[philipkglass]

Some countries have formally committed to eliminating power from nuclear fission. Most have not. Most countries have at least some regions that would be happy to accept the good, stable employment and tax base offered by a nuclear power plant in exchange for the very slight risk of accidents. So why aren't more fission power plants getting built in countries where they are not formally excluded? I believe the single greatest factor is that they take too long to build and require too large a lump of money, because a single modern reactor is too big.

Modern Generation III/III+ reactor designs have actually made this "chunkiness" worse; there's no modern reactor design under 500 MWe that's certified in the US, Canada, Japan, South Korea, or the EU. Designs in that lower power range were built decades ago, and a few still run, but now it's big-or-nothing. AP1000 and EPR are 1117 and 1600 MWe, respectively, and all projects using them are behind schedule/over budget. And since these large projects are leading to such poor outcomes for involved companies, it will be difficult to get follow-on orders that will benefit from the hard earned lessons of their initial builds.

Even if all the technical problems are solved and fusion proves capable of producing net electricity, fusion power risks hitting this same too-big-to-build problem afflicting fission if it can't scale down. If the only approaches that work are enormous tokamaks with an entry level price of $10-billion-or-more, then they'll be engineering marvels that hardly anyone builds. Maybe they'll someday supply 10% of China's electricity.

[continuations]

> Modern Generation III/III+ reactor designs have actually made this "chunkiness" worse; there's no modern reactor design under 500 MWe that's certified in the US, Canada, Japan, South Korea, or the EU

China has Generation IV pebble bed reactors coming online this year that are smaller (210 MWe) [1]. They would be the world's first Gen IV reactors in production. Would be interesting to see how that goes.

[1] http://www.sciencedirect.com/science/article/pii/S2095809916...

[MrBuddyCasino]

There was an experimental pebble bed test reactor in Germany. That didn't end well, it will be interesting to see if China does it better.

[honestoHeminway]

A collague of my grandfahther spend most of his lives work errecting that reactor, and then building it back until there was only green fields left where the building did stand.

Strange thing to do.

[hsod]

For anyone else curious, MWe stands for "megawatt electric". The average American home uses very roughly 1000 watts on average [0], so a 500 MWe plant serves very roughly 500,000 American homes.

[0]: https://www.eia.gov/tools/faqs/faq.php?id=97&t=3

901,000 watt-hours per month / (30 * 24) ~= 1,250 watts

[beojan]

An average isn't very helpful here. A single kettle / oven / heater will use 1 kW by itself.

[theandrewbailey]

Not every house will have one of those on at all times. Sure, a house could use many multiples more than 1 kW for a bit, but will go back down eventually.

[beojan]

Yes, that's why an average isn't too meaningful. The times these things are on aren't the same for every house, but they are correlated.

[acidburnNSA]

The South Koreans are building large, Gen-III APR-1400s on time and on budget. They picked a standard design, perfected the project management of constructing them, and can now sell them internationally.

[sanxiyn]

South Korea already sold APR-1400 to UAE. Four of them are currently in construction.

[Jedd]

> Some countries have formally committed to eliminating power from nuclear fission. Most have not.

Most countries have not formally committed to eliminating power from fossil fuels, either - though there's obviously more of a trend in that direction.

Point being that past lack of commitment is no indicator of future direction.

> Most countries have at least some regions that would be happy to accept the good, stable employment and tax base offered by a nuclear power plant in exchange for the very slight risk of accidents.

Given there's 195 countries - 'most' may be misapplied here. I think it's optimistic to believe that there'll be stable employment for more than a handful of people per nuclear fission power plant.

Tax legislation comes and goes. And as to slight risk .. I think that's misrepresenting things too. A risk matrix typically takes into account likelihood vs effect. You may be suggesting there's a 'small likelihood', while ignoring the massive potential effects of an incident -- equating to significant actual risk.

A review of the history and progress <sic> of the Hinkley nuclear fission power plant in the UK - a country that certainly meets the pro-nuclear-fission requirement - is sobering.

This article (from late 2015), and apologies for Telegraph link [1] is highly depressing.

The wikipedia page [2] has a few more recent details.

[1] http://www.telegraph.co.uk/news/earth/energy/nuclearpower/11...

[2] https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_...

[dest]

Maybe someday those power plants will be available in smaller sizes, e.g. for fusion with the discovery of better confinement techniques, or high magnetic field materials

[milcron]

Nuclear plants were originally designed small, as for submarines. For some reason the design trend has been towards larger plants. Perhaps for economies of scale?

An interesting watch is Adam Curtis' documentary "A is for Atom", which covers the history of nuclear power.

It's usually pretty easy to find Adam's content on youtube: https://www.youtube.com/watch?v=S3i9WHHl3qA https://www.youtube.com/watch?v=-FDrA7yUdFc

Summary on Wikipedia: https://en.wikipedia.org/wiki/Pandora%27s_Box_(TV_series)#Pa...

[philipkglass]

The larger reactors were supposed to improve the economics of nuclear power once they reached series production. Making the minimum size larger seems to have impeded progression toward series production, zeroing out (and then some) the expected reduction in cost-per-megawatt from larger units.

For example, the 1117 MWe AP1000 design that's driven Westinghouse to bankruptcy derives from the AP600 design of only 600 MWe. After-the-fact criticism is too easy, but it seems to me that Westinghouse and its US partners would have suffered less if they were trying to build 4x 600 MWe reactors, and the projects fell behind schedule/over budget, than in their actual situation where they're behind schedule/over budget on 4x 1117 MWe reactors.

[otp124]

Would it be possible to make them modular? As in, build X number of the best submarine-size reactors and drop them in side-by-side in essentially secure concrete warehouses cooled by water pumped from a nearby lake? Kind of like racking servers is what I'm picturing.

[philipkglass]

That's the "small modular reactor" concept, under development by a few companies. I think that the concept has considerable merits. I hope that it will be tested in practice. I'm pretty bullish on renewables but nuclear power is very low emissions over its life cycle, safe enough (IMO), and nuclear construction capability should be (again IMO) maintained as a complement/alternative to renewable generation for the post-fossil era. For continued viability new nuclear construction needs lower, more predictable costs or steady long-term support from governments; I think that cost improvements would be the better of the two.

[otp124]

Sounds like a viable model, thanks!

[acidburnNSA]

One advantage submarine reactors have is that they can use fully enriched uranium (bomb-quality, >90% U-235). You can still make pretty small reactors with non-weapons uranium (defined as <20% enriched) but they are not as performant or long-lasting as the pure U-235 submarines.

[otp124]

Interesting. So can the core of uranium weapons be re-used for nuclear reactors? (Assuming there is a big arms reduction event)

[Spooky23]

Money. Fission is both expensive and risky.

[_Codemonkeyism]

With the growing understanding of externalizing costs and risks is a bad thing for communities and countries making externalized costs explicit to companies more and more, in many countries energy companies fear they would need to insure their plants which together with sinking costs of solar would make fission too costly and non-competing.

[21]

Fission was killed by Fukushima. While in theory we can build safe reactors, practice shows otherwise.

[coddingtonbear]

It's important to keep in mind the reactors at Fukushima were designed in the 1960s -- literally fifty years ago. During that same time period, we were doing things like designing machines for efficiently burning down the rainforest so we could open up land for farming, and pouring miscellaneous mixed nuclear waste into underground vats for future generations to figure out.

It might be debatable whether humanity is sufficiently responsible to maintain any dangerous machinery requiring generation-level planning, but it is important to keep in mind how far removed we are from the era in which every nuclear reactor ever having experienced a major nuclear incident was designed.

[acidburnNSA]

Give it a second look. Practice shows that nuclear reactors are actually extraordinarily safe compared to most other options. In fact, by 2013 they saved 1.8 million lives simply by displacing air pollution deaths, and prevented 65 billion tonnes CO2-eq. [1]

[1] http://pubs.acs.org/doi/abs/10.1021/es3051197?source=cen

[philipkglass]

Construction on the 4 new US AP1000s that bankrupted Westinghouse went ahead after Fukushima. There were regions of the US perfectly willing to take new reactors even after Fukushima, and power companies willing to buy the power from those new reactors. But construction is years behind planned schedules and billions over planned budgets. "Overruns" more than "Fukushima" seems like the greatest barrier to future US orders for more AP1000s.

[Aron]

We've certainly shown that reactors designed in the 60s and built in the 70s are problematic in very rare cases.

[21]

So you're saying that the nuclear industry was saying before that "yes, we fully agree that we could have a full blown meltdown at one of the older reactors in very rare cases".

BTW, how it's a tsunami a very rare case on a decades long horizon? There were warnings about the possibility of a big tsunami. It wasn't a black swan, it was a white one.

> A spokesman from the Tokyo Electric Power Corporation [TEPCO] told FRONTLINE that the company was aware of Minoura’s work and was in the process of considering plant modifications in case of a massive tsunami.

Fission fails deadly, and humans are really bad at managing these kinds of things (ie: I don't want to pay 200 mil to build a better tsunami barrier for a 1 in a million chance...)

[logicallee]

Like the sibling commenter, I can't parse your sentence and intention except as "we've certainly shown they're not commonly problematic" which is quite a weak statement.

Is your sentence a positive sentence (do you mean we've certainly shown they're problematic essentially never) or is your sentence negative (we've certainly shown they're problematic definitely sometimes - which is unacceptable)?

I tried to read your attitude but failed.

[Aron]

I was mostly rebuffing the structure of the argument "While in theory we can build safe reactors, practice shows otherwise" which has two different reference classes for reactors (modern safer ones, and less safe fukushima ones).

[logicallee]

Oh, okay! It was confusing because it sounds like "they're rarely problematic" which is not your meaning. Thanks for the clarification!

[lz400]

I don't think so. Fission is being killed by its lack of competitiveness and how long reactors take to build in a world where renewables are getting better so quickly people think twice about locking themselves up in a multi billion nuclear power plant project that takes 10 years to build.

[atemerev]

Fission was killed by cheap oil and fracking revolution. Thankfully, these things don't last forever.

[lz400]

Very cheap oil is very recent and it's an artificial situation created by pumping by OPEC countries. Fracking was only worth it because of how expensive oil was before, now it's not worth it anymore while oil stays this cheap. I think current fission nuclear tech is competitive with fossil fuels in the "fracking era". It's not competitive with current oil but that wouldn't kill it, just at most put it on hold until inevitably oil goes up again. Not to mention environmental issues, which massively favor nuclear over fossil fuels.

IMHO it is competitive renewables that is a death sentence for current reactor tech. Nuclear needs to go back to the drawing board and come up with more efficient and realistic designs.

[atemerev]

There are no competitive renewables. Photovoltaics can't survive at scale without massive government subsidies or screwing the environment like in China (choose one), wind power doesn't scale, and don't even get me started about biodiesel.

[lz400]

PV are pretty much competitive now without subsidies in current studies and getting better all the time. The "renewables are not competitive" is a thing of the past.

[int_19h]

Did China stop building after Fukushima? As I recall, they had by far the longest list of planned new reactors.