Nuclear takes too long?

Industry Discussions Nuclear Power
1

I am often told on this and other TMF discussion boards that nuclear takes too long, and therefore we shouldn’t even try building any new power plants. I am told by the people who disparage nuclear power that building a plant takes 10 or 15 or 20 years. Their implication is that every plant will always take that long. I think it is a weak argument.

China has a large, dedicated nuclear construction organization. They have gotten good at building plants efficiently and quickly, in about 6 years from the first safety-related concrete pour to plant operation. The more you do something in a repetitive manner, the better you get at it. This is true for many aspects of life.

Today’s nuclear power plants are built to operate for at least 60 years, and it is becoming increasingly obvious that a well-maintained plant may last for 80 years or more. If it takes a few years to build the plant up front, then so be it. Start the planning process early, so the plant will be ready to supply power when you need it.

~ ~ ~ ~ ~ ~ ~

Antares Nuclear just started up a small, test reactor in Idaho. It took less than a year from beginning to reactor criticality. Admittedly, this is just a low-power test reactor. It is not a working power plant and will not generate electricity. However, it is a real, operating, nuclear reactor that is sustaining a nuclear chain reaction.

The demonstration took place at Idaho National Laboratory (INL) under US Department of Energy (DOE) authorisation, with Antares saying it is the first private company to bring an advanced reactor to criticality under the DOE Reactor Pilot Program. Antares was one of five companies selected last year by the DOE for support under the programme to expedite the testing of advanced reactor designs. The pilot programme is part of the Reforming Nuclear Reactor Testing at the Department of Energy executive order signed by Trump in May last year, with a goal to “to construct, operate, and achieve criticality of at least three test reactors using the DOE authorisation process by July 4, 2026”.

“We said criticality in 2026, electricity production in 2027, and power to the warfighter in 2028. Today is the first of those commitments delivered on the schedule we set. The President and DOE set an ambitious timeline for reactor testing, and we met that challenge,” Bramble said.

“We went from concept to a critical reactor, safely, in less than 12 months,”

~ ~ ~ ~ ~ ~ ~

The US Army is evidently the customer, which plans to use this reactor type for supplying electric power to either existing bases, or perhaps to remote locations.

I can’t find much detailed information about this microreactor design, other than it uses sodium for cooling, TRISO fuel (see other recent post about this fuel type), and heat pipes to remove core heat.

There are reportedly other companies at the Idaho National Lab that are also installing and starting up their own reactors. They have a little less than a month to meet the July 4 deadline, although I won’t consider it a failure if they are late. It is better to do it right, than too quickly.

_ Pete

5 Likes
2

Cookie cutter plants should also benefit from the learning curve making them faster and easier to build.

It’s tempting to optimize by building improvements into each new plant. But that adds to costs and risks delays.

Didn’t France use the cookie cutter approach?

2 Likes
3

Yes. They did one design, and cranked them out in volume. We should be doing the same.

The only really valid argument against nuclear power is waste management. Which is not an intractable problem. But it does need to be taken seriously, and planned for. Other than that, it’s all benefit.

2 Likes
4

Well, there’s a teensy bit of risk, I’d say. Of course that’s true of any power plant, but with a nuke it might be an order of magnitude larger.

5

That sentence is instructive for what it omits more than what it says. It probably wouldn’t take too much longer from the beginning of construction to operation here in the United States as well - or thereabouts. Maybe ten years (that’s what happened in Vogtle), but certainly not twenty.

The “takes too long” in the U.S. wouldn’t be for the construction. It would be for the approvals and permitting:

When plans for the SunZia project—a massive power line and wind farm in the Western U.S.—were first laid in 2006, Taylor Swift had just released a debut album and George W. Bush was president.

The project finally received a crucial permit this week.

That 17-year process underscores why some in Washington say there is a pressing need for an overhaul of the country’s rules for approving infrastructure projects.

The Interior Department’s Bureau of Land Management gave the green light on Thursday for a high-voltage power line. The permit allows the developer Pattern Energy to build the country’s largest wind energy project across three counties in rural New Mexico and deliver that electricity to large markets in Arizona and California.

Developers first applied for federal approval in 2008; environmental reviews started in 2009. In 2011, the project was “fast-tracked” by the Obama administration.

Exclusive | ‘Fast-Tracked’ SunZia Power Project Gets Approval 17 Years Later - WSJ

Obviously a 17-year approval timeframe is an outlier. But I would be surprised if you could get through the review process and the inevitable litigation that followed for a brand new nuclear plant at a new location in too much less than a decade. And that would be measured from when you apply - it will take you years to assemble the studies and expert reports and other materials you would need to even walk in the door.

IOW, if you found a piece of property and said, “I’d like to build a nuclear plant at this location, let’s get started,” it would indeed probably be close to twenty years before you began generating power from that site - even if the construction phase was only six years.

2 Likes
6

I have seen some changes lately occurring in the industry, which will hopefully cut down on your 20 year estimate. Below are two examples, but there are others which give me some hope.

#1

Bill Gates’ TerraPower is now building the first-of-its-kind Natrium reactor in Wyoming.

The link above gives a timeline on the progression of the Kemmerer project. The location was officially announced in 2021. The first Natrium plant will be built at an existing coal plant site, which is owned by one of Warren Buffett’s companies.

"In November 2021, the company officially selected Kemmerer as its preferred site. Kemmerer is near Naughton power plant, a 448-MWe coal-fired facility that is on track to be fully converted to a natural gas plant by 2026."

With construction now underway, they are targeting 2031 or so for operation. Since this is a first-of-a-kind plant, and is TerraPower’s first major construction project, they will undoubtedly run into delays and unforeseen problems. But there has to be a first experience, in order to find out where those problems lie.

~ ~ ~ ~ ~ ~ ~ ~ ~

#2

Dow Chemical is planning to install a high temperature gas cooled reactor in Texas to supply heat and power for one of Dow’s production plants. The NRC recently completed a major environmental review of the plan, and found no significant problems.

From the link:

According to X-energy, the NRC completed the environmental review in under a year, marking the first time a U.S. commercial advanced reactor project has completed its National Environmental Policy Act (NEPA) review through the more efficient EA pathway, rather than a full environmental impact statement (EIS), in a Part 50 construction permit proceeding.

The NRC’s review—completed in under 12 months and ahead of schedule—clears a key prerequisite in the agency’s review of the construction permit application filed by Long Mott Energy LLC, a Dow wholly owned subsidiary, in March 2025. The application covers a four-module, 800-MWth/320-MWe Xe-100 high-temperature gas-cooled reactor facility at Dow Chemical’s UCC Seadrift Operations site on the Texas Gulf Coast.

According to the NRC’s project dashboard, the agency will now work to complete its final safety evaluation by November 2026—within the 18-month timeline required by Executive Order 14300 —after which it could issue a final construction permit decision that will authorize construction of the facility.

~ ~ ~ ~ ~ ~ ~

From my perspective, there seems to be a genuine good faith effort from the NRC to make their reviews and approvals go faster than those activities have gone in the past.

_ Pete

1 Like
7

They did several designs as described below:

900 MWe class (CP0, CP1 and CP2 designs)
Based on the Beaver Valley Nuclear Power Station and later the North Anna Nuclear Generating Station, the CP series was the first Westinghouse-type reactors to be built in France.[87] There are a total of 34 of these reactors in operation; most were constructed in the 1970s and the early 1980s. In 2002, they had a uniform review and all were granted a 10-year life extension.

With the CP0 and CP1 designs, two reactors share the same machine and command room. With the CP2 design, each reactor has its own machine and command room. Apart from this difference, CP1 and CP2 use the same technologies, and the two types are frequently referred to as CPY. Compared to CP0 they have an additional cooling circuit between the emergency system that in case of an accident allows to spray water into the containment and the circuit which contains river water, a more flexible control system and some minor difference in the layout of the building.[88]

This three loop design (three steam generators and three primary circulation pumps) was also exported to a number of other countries.

1300 MWe class (P4 and P’4 designs)
There are 20 reactors of this design (four steam generators and four primary circulation pumps) operating in France. The P4 and P’4 type have some minor difference in the layout of the building, especially for the structure which contain the fuel rods and the circuitry.[88]

In July 2025 French regulator (French Nuclear Safety and Radiation Protection Authority) has decided EDF can operate 1300 MWe reactors beyond the initial 40 years lifespan.[91]

1500 MWe class (N4 design)
There are only four of these reactors, housed at two separate sites: Civaux and Chooz. Construction of these reactors started between 1984 and 1991, but full commercial operation did not begin until between 2000 and 2002 because of thermal fatigue flaws in the heat removal system requiring the redesign and replacement of parts in each N4 power station.[92][93] By 2002 the reactors had been uprated from 1450 MWe to 1500 MWe.[94] Serious stress corrosion cracking in the stainless steel safety system piping was discovered to 2021, requiring shutdowns for inspections and repair.[53]

1650 MWe class (EPR design)
The next generation design for French reactors is the EPR, which is also intended for foreign markets. The EPR was originally developed as a German-French joint project to incorporate the advantages of the highly reliable German Konvoi design as well as French experience at mass construction of relatively “standardized” nuclear facilities. The design was intended to be built in both Germany and France as well as various export markets. However, the German nuclear phase-out precluded any construction of EPRs in Germany and ultimately led to Siemens selling its shares in the joint venture (see below). Two EPR units are in operation at Taishan in China, one at Olkiluoto in Finland and one at Flamanville in France. In the United Kingdom, two units are under construction at Hinkley Point C and two are planned at Sizewell C.

In June 2023, EDF announced it was starting the authorisation process to build two EPR 2 reactors at the Penly Nuclear Power Plant, anticipating that site preparatory work would begin in summer 2024 and construction would begin about 2027.[95]

The reactor design was developed by Areva contributing its N4 reactor technology and the German company Siemens contributing its Konvoi reactor technology. In keeping with the French approach of highly standardized plants and proven technology, it uses more traditional active safety systems and is more similar to current plant designs than international competitors such as the AP1000 or the ESBWR.

In 2013, EDF acknowledged the difficulties it was having building the EPR design.[96] In September 2015, EDF’s chief executive, Jean-Bernard Lévy, stated that the design of a “New Model” EPR (later called EPR2) was being worked on, which will be easier and cheaper to build, which would be ready for orders from about 2020.[97] The EPR2 design maintains the post-Fukushima safety measures of the EPR, but simpler construction and lower capital costs are exchanged for the ease of maintenance of the original EPR design that are due to the “two-room” containment and fourth safety train.[98][99] In 2016, EDF planned to build two New Model EPR reactors in France by 2030 to prepare for renewing its fleet of older reactors.[100] However following financial difficulties at Areva, and its merger with EDF, French Energy Minister Nicolas Hulot said in January 2018 “for now [building a New Model EPR] is neither a priority or a plan. Right now the priority is to develop renewable energy and to reduce the share of nuclear.”[101]

In March 2025, the Nuclear Policy Council agreed that a subsidised government loan should be made available to cover at least half the build cost of six EPR2 reactors. A Contract for Difference at no more than €100 per MWh would cover the remaining costs. The first three double EPR2 reactors are proposed for the Penly, Gravelines and Bugey sites, with construction starting in 2027.[102](Flamanville Nuclear Power Plant - Wikipedia) in France. In the United Kingdom, two units are under construction at Hinkley Point C and two are planned at Sizewell C.

3 Likes
8

Nuclear power plants must compete with Natural Gas, Wind and Solar on cost and schedule. Currently all three are cost and schedule winners over nuclear. But the Iran war has added new concerns reliability, cost and schedule concerns for Natural Gas power plants around the world except in the US. The jury is still out on nuclear power cost and schedule, and nuclear waste costs.

9

China has 60 nuclear reactors in operation and 35 in construction.
But nuclear power is not making headway against other sources of power generation in China. Nuclear power generation was 5% of total China generation from 2019 to 2023, but only 4.5% in 2024.

Overview

Table 1. China energy indicators, 2023

10

Yes, those all cost a bit less. Wind and solar have the disadvantage of not being reliable (no sun at night, and the wind doesn’t always blow). And, as you point out, gas isn’t reliable, either, due to geopolitical concerns.

IMO, nuclear should be used as a stop-gap measure for wind and solar. When those aren’t working, nuclear picks up the slack. We shouldn’t be relying on anyone else on the planet for our energy. It’s strategically unwise.

As for French designs, apparently I misremembered a documentary I saw several years ago about nuclear programs. I double-checked, and the French use three designs (and just started-up a fourth in 2024). Out of 56 reactors, three designs plus 1 reactor of the latest design. The reason for the multiple designs is increased power output of the designs. Compare and contrast to the USA where every reactor in the country is unique. So, we have R&D costs for every one of them. In France they turned out 30 of their first reactors, and only had to R&D once.

1 Like
11

That’s not how nuclear works. It isn’t something you switch on and off light a lightbulb. It’s baseload, because it takes a (relatively) long time to ramp up or down the power output, being more like a coal plant which takes time to come up to temperature or cool back down. (The newer, smaller ones are technically able to ramp up and down better than the earlier ones, but because of the immense cost and potential output, it’s better to run them continuously as base load.)

Gas turbines and other peaker plants are what you want. You can do it with hydro, and of course industrial grade battery is the fastest of them all.

12

Interesting, and great news. From what I could tell, though, neither of those projects was challenged by opponents - which is what often contributes to the longest delays. And that’s itself a positive sign, since I think there’s been some softening among environmental advocacy groups in their staunch opposition to nuclear power. It has its own problems, but at least it’s not fossil fuels. If it’s reaching the point where companies can get through the licensing/approval process for projects without an extra decade in court, that would definitely help. Though a decade from site selection to power generation is still a fair amount of time.

13

It kinda is, though. Sure, it isn’t instantaneous response. But it’s not a novel concept among power engineers. Solar is the easy one because you can predict solar. Overall, they deal with averages. Here’s a decent article (first hit I got, actually):

https://www.iaea.org/newscenter/news/five-reasons-the-clean-energy-transition-needs-nuclear-power

14

In some ways yes, but clouds (and snow) are only predictable in the short term. You can predict the climate with fair certainty, but weather is rather important too.

15

It is a common misconception that nuclear plants must be operated in baseload only and they are not capable of adjusting power up and down.

In reality, nuclear plants are often used in some areas for load following. Since France relies so much on nuclear power (~70%), the French nuclear plants routinely load follow, as the system demand rises and falls throughout the day.

In most other countries, it is usually most economical to operate the nukes as baseload only. Other generators, such as natural gas fueled plants, are the load followers. But it doesn’t need to be that way.

From the link:

The high reliance on nuclear power in France thus poses some technical challenges, since the reactors collectively need to be used in load-following mode.

And:

There are two ways of varying the power output from a PWR: control rods, and boron addition to the primary cooling water. Using normal control rods to reduce power means that there is a portion of the core where neutrons are being absorbed rather than creating fission, and if this is maintained it creates an imbalance in the fuel, with the lower part of the fuel assemblies being more reactive than the upper parts. Adding boron to the water diminishes the reactivity uniformly, but to reverse the effect the water has to be treated to remove the boron, which is slow and costly, and it creates a radioactive waste.

So to minimize these impacts since the 1980s EdF has used in each PWR reactor some less absorptive ‘grey’ control rods which weigh less from a neutronic point of view than ordinary control rods and they allow sustained variation in power output. This means that RTE can depend on flexible load following from the nuclear fleet to contribute to regulation in these three respects:

PWR = Pressurized Water Reactor. This is the most common reactor type used in the world today. Boiling Water Reactors are also capable of load following with control rods, and also by adjusting the recirculation water flow through the core. This changes the amount of boiling, and therefore changes reactor power. The Columbia Generating Station in Washington state, for instance, is often used in this way.

~ ~ ~ ~ ~ ~
Most of the newer SMRs are specifically designed for load following. The Natrium Kemmerer plant being built in Wyoming will have a unique molten salt heat storage system. This will allow the reactor to remain at full power most of the time, but electrical output from the main turbine can be varied up and down, as more or less steam is produced from the heat in the molten salt.

_ Pete

4 Likes
16

In the example I gave, that is a first-of-a-kind plant that needs to go through the NRC design reviews and approvals. Once those approvals are given, then the design will be standardized, and subsequent projects should go faster. Also, the construction organization will learn lessons along the way, so the building process should go smoother, after the first plant is in operation.

_ Pete

17

I didn’t say that, nor do I think that. I do know that the great majority of nuclear plants are used in base load configuration, because it’s the most efficient given the costs and the maintenance and other variables that come into the equation.

There may be newer designs that are more adept at load following, but most of the plants in the world are used full out, with minor variations. They can ramp up and down, but more slowly than other options, and as a general rule they don’t.

1 Like
18

It has been standard knowledge in the power generation industry that nuclear power is economically feasible ONLY when they are operated as near 100% as possible. The capital cost of nuclear power is so high that operating them in the load following range of 40% to 70% would extend their payoff from 30 years to 60 years.

The real solution to solar and wind downtime has been already established. The solution is batteries and other energy storage methods. Currently utilities all over the world (including US) are scrambling to add more batteries/energy storage to the electrical generation systems. Read the following:

Battery storage is the fastest growing power technology today. In 2025, 108 GW of new battery storage capacity was deployed worldwide, 40% more than in 2024. Installed capacity is now eleven times higher than in 2021. Lithium‑iron phosphate (LFP) batteries now account for around 90% of deployments; while less energy‑dense than rival chemistries commonly used in EVs, LFP batteries are typically cheaper and better suited to more frequent cycling. Just five years ago, the market share of LFP batteries in deployments was well below 50%.

Around 80% of new battery capacity in 2025 was utility‑scale. The remainder was behind-the-meter capacity installed by commercial and residential consumers. Battery storage durations are gradually lengthening. While most projects still cluster around two hours, an increasing number can be deployed for four hours or more, reflecting the growing value of flexibility in systems with rising shares of PV.

China continued to lead battery deployment in 2025, accounting for around 60% of global additions, followed by the United States and Europe. However, deployment is widening beyond the largest markets, with strong momentum in Australia and parts of the Middle East, where storage is increasingly seen as a key building block for electricity security and renewables integration.

Battery-based uninterruptible power supplies (UPS) – primarily in data centres – also saw significant growth, with capacity additions rising 30% to 45 GW in 2025. However, unlike battery storage systems, UPS generally only provide short‑duration backup to bridge outages until other backup sources start.

1 Like
19

A bit more information on this Antares reactor. The power production equipment uses a Brayton cycle, with nitrogen gas as the working fluid. The Brayton cycle is the same thermodynamic process that jet engines and other combustion turbines use. Here, the reactor supplies the heat, rather than a combustion chamber, and the whole thing is a closed system, rather than open like a jet engine. I like the idea of using nitrogen as the working fluid. Other designs I’ve seen sometimes use helium, which has its disadvantages in some respects.

I am not aware of any nuclear power plant anywhere using the Brayton cycle for power generation. Even China’s HTR-PM SMR currently in operation uses a standard Rankine steam cycle. Using Brayton does make building a small, portable nuclear power plant somewhat simpler, since they aren’t boiling water to make steam.

From the Antares website:

Screenshot 2026-06-12 at 15-12-08 Antares Nuclear Factory-Produced Fission Microreactors for Strategic Energy
Screenshot 2026-06-12 at 15-12-08 Antares Nuclear Factory-Produced Fission Microreactors for Strategic Energy1268×735 120 KB

In the diagram, it is interesting they call the electrical generator an alternator. I’ve never heard of a power plant AC generator referred to in that way. “Alternator” is more of an automotive term, although the function is pretty much the same.

_ Pete

3 Likes
20

A second microreactor has achieved initial criticality as part of the DOE reactor pilot program. This second design is from Valar Atomics, with its Ward 250 reactor at a test site in Utah. Again, this is a TRISO fueled reactor.

From the link:

“Ward 250 reached these milestones as a complete, fully integrated system configured for power operations. The reactor has now begun non-commercial power ascension, and Valar will share further milestones as they are achieved,” Valar said in a statement. Valar noted that Ward 250 will continue “a series of experiments in the weeks ahead,” and that Valar expects to share milestones as they are achieved.

~ ~ ~ ~ ~ ~ ~

It sounds like this project will produce thermal heat at least, and might even generate electricity.

A bit more information:

Ward 250, an HTGR rated at 100 kWt initial test power and scalable to 5 MWe, uses helium coolant and TRISO fuel particles in Advanced Gas-Cooled Reactor (AGR) compacts, according to an October 2025 Valar quality-assurance program description. The document says Ward 250 incorporates passive safety features and builds on WardZero prototype technology, while the co-located Valarin Fuel Fabrication Facility is designed to manufacture TRISO-coated particle fuel embedded in graphite compacts using German HOBEG technology with modern process improvements. Kiewit Nuclear Solutions served as the project’s engineering, procurement, and construction contractor.

And:

“Criticality proves the physics; power operations prove the engineering,” Valar said. “Power operations require cooling, instrumentation, controls, advanced shielding, and power conversion to function together as a complete system. Ward 250 went critical with those systems integrated and in place.”

_ Pete

1 Like