NuScale announces a major step in the NRC’s review of its passively safe SMR
NuScale, a leader in the increasingly competitive field of advanced nuclear reactor design, has announced that the U.S. Nuclear Regulatory Commission (NRC) reviewers issued a document that formally agrees that their design does not require any electrical power to achieve safe shutdown.
It’s difficult to explain the importance of that regulatory position. People with experience in the costly hurdles associated with designing, procuring and maintaining Class 1E electrical power systems have a better chance of understanding.
Borrowing an overused, but appropriate descriptor, this step in the licensing process for NuScale is “yuuuge.”
It means that if a NuScale power station had been sitting on the coast of Japan and was inundated with a 15 meter high tsunami, it would not have mattered where its emergency diesel generators were installed. It would not have mattered that emergency connections for backup generators were also underwater or that the batteries were flooded.
If a NuScale power station was on the coast of Massachusetts and one or both of its grid connections failed, it wouldn’t have required a shutdown. The plant could have kept operating at low power in an island mode, ready to return to full power as soon as the grid connection had been repaired.
It means that NuScale power stations do not need to have a building full of storage batteries on site and in a seismically safe building. It means that the plant owners will not have to go through contortions to ensure that the Class 1E systems are always available with suitable redundancy even during repair periods.
The plants will still operationally benefit from reliable electrical power. There will still be codes and standards that apply to the electrical power systems to ensure conventional safety and uninterrupted ability to function and produce a valuable product. There is, however, a major cost differential between identical parts that carry a Class 1E classification with its associated pedigree and those that do not.
The specific action that the NRC took was that it issued a Safety Evaluation Report (SER) that concurred with NuScale’s Licensing Topical Report titled “Safety Classification of Passive Nuclear Power Plant Electrical Systems.”
That document describes NuScale’s approach and analysis ensuring that their reactors will be safe without reliance on any safety-related electrical power.
The NRC’s approval of a safety classification topical report that does not contain any Class 1E electrical systems is limited to the NuScale design.
The decision, however, establishes that the NRC is open to well-justified design choices that follow a path that deviates from the traditional prescriptive assumptions.
The NRC’s review of NuScale’s Design Certification Application remains on track to be completed by September 2020.
Disclosure: Nucale subscribes to Atomic Insights.
Does this hint at anything about the rest of the balance of plant? It would be a major cost-saver if none of the pumps, valves, turbines etc. had to be NRC-certified and could just be COTS, the turbine building done by an ordinary architect, etc.
What does “COTS” mean?
If it means what I think it does (basically everything is done to more typical industrial standards – like gas plants are) then it should be a no-brainer that the balance of plant should be done that way, for NuScale, given its inherent safety. More generally, it is imperative that *everything* at the site (i.e., everywhere other than the module assembly line plant) be done to typical industrial standards.
I *might* be able to believe that the highly experienced, dedicated staff at the module factory, who are just building large numbers of copies of the same thing, will be able to deal with NQA-1 (and “nuclear grade” requirements in general), w/o it resulting in a huge increase in cost, but I’m not willing to believe that anyone else in the process (the utility, on-site staff) would be. History shows this pretty clearly.
COTS = “commercial off-the-shelf”
Interestingly a problem with all PWR (and more generally low temp nuclear power plants) is the cost of the steam cycle. The turbines and associated components are pretty much bespoke because with the exception of nuclear no one uses low temperature wet steam turbines. Their cost due to very low numbers produced is therefore very high, especially compared to steam cycles at conventional 550 deg C superheated TG sets found on almost all combined cycle power stations. I have had this confirmed by Ian Scott at Moltex energy
I would like to see the projected benefits of Lightbridge fuel rods added to the NuScale design, by an expert. Enhanced convection with no spacers between the rods to block the water flow. Higher power levels. Better economics. Better safety margins.
Lightbridge fuel is designed to work in SMRs. Seth once hinted that they have at least talked with Nuscale about the potential synergy and furthermore Nuscale works closely with Framatome (New Areva name) It’s not something that’ll happen this year but it could happen. I’d love to see it.
Increasing the power means a higher decay heat load at shutdown. This might require the cooling pool size to be increased so as to dissipate this extra heat passively, and there will be a limit as to the total decay heat power which can be handled through passive dissipation mechanisms. This will put a ceiling on the power a NuScale can handle and remain passively safe.
This has implications for light-water breeders as well. Pa-233 has a decay energy of 570 keV and a half-life of nearly a month. At shutdown from steady-state operation at 160 MW(t) the initial decay heat from Pa-233 would be about half a megawatt and it would still be ~250 kW a month later.
If I remember correctly, for natural circulation systems coolant flow increases with the 2/3 power of core power. So the flow/power ratio decreases with increasing power. Ultimately you would get into a region of thermohydraulic instability where heat removal decreases, fuel temperature rises, power decreases, heat removal increases, power increases …. etc.
Better economics? Hard to imagine it will have the same uranium atoms per CC with it being an alloy of U and Zr with a central Zr spacer and Zr outer “clad”. You would have to enrich it beyond 5% to get the equivalent fuel load with those characteristics. The fuel doesn’t carry as much uranium.
*Unless the U-Zr alloy is at least 15g/cc or so.
Shutdown Rover 2JV
(Doo Dah Doo Dah)
Start the Diesel Gag XC
(Oh Doo Dah Day)
Wow……does that bring back memories…….SSBN632…1982-1986!!
By the way..I went to the S1C Prototype and it too has a giant tank of water on top to be used as XC if needed. Wow!!
Which crew? I was Gold crew Engineer on Von Steuben from 1987-1990
I was Gold Crew RC Div ET2 from 1982-January 1986. What a small world! I did 7 patrols on the Von S. I came on board in the shipyard in Newport News and took her to Goose Creek for a year and then moved her to Kings Bay.
Definitely a small world. Sure we shared a number of shipmates. I arrived in Oct 87.
So far so good. But what will the fossilfuel industry and anti-nuclears do to prevent the Nuscale to be built and deployed I wonder. They will not like this.
NuScale is a great design, but it’s too expensive.
The company’s own literature puts the LCOE of a 12-pack at $90 per MWh, nth of a kind. No way that competes with US wholesale prices at $40-50 per MWh, or with new gas (with wind greenwashing). (NuScale might have a brighter future in the UK, where wholesale prices are higher and the government is giving systematic incentives to new nukes.)
The problem is that the economies of “modular” in construction are outweighed by the diseconomies of “small” in operations. At the end of the day, a NuScale 12-pack is just a 570-MW (net) nuclear plant—exactly the sort that have been going bankrupt lately despite having paid off their construction costs.
NuScale doesn’t even compete with other nuclear options, like the APR1400. The Barakah plant’s LCOE will be about $70 per mwh, Korean builds are even cheaper. A big part of the cost advantage of Chinese and Korean nuclear is that they build huge plants of 4-6 gigs—lots of kwh to spread the costs over. NuScale should be thinking in terms of 120-pack plants, not 12-packs.
I hope I’m wrong, but the SMR movement seems like a dead end. Economies of scale really matter when it comes to power-plant economics.
I’d have similar concerns but consider these facts
1, NuScale’s target market is retiring coal plants. These are typically 200 – 600 MW sized units with appropriately sized grid connections
2. A 600 MW NuScale involves a Capex ~ $3Billion. For this quantum you might build a 1,000 MW ultra super critical HELE type coal plant or , of course, you could build a lot more CCGT’s .
3, Both your coal and gas options are facing uncertain carbon taxes/environmental risks , and in the latter case, price volatility .
4, Your $3B capital outlay can be phased- there’s probably a minimum spend of ~~$2B or so but this means it’s going to be much easier to finance
5, ( all ) NP are likely to receive low low-carbon credits / recognition going forward. There’s also the fuel diversity issue to factor in
Diarmuid, I’m not hopeful.
I assume the NuScale LCOE estimates are a best-case scenario with phased construction to ease financing costs, but they are still $90 per mwh, NOAK. And if those estimates prove optimistic (anyone want to bet against that?) the LCOE heads further north.
I’m not sanguine about new nukes getting credit and subsidies in the US with Trump having axed the Clean Power Plan and green lobbyists pushing popular state RE incentives instead. (UK a different story). And all the forecasts see NG staying cheap for decades. NuScale might insure against price volatility 20 years out, if that happens, but in the meantime it’s a money pit compare to gas. That’s a gamble utilities aren’t willing to take anymore.
If gas prices soar and carbon taxes happen then yes, NuScale might become competitive. But so will every other kind of nuclear plant, which might well have lower costs using successful models like the APR1400.
The deeper question is exactly what advantage SMRs have over conventional LWRs for the nuclear industry.
The NuScale LCOE suggests that there are no cost advantages over GW-scale reactors, and major disadvantages compared to well-managed large-scale installations with good designs like the APR1400. Those cost disadvantages are likely intractable because they are inherent in small-scale plants.
And yes, NuScale has a great safety case, but Gen II is already plenty safe. In a sense, NuScale is just another capitulation to the irrationally strict nuclear safety standards that have crippled the economics of the industry.
SMRs can’t fix what’s killing the industry: political opposition and crazy regulatory standards that drive up costs.
I know some of the key leaders at NuScale pretty well. They are good engineers who tend toward humility and conservatism.
They wouldn’t be comfortable working for Musk, for example.
Under promise and over deliver. Strive towards making steady improvements that eliminate cost drivers, but do NOT assume that they have the desired effect.
In other words, I think their cost estimates aren’t optimistic.
A good part of the world and 25% of the US, like Energy Northwest and TVA are public power with finance rates in the 3% range putting a $3B Capex on a 600 MW plant in the 4 cent a kwh range – Energy Northwest’s current cost and lots cheaper than the all in cost of gas even at today’s price.
Real numbers from US and Canada fracking operations are telling us that few of them are close to doing more than breaking even at current gas costs with capital being used as cash flow. With LNG builds coming rapidly, betting on low gas rates is putting the investor in for a world of hurt.
Outside of the US, where corruption is much less of an issue, public power operations are common, and coal plants are prolific, gas prices are much higher giving nukes at these prices a significant cost advantage.
On Nuscale’s ‘Value Proposition’ PDF, there is a LCOE break down for a FOAK unit with 40 year, 3.5 municipal debt financing as following:
Capital – $30/MWh
O&M – $26/MWh
Fuel & Fuel Waste – $16/MWh
Total – $72/MWh
Fuel costs seem high compared to traditional smaller single nuclear units (per NEI estimates), while the O&M costs are comparable. Considering the capital costs are for a FOAK unit and potential design innovations will hopefully lead to lower operating costs, there maybe room for further cost improvements. Additionally, if these plants are able to operate 60 years and beyond – they will have a place on the grid. Potentially, the greatest value they may offer is if they could be built on time and on budget – as it would reinstall the public’s confidence in nuclear given the current trouble.
“And all the forecasts see NG staying cheap for decades.”
Unless we go into a long, secular economic slump, I think this is dubious. And if that happens, we (US) won’t need much in the way of electricity capacity expansion.
Fracking rapidly depletes wells requiring a rapid pace of e ploration and drilling. Further, the US will be exporting increasing amounts of LNG.
I think NuScale could be available at just about the time NG prices are significantly higher than today. Assuming they operate at equal or better capacity factors than current nuclear plants and considering the predictability in price and schedule, they could be viable.
I wouldn’t count on the climate change meme being around in a few years. Nobody really believes it except the social justice types. And if they come to power … well how’s that outreach to the environmentslists working out for ya?
I don’t believe the control room staffing or evacuation zone issues have been resolved and they will be significant.
I like and respect operators but I hope NuScale could do more in the way of automation/AI. It’s like watching Star Trek and seeing the Enterprise throttled with a stick shift.
‘I wouldn’t count on the climate change meme being around in a few years. Nobody really believes it except the social justice types.’ (FermiAged )
The climate scientists are mostly on board too. And the people at this end of New Zealand having the hottest days ever recorded aren’t arguing much.
Weather is not climate. Unless it’s used to support global warming. Then global warming explains everything.
Will – Nuscale out competes a AP1400 on risk of capital. The $20B or so going into a pair of AP1400’d can break both builders and owners, see Westinghouse with VC Summer.
With NuScale modules built one or two at a time, if there is some mismanagement scandal or the election of some corrupt Governor who wrecks the project on behalf of his gas interests, the utility never misses the cash on a NuScale module. Nor does the utility have a huge capacity hole to fill if an SMR project goes wrong.
Falstaff, yes, the overruns on the *AP1000* projects bankrupted Westinghouse. But KEPCO’s *APR1400* projects have gone well, at an all-in-cost of $4000-4600 per kw. $6-7 billion for an APR1400 won’t bankrupt a large utitlity or KEPCO, and if the build goes wrong it can be halted well shy of the full construction cost, as at VC Summer. It’s not a drastically larger bankruptcy risk than the $3-4 billion all-in cost (assuming no overruns) of a NuScale 12-pack that produces only 40 percent as much electricity.
Anyway, I don’t think the smaller investment required for a NuScale makes up for the virtual certainty that its high LCOE will make it uncompetitive in the market:
“If we build this plant we’re going to lose money on it!”
“That’s OK, we can afford the loss.”
Not a compelling argument to a utility or a PSC.
You’re focusing on average market prices in a market where there are large differences between prices in different locations depending on numerous factors.
If NuScale can be built on time and on budget, it is going to be an attractive option in places where electricity is expensive due to various constraints on fuel supplies, emissions requirements, or electrical transmission capacity.
Plants that experience enough cost challenges and delays that they are cancelled become a dead loss that never produces anything of value to anyone.
A completed plant with a low marginal cost of operation might not be able to pay back loans immediately and may need financial restructuring, but it can keep generating power and successfully produce revenue. It will most likely be cash flow positive before considering capital expenditures. How many years did that description apply to companies like Amazon and Tesla?
“Plants that experience enough cost challenges and delays that they are cancelled become a dead loss that never produces anything of value to anyone.”
Right, but a NuScale that’s radically downsized in mid-build as Falstaff suggested would produce just a trickle of electricity that can’t be economically viable given the huge overhead. It would be cancelled outright just like the AP1000s.
Let me push back on the underlying idea here that SMR plants can be developed a few modules at a time to minimize investment risk.
The problem is that construction costs won’t scale downward with the downsized capacity if you halt a 12-pack build short of 12 modules. To generate power, a truncated 2-pack build, say, would cost almost as much as the whole 12-pack while producing drastically less revenue.
A NuScale build is planned as a 12-pack build, not “a 2-pack then see how things go” build. The reactors may be modular, but everything else—the reactor building and BOP—is not modular. They won’t first dig out and pour the foundation for 2 modules, then come back and dig out and pour the foundation for the next 2 modules etc.; they do it for all 12 modules at once. (Doing it piece-meal would be fantastically expensive.) Likewise, they won’t build one-sixth of a turbine hall and one-sixth of an admin building and one-sixth of a testing lab for each 2-module increment. Probably the only construction savings from downsizing a build is the reactors themselves, which are a small part of the overall capital cost of a 12-pack. Operating costs won’t scale down either with a downsized build; NRC won’t let a 2-pack get by with one-sixth the security guards of a 12-pack.
Again, modularity cannot overcome diseconomies of reduced scale. The reason NuScale is selling 12-packs rather than 10-packs or 4-packs is that the overhead, both capital and operating costs, must be hopelessly uneconomic below 12.
All of this means that a NuScale 12-pack is a unit; it lives and dies as a 12-pack. The notion of bootstrapping it module by module is fantasy. The only way to build it economically is fast, all at once, and at full size. A NuScale build that’s halted short of full size will be dead as a doorknob—even if it could produce some power, the kwhs would be so expensive that the owner would just pull the plug.
So IMHO NuScale offers little meaningful reduction in investment risk compared to a well-managed large-scale nuclear project.
“You’re focusing on average market prices in a market where there are large differences between prices in different locations depending on numerous factors.
If NuScale can be built on time and on budget, it is going to be an attractive option in places where electricity is expensive due to various constraints on fuel supplies, emissions requirements, or electrical transmission capacity.”
Maybe. My guess is that the best niche for SMRs would be to put them on barges to service ports with high-priced, underdeveloped grids (but that’s not the 12-pack model). Otherwise, I think there are very few markets where NuScale’s LCOE is competitive. That’s why NuScale is focusing on a municipal financing model with large implicit subsidies.
Which is OK by me. NuScale’s costs, while not really competitive with FF for the foreseeable future, are not outrageous. I think we should pay a little more for clean, reliable power, and I hope many NuScales get built.
My point is that we need to question the notion that NuScale and SMRs in general represent a breakthrough in cost that will redeem the economics of nuclear power. NuScale’s economics are exactly the same as the economics of any small-ish, conventional LWR plant—i.e., not viable in most places without government support. And they are substantially worse than the economics of well-managed large-scale LWR projects. So people should think hard about whether SMRs are the right path for the industry to follow or another blind alley like Gen III+.
Grid stability studies:
One of the advantages of the Nuscale 12pak is, according to a Nuscale study, the ability to load follow renewables. This is faster response to changing conditions than CCGTS are capable of, explaining the large market for open cycle gas turbines, OCGTs. Nuscale modules are surely more cost efficient than OCGTs, even with current low prices for natgas.
NuScales for wind-following? I’m skeptical.
That just means the plant is constantly ramping down to accommodate wind (or solar) surges, so its capacity factor plummets and the per-kwh LCOE soars. Since NuScale has hugely greater cap-ex and fixed operating costs than an OCGT, gas prices would have to be very high indeed before the cost comparison favors NuScale.
I agree that load following to accommodate wind makes little or no sense. Actually operating in that mode without compensation would be detrimental to revenue generation.
I can think of actual examples where a plant the operates at a steady 400-500 MWe output with the ability to quickly ramp to 600 MWe when needed, for as long as needed, can produce more revenue than one that operates on a long term contract for 100% of the capacity.
Will, using the Pacific Northwest capacity factor of 0.28 for wind power, to completely balance that requires a dispatchable generator with a capacity factor of 0.72. Using Nuscale’s estimated LCOE but with the lower capacity factor I estimate a cost of $115/MWh. From EIA estimates for 2020 open cycle combustion turbine average $142/MWh.
David, I don’t see how you arrived at 72 percent for a canonical NuScale capacity factor in wind/solar-following operation. It seems like the capacity factor of a balancing plant can approach zero, at least in theory, as arbitrarily large amounts of intermittents enter the grid.
I’m not sure what data you are looking at, but EIA’s “Levelized Cost and Levelized Avoided Cost of New Generation Resources in the Annual Energy Outlook 2017” report (https://www.eia.gov/outlooks/aeo/electricity_generation.php) puts the 2022 weighted average LCOE for gas combustion turbines at $87.10 to $100.70 per mwh. That’s for a typical CF of 30 percent.
So operating at a 72 percent CF gas would still have a substantial cost advantage over NuScale as a balancing plant, but the margin with NuScale’s $115 per mwh is small enough that a permanent 20 to 40 percent rise in gas prices might tip the comparison in NuScale’s favor.
The problem is that if wind and solar penetrations grow from that point and the CF of the balancing plant shrinks, then the NuScale LCOE skyrockets while the gas LCOE stays stable because it’s mainly marginal fuel costs.
So at a 30 percent CF, which EIA thinks is typical for OCGT in balancing service, NuScale’s LCOE would rise to $270 per mwh. Gas prices would have to roughly triple in order for NuScale to be competitive in that scenario.
Looking at comparable service profiles and CFs, it’s hard to see NuScale beating gas as a balancing generator. And given the much greater fixed investment and the much greater potential downside, why does a utility bet on NuScale instead of gas?
The AP1000 was intended to play nice with unreliables. If they go into operation, that option will prove more troublesome than its worth.
Will — For simplicity, assume a constant demand, the load which must always be met irrespective of the wind state. Then as
that is the required capacity for the balancing agent. If more wind farms are added, so must more balancing agent be added. For simplicity, we ignore CCGTs to consider just open cycle combustion turbines versus Nuscale 12paks @$115/MWh. The planning horizon is just 30 years, again for simplicity. So using constant dollars, that figure is the for sure cost of Nuscale 12paks as balancing agents.
What is the LCOE for OCGTs for the next 30 years? This heavily depends on the fuel price and the price of natgas has traditionally been highly volatile. Even predictions of it are. I gave the EIA figure from a 2015 document. You gave the EIA guess from a 2017 document. Both are but 5 year projections.
So if one wants current low prices but known future volitility natgas is the way to go. Incidentally, look at what happened to natgas prices, even availability, in the Northeast during the recent cold snap.
If predictablity is preferred, choose the Nuscale 12paks.
Actual power planning is much more complex than will fit in a blog post. I am only pointing out that Nuscale is not wrong to suggest their modules as potential balancing agents. After the demonstration 12pak is successfully running at Idaho Falls, a replacement for retiring coal burners, maybe some utilities will consider the balancing agent role for these SMRs.
Still not following you, David. There is no reason why nuclear plants would always maintain a CF of 72 percent or above in balancing wind/solar with CF of 28 percent.
For simplicity, consider an islanded grid whose only power source is 10 GW of nuclear, with a perfectly flat load curve and total yearly demand of 79 TWh, so a constant power draw of 9 GW; the nuclear capacity factor is then 90 percent. Now install 4.5 GW of wind and 4.5 GW of solar, each with CF of 28 percent; they have priority access to the grid, so no curtailment for them and the nuclear plants do all the balancing. Wind and solar will then generate 22 TWh per year, bringing yearly nuclear output down to 79 – 22 = 57 TWh per year, for a nuclear capacity factor of 65 percent.
Now add another 4.5 GW of wind. There may then be some curtailment of RE in daytime hours, but none at night. The extra wind capacity will therefore contribute at least 5 TWh yearly. That will bring the yearly nuclear output down to 57 – 5 = 52 TWh, for a nuclear CF of 59 percent. Adding more and more wind and solar will continue displacing nuclear output and thus reduce the nuclear CF, though not by a linear proportion since increasingly the new RE will displace old RE through curtailment.
It’s true that the nuclear *capacity* needed to balance the RE will remain at 9-10 GW, since grid planners have to bank on all the RE going dark at once. But the nuclear capacity *factor* will continually decline as more RE is added; since the nuclear LCOE is all overhead it will continually rise.
In that dynamic of rising RE penetration gas balancers would not see their LCOE go up significantly because theirs is almost all fuel cost, which declines linearly with declining output. That’s why gas makes a smarter bet as a balancer if you’re expecting substantial and rising wind/solar penetrations. Gas prices have to hugely and permanently skyrocket to offset that advantage, and there is no prospect of them doing so. Even if they did, the sunk cost of a bankrupt gas plant is small enough that utilities and PSCs don’t worry too much about that potential volatility many years down the road.
@ Will Boisvert
“they have priority access to the grid, so no curtailment for them and the nuclear plants do all the balancing. ” WHY???
Nuclear power plants need to schedule outages months -years in advance. With 12 “units” the scheduling becomes even more rigid. Giving priority access to Renewables wastes nuclear fuel, making the “free” green energy (which is no greener than nuclear) actually more expensive on the bottom line of the accounting sheet. Since this is an “island” grid you will always need 100% nuclear regardless of the percentage of renewables, thus the whole program is flawed.
Strange, since 1970, more than 40 years, we have been pushing renewable energy, yet the amount of fossil energy has not decreased – flatlined. Why? Recent studies show that in areas where the use of LED lamps have increased, the electrical usage has also increased. Evin I have replaced all of my lamps with LEDs (other than those that are rarely turned on) and no decrease in my electric usage.
If the objective is to reduce CO2, the recipe is wrong. Batteries, Pumped Storage, etc. all add wasted expenses doubling and tripling the cost of electricity. Or, is that the objective?
Of course increasing costs is an objective of some programs.
After all, someone’s cost is another’s revenue. That’s the way an economy works.
Rich, I agree with you 100 percent–it makes no sense to curtail nuclear to make way for wind and solar.
My modeling exercise was just addressing David Benson’s suggestion that NuScale can balance wind and solar cheaper than gas plants can. That’s why I assumed that wind and solar get priority grid access–that’s what it means for nuclear to act as a balancing generator. The company is also touting wind/solar-balancing as one of the “advantages” of NuScale. I agree it’s a bad idea, from the standpoint of market competition with gas as well as all the reasons you bring up.
I watch commercials regularly that highlight capabilities of cars and trucks that 99% of all buyers will rarely, if ever, use.
That doesn’t mean that the marketers and designers are wasting time and money by ensuring those capabilities exist and that buyers are told about them during the effort to close a sale.
Nuclear may be able to fill the gaps left by wind, but why would you install the wind turbines in the first place if you allow nuclear? To save a bit of fuel? Uranium is cheap.
With storage, the equation changes a bit and in low latitudes solar+storage could be a good partner for nuclear baseload.
We have a lot of intermittent RE coming in and next-gen nuclear is now kindof late to the game, so this means that power prices will be close to zero intermittently. That’s the problem, and whether the new nukes curtail at times of high RE production or not is not very important economically as they will get zero revenue either way. However, as Rod says, the ability to load follow might be an additional selling point. The important thing is for new nukes to get a foothold, build more experienced supply chains, have industrial learning set in and establish reliable costs and schedules. Then RE will be squeezed out in due time.
Will the instrumentation also be passive and not require class 1E classification? Seems to me that if there is an accident that people will ask a lot of questions about post accident monitoring. How about the Met Tower thing? Is it still going to need that and all of those emergency sirens? I’d guess the controls will still need Appendix R separation. I’d guess the requirement for the safe shutdown panel would be gone.
Eliminating the redundant offsite power equipment not only eliminates a lot of capital cost, it eliminates a lot of testing as well. It probably won’t need emergency diesels at all. To keep instrumentation going a small propane powered backup generator would probably do it.
I have mixed feelings. That stuff which may be eliminated gave me a lot of billable manhours in years gone by.
It’s not all about you.
Think about how many billable hours you could get from a 5x expansion of nuclear energy in the USA, even if you’d be doing something different.
Mr. EP that was a joke. I doubt whether I’ll have another nuke job. My comment on the instrumentation was intended to be real. The plant operators will still need “eyes and ears” to monitor the process in the event of some sort of design basis accident. There will still be some piping and I suppose people will be cogitating regarding a double ended guillotine pipe break or something.
I think either E-P missed Eino’s [/irony] tag, or I missed his… 🙂
Your NuScale LCOE estimate of $72 / mwh is for a municipal financing model (i.e., public ownership) and the cost reductions are entirely due to large implicit public subsidies: financing with tax-exempt municipal bonds (hence the low interest rate and low capital cost inclusive of interest and ROE); and low operating costs because they don’t have to pay property taxes and other taxes, which alone reduces operating costs by $18 per mwh. Without those implicit public subsidies, the NuScale LCOE under regulated utility financing is $106 / mwh, according to the “Value Proposition” document you cite. (http://www.nuscalepower.com/pdf/nuscale-value-proposition.pdf p. 28)
I’m all for public financing, but those subsidies would accrue to *any* municipally owned nuclear project and lower the LCOE by a comparable amount. So that $72 / mwh figure does not indicate a true cost advantage of NuScale over conventional nuclear (or gas); large-scale nuclear projects like the APR1400 builds would still be substantially cheaper than NuScale on an LCOE basis.
@ Will Biosvert
I’m all for the development of large scale LWR nuclear projects, as the cost & economics will always to be in their favor – just trying to highlight how the SMR might provide reasonable cost. For context, the EIA LCOE/LACE 2017 Annual Output report previous linked to, lists the AP-1000 advance nuclear reactor’s LCOE at $96.2/MWh using 30 year, 7.8% financing with overnight capital costs of $5,880 per KW. The latest Voglte estimates of $25 billion put it at >$11,000 per KW (if I am reading the media reports correct). Given NuScale’s Nth of a kind estimate of $5,078 per KW – and its modular, subterranean, 7 layer safety design that hopeful will insulated it from ‘big project’ risk – it might be attractive to certain utilities.
@Tom d says January 14, 2018 at 10:10 AM
“I’m all for the development of large scale LWR nuclear projects, as the cost & economics will always to be in their favor –”
In actual practice this is a myth. In the total history of nuclear builds one has never been completed without a typical 200%-300% cost and schedule overrun. So stop saying economics will be in their favor. Do you understand what ‘It’s never been done’ means? The Vogtle (and VC Summer) experience is normal. The final actual cost is always hugely over the initial estimates, so the initial estimates are worthless (including NuScale). If you want to solve this problem, start with that fact as the truth… and work backwards from there to address the root cause(s)… but don’t change the real world actual experience to ‘textbook myth’ as the starting point.
“Given NuScale’s Nth of a kind estimate of $5,078 per KW – and its modular, subterranean, 7 layer safety design that hopeful will insulated it from ‘big project’ risk – it might be attractive to certain utilities.”
The current plant killer of the current (merchant) fleet is O&M cost… after all, they are all “paid for”, except for NRC required ‘backfit’ capitol costs due to expanding the original Design Basis requirements. If a plant is never really “paid for”… how can you do long range financial planning or estimate the true capitol cost of the plant? Several of the current “paid for” plants are in financial trouble and will probably be added to the list of early decomm US nukes. This problem doesn’t belong to NuScale… it is a buyer problem, and a buyer problem to solve (or even understand). If a “paid for” plant can’t make it (reality)… start from there and work backwards to find the root cause of economic non-competitive O&M costs. (Exception… ‘regulated’ utility plants can make it because they can pass O&M overhead to the customers. If 2 identical plants exist, 1 regulated, 1 merchant and the merchant plant can’t make it… seems to be a clue there).
The US AP1000s were NRC Certified Designs of FOAK “paper reactors”. Builders need engineering details (and experience) to build them. NuScale doesn’t even have a Certified Design paper reactor yet. And the process of NRC Design Certification doesn’t include a potential buyer’s continuing O&M costs, which is in fact the current plant killer.
Actually the last 7 Candu’s #20 to 26 were all built on time in 4 years and less and and on budget at $2.5B/GW in $2018. Similar results are common in other countries with n of a kind builds.
State entities borrow at low interest rates since there is little or no risk to the borrower. There are very little property taxes and municipal costs incurred in building nukes so these are not a subsidies but competitive advantages.
According to Ron’s work “nuclear-energy-is-cheap-and-disruptive-controlling-the-initial-cost-of-nuclear-power-plants-is-a-solvable-problem”
The 2008 cost of running a nuke fuel + OM is 1.86 cents a kwh. Other countries no doubt have lower cost.
Energy Northwest claims its Tricities Plant has an all in cost of 4 cents a kwh.
Nuscales 4.2 cents a kwh O&M cost for a modern reactor is a bit of an outlier – something they need to work on.
Seth, your information regarding the current costs of power from Energy Northwest’s Columbia Generating Station, on the Hanford reservation, appears to be dated. Last I saw was $46/MWh, pricey compared to BPA’s quoted hydropower at just over $35/MWh for 2017.
I just checked a report which states that the ‘variable running costs’ of Columbia Generating Station in 2015 were $50.50/MWh.
As mentioned during last weeks all hands meeting, Its high compared to a lot of nukes because all corporate, business services, wind and hydro costs are charged to our workhorse……the reactor.
Much public land exists in the US where it is not appropriate to install a fossil power plant along with the transit of a gas pipeline or daily coal trains, but a zero emissions power plant with no fuel transits is appropriate.
Interesting that nobody linked to the NRC SER on this issue. It’s publicly available. I would think that instead of congratulating yourselves on a pro-nuclear victory, you all would actually read what the NRC wrote. Hey, some of you guys are real engineers, right? I’d be interested in your opinions on this report because I am no electrical expert – just a layman.
Yup – did a simple Google search and found it. I will give you a hint: ML17205A380.
And now to correct my own mistake in my comparison between NuScale costs and those of the APR1400 or Chinese builds, per Tom D’s point about municipal financing lowering NuScale costs.
So it’s likely that the Korean projects, the Chinese projects, and Barakah all benefit from implicit public subsidies because of public or quasi-public ownership, access to cheap state financing, tax exemptions etc. (Hard to know exactly because details are murky and unreported.) The cheaper NuScale municipal financing scenario is therefore probably closer to the cost structure that large-scale Asian nuclear projects face than the more expensive regulated-utility financing model. So the large Asian LWR projects are probably closer in actual cost to NuScale than I’ve suggested above.
I still think NuScale is no cheaper (and probably more expensive) than successful large LWR projects, but the difference is likely smaller than I reckoned and NuScale more competitive with conventional nuclear.
I largely share Will’s concerns. In order for SMRs like NuScale to be competitive, we can’t just hope that mass-scale (modular) construction will more than offset economy of scale with respect to final overall cost. That isn’t even the main thing we’re trading away economy of scale for. We’re trading it for a massive increase in inherent safety (i.e., drastically reduced meltdown possibility and greatly reduced potential source term even if one were somehow to occur). For them to be economical, we need to “take credit” for that inherent safety.
I’ve heard NuSCale developers say that not only could all (active?) components fail and a meltdown still wouldn’t occur, but even if one did the release would be so small that dose rates above the natural range would not occur anywhere outside the site boundary. Basically, this reactor is incapable of harming anyone, and it needs to be regulated accordingly, if it is to succeed.
It’s great that NRC is appearing to be reasonable with respect to electrical power systems, but this is nowhere near enough. Are they also going to do away with emergency planning, entirely? Or, at least, are they willing to abandon the notion of requiring rapid evacuation (as opposed to shelter in place w/ iodine pills)? Are they going to eliminate the massive, burdensome security requirements than are only required for the nuclear industry?
EP asks if the balance of plant could not be done to typical industrial standards. I find myself asking why the whole plant couldn’t be, given the lack of potential harm. It’s time for nuclear exceptionalism (esp. baseless exceptionalism) to end. NuScale provides the justification.
Oh, and another thing. We need to have the massive module assembly line be in China. We’ll need to take advantage of the cheap labor, just like consumer products and solar panels do. Also, the assembly line is more likely to be there anyway since their demand for the product will be far greater. One more advantage of having the entire NSSS be a transportable module is that you can site the factory where fabrication costs will be lowest.
“We need to have the massive module assembly line be in China.”
I understand what you’re driving at but this is the equivalent of giving the Chinese the design. Assuming they haven’t already stolen what has been designed so far.
We have no idea what the state of relations between the US and China will bein the next few years.
I would prefer that cost reduction come from automating operation as much as possible and perhaps using additive mfg. technologies.
I would like to see the government purchase the prototype and use it for a series of tests that would justify greatly reduced emergency planning/staffing requirements. NuScale would get a capital infusion and future customers would get a needed sense of predictability.
“Are they also going to do away with emergency planning, entirely?”
I doubt that. Even the hydro plants I worked at had an emergency plan required by some Government branch.
NRC emergency nuclear plan requirements include panic inducing evacuation of major cities 30 miles away. Down converting that to sound-the-alarm hydro emergency plans would be an enormous improvement.
My understanding is that both construction and operation costs for nuclear, many decades ago, were ~1/3 of today’s even in inflation adjusted dollars. They may have had cost overruns back then too, but the initial estimates were that much lower (i.e., ~1/9 of today’s actual costs, I suppose, then they overran up to 1/3). I think the initial estimate for the ANO plant was $100 million!
Also, according to a Northern States Power report, staffing levels at their plant were also only ~1/3 of what they are today, the increase being “due to increased requirements”.
This all shows that you can build, staff and operate nuclear plants at ~1/3 of today’s costs/levels, safely and effectively. We did it before! Most of the increase is purely due to unnecessary regulations and requirements. The hope is that NuScale’s inherent safety could provide the justification to turn back the clock on most of this BS.
I also find the operations costs estimates for NuScale to be concerning, but the NSP report provides hope that, w/ sane requirements (like they had in the old days), staffing levels and other sources of operational cost could be cut way back.
As for capital cost, I agree that ~$5,000/kW for nth of a kind won’t be good enough. It will need to be closer to $2,000/kW. Reasonable requirements, and cheap labor, etc., may make that possible. Also, to end on an optimistic note, I don’t think anyone thought that the cost of renewables (which involve mass-scale construction of small “modules”/units) could fall as far as it has. When you’re making something on mass scale, and gain experience, you may be surprised at how many ways one may find to reduce costs. Especially, if requirements and fab QA standards are reasonable.
@JamesEHopf says January 15, 2018 at 9:04 PM
I’m a ‘plank owner’ of DBNPP. Our PR announced initial cost in ’69 was about the same as ANO. Went commercial in ’78 for ~$380M. TOTAL on-site staff <250; included no Engineering (not implying that staffing was enough… clearly it wasn't as DBNPP early history shows).
"The hope is that NuScale’s inherent safety could provide the justification to turn back the clock on most of this BS."
"HOPE" is not going to change NuScale's paper reactor into a functioning commercial reactor. Congressional action will… by getting rid of the current method of regulation (and do it the way the USN Naval Reactors regulates, via a proven system for both New and Operating plants. New plant designs are designed, built, tested, and 'certified' at National Labs). How likely is that?
"I also find the operations costs estimates for NuScale to be concerning, but the NSP report provides hope that, w/ sane requirements (like they had in the old days), staffing levels and other sources of operational cost could be cut way back."
"HOPE" isn't going to solve the Staffing O&M problems taking down plants. (When some sites have hundreds of just Security personnel). Congressional action will. "HOPE" isn't going to solve O&M Staffing levels of one person/MWe. But getting rid of INPO will (along with NRC "BS"). Takes Congressional action to change the Insurance requirement of must participate in INPO to get Insurance. Are potential NuScale buyers even aware of this, and how it impacts Staffing? A one Module (4-unit) NuScale plant will likely require a larger Training Staff (License certified), just to get an INPO 'accredited' Training Program (remember almost every job on site requires INPO 'certified' training). How likely are these changes?
Even if these changes were understood, the time to start working on them was way before "yesterday."
Not all of us are in-group.
Thank you in advance.
Davis Besse Nuclear Power Plant
Institute of Nuclear Plant Operators
“As for capital cost, I agree that ~$5,000/kW for nth of a kind won’t be good enough. It will need to be closer to $2,000/kW.”
Right–smallness aside, NuScale’s modularity is not delivering meaningful cost reductions. The construction cost estimate is $5,078 per kw overnight costs; financing and escalation will raise it to $ 6-7,000 per kw. And NuScale’s LCOE estimate omits owners costs, which they say would raise it another $6 per mwh.
By comparison, the latest APR1400 projects in S Korea announced overnight costs of about $2700 per kw; add 50 percent for financing, escalation and owners costs and a reasonable all-in capital cost for a NOAK APR1400 build is $4000 per kw, unsubsidized. Barakah’s all-in cost, including first fuel load, is about $4,600 per kw. Barakah’s huge output means operating costs will be $30 per mwh or less, compared to roughly $50 per mwh for NuScale. And the APR1400 is a mature design with an experienced contractor, a proven track record and a robust supply chain, so its costs are a surer bet than NuScale’s projections.
In short, there’s just no way NuScale can match the price of best-practice, large-scale LWRs.
Look, I hope NuScale succeeds. The price is still reasonable for reliable clean energy, and I would like to see thousands of them get built.
But conventional nuclear done right is still a great bargain. The APR1400 has a license application before the NRC, and it’s crucial for its prospects that it get approved without being EPR-ized with the double containment buildings etc that have blown out the costs of other Gen III models. While the pro-nuke community is cheering on NuScale, it should also be lobbying NRC to approve the APR1400 as is. Building more APR1400s is at least as important as SMRs are.
It’s not revolutionary designs that nuclear power needs right now, just good execution by the industry and sane regulation by governments.
Oh, I forgot to mention that Will Boisvert’s analysis of the capacity factor of a balancing agent for wind and solar is much closer to correct than mine.
The original point was that the Nuscale SMR is able to adjust the power level fast enough to do so. This might be economic in some situations. After all, France does load following with their nuclear power plants.
Trump just st imposed 30% tariff on imported solar panels.
I AM INTERESTED IN THE SCIENCE OF SHUTTING DOWN A NUSCALE REACTOR WITHOUT ELECTRIC POWED. DOES IT USE GRAVITY
Has someone hacked Susanne E. Vandenbosch’s account? This comment is quite … unusual (the politest term I could find).
MY HUSBAND SAYS HE HOPES HE IS NOT AROUND WHEN THEY TEST SHUTTING DOWN THE NUSCALE REACTOR.
YOU DON’T HAVE TO SHOUT.
They tested shutting down the EBR-II from full power by turning off its cooling. It was a non-event, because it was designed to be.
The Nuscale reactor is contained in an arrangement so that an ample supply of cooling water is always available while the uranium oxide in the pins cools. NRC agrees that this works, the substance of this article by Rod Adams.
Recent Comments from our Readers
The Clinton Nuclear Plant also in Illinois was shutdown essentially for almost 2 years before it was taken over by…
Good Podcast – Very informative One thing that was not discussed is how to deal with a particular fear that…
Renewables people are masters in marketing. Unreliable intermittent generators whose output is all over the place, and usually badly correlated…
Looking at their lineup, Westinghouse seems bound and determined to keep Gen IV in its “place” which is apparently the…
So they are developing a scaled down version of the AP1000, which is a scaled up version of the AP600,…