This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


What advanced nuclear reactor designs are likely to be successful and why?

Dan Yurman's picture
Editor & Publisher, NeutronBytes, a blog about nuclear energy

Publisher of NeutronBytes, a blog about nuclear energy online since 2007.  Consultant and project manager for technology innovation processes and new product / program development for commercial...

  • Member since 2018
  • 1,629 items added with 1,300,338 views
  • Nov 15, 2018

Recent studies by the Third Way, a DC think tank, point to 40 separate efforts to develop advanced nuclear reactor designs. 

As some readers may recall, US Navy Admiral Hyman Rickover pointed out that paper reactors are easy to design. The hard part is building one.

This axiom was proved recently as Transatomic Power folded its high profile tent saying that their design would not be commercially viable.

Examples of advanced designs include TRISO fueled systems, molten salt, sodium cooled, lead cooled, etc. It's a long list.

So what are the success factors that will bring any of these designs to market and what firms are more likely to master them than others?  Please post your replies here. 



Your access to Member Features is limited.

I have been researching and writing about nuclear power alternatives to the conventional power plants that have been built and designed in the past. I'm happy to share with you in this discussion group what I have discovered. See which describes the U-Battery, a small-scale modular encapsulated reactor that could serve a wide variety of power applications.

The pursuit of fusion-fission reactors is described in a posting at

A company in Ontario is developing an integrated molten salt reactor with the assistance of the Oak Ridge National Laboratory. See



Eric Lane's picture
Eric Lane on Dec 1, 2018

Len, I read all three links.  Interesting.  When it comes to nuclear, my first concern is always the backend, waste.  It is the presumption of arrogant ignorance that believes we can safely store nuclear waste for 10,000 years or longer.  We can't.  Period.  That's a fantasy that greed creates.  For example, the IMSR is an interesting concept but would create a huge stream of waste.  We have to realize that each nuclear concept wants to become a major player in nuclear power porduction.   If the IMSR goes into full production, you could have tens of thousands of these units all over the world and every seven years needing to be replaced and the spent units stored.   This is the true weakness of nuclear power.  Wind and solar are the only energy forms I am aware of that do not create a waste stream.  If we have learned anything about our dependence on fossil fuels, it is that there are serious consequences to the waste created.  In the fossil fuel case, it's climate change.  There may be other consequences that we aren't even aware of yet.  Since the dawn of the nuclear age, we have gone along merely on the assumption that nuclear waste will somehow magically take care of itself.  I hasn't.  It has become another serious environmental threat.  My argument is really quite simple; until we solve the nuclear waste issue and I don't mean the storage issue but the due no harm principle, nuclear has been and remains a serious threat to human safety.  Why would we replace a fossil fuel industry that presently poses a serious threat to human existance with a nuclear industry that would pose an equally serious threat to human existence as presently construed?       

Bob Meinetz's picture
Bob Meinetz on Dec 12, 2018

Eric, nuclear "waste" is a non-issue:

1) Spent fuel from power plants has been stored safely, without fatalities or injuries, for over half a century.

2) Spent fuel from power plants is stored with 95% of its fissile material unused. By recycling it as France does, we could get twenty times the energy out of the same amount of uranium.

3) Spent fuel's radioactive danger, within a few hundred years, declines to levels below natural background radiation. Today, you are at more danger from atomic testing in the 1950s and 1960s, the radioactive plutonium from which surrounds you day and night (there's nothing you can do about it, but don't worry - in proportion to the threat of being struck by lightning or being stung to death by bees, it's harmless).

We don't need spent nuclear fuel to "magically take care of itself", but humans to magically educate themselves - the biggest challenge of all.

Bob Meinetz's picture
Bob Meinetz on Dec 12, 2018


I do not see any of the existing designs to be economical successfull, when even the prices the designers hope to reach in far futer prices that system out of the market.

If there is more than a few €/MWh price difference betweeen wind/solar on one side and nuclear on the other side, it will be cheaper to expand the grids and remove the local volatility of wind and solar. Just make room in your head and undersatnd that the latest UHVDC Pover line built in china (12GW per system) is long enough to make a direct connection from Ireland to Canada, or west Africa to Brasil.

And that even now when we discuss this, it is already possible to exchange some GW electrical power  between Saigon and Lissabon. It is technically possible (see above, and 12GW is just the beginning, not the limit) to expand the grids to the amount of power transfer needed.

I do not know how much power transfer will be needed - Entsoe thinks they can handle 100% renewables with a quite small expansion of grids just inside europe - but I am sure that there is not much night within the grid when your grid starts in Los Angeles and stretches from there eastwards till Kamtschatka. Also I see no likelyhood that the wind settles on the whole planet at the same time. Which would be the worrst case grid size, but still fully economical viable.

Think about the following fact when you expect a nuclear renaisance from china: China stoppt to start new nuclear power plant constructions durin the last two years (only existing construction sites are finished) in parallel state Grid china and other companies started buying grid companies all over the world (while publishig that they want to interconnect the world with a HVDC Overlay grid) I see the grid as the winning technology, and I would invest in expanding the grids and not leave this - and the control of the grids - to China.

The West has already abandoned its leadership in nuclear energy; for the future of nuclear power we must look to China and India.

Despite the renewable hype, both of these countries have coal-dominated power systems, both for the exising fleets and for new construction (China has unusually large hydro resources and deployment, but the other non-fossil energy sources are small).  

Both countries boast of large renewables investments.  Both are continuing to build coal-fired power plants.  But also, both are building new nuclear, at about the same rate as solar or wind.  Most reports seems to say the power cost of renewables is less than that from nuclear, but the lower integration cost (including fossil fuel backup) makes nuclear more economical in these countries.

According to Chinese policy, large light water reactors are the technology of choice, with new builds switching to sodium cooled fast reactors at some point in the future, ultimately yielding a mixed fleet.  In India, nuclear builds are split between domestically designed large heavy water reactors and imported (Russian) large light water reactors.  Both are already building their first-of-a-kind fast reactors.

There don't seem to be any serious analysts who think storage will become so cheap that it will fix the integration/backup cost of renewables.  A large boom in fossil gas production would help reduce renewable system costs, but they have a very long way to go to reach that level of gas production.

Small reactors (SMRs) are partly interesting to US utilities because our national grid is divided is many separate jurisdictions.  China and India don't have that situation, so 1GW class reactors are just fine.  The US grids are being interconnected more and more (in part due to windpower deployments).  This new level of interconnection will make it easier for us to use large reactors also, so we may eventually follow in the footsteps of China and India.  The SMRs do help build industry experience quickly, but that is a short term effect, if you goal is to replace the entire fossil fuel generation fleet over a 40 year period (i.e. we need 12 GW per year of new construction, adjusted for capacity factor).

Eric Lane's picture
Eric Lane on Nov 26, 2018

My argument to this post, Nathan, is Fukushima.  When you add in that disaster to the cost of nuclear power to Japan and the fact that it may be 40 to 60 years before they can contain the nuclear rods all the while piling up radioactive waste, the cost of nuclear power in Japan has gone through the atmosphere.  I just find it hard to believe that we can't put our American resourcefulness and come up with non-toxic forms of energy production.  We better do it soon or humanity won't be around for the long term. 

Jesper Antonsson's picture
Jesper Antonsson on Nov 30, 2018

Fukushima proved how mild nuclear disasters are. The run-rate costs of Fukushima is about one third of that of the German Energiewende, and that is despite the fact that Fukushima costs could've been cut to a fifth or less by simply having more rational policies in place.

Nuclear power, from a life cycle perspective including accidents, is less toxic than RE alternatives even before considering the RE externalities such as extremely costly grid batteries, natgas balancing, continent-wide grids and so on.

No one seems to be addressing the major problem with nuclear power since its inception.  It's not necessarily the cost of construction and production.  We might be able to live with nuclear accidents although folks at Fukushima would give anything for a do over and a forget about it.  No, the real problem with nuclear is the waste.  It always has been and until that problem is solved, nuclear is simply too dangerous as a power source.  If you look at the entire life cycle of nuclear power from construction to dismantling and storage of waste for up to 10,000 years (!), I believe this puts nuclear in the crazy expensive catagory.  Imagine if nations became dependent on nuclear power and thousands of nuclear power plants were built, we would be replacing the consequences of fossil fuel based climate change with nuclear waste based extinction. 

Nathan Wilson's picture
Nathan Wilson on Nov 22, 2018

Lots of activists tell us this (to fear nuclear waste), but we never hear this from real scientist; it's simply not true.  Geologist tell us there are several good ways to dispose of spent nuclear fuel; temporary storage followed by recycling is my favorite.

The fact that activists have convinced the public to fear nuclear waste in spite of the science, and the fossil fuel industry has benefited as a result should be of serious concern to anyone who wants to see science-driven public policy.

Eric Lane's picture
Eric Lane on Nov 26, 2018

Sorry for not responding sooner.  I only now received notification of your response.  First, name calling is rather juvenile in my book.  It would be like me calling you a right wing wacko bird extremist who doesn't give a rat's fanny about the consequences of his actions.  But I won't. 

I'm not an activist.  I consider myself a rationalist guided by common sense.  Obviously, if someone is earning their income in the nuclear industry, it's harder to be neutral or rational.  I have made the statement that nuclear waste is the Achilles heel of nuclear power.  I want you to imagine if we made the world dependent on nuclear power what that would mean for nuclear waste.  Your argument, Nathan, puts the cart before the horse.  First solve the problem of nuclear waste, then we can consider nuclear power.  The other way around is nothing more than wishfull thinking.  Something the nuclear industry has been doing since its inception.  Is this idea really that hard to grasp?  Can it be so easily dismissed by calling me an activist?  Do you realize that some of the nuclear waste has a longer half life than all of recorded history?  We have a hard time finding Sumarian villages and Egyptian tombs and we are going to burry waste for a minimum of 10,000 years!  To be honest, I think that's kind of crazy, activist or not.  By the way, I do know nuclear scientists and academics who oppose nuclear power for an array of reasons. 

Jesper Antonsson's picture
Jesper Antonsson on Nov 30, 2018

Recorded history is an irrelevant time frame for nuclear waste, which should be compared to geological time frames.

Do you know why we enjoy natural gas and oil? These are extremely volatile compounds that very easily moves around. Yet we can use them exactly because geology has locked them up for tens of millions of years.

Nuclear waste is composed of ceramical pellets and heavy elements that doesn't dissolve easily and doesn't move much. They need to stay buried and not leak much for only tens of thousands of years. That's a thousandth of the gas/oil time frame, and we need not leave it entirely to nature. We can choose site and design its environment. It's ridiculously easy and safe to do.

So this is simply not a significant tech problem. It's only a significant problem for activists and the politicians they push in front of them. Also, to be blunt, nuclear waste storage is not a task we can choose to not have at this point, and producing 10x more or 100x more won't make a difference as long as we collect it at the same repository sites. Saying "ok, let's do a global climate disaster because we don't want to put 10x as much waste in Yucca Mountain", that's quite irrational.

The decison will be based on ecponomics - and one of them will do.  The EM2 design is elegance itself, General Atomics have a solid background, they have built gas cooled reactors and the silcon carbide fuel claffing is useful in LWR.  With their cost estimated it is already competitive in much of the worls - and in the US at a gas price of $7.99/MMBtu.

The answer may be more synergistic




When I was a member of the Board of UKAEA, I used to believe that Nuclear energy was the answer to the world’s insatiable need for energy. I am not now convinced that it is either economically viable nor environmentally sound.

Times change and things move on but I continue to believe that nuclear fusion has a vitall important role in the future, if we can convince ourselves to make the massive but necessary investment.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

When I first became concerned about climate change 20+ years ago I also thought nuclear energy was our future.  At the time wind and solar were much more expensive than nuclear, simply unaffordable.

About 30 years ago wind cost $0.39/kWh.  Solar panels sold for about $100/watt.

Now we're at a time when wind has dropped under $0.03/kwh and solar panels can be purchased for under $0.50/watt.

Back then I was very concerned about how we would possibly get people to accept higher electricity prices in order to deal with a problem that wouldn't hit us for some time.  New nuclear would have been more expensive than coal.  The market wouldn't have supported the change away from fossil fuels.

Now we've reached a point at which wind and solar are our least expensive ways to generate electricity.  Electricity from turbines and panels is now less expensive than from paid off coal plants.  Even before we factor in the very high external costs of coal.

We've reached a point at which we can minimize extreme climate change by switching to wind and solar and lower our cost of electricity at the same time.

Between 2010 and 2017 renewables saved Texans about $5 billion in electricity prices.



Nathan Wilson's picture
Nathan Wilson on Nov 21, 2018

"...switching to wind and solar and lower our cost of electricity..."

Advocates keep telling us this, but someone keeps convincing the government to extend the subsidies and policy support for renewables.

Meanwhile organizations like the IPCC and even the (traditionally anti-nuclear) Union of Concerned Scientists are starting to admit that renewables are not displacing fossil fuel fast enough, and in particular, lack of policy support for exisint nuclear is making the situation much worse.

"What Advanced nuclear reactor designs are likely to be successful and why?​"

To answer this question the timeframe needs to be defined.

The disruption to the learning and deployment rates that began in the late 1960s and has continued since has delayed progress and increased costs enormously. The time lost can never be recovered (like a lost golf stroke cannot be recovered).  All we can do, is to remove the impediments to future progress as quickly as possible.  If we could remove all the public perception, legislative and regulatory impediments to progress overnight, it would still take around 80 years to replace the last of the existing and committed reactors with newer technologies.

But it will take much longer to get to where we would have been now if not for the disruption. To put some figures on this, the overnight construction cost (OCC) of nuclear power reactors was decreasing at around 25% per doubling of cumulative global capacity from 1951 to 1967. At that stage the cumulative global capacity of construction starts was about 64 GW and doubling was occurring about every 3 years.  If that rate had continued, the OCC would now be around 10% of what it is. However, the situation is quite different now.  The cumulative global capacity of reactor starts is now around 500 GW.  Therefore, even if we could instantaneously increase the learning rate to 25%, we would need to double the cumulative global capacity of reactor starts to 1000 GW in order to reduce overnight capital cost by 25%, and to 2000 GW to reduce it by another 25%.

For more on this see What Could Have Been – If Nuclear Power Deployment Had Not Been Disrupted, and/or Lang (2017)

The point is that it will take a long time to substantially reduce costs.

Learning and deployment rates can increase fastest with small reactors. But it will still take probably decades to return to the learning and deployment rates that existed in the 1950s and 1960s.

Picking specific technology winners is the wrong approach. The correct approach is to remove the impediments that are blocking progress so that free and fair competition can thrive.  In this environment, innovators, entrepreneurs, investors, and vendors will compete and the market will find the best technologies.

Helmut Frik's picture
Helmut Frik on Nov 26, 2018

Price reduction die to economy of scale does not go down to zero, but just to the raw material prices (prices of pieces comming from much bigger markets or from real raw materials), on this price level the price decline ends.

Which means for example that the price for all nthe non nuclear part of a nuclear power plant will not fall unless you buy several TW of new capacity, because the same parts are used in coal pwoer stations, of which several TW have already been built. Concrete and steel will not become cheaper by building more nuclear power plants, because nuclear power plants, even if you build many of them, just represent a small share of the market.

Same limits are included in the cost projections of solar and wind power. Both use moderate amounts of cheap materials from bigger markets, (which stay constant in costs) while the expensive parts are the ones which are used for wind and solar only, which are dropping in cost due to economy of scales.

In solar power now costs for ALuminium, glass and plastics slowly becomes relevant for the module prices. But this happens at module prices which are partly below 20ct/Wp already. Below 10ct/Wp these components will start to dominate the prices (unless efficiency does not rise) which will bring cost reductions of solar modules to a stop somewhere in this region. Which will not help nuclear any more.

Peter Lang's picture
Peter Lang on Nov 26, 2018

Helmut Frik,

Thank you for your comment. However, I suspect you may not have read the paper, or perhaps misunderstood that 'learning rate' can continue indefinitely, through all generations of a technology (e.g since the first nuclear power plant started operation in 1954), and the cost never reaches zero.  The cost reduces as a percentage cost reduction per doubling of cumulative capacity ever built.

Helmut Frik's picture
Helmut Frik on Nov 27, 2018

Seems you have not understood what I wrote.

The cost does not approximate zero, it approximates the raw material costs, and can not become lower no matter how much learning happens.

If you have Iron for 1000€ in a machine, which starts at a cost  of 10.000€, and learnin curve reduces costs to build that machine by 50% every day, even in 100 years your machine will cost slightly above 1000€, and not slightly above 0€ because of the 1000€ Iron which go into it. (Assumed the learning curve of the iron production is flat. It is almost flat today due to the long time iron has been pruoduced and the hiuge amounts of iron that have been produced so far).

Same happens with nuclear power. And especially with the steam power equipment nuclear power shares with other kinds of power generation.

Peter Lang's picture
Peter Lang on Nov 29, 2018

Helmut Frik, I did read your comment. Have you read the paper and do you understand what learning rates are?

Learning rates show the percent reduction in costs of the technology per doubling of cumulative capacity built to date.  The reason the costs reduce is because the technologies improve over time.  They use less material, less O&M, and less fuel per GWh of electricity generated as designs improve.  This has been demonstrated bymost technologies from the time they were first invented until they were superseded (e.g canals replaced by rail, etc).  Examples are coal power generation, hydro electric, gas turbines, solar PV over 65 years, wind, batteries over 218 years, cars, computers, and more. Nuclear had a 25% learning rate until it was disrupted in the late 1960s.  Read the notes in Appendix B and the references cited to understand the cause of the disruption to nuclear power learning and deployment rates. If not for the disruption, the cost of nuclear power would now be around 10% of what it is.

Jesper Antonsson's picture
Jesper Antonsson on Nov 30, 2018

Peter, the learning curve maths doesn't include the idea of disruptions that we've had in nuclear. Thus we don't need to double cumulative global nuclear capacity to get a 20% (or so) learning.

If supply chains are dead, which they basically are, we get 20% learning from the first pair we build to the next. Then another 20% for the next four. This is regardless of the amount of legacy plants built with now-extinct supply chains in different regulatory environments.

What advanced nuclear reactor designs are likely to be successful is a secondary question. Our power civilisation, of many twists and turns, nooks and crannies, has run on fossil since about 1600, entailing growing technical prowess, never dreamed of, with little questioning of the suitability of this, fossil's hard limits, until recent times. As the foregoing discussion highlights, the factors to be considered for nuclear alone, are weighty - considerable. There are many technologies that relate to power generation, and appraising solely in terms of certain technical features would be inadequate. As power is the bedrock of our existence, there is no scope to head down the "wrong" road. There should be energy / power think tanks for each nation. So, the concept of electric road vehicles is appealing but the competing hydrogen fuel cell shows the real task involved. Paradoxically, in a very technically able age, there are many reasons, not just technical, but for security reasons, for example, in power sources being as widely spread as possible. And in a very polluted world to minimise waste and obsolescence. Thank you for the opportunity in writing.

Without advocating for any specific nuclear reactor design, I want to echo a sentiment that Hal Harvey of Energy Innovation said at a climate panel discussion when I asked him about how nuclear energy fits into climate policy solutions just last night: 

People have a recent love affair with Gen 4 modular nuclear reactors, and other ideas, innovations, and potential designs, and by all accounts the possiblity of those technologies are great. BUT, for now that's all they are-- ideas and designs. The build out of nuclear innovation technologies are not yet being built on the scale to truly test and eventually deploy these technologies, and unless at least 2 or 3 governments in the world truly commit to supporting them then they are likely to remain more ideas than actual solutions. Due to the physics of existing reactors dying and the difficult economics of potential future reactors, without such new designs and buy-in from governments then nuclear will be trending down and going away in the coming decades. 

Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

"Due to the physics of existing reactors dying..."

Matt, you're confusing physics, which is timeless, with public sentiment. Despite U.S. pressurized water reactor technology being the safest way to generate grid electricity - bar none - nuclear is struggling with an irrational reaction to the Fukushima accident, grounded solely in public misperceptions.

Step on an anthill - at first, the ants run around madly in random directions, trying to figure out how to react to a cataclysmic event they don't understand. Wait a few minutes, and the ants gradually settle down; in an hour or so activity around the hill is the same as it was before. In my life I've witnessed exactly the same phenomenon after Three Mile Island, Chernobyl, and Fukushima but with people, and the time scale stretched to years instead of minutes.

Gradually the world public is coming to its senses after a 1,500-year earthquake presented them with a situation they didn't understand. We desperately need to be able to appreciate the relative risks of nuclear energy and climate change - and it can't happen soon enough.

Matt Chester's picture
Matt Chester on Nov 16, 2018

Respectfully, Bob, I don't think I'm confused here. I agree that nuclear can be and already is among the safest energy sources we have-- especially when you consider the public health effects of pollution from fossil fuels.

They physics I was quoting Mr. Harvey on was the fact that there is a physical life behind each nuclear reactor. We've extended many of them beyond the initial 40 year lifetimes, which is terrific, but eventually they do reach their end of life-- that is a fact of physics. The public sentiment (and economics) that you mention then come in when it comes to retrofitting/rebuilding those reactors, building new reactors, etc. All I meant to say there is a nuclear reactor will reach its end of life, not that nuclear power as an energy source will die.

I hope that clarifies my original comment!

Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

Matt, my misunderstanding. That said, the NRC's 40-year license period dates back to the 1954 Atomic Energy Act, and was based on economic and antitrust considerations, not on limitations of nuclear technology.

Nuclear plants in the U.S. undergo a mandatory safety review every year undertaken by independent safety commissions. Because they're continuously maintained and upgraded, there is no physical reason a nuclear plant can't last indefinitely. Their biggest threat is not any inherent longevity limit but the irrational fear of anti-nuclear activists...seriously. They alone are responsible for the $billions in decommissioning expenses of all but a handful of U.S. nuclear plants.

I urge anyone with doubts to tour a nuclear plant. In the case of Diablo Canyon, they will find a meticulously maintained facility, more like a surgical suite than an industrial complex. There is a rising groundswell of opposition to its 2025 shutdown among people who work there and others, who are challenging it in the California Court of Appeals.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

Bob, we could extend the life of nuclear reactors past their 40 year design lifetimes but that comes at a cost.  At some point the metal exposed to daily radiation bombardment becomes brittle and would need to be replaced.

For a technology that is already struggling to exist in the form of paid off reactors that extra cost would be fatal.

France has discovered that several of their 30 year old reactors have serious corosine problems.  The cost of replacing the pipes was unaffordable so they wrapped the pipes with carbon fiber 'bandages'.

Fort Calhoun was offline for a year for repairs.  After returning to service it was realized that the cost of repairs made it noncompetitive.

SONGS closed due to repair costs.

Diablo Canyon reactors are closing because it would cost too much to refurbish them for operation past 2025.

As machines age they suffer more problems.  Old things break.

As for operating "forever", at some point the concrete in the reactor would become so hot, so radioactive that humans couldn't work inside the reactor to make repairs.

It's not nuclear activists who are closing nuclear plants.  It's the owners of those nuclear plants who aren't willing to keep losing money. 

It's not activists protesting.  It's accountants working the numbers.



Bob Meinetz's picture
Bob Meinetz on Nov 18, 2018

Bob, perhaps you didn't read my post. The idea nuclear reactors have "40-year design lifetimes" is a myth which anti-nuclear activists, driven by irrational fear, are only too happy to regurgitate ad infinitum to help them sleep at night.

As are the myths nuclear energy is uneconomical, that utility electricity is a competitive enterprise, that Diablo Canyon's "reactors are closing" because it would cost too much to refurbish them, that SONGS closed due to repair costs, that concrete is a component of any nuclear reactor ever made, that old machines can't be repaired/upgraded, yada yada yada...I would urge you to attempt to verify some of this mythology instead of repeating it, but in general anti-nuclear activists are loathe to take me up on the suggestion.

Confronting one's own irrational fears is, in itself, a frightening prospect - for years I refused to fly on commercial airline flights after an engine caught fire on a McDonnell-Douglas DC-10 on which I was traveling. The airplane was analogous to Chernobyl's RBMK reactor - an accident waiting to happen. And happen it did - on May 25, 1979 American Flight 191, a flight I myself had taken several times, crashed on takeoff in Chicago, killing everyone on board.

Soon thereafter DC-10s were removed from service. Commercial aviation improved, and now flying for me isn't a big deal. But Chernobyl was 32 years ago, and with climate change we're running out of time. So in the interest of promoting realistic solutions, opponents might seek psychological counseling, or help from prescription medications.

"Nuclear energy paves the only viable path forward on climate change." - Internationally-recognized climatologists James Hansen, Kerry Emanuel, Ken Caldeira, and Tom Wigley

Helmut Frik's picture
Helmut Frik on Nov 26, 2018

You did not read the post above? It were newer crowds of protesting people in front of power plants that closed them down, but accountants unwilling to bear further losses.

Beside radiation, all the interfaces and tecnologies used in a old machine are dying, so spareparts which fits into that machnine are not available any more from the market, and must be hand made in some cases, or can not be made any more in other cases, so the whole machine must be reconstructed to fit to existing new parts. This kind of costs is continuously rising the older that machine gets.

The reliability of nuclear power e.g. in swizerland, belgium, france is falling. They stop working for increasing numbers of causes. It's the classical old bathtub curve. The old nuclar pwoer plants are not any more at the flat bottom of the bathtub, nut at the rising end, heading for the end of their ecoomical lifetime.

Second problem nuclear is facing is that it shares the supply chain for the non nuclear part with coal power. Coal power starts dying, which causes these parts, even if they are still o the market, to rise in cost due to the inverse economy of scale. Even if they are still available.

Matt Chester's picture
Matt Chester on Nov 16, 2018

Respectfully, Bob, I don't think I'm confused here. I agree that nuclear can be and already is among the safest energy sources we have-- especially when you consider the public health effects of pollution from fossil fuels.

They physics I was quoting Mr. Harvey on was the fact that there is a physical life behind each nuclear reactor. We've extended many of them beyond the initial 40 year lifetimes, which is terrific, but eventually they do reach their end of life-- that is a fact of physics. The public sentiment (and economics) that you mention then come in when it comes to retrofitting/rebuilding those reactors, building new reactors, etc. All I meant to say there is a nuclear reactor will reach its end of life, not that nuclear power as an energy source will die.

I hope that clarifies my original comment!

Seriously, there doesn't seem to be any realistic contenders.   The industry has been struggling for decades to come up with anything better than the conventional reactors we see around the world.  The catastrophic failures of the two U.S. projects (Summer is abandoned, Vogtle continues but can never recover its cost) highlight the problems that cost imposes.

This was during an era when average wholesale prices were 5 - 6 cents per KWh and expected to rise.  But today wind and solar are coming in under 2 cents in some places, and firmly below the average wholesale cost of fully amortized coal and natural gas plants.  Utilities need much less in the way of compensating resources than is generally believed, since a thousand wind turbines function a lot more like a coal or nuclear plant than a single turbine.  And the economics dictate a lower premium value for baseload, and a higher one for dispatchability, which favors natural gas in the short term, and dispatchable load and storage in the slightly longer term.

The real dynamic is due to the extreme time needed for a nuclear construction project.  A new nuclear plant initiated today can be displaced with cheaper wind and solar before it is complete.  Investors can't tolerate that risk. 

Of course we should never say never, but the question is which technology rises to the surface, and the answer is "not nuclear".  If you hypothesize a world where new nuclear plants are competitive and reasonably priced, you must also hypothesize a world where a whole lot of other things we don't think about much come in to play.  The most likely is using solar and wind-powered electricity to produce liquid hydrocarbons to use in the existing combined cycle plants.  But there are dozens of other concepts which address the intermittence of wind and solar and are already cheaper than new nuclear plants. 

Michael Keller's picture
Michael Keller on Nov 17, 2018

We do have a lot of natural gas and gas turbines are really efficient, but they use prodigious amounts of fuel. Renewable energy takes up vast amounts of land. Doubt using renewable energy to create gas turbine fuel would come close to being financially viable because a lot of energy is required to make hydrogen or similar fuel.

However, you could use a nuclear gas  reactor to drive a gas turbine’s air compressor. Drastically reduces the amount of fossil fuel needed and about doubles the gas turbines electrical output. Should come dam close to beating the production cost of the gas turbine as a result to stunning economies of scale coupled with the low cost of nuclear fuel relative to natural gas. Build cost appears to be around 80% less than conventional nuclear plant.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

 "Renewable energy takes up vast amounts of land. "


(screwed up the quote function)

That is simply incorrect.

Wind farms can cover a large area but the percentage of land used for tower foundations, access roads, and ancillary buildings is less than 2% of the area.  The remaining area is available for the original use be that grazing, agriculture or wildlife.

Here's a map of the amount of land that would be used if PV solar provided all of the world's energy.  Obviously we would get a maximum of around 40% from solar with the rest being supplied by wind, hydro and other renewables.  So cut the small rectangles by 60%.

And the rectangles used fixed mount solar.  Single axis tracking adds roughly 50% output so make the tiny yellow rectanges smaller yet.

Then we're moving to bifacial panels which should give an extra 10% to 20% output.  Trim some more off the rectangles.

Finally, realize that a large portion of solar will be installed on buildings and over parking lots so those panels will take zero land out of use.

Solar will use so little land that it is not worth considering.  


The challenge has always been demonstrated economics. This applied to previous generations and well as any new ones.  Thus far, or at least in the last 40 years, those economics have not been demonstrated.  In every instance, the technology was deployed as a traditional commodity electricity generator product (a new way to boil water).  Current buyers in that market are going to be extremely risk averse to ordering and installing a unit. So what any developer needs to do is rethink what markets in which it might initially participate, markets that want some of the unique aspects of this technology that go beyond a grid kWh, particularly one with a higher value proposition - they need to sell the product of the product, so to speak. An example- powering remote desalinization plants. Once up and running, and assuming the economics are then demonstrable, there might be a chance in the traditional market.

Michael Keller's picture
Michael Keller on Nov 16, 2018

I think an alternative approach is to provide more than a single revenue stream by deploying a highly versatile technology. That is highly unlikely for traditional nuclear power plants that realistically can only generate electricity due to their relatively low operating temperatures. To me that points to some form of gas reactor or liquid salt/fuel type, although the latter two may have a really rough (expensive) sled ride through the regulatory process.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

An 80% installed cost reduction.

Sometimes the mountain is simply too high to climb.


Economics will be the driver with the ability to compete in the market the biggest issue. Generally, smaller plants are more costly on a unit production basis than larger plants, which is exactly what occurs with gas turbine (combined-cycle). That means small reactors will be more costly to operate than their larger reactors.

In order to compete, advanced water reactors need to be much more efficient than the current fleet. Small water reactors are significantly less efficient than their larger cousins. Water reactors are a mature technology in the twilight of their lifecycle. Advanced reactors that operate at much higher temperatures than water reactors will have a significant edge.

Further, the build cost ($/KW) needs to be significantly less than the roughly $5500/KW level of conventional water reactors and more in line with the roughly $1000/KW of the combined-cycle plant.

Advanced reactors will need to be absolutely passively fail safe to protect both the public and the investment. The ability to earn a good profit is paramount to overcome the perceived risk associated with owning a nuclear reactor. Small reactors do not make a very big profit.

Fast reactors (makes more fuel) are not economically viable, as the costs to processs and re-use fuel vastly exceeds once-thru fuel cycles.

The current regulatory process places a massive financial burden on the development of advanced reactors. The current “risk-based” attempts to reform the license process makes matters worse. Attempting to tie regulatory requirements to uncertainties creates massive uncertainties into what is or is not acceptable.From a financial standpoint, that creates significant uncertainty as to cost and does not inspire investor confidence.

Also, jumping on the “global-warming” bandwagon is unwise because competitiveness is the real issue.

Bottom line, currently way to many obstacles to successful commercial deployment of advanced reactors in the US.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

"build cost ($/KW) needs to be significantly less than the roughly $5500/KW level of conventional water reactors and more in line with the roughly $1000/KW of the combined-cycle plant"

What you are saying is that there somehow would have to be a ~80% decrease in the build cost of reactors in order for them to be viable in the market.

That's such an immense cost decrease that it says nuclear, as we know it, has no future.


Michael Keller's picture
Michael Keller on Nov 16, 2018

Yep, that is pretty much what I have concluded based on in-depth financial analysis.

Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

"Economics will be the driver with the ability to compete in the market the biggest issue."

Michael, let's end the mythology there is some kind of competitive market in utility electricity right now.

Utility electricity is a natural monopoly - there is no competition. Utilities are compensated on a cost-plus basis, so no matter what kind of generation they choose - nuclear, renewables, hamster wheels - if generation is approved by their state public utility commission, utilities have a guarantee it will be profitable.

Now, some activists maintain renewables are more "competitive" than nuclear, without being able to provide a single instance of a U.S. nuclear plant being replaced by gas + renewables which has resulted in a price discount for consumers. De facto proof: gas + renewables is more expensive than nuclear for customers, the people who have entrusted elected officials to serve their interests.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

"Michael, let's end the mythology there is some kind of competitive market in utility electricity right now."

But there is.  Look at the coal and nuclear plants which have closed because they could not compete with cheaper NG and renewable energy.

Look at the paid off US nuclear plants that are asking for subsidies in order to stay in operation.

Look at what is being installed.  Other than Vogtle nuclear is dead.  Coal is dead.  Wind and solar are outpacing CCNG.



Michael Keller's picture
Michael Keller on Nov 16, 2018

Bob has a point. Most regulators will not go along with massive increases in the cost of power, with the exception of Georgia and the Vogtle financial catastrophe.

As for the “uncomptiveness” of older single unit plants, there is something out of whack financially. Those unit, while somewhat expensive for their time should have a major advantage because their debt is paid off and the return-on-equity (profit) should not be that great (by today’s standard, the investment was relatively palatable). I strongly suspect that the assets were sold off at an excessive price by the utility to their own sub- entity. The original utility made a huge profit, while the new entity took on a massive debt, making the nuclear unit uncompetitive. 

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

I don't think you're right about the jacked up sales price, Michael.  My impression is that most reactors are owned by the companies and utilities that built them.

I think there may have been some sales as companies attempt to remove the liability for decommissioning.  But there's no reason a company would have overpaid for an asset with little residual life.


Michael Keller's picture
Michael Keller on Nov 17, 2018

As near as I can tell, the paid for or nearly paid for  assets were spun off and bought by an entity that borrowed money, anticipating they could make a profit. They paid too much and borrowed too much. The original debt-free asset would likely be competitive.

That, in a nut-shell, is what I believe is one cause of the demise of Some of the smaller units. However, it is very difficult to find out the actual financial position of the plants. I am pretty sure there were financial shenanigans going on, including politicians demanding generating assets be spun off, with utilities happy to go along with the scheme because they end up with a big chunk of change and no risk. 

Helmut Frik's picture
Helmut Frik on Nov 26, 2018

This does not explain that the systems go offline. If they go offline noone can pay the debt on it, so you can forget about any credit, they are all sunk costs of the question of closing the plant is in the room.

The only cause to close the plant is, if producing a kWh with it costs more than the price for which it can be sold. this is the only relevant question.

Everything else can force the owning company in bacrupt and similar things, selling the plant then for a apple and a egg to a new owner, but as long as it can earn more per kWh than it costs to produce one, the power plant will stay open.

Jesper Antonsson's picture
Jesper Antonsson on Nov 30, 2018

The example of renewables and the financial difficulties of large projects in general (be it bridges or powerplants) has proved that the nuclear industry went wrong with huge reactors. It was understandable, but wrong.

Nuclear needs industrial learning and it needs certainty, so that interest required goes down. As you say, it needs to be passively safe, but not to protect anyone, but to allow the NRC to back down a bit and allow less staffing and less redundancies.

Here Nuscale has a good offering overall. They'll mass produce 70 MW modules and operators can be sure of what they'll pay and they'll enjoy cheaper, simplified O&M. I think this is the last chance for the US, actually. There's a number of interesting advanced R&D projects going on in the US, but China will beat them to it if Nuscale doesn't make it.

My thinking is that nuclear reactor technology has to follow the same sustainability criteria as other systems to be viable. The biggest issues with power production systems is scale. Big systems require huge resources and follow a "big is better" dictum. For sustainable systems, this anthema as most sustainable systems adhere to a distribute resource mentality. For example, look at how China is addressing its energy needs,  https://The Global Ripple Effects of China's Renewable Energy Efforts . Distributed resources, particularly for energy, is a must.

Making nuclear energy distributed has been a challenge, but there are encouraging signs of progress. A recent piece, https://Big Gains for Tiny Nuclear Reactors , by Sonal Patel in Power Magazine, outlines the technical challenges for shrinking reactor sizing and projections for commercialization schedules.

One thing for sure in the sustainability space, never say "Never!" In a book I wrote a few years book, I dismissed the idea of nuclear energy having any role as part a future renewable economy. I happily note that I was premature in my assessment. In plain English, I was wrong and, in this case, I am glad I was.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

If we look at the nuclear plants that are failing for economic reasons in the US what we see is that they are mostly single reactor operations.  Often smaller reactors.

The nuclear plants which are most competitive typically have one or more larger reactors which allows fixed opex to be spread over more MWh.

Spreading small reactors around the grid goes in the opposite direction.  The necessary technical, security and administration costs would be charged to a much smaller output.

As for China, I don't think China has ever seen nucear as a major source of electricity.  And what has happened over the last few years is that wind generation has caught up and passed nuclear.  Solar is rapidly gaining on nuclear and likely to generate more electricity per year by 2020. 

Furthermore, China has failed to start construction on any new commercial plants for about two years.  It's not clear whether China is waiting for a new reactor design to be completed or rethinking their investment in nuclear energy.




Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

Bob, China has 13 new reactors under construction, for a total capacity of 14 GW.

Just last week, Sanmen Nuclear Power Station went online with its second 4th-Gen Westinghouse AP 1000. Units 1 and 2 are now generating 2.3 GW of clean electricity day or night, windy or calm.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

Constrtuction on those plants started some years back.

As far as I can tell China has not started construction on any commercial nuclear reactors in the last two years.  Three were suppose to start constuction in 2018 but the year is mostly gone with no news of construction underway. 

A year after the Fukushima disaster China put a halt to new nuclear construction starts.  Then a number of new reactors started construction.  But starting in 2016 there was a large slowdown in new construction starts.  IIRC construction starts were expected on three reactors this year with no news of any starts to date.


Best to not mislead yourself by thinking that nuclear reactors are highly dependable.  They are not.  Unannounced shutdowns occur much more often that most people realize.



Bob Wallace's picture
Bob Wallace on Nov 16, 2018

According to Wiki China began construction on only two reactors in 2016 and one in 2017.  The 2017 start was a largely experimental, smaller sodium cooled fast reactor.

Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

Bob, nonsense. U.S. nuclear reactors are more dependable than ever, with capacity figures exceeding 90%.

Solar (avg. ~20%) and wind (avg. ~30%) don't even come close.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

Utility solar is 30%.  Wind farms coming online since 2013 have average CFs in the mid 40% range.


The DOE has identified sites around the country where with the use of 140 meter hub heights we can expect 60% to 70% CFs  for onshore wind.

Since the wind tends to blow more when the Sun is not shining CFs of 30% and 60% mean many hours when we can generate the electricity with only wind and solar direct.  Storage can fill in the remaining hours.

Overbuilding wind and solar along with the use of dispatchable loads will make it even easier to achieve a 100% renewable energy grid.



Michael Keller's picture
Michael Keller on Nov 16, 2018

I generally do not believe anything the DOE puts out. Too many instances of “thumb-on-scale” associated with the prevailing politically driven agenda. After all, the primary objective of the DOE is to line their own pockets and those of their academic buddies. Commercial success of a technology is not really a consideration because the DOE has no “skin-in-the-game” relative to making a profit.

Harsh assessment to be sure. However, the DOE has spent billions on advanced nuclear with no commercial successes. Need a different model to support development by the private sector of cost effective and strategically valuable energy production facilities. Bureaucrats are dismal predictors of what is economically viable.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

I have problems with the DOE's, actually the EIA's predictions.  They generally fail to include changes that are clearly underway.

I've never seen a reputuable person in the energy field throw doubt on the EIA's data.  

The wind charts I mentioned are very highly likely to be very reliable.  The scientists who collect data take their jobs seriously.

And that data is supported by findings in other countries.  We've been using too short towers.  The best wind is not a 50 meters where we set our first generation of turbines nor at the more common 80 meter height.  But, depending on the site, we find extremely good winds in the range of 100 meters to 140 meters.  

Moving to higher hub heights as other countries are already doing can create a 50% increase in harvested energy with no great deal more investimet.



Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018


Bob Wallace's picture
Bob Wallace on Nov 16, 2018

Here's a time sample of nuclear outages..

September 2017

V.C. Summer down due to problem with main transformer. (1)

Turkey Point and St. Lucie reactors were shut down due to approaching Hurricane Irma. (2,3)

Florida’s St. Lucie nuclear plant was forced into a partial shutdown to clean salt buildup from its transmission yard after Hurricane Irma in September. (4)

Cook Nuclear Plant Unit 1 reactor was shut down after a valve that's part of its reactor coolant supply system opened. (5)

Columbia down due to vacuum leak (6)

November 2017

Beaver Valley offline due to electrical problem (7)

Indian Point down for generator repairs. (8)

Grand Gulf was shut down at the same time due to fluctuating power output and an apparent lack of knowledge of staff about how to deal with the problem (9)

December 2017

North Anna Unit 2 shut down due to water leak in cooling system. (10)

Clinton down due to a transformer problem (11)

Unit 2 at the Indian Point nuclear power plant was offline for two weeks due to leakage from a reactor coolant pump (12)

Columbia shut down due to transmission problem.  Didn’t turn up in the news until March 2018. (13)

January 2018

Pilgrim Nuclear Power Station shutdown after one of the two main 345-volt lines that provide off site power to the plant “became unavailable,” (14)

February 2018

Unit 3 at Indian Point shut down when a generator failed. (15)

March 2018

Browns Ferry Unit 1 taken offline due to problems with turbine control system. (16)

Pilgrim Nuclear Power Station second time in 2018 due to leak and transformer problems.  (17)

April 2018

Fermi 2 down due to a plant transformer malfunction. (18)

Pilgrim Nuclear Power Station down for third time in 2018.  Valve problem where water flows into reactor. Pilgrim has been offline 53 of the last 120 days. (19)

May 2018

Columbia down for unknown problem (20)


Over a 7-8 month span 19 of 98 or 19% of all US reactors were offline for reasons other than scheduled refueling and maintenance.  A greater than 30% failure rate on an annual basis.

The number of shutdowns may have been higher.  Sometimes it’s been months after the shutdown that I find the news.  There seems to be no site that reliably logs nuclear outages.


John Oneill's picture
John Oneill on Nov 19, 2018

'Over a 7-8 month span 19 of 98 or 19% of US reactors were offline for reasons other than scheduled refueling and maintenance. A greater than 30% failure rate on an annual basis.' In any one month, no more than 6 were offline, and the average was about two. Considering that most of the closures were less than a full month, that's more like a one percent failure rate. Many of the failures were also not connected to the nuclear part of the plant, but generator or line problems that could happen to any power source - wind overall has about a 4% non-availability rate, which is lost in the detail of its ~60% 'shortage of fuel' failure rate. Can you provide a breakdown of which countries are building wind without renewables mandates ? I know that New Zealand is much better suited for wind than most places - we span the Roaring Forties, and have copious hydro to fill in the doldrum periods. Yet the percentage of wind in our generation is only about 5%, and most of that was built back when the government owned most of the gentail companies, and mandated no new fossils. Since then some of our big four companies have been building new wind farms - but only in Australia, where there are renewables mandates.

Dan, I attended NGO Day at NuScale Power in October, a promotional event to which the company invites reps of non-governmental organizations. Even though the cores in NuScale reactors use a more-or-less traditional design, there are significant differences. Among of the most important: they have a significantly lower operating temperature (~500ºF) and pressure (1650 psi), allowing engineers to create a passively-safe design without reliance on any external power or water pumps. Each modular unit is capable of delivering 70MW, and up to 12 can be linked together at the same facility.

They are aggressively marketing NuScale Power Modules (NPMs) for use with local micro- or mini-grids which currently are reliant on diesel generation, and the delivered purchase price will be somewhere in the neighborhood of $800 million. Utah Associated Municipal Power Systems (UAMPS), their first customer, is expected to have a NuScale facility online by 2025.

NuScale views their traditional PWR approach not as a drawback but an asset. Because the technology of PWRs has the experience of millions of operating hours to draw from, their behavior is well understood. Probabilistic risk assessments estimate the chance of a core failure at 1.3/10^9yr - one every 1.3 billion years. The company expects their passively-safe design and relatively low cost to be huge selling points, and they make a pretty good case for it.

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

On NuScale's website they claim a cost of $0.08/kWh once they have built enough reactors to benefit from the learning curve and work out inefficiencies.  That is not an acceptable price.

With wind and solar closing on $0.02/kWh (unsubsidized), the cost of storage dropping, and low cost NG available for fill-in NuScale won't find a market.  Especially for the first several reactors they would need to build in order to reach their target price.



Bob Meinetz's picture
Bob Meinetz on Nov 16, 2018

Bob, wind and solar have never cost "$0.02/kWh" and never will, because they require backup gas generation to be viable on an electrical grid. Add requisite gas generation into the mix, and EIA finds the combination of the two is 29% more expensive than existing nuclear. Though you may find $0.08/kWh unacceptable, utilities disagree:

"UAMPS, a public power agency that provides electricity at wholesale to more than 40 community-owned electric utilities in the Intermountain West, is working with NuScale Power on a project that involves the installation of 12 small modular reactors in Idaho. 

'The business model is perfect for public power,' Doug Hunter, CEO of UAMPS, says. Each of the 12 modules will be 60 MW, which Hunter calls 'a great size for a public power utility.' At that size, he says, member utilities can buy output by the module. Hunter sometimes refers to the project as a '12 car garage you can park your nuclear module in.'"

UAMPS Sees Cost and Safety Benefits With NuScale SMR Technology

Bob Wallace's picture
Bob Wallace on Nov 16, 2018

First, I said "closing on $0.02/kWh"  Lazard's 2018 LCOE sets the last year minimum for onshore wind at $0.029 and PV solar at $0.037.  Recently a contract was signed for unsubsidized PV solar for $0.024.  

Second, let's be clear.  There is no way to build "existing nuclear".  We would have to build new nuclear which would mean about 30 years of very high electricity prices before we reached "existing" prices.

Current nuclear electricity prices are running weill above $0.10/kWh (Vogtle and Hinkley Point).

Furthermore, a combination of wind, solar and CCNG is cheaper than more than half of our existing nuclear.  Were that not the case then we wouldn't see roughly 60% of nuclear plants losing money.  The market would pay them what they need to stay in operation.

If UAMPS wants to give it a go with NuScale that's up to them.  Utilities in South Carolina and Georgia gave it a go with the AP1000 which was suppose to be the answer.  Just because someone thinks something will work does not make it so.

Before a new reactor can be built we should expect unsubsidized wind and solar to be $0.02/kWh, possibly lower.  But let's put a fat thumb on the scale in favor of nuclear and run some numbers.

Wind 3 cents

Solar 3 cents

CCNG 6 cents (Lazard average)

40% Wind + 40% Solar + 20% NG = 4 Cent Electricity

Feel free to use any distributions of wind, solar and NG you like.  100% NG would be 6 cents which is lower than NuScale's "hoped for " 8 cents.





Bob Meinetz's picture
Bob Meinetz on Nov 19, 2018

Bob, Diablo Canyon Nuclear Power Plant is generating 2.1 billion watts of clean electricity, at this very moment, at a cost to Pacific Gas & Electric of 2.7¢/kWh - according to the utility itself. Why would anyone care what numbers Lazard, or Deutsche Bank, or any other investment bank uses to promote its $billions in renewables investments?

Georgia Power is betting $25 billion on Vogle 3 & 4, and the project is progressing by leaps and bounds - much to the chagrin of displaced renewables entrepreneurs.

We both know "40% solar" is a fantasy, that it's never been possible anywhere in the world, and never will be - there's too much nighttime and weather getting in the way. But am I seeing  "100% NG" is now part of the conversation? It's almost as though renewables were only invited to the party to greenwash the replacement of nuclear with natural gas (because both are dispatchable, at least it would make sense).

Bob Meinetz's picture
Bob Meinetz on Nov 18, 2018


Michael Keller's picture
Michael Keller on Nov 16, 2018

The financials do not support $.02/kWh for renewables. What is likely going on is taking credit for a $25/MWh production tax credit, tax credits associated with building the facilities, financial inducements of various sortts and forced use of renewable energy. True cost is somewhere north of roughly $125/mWh. New nuclear is above that, but how much higher is really hard to guess.

The NUSCALE claims are also quite suspect. True cost is somewhat similar to renewables. However, like renewables, appears to be major reliance on tax payers and consumers being forced to provide huge subsidies.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

Sorry Michael but you are simply wrong.

Recent onshore wind PPAs are running under  $0.02/kWh with subsidies.  Tease out the subsidy and US onshore wind is now under $0.03/kWh.  As we move to higher hub heights with higher CFs the cost will drop furture. 

Solar is already under $0.04/kWh unsubsidized in the US with every sign of dropping further.  Jinko solar just bid a solar farm in the UAE at $0.024/kWh.  That's under three cents, unsubsidized.

".True cost is somewhere north of roughly $125/mWh"

Is simply a gargage statement.  The cost of wind and solar are what they are.  There will be some other costs involved in creating a 24/365 renewable grid just as there would be other costs in creating a 24/365 grid from nuclear or any other energy source.  But to say that those costs would be high enough to bring the cost to $0.125 is simply silly.

NuScale claims $0.08.  After they build several more expensive reactors.  Reaching economies of scale is not achieved with manufacturing one or two units.  Carlos Ghosn of Nissan has stated that EVs will need to reach a scale of 500,000 per year before they become similar prices to ICEVs.  There are many costs to work through.

It probably wouldn't take hundreds of thousands of SMRs to hit scale but it could take dozens.  And there would need to be someone willing to pay a very high price for electricity simply to move NuScale into production.  It makes no sense to subsidize very expensive nuclear just to end up with nuclear still too expensive to compete.  

Kewaunee closed with years left on its license and no repair needs.  The electricity that reactor produced needed a $0.05 price to stay in business and the market would not pay it.  That's amost half of NuScale's target price.

Plus, let's be honest.  Nuclear has a strong track record of promising affordable and delivering unaffordable.  We have to leave room to assume NuScale will never be able to hit their price point.



Michael Keller's picture
Michael Keller on Nov 17, 2018

Bob, at installed costs of around $1500 to $2000/KW claimed by DOE and realistic capacities of around 30%, the production numbers (including debt repayment and profit) are around $125/MWh. This is a straightforward calculation. The only way to hit $20/MWh is to drastically reduced the build cost and drastically increase the capacity OR have somebody else actually picking up the cost (as in the hapless taxpayer and consumer). By the way, these are not Level Cost of Energy Numbers, which are lower. Also, you have to look at average conditions, not the best locations because to produce a lot of power requires a lot of land. The best spots are relatively limited in terms of unfettered access.

As for NUSCLE, you have to ask yourself why a utility would build the plants when significantly cheaper and less risky options (combined-cycle) are available. Makes little financial sense and most regulated utilities have a lawful obligation to provide reasonably priced power. Public power companies have similar fiduciary obligations but have a little more leeway but the managers can be voted out of office by the public.

Brian Welch's picture
Brian Welch on Nov 17, 2018


I think the item there missing is that 30% capacity number. I don’t know what itis for wind, but for solar it’s right around that. If you’re saying solar is 2 cents then you need to get 70% more paneling bringing the cost up to 7 cents. Plus land. Plus batteries. That’s what they miss. I would say throw up a bunch of nukes, but all that won’t be necessary next year. 


John Oneill's picture
John Oneill on Nov 20, 2018

No solar array on this planet gets a 30% capacity factor, and in places with a winter, it's seasonally far less.

Brian Welch's picture
Brian Welch on Nov 17, 2018


I think the item there missing is that 30% capacity number. I don’t know what itis for wind, but for solar it’s right around that. If you’re saying solar is 2 cents then you need to get 70% more paneling bringing the cost up to 7 cents. Plus land. Plus batteries. That’s what they miss. I would say throw up a bunch of nukes, but all that won’t be necessary next year. 


Helmut Frik's picture
Helmut Frik on Nov 26, 2018

One axis solar power in the sun belt of the earth gets above 30% capacity factor in deserts. Costs for ground is around zero in deserts.

System costs for solar power in germayn on large roofs or on ground in utility scale is at 600-700€/kWp all included. US has extremely high system prices which are not payed elswhere in the world beside maybe japan.

Bob Meinetz's picture
Bob Meinetz on Nov 19, 2018

@ Michael -

"As for NUSCLE, you have to ask yourself why a utility would build the plants when significantly cheaper and less risky options (combined-cycle) are available."

Short answer: 29 states have adopted Zero-Carbon Portfolio Standards, deciding regulated utilities also have an obligation to provide carbon-free power.

An 840MW combined-cycle plant generates ~5.6 megatonnes of CO2/yr; NuScale modules generate none.

Michael Keller's picture
Michael Keller on Nov 17, 2018

As for your PPA’s, you need to be able to read what is in them and the utilities will not allow that. Also need to see the financials for the renewable assets. Who owns the asset, including tax credits. Is there some form of consideration associated with helping a utility meet a mandated use of renewable energy? Does a utility, or sub-entity actually own part of the renewable asset and is using that for some type of tax write-off. Does the utility pass thru and make a profit on costs associated with meeting renewable mandates? This stuff gets really complicated to figure out, particularly when everything is “confidential”.

What I am driving at is there is likely more behind the curtain than anyone will own up to. Relying on a number without knowing what is in the contracts can lead to unwarranted conclusions.

if a relatively straightforward financial analysis shows a massive disparity in numbers, then it is likely there are other considerations in play that are not being owned up to by the involved parties.

Michael Keller's picture
Michael Keller on Nov 17, 2018

Bob Wallace,

Did a little more thinking on PPA and odd wind energy price.

Normally, a Power Purchase Agreement (PPA) is for so much energy to be delivered at a time certain at a set price to meet the utility’s grid energy demand. However, in the case of wind, the driver is to meet a politically created requirement (of suspect need) to use a prescribed type of generated energy. Actual load demand is irrelevant.

In a normal market, the load demand dictates price. Provide energy when it is not needed and you pay to put the energy into the grid because others are damaged by your acrions. In the Midwest, wind is generally more prevalent in the morning hours when demand is low. Not that much wind in the day in the summer when demand is high.

However, if the utility simply accepts the normally unneeded wind energy whenever it shows up and gets compensated for their net higher costs, then I think therein lies an explanation for the odd wind number. Only losers in this are the consumer, the tax payer and any “merchant” type power plant, which is exactly what a number of nuclear plants actually are. Viola, that is what is occurring.

Logically, if reducing CO2 is the goal, then the nuclear plants should be subsidized like renewable energy. Please note, I object to all subsidies/mandates. Let market forces determine the most cost effective source of power.

Actual pollution issues are the province of the Congress and implementing regulations. The state power regulators should limit themselves to insuring cost effective, reliable electrical energy is available to the residents and businesses of the state. Can renewable energy be a part of the mix? Sure, if helps provide cost effective and reliable energy to the consumer.

Bob Wallace's picture
Bob Wallace on Nov 17, 2018

Michael, a PPA is a power purchase agreement.  It includes all the costs of producing electricity along with whatever profit the owners take.  The cost may be lowered by subsidies.  Not all PPAs are made public but those written with public utilities are public information.

You can tapdance all you like trying to argue that wind and solar cost several multiples of what they do but that's just bunk.



Bob Wallace's picture
Bob Wallace on Nov 17, 2018

This is intended to follow the wind PPA chart I posted.  The software for this site is pretty clunky so this may appear out of sequence.

The costs there are the contracted delivery price for the wind PPAs that have been made public.  They include federal subsidies which can be teased out by adding in 1.15 cent for pre 2018 prices.  The federal subsidy was 2.3 cents per kWh produced during the first ten years.  Since PPAs are generally 20 to 25 years a bit over a penny is a reasonable estimate.  It's also roughly correct for a owner who takes the 30% ITC rather than the PTC.

The price also includes profits for owners.  I suspect we should assume something in the 15% to 20% range.  Start with 1.8 cents, add 1.15 cents to make the unsubsidized price 2.95 cents.  Then reduce that by the owner profite and we end up at roughly 2.5 cents as the real cost of production.

Wind and solar do not contract to deliver at specific times (except for, perhaps, delivering a firm 15 minute block).  They contract to delived when when the wind blows or the Sun shines.  Utilities then use other sources such as CCNG or storage to supply the other times.


Bob Wallace's picture
Bob Wallace on Nov 17, 2018

Let's talk about wind and solar paying to put electricity on grids.  That is not what is happening.  Nuclear and coal sometimes pay in order to avoid cycling down and back up.  Nuclear and coal simply cannot cycle quickly.

This is what is killing nuclear and coal.  Since there are contracted supplies of wind and solar thermal plants have to bid in lower than the contracted price in order to force renewables to curtail.  

Over half of US nuclear plants need to average around 4 cents per kWh - around the clock.  If they have significant periods where they sell for less then they have to earn more during other hours or they lose money.

Because CCNG can sell at a price just above wind and solar gas will bid in below the price nuclear and coal need.

We've reached a point at which nuclear and coal are no longer economically viable.  And it's only going to get worse.  Storage prices are rapidly falling which means that we can "overbuild" wind and solar so that wind and solar supply more of the hours.  Then the generation in excess of current demand will get stored away and sell in at a price below thermal plants.

It's a new energy world.  The old is being replaced just as computers replaced ledger books, adding machines, and typewriters.




Bob Meinetz's picture
Bob Meinetz on Nov 19, 2018

@ Bob Wallace -

"Let's talk about wind and solar paying to put electricity on grids. That is not what is happening."

Bob, that is what is happening - California ratepayers are paying $1 billion / year in negative pricing to direct afternoon solar overgeneration (the back of the Duck Curve) to Arizona, Nevada, and Oregon (video).

Afternoon solar energy in California has moved beyond worthlessness, and become a liability.

Helmut Frik's picture
Helmut Frik on Nov 26, 2018

Ni, it i not solar which forces the prices negative, it is mainly gas and other fossil generation which does not reduce output when nbody needs their power - either by aoutdated power plant design unable to ramp, or by stupid contract construction which forces the power plant to run and someone else to buy that power even wehn it is not needed. There was no minute so far when wind and solar generation exceeded power consumption in California according to the duck curve you reference.

Beside that, it is only a question to strengthen and expand grids to find buyers who like to buy californain solar power and do somethig useful with it when california builds so much solar power that the export it regulary after suppluying their own demand. The world is changing, and especially the electricity world is changing.

Ned Ford's picture
Ned Ford on Nov 26, 2018

The "negative" pricing you speak about is not discussed in the link you provided.  It does exist, but only for a few hours per year, and it represents profoundly lower actual costs per KWh than would otherwise exist.

I took MISO hourly wind data and hourly total consumption data for 2017, and increased the wind generation to the point that there started to be actual excess generation by wind above total consumption.  MISO doesn't have the transmission network to handle such large quantities of wind generation yet, but about six times current wind generation, they get 50% of total electricity around the year from wind, with only 0.7% of that electricity above consumption.  At current prices, with or without subsidy (which will be gone at the end of 2019) wind developers could easily afford to curtail generation for 0.7% of total MISO generation.  They won't have to because customers will find uses for power by accepting lower rates for those few hours per year.  Storage is one option, but it is not the only option.

As wind and solar costs drop, the places where they are economic increase enormously.  It now makes sense to build solar anywhere in the U.S.  It is not only cheaper than fossil or nuclear generation, it is also cheaper than wind and solar in better places, plus the transmission to get the power to other places.

We already have enough natural gas generation to fill in all the spaces between a large wind and solar resource.  It will still be the most expensive part of our generation, and I'm looking at people developing renewable liquid and gas fuels, by synthesizing hydrogen and taking carbon from the atmosphere or other sources.  We just saw the first transatlantic commercial jet fueled this way, although I don't know if it was economic.  I think it will be in the future because electricity from renewables is now cheap enough to permit that. 

Bob Meinetz's picture
Bob Meinetz on Dec 1, 2018


"The "negative" pricing you speak about is not discussed in the link you provided."

In the video on the page I linked to, CA Assemblymember Brian Dahle: "We're producing too much energy during the day, and not enough at night when the sun isn't shining and the wind's not blowing...the ratepayers in California are paying a billion dollars a year, right now, to offset the cost going to other states..."

"I'm looking at people developing renewable liquid and gas fuels, by synthesizing hydrogen and taking carbon from the atmosphere or other sources..."

I've been hearing about people looking at people developing renewable liquid and gas fuels by synthesizing hydrogen for at least two decades. Know of anyone who's doing it?

"It now makes sense to build solar anywhere in the U.S."

Yet somehow, only one-third of one percent of U.S. households have PV panels on the roof. Evidently, the other 99.66% of homeowners disagree with you.

We just saw the first transatlantic commercial jet fueled this way [burning renewable liquid and gas fuels, by synthesizing hydrogen and taking carbon from the atmosphere or other sources...]"

Got a link? If you're referring to Branson's Virgin jet, you're way off the mark - hydrogen was the product of steam reforming methane, and 50% of the fuel mix was good ol' fossil fuel kerosene.

Willem Jan Oosterkamp's picture
Willem Jan Oosterkamp on Nov 19, 2018

All nuclear reactors are inherently unsafe.

risk of proliferation

risk of loss of coolant

reactivity transients

Risk acceptance differs from person to person and from goverment to goverment. Acceptance in probalistic risk assessment differs also. e.g. the fact that during the Fukushima accident safety equipment were shut off by personel according to procedures is as far I know not covered in probalistic risk asessment.


Non proliferation consideration exclude al reactors that have enrichment over 20 %, all reactors that use graphite and heavy water and all reprocessing. The thorium cycles produce plutonium and U-233, both of which can be used for nuclear bombs.

I think only light water reactors in once through cycles have a low proliferation risk.

Existing light water reactors suffer from the chemical reaction of zircaloy with steam, that produces heat and hydrogen. aound 50 % of the energy released in the Three Mile Island and Fukushima accidents came from the chemical reaction. The last thirty years have been lost as only recently molubdinum cladding for fuel elements are being developped, whith a high resistance agains corrosion and formation of hydrogen. Of course the steel vessels can react with steam giving of large amounts of hydrogen.

Pressurised water reactors should eliminate the use of borated water for reactivity control as deboration transient (where water of the steam generators enters the rreactor) must be avoided at all costs. Boiling water reactors can depressurise when control rods fail to insert, reducing the power level significantly.

Excisting reators have been designed with the defence in depth philosophy. Fission products are contained in a uranium oxide matrix, contained in fuel cladding, contained in a reactor vessel and the vesel is contained in a containment building. The risk can be further reduced by placing the reactor building underground and in areas with a low population density as it was done in the early days (Hanford and Idaho). 

John Oneill's picture
John Oneill on Nov 20, 2018

'All fossil fuel uses are inherently unsafe - certainty of climate change, proven effects on health, frequent serious fires and explosions right along the fuel chain.' North Korea and Israel developed nuclear weapons without having any power reactors. For that matter, so did the US, the USSR, the UK, South Africa, and China. It's certain that a country, if sufficiently motivated, can build a weapon. Since that risk is not avoided by basically banning all reactors except PWRs, why not address the much more pressing risks of climate change instead ? PWRs still need an enrichment industry, arguably a bigger proliferation risk than plutonium from high burn-up spent fuel. High temperature reactors, whether salt, metal, or gas cooled, can get much higher Carnot efficiency than LWRs, with lower pressures, and could be used for industrial heat and synthetic fuel production. They should also be able to close the fuel cycle eventually, nearly eliminating mining and waste. 

Bob Meinetz's picture
Bob Meinetz on Dec 1, 2018

Willem, to support your claim that "all nuclear reactors are unsafe", you cite not statistics but "risks" which, one must assume, are a product of your imagination.

Science doesn't work that way.

Wayne Lusvardi's picture
Wayne Lusvardi on Nov 30, 2018

Duh, that is false advertising and Lazard's does not reflect the weighted price for a 24 hr. cycle.  

Ex: $0.02 x 8 hrs. (maximum) solar

Plus, Say: $0.08 x 8 hrs. backup spinning reserve

Plus: Say, $0.06 x 4 hrs. energy imbalance market (hyrdropower)

Plus: Say, $0.04 x 12 hrs. off peak (wind)

The above is oversimplified and suggestive. 



Tap Into The Experience of the Network

One of the great things about our industry is our willingness to share knowledge and experience.

The Energy Central Q&A platform allows you to easily tap into the experience of thousands of your colleagues in utilities.

When you need advice, have a tough problem or just need other viewpoints, post a question. Your question will go out to our network of industry professionals and experts. If it is sensitive, you can post anonymously.