Small nuclear power: US and Canada are back in the game!

[sketerpot]

Babcock and Wilcox just announced the mPower reactor, a 125 MWe light water reactor meant to lower capital costs. Linky!

Here is a bullet list of the key features of the mPower as I see them:

* Pressurized water reactor (PWR) 17 x 17 fuel bundles (Shorter than normal, but otherwise standard)
* Five year refueling schedule
* Fuel storage pool large enough for 60 years worth of fuel
* Adaptable to advances in LWR fuel (MOX or thorium)
* Below grade construction in most locations
* Air cooled condensers
* Tall, thin pressure vessel
* Passive cooling
* Manufactured system with rail delivery to site
* American engineering and manufacturing (avoids queue at Japan Steel Works)
* 125 MWe of electrical power output (I admit, I was wrong yesterday with my prediction of an even smaller system, but 125 MW is about 10th the size of the AP-1000.)

Knowing what I know about the company's performance for a very demanding customer, I feel reasonably confident that the timelines announced yesterday (design certification application in 2011, COL application in 2012, full license in 2015 and commercial operation in 2018) are not "stretch goals", but are reasonably achievable with some margin for the inevitable obstacles. Though the company made it very clear during yesterday's press conference that they are not going to be quoted on an expected cost - there are still way too many variables in the equation all with uncertainty ranges - company leaders did state that they expected to be able to produce their 125 MWe modules for a competitive price per kilowatt of capacity.

I have to admit something. While listening to the ten speakers (five from industry and three senators and two congressmen), I got a bit choked up. It was refreshing and even inspiring to hear about manufacturing reliable, affordable, zero emission power plants in North America (B&W maintains some of its manufacturing capability in Canada).

I kept thinking about all of those Americans who love to MAKE THINGS and who love to go home at the end of the day knowing that they have produced something more useful than paper, presentations or spreadsheet figures. I though about all of the people I know who enjoy bending metal and creating items of real precision and beauty. I thought about the importance to the entire economy of having reliable power systems. I thought of how nice it will be for electrical power suppliers to know that when their customers need more power they will be able to order a new unit, secure in the knowledge that it will be delivered on schedule, that it will work when it is needed, and that it's final price is the same as what the company promised.

The barriers to building nuclear plants are not just regulatory. These days it seems like all the plants are extra-large multi-gigawatt monsters, and those are swell, but they take several years and billions of dollars in up-front capital to build, and there are often delays and budget overruns, and local market conditions can change over the several years it takes from the time the first concrete is poured to the time when the plant comes online, so it has been very hard to find investors to back construction of new nuke plants.

This new reactor changes all that. It won't cost billions (B&W are vague on how much it will cost, but they're pretty clear on this point), it will have a predictable price and schedule, and the small size makes siting issues easier. Suddenly buying a nuclear reactor becomes vastly less risky, and at the same time it requires less up-front investment. These things will be manufactured in existing factories with plenty of experience making nuclear equipment, and delivered by train.

From an engineering standpoint, the design is pretty conservative. Unlike fourth-generation reactors like the various pebble bed projects, it doesn't try to introduce a new type of fuel. It's a light water reactor, so there's plenty of experience to draw upon (and it should be easier to get past the NRC). The passive safety features are nice, too.

I predict that, once over-strained electric companies can order a cheap(er) nuclear plant, guaranteed on time and for a fixed price, with a pool of trained operators and an established fuel supply chain, the floodgates will open and we'll finally see a real nuclear renaissance. If nothing else, it's way harder for the NIMBYs to protest ten little plants that are built quickly than one big plant that takes forever to finish.

[fnord]

All they need to do now is build a few outside the US to avoid the NRC's baleful influence.

[sketerpot]

All they need to do now is build a few outside the US to avoid the NRC's baleful influence.

That's what other small reactor makers are trying, but I think these guys are big enough and well-connected enough to actually bludgeon the NRC into something that bears a passing resemblance to almost sensible actions. I'm sure one of the reasons why they went with such a conservative design was to make regulatory approval faster and less expensive. The NRC doesn't really know how to handle anything but relentlessly conventional light water reactors, and they're too underfunded to learn quickly.

Now, some pictures! Here's the reactor itself:



Here's some reactor porn, a close up of the design. Warning: HUGE image file!

What this picture doesn't show is that the whole thing is underground, with generators in a building on top. The cooling pond is inside the containment structure, which kind of mitigates a lot of concerns about the security of spent fuel. It's downright pretty.

Reactors are delivered completely assembled. Expected construction time is three years. They have a letter of intent from the Tennessee Valley Authority, who are looking to build a plant based on mPower reactors.

[fnord]

A fellow Atomic Podcast listener.. not altogether unexpected.

How far can one of these suckers be pushed, burnup-wise? Can something like the MIT idea of annular fuel elements be used as well?

How long does a given core last? What happens at end of core life? Do B&W reprocess, refabricate and replace the core themselves?

Why aren't they going for a higher enrichment than U(5)? Commercial availability?

[sketerpot]

How long does a given core last? What happens at end of core life? Do B&W reprocess, refabricate and replace the core themselves?

The design lifetime of the reactor is 60 years.

Why aren't they going for a higher enrichment than U(5)? Commercial availability?

It's nice standard fuel, which B&W manufactures at their Lynchburg, Virginia factory. They're also talking about 10% enriched fuel for people wanting a ten-year refueling cycle.

[fnord]

That neatly explains the U(5). Veree smick.

I'm still interested in what they plan to do with the "spent" fuel at end of life, and if any core changes are proposed during reactor life. The mention of MOX capability may mean they intend to reprocess spent cores and top them up with enough U(5), or an additional TRU destruction capability - either way, that's a good thing by cutting down on Pu stocks.

It would be interesting if they've done analyses on running on U233(3), which has roughly equivalent reactivity - they mentioned ability to use thorium, which itself is a fertile material, breeding to U233 via Pa.

[Uraniun235]

I think Toshiba had a project going for small reactors in the 10-20MW range, but this sits very nicely between those and the big gigawatt reactors.

[fnord]

Toshiba's 4S design is a sodium cooled, fast neutron reactor which also pushes some edges - especially in liquid flow and equipment reliability. Sodium was apparently picked for its relatively high heat capacity, experience in use as a coolant, and it doesn't need to be pressurised, unlike the light water coolant/moderator in the B&W design.

Since it's a fast reactor, neutron moderation is a no no, meaning neutron reflectors are used to retain enough neutron flux to remain critical. Also looks like a high conversion ratio (ratio of fissile produced / fissile consumed), possibly at or above unity, helping to keep fuel needs down. I'm not sure what enrichment level Toshiba is looking at, but anything above the top end of commercial fuel production is a mongrel to obtain, generally coming from HEU downblending.

[tim31]

*pounds fists against the wall*

Why can't we have these down here.

WHY

It just makes you want to paint Greenpeace's door red.

[Patrick Degan]

The barriers to building nuclear plants are not just regulatory. These days it seems like all the plants are extra-large multi-gigawatt monsters, and those are swell, but they take several years and billions of dollars in up-front capital to build, and there are often delays and budget overruns, and local market conditions can change over the several years it takes from the time the first concrete is poured to the time when the plant comes online, so it has been very hard to find investors to back construction of new nuke plants.

This new reactor changes all that. It won't cost billions (B&W are vague on how much it will cost, but they're pretty clear on this point), it will have a predictable price and schedule, and the small size makes siting issues easier. Suddenly buying a nuclear reactor becomes vastly less risky, and at the same time it requires less up-front investment. These things will be manufactured in existing factories with plenty of experience making nuclear equipment, and delivered by train.

From an engineering standpoint, the design is pretty conservative. Unlike fourth-generation reactors like the various pebble bed projects, it doesn't try to introduce a new type of fuel. It's a light water reactor, so there's plenty of experience to draw upon (and it should be easier to get past the NRC). The passive safety features are nice, too.

I predict that, once over-strained electric companies can order a cheap(er) nuclear plant, guaranteed on time and for a fixed price, with a pool of trained operators and an established fuel supply chain, the floodgates will open and we'll finally see a real nuclear renaissance. If nothing else, it's way harder for the NIMBYs to protest ten little plants that are built quickly than one big plant that takes forever to finish.

Precisely. Once reactor design can be made a matter of assembly-line production, it then becomes possible to introduce economy-of-scale, especially as the cooling/containment pit would also be of standardised design and could possibly be pre-fabricated as well. It is also conceivable to adapt existing powerplants to these plug-in reactors by constructing the cooling/containment pits on-site while the plant is running on it's existing system, then make the switchover when the reactor assembly is ready to be installed.

[Memnon]

it then becomes possible to introduce economy-of-scale, especially as the cooling/containment pit would also be of standardised design and could possibly be pre-fabricated as well.

A nice part of this is that (as shown by France's nuclear reactors and by Canada's) parts become standardized to a degree as well, which lowers the cost of maintenance (another barrier to nuclear reactor construction).