This week we conclude our look at nuclear power construction costs. So far we’ve looked solely at conventional nuclear power stations that supply electricity to the grid for residential and commercial use. But civilian power stations aren’t the only sort of nuclear reactor that the US builds. The US Navy has been building nuclear reactors for 70 years as part of its Naval Reactors program. It’s built over
Great article! I’m a nuke sub officer, and I don’t think I can emphasize enough the focus on training for the navy nuclear power program. In addition to nuclear power school and prototype (where you qualify to operate a full-size reactor in a training context), the first six months or so of arriving at your eventual boat is focused on learning again how to operate a reactor. And then once you’re qualified, there is tons of continuing training - on my boat we did a training session three times a week, exams once a month, as well as smaller training sessions while on watch. And that doesn’t even get into the frequent drills.
One minor quibble about your article - technically, SUBSAFE doesn’t apply to the reactor systems. SUBSAFE is specifically for systems to ensure the dive/surface ratio stays at 1 - things like main ballast tank vents and pressure hull boundaries. Reactor systems have their own, separate QA programs, though the requirements are comparable.
Re the various costing data, military costing is fantastically complicated, but the $18 billion for the Seawolf is going to include a bunch of R&D spending that benefited the Virginias as well and wasn't tied directly to shipbuilding. I'd think the $3-3.5 billion is probably the right number to use. And I'm quite sure that the cost of maintaining the Naval Reactors program isn't included in the shipbuilding budget.
Also worth noting that Naval Reactors has always had a political program second to none (at one point Rickover was almost running a "vote for my program and you get a sub" deal) and there's still a law on the books stating that any new surface combatant over 8,000 tons needs either nuclear power or a waiver. The waiver seems likely for any new ships going forward, though.
There was no cost improvement on sending a kilogram into space for the entire time NASA had the monopoly. As soon as SpaceX came into play, the cost dropped. But 98%. Not a typo. Perhaps a similar thing could occur for nuclear power in ships using CMSRs?
The THRESHER was lost because of a reactor scram and the time/difficulty in restarting -- See Death of the Thresher (1964) by Norman Polmar, written with the assistance of RADM Dean Axene, the first commanding officer of the THRESHER and other, highly qualified submariners. As soon after the loss as possible Rickover convened a meeting of available nuclear submarine COs in his office to determine more rapid methods of restarting a reactor SCRAM. N. Polmar
Curious if there are any cost estimates for navy nuclear plants e.g. $/MWe? It seems to me like there's a regulatory opportunity here to contract the Navy to build civilian on-shore nuclear power plants (or just power for military bases). Naval reactors are basically tried and tested small modular reactor and given the high degree of public trust, regulatory arbitrage that comes from anything defense related, and a proven history+institution of safety, this seems like an easy way to introduce more nuclear capacity into a declining civilian nuclear energy fleet
I found a few estimates but they're not sourced super well, and they tend to contradict the more reliable, consistent information we have that a shipboard reactor adds a LOT of cost.
yeah makes sense, I'm guessing its hard to disentangle/account the cost of the reactor itself vs the design changes needed in shipmaking + labor cost to maintain.
Still, if they're hitting anything close to $1/MW - $5/MW, then thats cost-competitive with wind/solar/batteries except with rampable generation capabilities that cover any duration (hours, days, weeks, seasons).
In general I just find it ironic that there are all these civilian nuclear SMR startups with new technology (which is a great thing to encourage!) that are trying to get through NRC approval, but the navy has been building and deploying hundreds of nuclear SMR for decades now, we've just never brought it over to the civilian side
Yeah I didn't end up getting into SMRs, but that would be a question I would ask someone working on them - "Why do you think the navy can't build reactors for less than other power sources, and what makes your approach different."
There's one other important advantage for adopting the naval nuclear approach for civilian use.
When you've finished with them, you can move the reactor for disposal.
Lots of smaller reactors operating in parallel also may get them down to the size where they can be put on the back of a truck - which triggers the economies of scale point you made about housing.
Perhaps it is time to build nuclear reactors of a size that will fit in a shipping container.
Great article! I’m a nuke sub officer, and I don’t think I can emphasize enough the focus on training for the navy nuclear power program. In addition to nuclear power school and prototype (where you qualify to operate a full-size reactor in a training context), the first six months or so of arriving at your eventual boat is focused on learning again how to operate a reactor. And then once you’re qualified, there is tons of continuing training - on my boat we did a training session three times a week, exams once a month, as well as smaller training sessions while on watch. And that doesn’t even get into the frequent drills.
One minor quibble about your article - technically, SUBSAFE doesn’t apply to the reactor systems. SUBSAFE is specifically for systems to ensure the dive/surface ratio stays at 1 - things like main ballast tank vents and pressure hull boundaries. Reactor systems have their own, separate QA programs, though the requirements are comparable.
Re the various costing data, military costing is fantastically complicated, but the $18 billion for the Seawolf is going to include a bunch of R&D spending that benefited the Virginias as well and wasn't tied directly to shipbuilding. I'd think the $3-3.5 billion is probably the right number to use. And I'm quite sure that the cost of maintaining the Naval Reactors program isn't included in the shipbuilding budget.
Also worth noting that Naval Reactors has always had a political program second to none (at one point Rickover was almost running a "vote for my program and you get a sub" deal) and there's still a law on the books stating that any new surface combatant over 8,000 tons needs either nuclear power or a waiver. The waiver seems likely for any new ships going forward, though.
Relevant blog posts of mine:
https://www.navalgazing.net/Modern-Propulsion-Part-4
https://www.navalgazing.net/Military-Procurement-Pricing
There was no cost improvement on sending a kilogram into space for the entire time NASA had the monopoly. As soon as SpaceX came into play, the cost dropped. But 98%. Not a typo. Perhaps a similar thing could occur for nuclear power in ships using CMSRs?
The THRESHER was lost because of a reactor scram and the time/difficulty in restarting -- See Death of the Thresher (1964) by Norman Polmar, written with the assistance of RADM Dean Axene, the first commanding officer of the THRESHER and other, highly qualified submariners. As soon after the loss as possible Rickover convened a meeting of available nuclear submarine COs in his office to determine more rapid methods of restarting a reactor SCRAM. N. Polmar
Curious if there are any cost estimates for navy nuclear plants e.g. $/MWe? It seems to me like there's a regulatory opportunity here to contract the Navy to build civilian on-shore nuclear power plants (or just power for military bases). Naval reactors are basically tried and tested small modular reactor and given the high degree of public trust, regulatory arbitrage that comes from anything defense related, and a proven history+institution of safety, this seems like an easy way to introduce more nuclear capacity into a declining civilian nuclear energy fleet
I found a few estimates but they're not sourced super well, and they tend to contradict the more reliable, consistent information we have that a shipboard reactor adds a LOT of cost.
This says $100 million for a submarine reactor in 2001 dollars, but again, unclear where that comes from: https://mragheb.com/NPRE%20402%20ME%20405%20Nuclear%20Power%20Engineering/Nuclear%20Marine%20Propulsion.pdf
yeah makes sense, I'm guessing its hard to disentangle/account the cost of the reactor itself vs the design changes needed in shipmaking + labor cost to maintain.
Still, if they're hitting anything close to $1/MW - $5/MW, then thats cost-competitive with wind/solar/batteries except with rampable generation capabilities that cover any duration (hours, days, weeks, seasons).
In general I just find it ironic that there are all these civilian nuclear SMR startups with new technology (which is a great thing to encourage!) that are trying to get through NRC approval, but the navy has been building and deploying hundreds of nuclear SMR for decades now, we've just never brought it over to the civilian side
Yeah I didn't end up getting into SMRs, but that would be a question I would ask someone working on them - "Why do you think the navy can't build reactors for less than other power sources, and what makes your approach different."
There's one other important advantage for adopting the naval nuclear approach for civilian use.
When you've finished with them, you can move the reactor for disposal.
Lots of smaller reactors operating in parallel also may get them down to the size where they can be put on the back of a truck - which triggers the economies of scale point you made about housing.
Perhaps it is time to build nuclear reactors of a size that will fit in a shipping container.
Thanks Brian! Very informative.
Thank you for writing this
Super interesting articles! Thanks