or 3 times as many globally) would have brought that deadline closer by the inverse ratio, and we'd likely already have tensions on nuclear fuel sourcing. Agree with you on the role of government spending though, but there also needs to be a radical paradigm shift in how we view the economy and what needs to happen in each sector (not just within energy in fact).

There are 100 years of uranium reserves available. Nobody is looking for more. If they were, they're probably going to find it.

Exploration programs are expensive.

Yes, sure, we can go mine the bottom of the oceans and extract it by centrifuging ocean waters, that probably would give us marginally more than a doubling of margin. Point remains: reserves are exhaustible, an exponential consumption may not be desirable. https://en.wikipedia.org/wiki/Peak_uranium#:~:text=As%20of%2....

Recent studies found that re-using a design made nukes cost more & take longer to build. Reason being each site is different, and so the design always needs to be tweaked, and at this scale changes are more expensive than from-scratch.

SMR's made in a factory seem like the only way to address that, but I have no idea if we can ramp SMR's to national viability in the next ten years.

The very successful reuse of a few fixed designs on many sites in France seems to quite heavily counter this statement.

A good point of reference, but that was also 20-40 years ago.

In a healthy civilization, as technology improves, things get easier over time, not harder.

The only reason it would have worked in France 40 years ago but not today is institutional malaise and bureaucratic capture.

"Why Nuclear Power Has Been a Flop" [0] apparently does a good job of explaining exactly how our civilization has regressed in this sense over the past 50ish years. In short, it's the principle of "As Low (risk) As Reasonably Achievable". Well, they minimized nuclear risk, but sure as hell didn't minimize global warming risk in the process.

Roots of Progress has an interesting review here [0], which is what I base that on. (It's on my list to read, just haven't gotten to it yet)

0: https://gordianknotbook.com/

1: https://rootsofprogress.org/devanney-on-the-nuclear-flop

Reply on news.ycombinator.com