Thanks for the excellent discussion. A few observations: It took the nuclear industry a couple of decades of hard work to drive LWR fleet-average capacity factors from 60% to 90%. The problems overcome were a combination of developing the relevant operating and maintenance skills, plant design issues, and equipment reliability issues. The non-LWR technologies will face significant learning curves as well, probably dominated by materials reliability issues associated with the very high operating temperatures most of them seem to be adopting. (The British tried high-temperature reactors with their AGR stations, and after a generation of trying to make them reliable, they have opted for LWR technology, starting with Sizewell B.) Anyone thinking about investing in, developing, or operating non-LWR technologies for electricity generation needs to clearly define in their own mind what the benefits are for taking on the inevitable, associated first-of-a-kind and learning-curve costs. The difficulties pointed out about dealing with the significant size of the capital investments in nuclear power and the long timescales for cost recovery are critical issues for both large LWRs and small non-LWRs. Humans tend to be risk-adverse, probably as a matter of evolution and natural selection. Most people will regularly opt for small, more certain gains over larger, less certain gains. It is not surprising to me that it is some of the high-technology companies who are the survivors and big-time winners of the computer/software/internet field that are beginning to invest in the newer types of nuclear power. It remains to be seen whether their high-technology talents are transferrable from the internet technology field to the nuclear power field. Building a standardized fleet of large LWRs is pretty much beyond the resources of even the largest companies in electricity generation. Southern Nuclear, in partnership with a couple of municipal utilities, had the wherewithal to persevere through the learning curves for Vogtle 3&4: first reactor of a new design, first use of the Part 52 licensing process, first use of nuclear modular construction, and first new nuclear construction start in the United States in more than a generation. However, they really do not have the need for another two, four, or six reactors anytime soon, so it is hard for them to profit from the learnings through faster, more efficient construction of follow-on units. If there is a future in the U.S. for a significant number of standardized, large LWRs, it is probably through a regional joint-venture company, owned by a handful of large utilities, that constructs, owns, and operates the units, sharing the output and costs. This allows the learning costs to be shared and applied to near-term follow-on units, generating capacity to be added in smaller, more frequent increments per owner, and captures the benefits of the efficiencies of the follow-on units to offset some of the learning-curve costs. There is precedent for such an organization. A group of relatively small utilities in New England formed the Yankee Group and shared the costs of developing, constructing, and operating the first few units in New England in the early days of nuclear power. This approach might work today. Electricity is a commodity essential to most everything modern society does. It must be both reasonably priced on an energy basis and reliable, with reliability being paramount, for society to prosper. Two strong points for current LWR-based nuclear power are its reliability and its low CO2 emissions. The energy costs of nuclear are higher than most other alternatives. If reliability were the only advantage nuclear power could offer, coal- and gas-fired generation, which also can operate reliably, would win out. Low CO2 is the key potential differentiator. It remains to be seen how much society is really prepared to pay for reduced CO2 emissions. The bet that enough of society continues to believe CO2 is a significant concern and that people are willing to pay much to reduce CO2 emissions may be the most critical risk to investments in all the nuclear options, LWR and non-LWR.
TO ALL: Towards the end James mentions EPCs and the way they operate. I'm an Australian engineer who has worked for EPCs including one of those James mentions. I actually work in control systems including safety instrumented systems. So I understand James comments on mean time between failure (mtbf). For those who don't know EPC stands for Engineering Procurement & Construction. The joke is it stands for Extra Pain & Costs because of how they operate. Due to the time taken on these projects and the very long lead times involved its almost impossible for any major project in any industry to do fixed cost unless they put a premium on the price to cover changes as well as having an incredibly strict contract clauses on adjustments. So the most common way contracts are done is called cost+ and to illustrate this I'll use a 20% cost+ but be aware the actual percentage is also a negotiable number in a contract. So with a 20% cost plus for every dollar of expense the EPC has on the project the customer pays them $1.20. So if you have an employee on the project costing $100/hr the customer pays the EPC $120 per hour. If the EPC buys something for $1 million then the customer pays the EPC $1.2 million. There's a downside to this. The EPC never has the incentive to hire good people unless they get into real trouble with time and one of the few constraints on EPC are time clauses. I see this in mining because when they build a new mine there's a date on which the mine has to start production because they will have already signed delivery contracts. EPCs handle this by simply hiring more people and at those times they they also hire better people to avoid time penalties. Trust ne the EPC always wins. In fact you really have to try hard to not make money on a cost plus contract. EPCs are also incentivised to buy the most expensive equipment for a project. If you have a choice between a $100 million item and a $120 million alternative you buy that alternative irrespective because 20% of $20 million is $4 million an you just made your boss and extra $4 million with that decision. Also the best thing an EPC ever hears is "we are changing the........(whatever)" because it means you can hire more people, buy more stuff and charge heaps of money to do whatever the change is. This is why the first AP 1000 plant in South Carolina at the Summer plant blew out so horrendously. You might think that Smart customers would hire smart contract managers to make sure they aren't getting ripped off *BUT YOU'D BE WRONG!* As James explained their are ways the customer dumps that back onto the state. In the mining industry where I have been involved those costs are dumped back onto the rest of the population through tax right-offs and basically the state doesn't earn the money they might think they should earn from having this company in their state. Here in Australia we had the Gorgon Gas plant blow out from $55 Billion to over $70 Billion. Do you think Chevron paid that or do you think Chevron wrote that off against taxes? Yes they dumped all that onto the people of Australia, just like other companies do in other places.
Thanks for hosting this, Chris! I’m new to your podcast, and it’s great to see such a platform for discussion of these topics in a long form. If I could offer one idea for improvement, for the more technical stuff, please encourage guests to avoid acronyms, or to at least state what they are the first time they use them! Thanks again! Looking forward to going through your other vids.
I really like this podcast, I watched Vogtle episodes, SMRs episodes, uranium mining episodes and all of them are so interesting. Thanks for making this podcasts
A suggestion for another episode: What are the building blocks that we need to put in place to be able to permit and construct energy infrastructure projects in a timely manner; focusing on reforming permitting processes to better balance local and national interests. Regardless of your favourite technology, the time required to get formal permission to build any electrical infrastructure project is simply too long, and too easily derailed by special interest groups who have no local interests, who either insert themselves directly like Greenpeace, FOE, etc. or subvert local interests by propaganda and misinformation to intervene in support of some else's agenda. If we have any hope of achieving energy transition in a timely manner and at a price we can afford, we need clear, competent, and transparent permitting processes that run to clearly understood timelines.
You have put your finger on the key problem: the ability of special interest groups to derail projects by objections timed to maximize the delays and disruptions of projects they do not want to proceed. These groups know that timely completion of high-capital-cost projects is critical to their success and that late-coming objections that delay the project are effective weapons, irrespective of the merits of the objections. The economist Larry Summers once memorably characterized this problem as "the promiscuous distribution of the power to hold things up."
There isn't much to debate about because this isn't an either-or question. We need more nuclear energy and need to get away from oil. Doesn't matter small or large just build, build, build.
The "Does Reactor Size Matter?" teaser title made me think of nuclear fusion. It's often not appreciated just how *big* fusion reactors will be compared to fission reactors of the same power output. Given the recent PR about fusion, it would be nice if you could interview someone from the fission side to comment on this. Lawrence Lidsky at MIT has been dead for some time now, but back in the 1980s he made the point of just how large the discrepancy had to be.
I remember sitting cross legged in the gym as a grade 4 student listening to kids who had won provinicial public speaking awards give their speeches.... One bright kid did a talk about fission verses fusion in powering our future I remember she said the holy grail of fusion was likely 10 years away. That talk was given in 1969. more than 50 years ago. As such, I don't think Fusion will happen commercially in my life time.
@@alisonmcgillivray8008 In my opinion, Helion is the least dubious of the commercial fusion efforts, because direct conversion would allow the no-go power density argument to possibly be evaded, as it wouldn't be functioning as a heat source to drive turbines. It could also be a great source of fuel for conventional fission reactors, since by using DD fusion it has large amounts of excess neutrons. This also presents a proliferation risk, of course.
@@danadurnfordkevinblanchdebunk There are generic arguments that DT reactors must be large. This is because the neutrons must radiate through the boundary of the reactor vessel, and the blanket hast to be at least a certain thickness dictated by nuclear cross sections. Limits on power/area at the wall therefore imply a limit on power/volume in the reactor, and this gets worse as the power of the reactor increases (square-cube law). Contrast this to a fission reactor, where coolant flows through the reactor itself, and area for heat transfer is large. Lidsky's conclusion was DT reactors had to be at least 10x larger than fission reactors of the same power output. Today, ARC would be 1/40th the volumetric power density (and ITER 1/400th) of a commercial PWR (taking as volume that of the reactor vessel), which is consistent with Lidsky's analysis.
Controlled fusion has been pursued since the late 1950s. In the mid-1970s, I was an undergraduate in a nuclear engineering program. A representative from ERDA, the predecessor of the U.S. Dept of Energy, came and gave a presentation on the U.S. fusion program. It involved something like achieving sustained energy breakeven by the late 1980s and having a demonstration electricity-generation plant operating by the year 2000. A couple of the professors pointed out some of the key technical challenges for which solutions did not then exist, including maintaining a stable plasma and developing a reliable first-wall design, and termed the program schedule unrealistic. The presenter pushed back against what he characterized as "can't do" thinking and insisted the program was credible. Today, at the cusp of 2025, fusion has yet to demonstrate sustained net thermal energy gain for a time remotely compatible with electricity production, and the first-wall problem is still with us. Significant money has been, and continues to be, spent on fusion research and development and some progress has been made. However, nobody should be counting on fusion becoming commercially viable and deployable for electricity production in the next generation or two. If several major technical and economic breakthroughs happen before then, be happy and enjoy the unexpected boon, but in the meantime, make economic plans on the assumption fusion is not a viable option for the economically relevant future.
Great discussion, thank you gentlemen. My response to the question is, 'all of the above'. We need large standardized GW class reactors that can be rolled out on a serial basis allowing construction resources to go from one build to the next over time. We also need to create an environment in which SMR's and advanced reactors can be developed and rolled out on a serial basis. The use case for big units is for markets and grids that can tolerate large single generating units, assuming that the installed cost in $/kW are on target. That said there are many, many grids in the world that simply cannot tolerate large single unit generators. Alberta (AB) for example, the Most Severe Single Contingency (MSSC) is 466 MW. In simple terms, AB cannot connect a 1,000+ MW generator, because when it tripped off-line the AB grid would collapse. There are many grids in the world that also have this challenge. For AB, assuming no changes on the grid side, 300 - 400 MWe Reactors would be ideal and 300 - 400 MW does not create too big a disturbance in the real-time energy-only electricity market there. For me, 300 MW is about as small as I think we should go. Nuclear technology is amazing, but the cost per unit of production gets high really fast as your total annual production number gets smaller. While the technology is amazing the overheads and one off project costs are eye-wateringly high; having fewer MWh/year in the denominator makes the overall cost of production very high on a per MWh basis.
I was also surprised nobody brought up MSSC issues with regard to 100% all large reactors. I believe this is a point in favor of SMRs, because there is no point to maximizing cost/unit generation if that is not the constraint.
Loved this conversation. Thanks to all three of you! IMO, nuclear plants don't need to worry about load following at all, just run all out all the time. Grids of the future won't be trying to match supply to load, instead, it's going to work the other way around. Loads will follow supply (eg VPPs managing EV charging, thermostats, and home batteries all to match available supply). My guess is within a decade or two, a large percentage of customer premises will have several hours of BTM battery installed. They'll charge when supply is plentiful and grid utilization is lower and discharge when supply is constrained or grid infrastructure is more stressed. Locating storage close to end users has a triple benefit: 1) outage protection, 2) can optimize for highest average T&D utilization, 3) can optimize supply utilization. This change will be great for nuclear heavy grids because plants can just run all out and not have to worry about the headaches and economics of trying to follow load. If nuclear can get its cost low enough, there will be no limit to the percent of generation nuclear can satisfy on a grid.
1. Build out at the maximum possible rate. 2. Match capacity with resistive shunts. 3. Outlaw metering. Customers will figure out how to use the wasted power.
Indeed it is a waste of time to reduce output from Nuclear Power, and cause more wear on the hardware too, without much saving on cheap fuel. Many countries need heating- and hot water can be stored cheaply and in huge quantities compared to batteries. Hydrogen could be next, but it is more pricey than heat to make it, store it and use it.
We need to build 10s of thousands of reactors. We need SML reactors. Small, Medium & Large reactors! We know how to build large ones today, lets start building those large ones, and the small and medium ones will follow along when they're ready. In 10, 20, 30 years, we'll naturally land up with all sorts of funky reactors... As soon as we get a thriving nuclear industry and supply train up to speed. Climate change isn't gonna wait for us to deliberate. The sooner we've built enough reactors, the sooner we can actually turn off the coal, gas, oil power stations and industrial heat generators.
@danadurnfordkevinblanchdebunk small in size, but ginormous in cost, and astronomical in fuel cost. Naval reactors are just not good for generating electricity for a grid. They're just so different, so specialised to their job.
@@danadurnfordkevinblanchdebunk $/GW? What fuel do they use? What enrichment? There's a LOT about their design that is tailored to being able to ramp up from nothing, to everything, and from everything back down to nothing, in mere seconds (probably). They're also keen on being pretty much surrounded by the biggest heatsink on the planet, i.e. the ocean. And don't forget that their designs are highly classified, too. Unless you know the secrets? They're the wrong tool for the job. It sounds cool if we could use them, but they're just not what we need for grid/industrial electricity & heat demands.
@@RichHornerCheeseAndJamSandwich Sounds like an excellent design for load following. Some commercial floating nuclear reactors are powering coastal towns in the Arctic circle.
The old adage; "A bird in the hand beats two in the bush." You have only two options that have been built and are deployable today, CANDU 6 (Monarch is not ready) or the AP1000. Everything else will be a fight for regulatory approval, FOAK build, financing the a new FOAK, unproven Capacity Factors, etc.
For what it is worth, I think that one of the most constructive climate initiatives that the federal Canadian government could have done is to pay SNC Lavalin to update the CANDU 6 design to EC6. Then license that design through CNSC under the nuclear safety assessment part for use anywhere in Canada. Provide free technology use licensing to any province that want to build them and kick off a serial construction program across Canada. Some federal enabling legislation would also be a help, something to say any site that meets minimum national standards for environmental impact is deemed to be permitted by default as part of Canada's climate change initiatives as a streamlining of the normally tortuous EA process.
You forget the Korean APR-1400 (Barakah) and the Russian VVER (sure, a terrible partner, led by a war criminal, but they do provide some amazing financing).
@@factnotfiction5915 I don’t think that VVER would meet the‘Reliable Partner’ test, but they do seem to know how to build NPP’s. The Koreans on the other hand definitely know how to build, but also make better partners. I think that they have real potential.
Unless Canada develops the ability to enrich its own fuel, building an AP-1000 would be a terrible mistake. America is showing the world that they are an unreliable trading parent and cant be trusted, being dependent on them for fuel would be a stupid move.
The U.S. does not come close to supplying its own demand for uranium enrichment, let alone supplying foreign demand. U.S. enrichment capacity supplies about 40% of U.S. enrichment demand, with the remainder coming from abroad. If Canada adopted enriched fuel, all its demand for enrichment services would have to be from non-U.S. sources. Further, the U.S. depends on Canada for natural uranium. Canada is the largest foreign supplier of uranium to the U.S., supplying about 30% of U.S. needs, so the U.S. is much more dependent on Canada than Canada is dependent on the U.S. when it comes to nuclear fuel.
The best take away from this is that we should declare the debate over. (Well-scheduled conversation, Chris!) We need both of these present paths to come to successful fruition. Not in the usual "all-of-the-above, nod-nod to the wind turbines" sense, but because we are looking at two unique each-wonderful toolboxes for the future. Tech typified by the AP-1000 is mature and ready to give us the Terawatts (just a few random nickels from the gutter, please). Let's figure out the political economics, etc., and go forward. Meanwhile, advanced tech has so much to offer in terms of potential simplification of manufacture and new and unique applications. Let's support that, too, and go forward. It's too bad that this has even been ossified into a debate. The anti-water-it's-bad-old-school-and-it-can-explode meme should be buried right along side the it's-only-a-paper-reactor-you-clown-you-don't-know-what-you-are-doing thread. Let's get real, and work together!
I love confident predictions on schedule and cost for something that has never been built at scale. Still it is all you have as the stuff built at scale hasn't gone so smoothly
32:40 "The *_Loan Programs Office (LPO)_* provides loans and loan guarantees available to help deploy innovative clean energy, advanced transportation, and Tribal energy projects in the United States. Over the past decade, LPO has closed more than $30 billion of deals across a variety of energy sectors." (US Department of Energy)
Privatization of Nuclear Energy in US baffles me. Rosatom is a state corporation in Russia. You shouldn't expect private entities to take upon themselves these enormous multi-billion infrastructure projects, that take a minimum of 5 years to build if you're lucky, and it's more like a minimum of 10-12 years to build in the West in the atmosphere of hostility from the general public and general legal uncertainty from administration to administration.
Appreciate the detailed breakdown! Just a quick off-topic question: I have a SafePal wallet with USDT, and I have the seed phrase. (alarm fetch churn bridge exercise tape speak race clerk couch crater letter). Could you explain how to move them to Binance?
11:37 Hinkley Point? You mean the most expensive construction project in human history? at potentially 20 billion dollars per GW, if it ever actually gets built.
8:59 It's not even about building large. Russian RITM-200 SMR are 55MW. But you need the kind of regulatory environment and the kind of serious state company environment with international political/diplomatic connections to have a market and a will to make these things. Those were icebreaker power units. Old soviet technology. First they put it on a floating platform to power a town in the Far East of the country, for testing. Now they got a large order from Azerbaijan (and a few others) and they're beginning to "mass-produce" them. You can't expect a startup or even your "above average billionaire" to be able to do this on their own (at least not at their current levels of wealth which may sound large in terms of market capitalisation, but doesn't necessarily translate into that substantial of an RnD and manufacturing capability -- at least not compared to the task). Free market is not magic. It could be short-sighted. Someone needs to provide a long-term vision.
@@pauldietz1325*_"Efficient"_* always needs to be defined. Otherwise, statements containing the adverbial phrase *_"more efficiently"_* are meaningless.
Some 2/3 of industrial heat energy used in the US is below 300 C and could be provided by industrial heat pumps. Above that temperature, heat is quite storable (for example in SiC or nickel-doped chromia, which is its own resistive heater, or even sand at 1200 C), so electric resistive heat could exploit peaks in renewable output when prices are low. Former nuclear vendor Babcock & Wilcox has been working on a system for storing heat in sand at 1200 C using very compact fluidized bed heat exchangers.
@@aliendroneservices6621 > Electricity is a simpler way to provide process heat. Heat is the simplest, and most efficient, method of supplying process heat. Converting high-grade electricity to low-grade heat is inefficient, and doubly inefficient if heat was used to generate the electricity in the first place.
Chris I am curious as to WHY have you kept Canada's CANDU heavy water reactor option out of this large scale light water or small scale non-light water debate? I am especially curious because the Candu refurbishment teams are delivering on time and on budget?
I suspect the reason is that CANDU is more expensive up front than light water and because they are A, American so not likely to work with foreign designs and B, because the biggest constraint they are trying to argue around is initial financing.
@@jonmce1 is that true ? is a new build candu delivered on time and on-budget more expensive than a time overrun, cost overrun alternative? Or it it like the Avro Arrow? Clearly Superior but it sticks-in-the-craw being better smarter non-american technology?
The mission timespan directly leading to given quantity of iterations over some duration of time is a damn good argument. If anyone was winning it was Krellenstein although maybe Ted is right that everyone is underestimating the political question but i cant judge that. Unfortunately we've attacked and sued our reactor operators for decades. The damage of this may never be able to be undone and so a technological paradigm shift may be necessary for that reason even though ironically the technology gets possibly worse.
Nuclear power is finite. Uranium supply would be consumed within decades if it was producing equivalent power to Oil(cars, planes, trains, boats, tractors, factories, ...)
How cheap is the renewable/storage competition going to be if the putative nuclear transition is extended? Nuclear has to be sprinting down an experience curve if it's going to have a chance.
I thought this a weird debate. It seems illogical to be talking either/or when it is logical to do both depending on location and capacity requirements. The debate seems to debating whether something that is already in work should be done. In Ontario the first BWRX 300 is underconstruction with 2 more planned. They are being constructed at the Darlington nuclear site so both environmental decisions and depth of experience exist already. It will be a model for other provinces to gain experience in such technology. More expansive use is expected to be used at old coal generating sites where environment and grid issuea are already established. A new large unit hopefully CANDU is planed to built at another existing power generating location.
We need to build new and established reactor design. Gen 2 gen 3 and Four. Both paths must be built out. We need more electricity and clean high temperature heat.
At this point, CANDU increasingly looks like the "safe bet" of a medium sized reactor that can always be built. Oddly, the shortfall in know how in the nuclear industry seems to be control systems these days, not so much the nuclear physics. But, regardless of technology, the stupefying amount of paperwork required to build and operate a reactor needs to come down as does the endless circus of continuous audits. There are way too many overheads. These overheads are prohibitive for any SMR design and they can't just be wished away easily.
Only low-pressure/high-temperature nukes can compete with combined-cycle natural gas, and the NRC will never let that happen. So, let's stick with gas until we can replace the NRC with a regulator that places equal value on safety and affordability.
Nothing can compete with CCGTs One was built recently in the UK 2022 at a cost of $0.50 / KW and it has a very high efficiency of ~63% Vs the reactors under construction in the UK at ~$20/KW Thats 40 x cheaper And they need ~15x fewer staff Wind solar nuclear can't compete with that, so we either build CCGTs and accept the carbon emissions or subsiside and guarantee Wind/Solar or Nuclear There is no shame or failure in nuclear highlighting it needs state guarantee and support because wind and solar do too
I'd like a link. AFAIK, CCGTs are more like $1/W rather than $0.50/W (which is what you meant). The latter is more like the cost of simple cycle gas plants. Did you leave off something from the cost, maybe the cost of the grid connection?
"That's 40 x cheaper" Then Singapore (95% natural-gas-fired) should have the cheapest electricity. Let's check: Singapore: $0.240/kWh Taiwan: $0.090/kWh Taiwan's electricity is: Coal: 43% Natural-gas: 40% Uranium: 6%
@aliendroneservices6621 I was talking about the construction cost of UK nuclear vs UK CCGTs and it is about 40x cheaper to build 1 watt of CCGT in the UK you can look up both costs Google Keadby 2 CCGT cost Hinckley Point C cost
@@kaya051285I was unable to find solid numbers for Keadby 2 anywhere. All I could find was SSE playing its usual game of *_"hide the numbers"._* From the pdf "Powering Progress" (October 2020): "Keadby 2 is one of the largest infrastructure projects to be built in the North Lincolnshire region. The investment by SSE Thermal associated with the project, from development through construction and to the end of its operational life, is expected to total £1.2bn" How much was the total project cost? No answer. *_£1,200M / 893MW = £1.34/W_*
I would not be surprised if the reactor control system can make small, incremental reactor-power adjustments at that rate for the purposes of holding reactor power stable around a desired power level, but I would be surprised if the power conversion cycle (i.e., the main turbine generator, feedwater, and feedwater heating systems) can tolerate a sustained rate of power change that large for more than a few seconds.
@@thomasgreene5750 The French have been exercising their NPP fleet (including generators and feedwater) across large % changes every day for decades. This is basically old tech that probably deserves upgrades as improvements in control and prediction are made. What we really need is large thermal storage/off-take to balance it out with less % changes on the electricity generation side.
@@factnotfiction5915 I'm aware that the existing generation of plants can load follow over a wide range and that they can make small changes rapidly. What I am finding surprising is the implication that they could run from 60% to 100% at 1% per second. That implies a ramp time of 40 seconds from 60% to 100% or from 100% to 60%. I find myself wondering how long the pellet-cladding interface can tolerate up-ramps that fast, and in my experience, down-ramps that large in power, but slower rates, say on the order of 3 or 4 minutes instead of 40 seconds, have run into feedwater heater water-level issues and feedwater pump NPSH issues.
@@thomasgreene5750 I see where you are coming from now. > I find myself wondering how long the pellet-cladding interface can tolerate up-ramps that fast There probably isn't much that can be engineered there, given one wants to use existing fuel formulations so I would presume that was not the limiting constraint. > down-ramps that large in power, but slower rates, say on the order of 3 or 4 minutes instead of 40 seconds, have run into feedwater heater water-level issues and feedwater pump NPSH issues.' But it may be that the new designs and new pumps can handle this feedwater constraint. > the implication that they could run from 60% to 100% at 1% per second. That implies a ramp time of 40 seconds from 60% to 100% or from 100% to 60%. We probably only have half the story here. Sure, between 60% and 100% power, the instantaneous ramp rate is 1% per second (which we need to remember is on the order of 10 MW for a 1000 MW reactor, so not insignificant). There are probably secondary, understated, constraints, among which is that the instantaneous rate and the sustained rate are somewhat different.
At what battery prices will solar and batteries enable an affordable load following profile? So AP1000 is 1.1GW and roughly 8.8TWh annually What would be the cost of a solar farm that can output 1.1 GW for 14h a day followed by ~0.8 GW for 10h a day. This would be better than baseload Somehting like 1.1 Grid connection 4 Gw solar 12 GWh batteries Would allow solar load following power station Worst months Dec/Jan would be ~60% capacity factor (at the grid interface) And then 8-10 months it can do the ideal load following of 1.1gW for 14h followed bt 0.8GW for 10h If this load following solar plant can be built cjeaper than Ap1000s then all sunny regions will do the above This is also super modular Can build it 1000x smaller than an ap1000 1.1MW grid connections 4MW solar panel 12MWh battery No good for non sunny locations but probably future for sunny regions
what a useless yapfest. How many times is the bald guy going to say that nuclear has a 92% capacity factor, even when computer programmers work during the day and solar works during the day. There was nothing resolved here, just a bunch of yapping, no facts, no figures, The guy who likes big nukes didn't explain how they would be cheaper, and the little nukes didn't explain how it would be cheaper. Maybe we should just admit that the future is Natural Gas. Since you started this podcast, the entire world has buillt a natural gas and refridge and defridge terminal, and we sent both natural gas and other liquid fuels around the world.
Thanks for the excellent discussion. A few observations:
It took the nuclear industry a couple of decades of hard work to drive LWR fleet-average capacity factors from 60% to 90%. The problems overcome were a combination of developing the relevant operating and maintenance skills, plant design issues, and equipment reliability issues. The non-LWR technologies will face significant learning curves as well, probably dominated by materials reliability issues associated with the very high operating temperatures most of them seem to be adopting. (The British tried high-temperature reactors with their AGR stations, and after a generation of trying to make them reliable, they have opted for LWR technology, starting with Sizewell B.) Anyone thinking about investing in, developing, or operating non-LWR technologies for electricity generation needs to clearly define in their own mind what the benefits are for taking on the inevitable, associated first-of-a-kind and learning-curve costs.
The difficulties pointed out about dealing with the significant size of the capital investments in nuclear power and the long timescales for cost recovery are critical issues for both large LWRs and small non-LWRs. Humans tend to be risk-adverse, probably as a matter of evolution and natural selection. Most people will regularly opt for small, more certain gains over larger, less certain gains. It is not surprising to me that it is some of the high-technology companies who are the survivors and big-time winners of the computer/software/internet field that are beginning to invest in the newer types of nuclear power. It remains to be seen whether their high-technology talents are transferrable from the internet technology field to the nuclear power field.
Building a standardized fleet of large LWRs is pretty much beyond the resources of even the largest companies in electricity generation. Southern Nuclear, in partnership with a couple of municipal utilities, had the wherewithal to persevere through the learning curves for Vogtle 3&4: first reactor of a new design, first use of the Part 52 licensing process, first use of nuclear modular construction, and first new nuclear construction start in the United States in more than a generation. However, they really do not have the need for another two, four, or six reactors anytime soon, so it is hard for them to profit from the learnings through faster, more efficient construction of follow-on units.
If there is a future in the U.S. for a significant number of standardized, large LWRs, it is probably through a regional joint-venture company, owned by a handful of large utilities, that constructs, owns, and operates the units, sharing the output and costs. This allows the learning costs to be shared and applied to near-term follow-on units, generating capacity to be added in smaller, more frequent increments per owner, and captures the benefits of the efficiencies of the follow-on units to offset some of the learning-curve costs. There is precedent for such an organization. A group of relatively small utilities in New England formed the Yankee Group and shared the costs of developing, constructing, and operating the first few units in New England in the early days of nuclear power. This approach might work today.
Electricity is a commodity essential to most everything modern society does. It must be both reasonably priced on an energy basis and reliable, with reliability being paramount, for society to prosper. Two strong points for current LWR-based nuclear power are its reliability and its low CO2 emissions. The energy costs of nuclear are higher than most other alternatives. If reliability were the only advantage nuclear power could offer, coal- and gas-fired generation, which also can operate reliably, would win out. Low CO2 is the key potential differentiator. It remains to be seen how much society is really prepared to pay for reduced CO2 emissions. The bet that enough of society continues to believe CO2 is a significant concern and that people are willing to pay much to reduce CO2 emissions may be the most critical risk to investments in all the nuclear options, LWR and non-LWR.
TO ALL: Towards the end James mentions EPCs and the way they operate.
I'm an Australian engineer who has worked for EPCs including one of those James mentions. I actually work in control systems including safety instrumented systems. So I understand James comments on mean time between failure (mtbf).
For those who don't know EPC stands for Engineering Procurement & Construction. The joke is it stands for Extra Pain & Costs because of how they operate. Due to the time taken on these projects and the very long lead times involved its almost impossible for any major project in any industry to do fixed cost unless they put a premium on the price to cover changes as well as having an incredibly strict contract clauses on adjustments.
So the most common way contracts are done is called cost+ and to illustrate this I'll use a 20% cost+ but be aware the actual percentage is also a negotiable number in a contract.
So with a 20% cost plus for every dollar of expense the EPC has on the project the customer pays them $1.20.
So if you have an employee on the project costing $100/hr the customer pays the EPC $120 per hour.
If the EPC buys something for $1 million then the customer pays the EPC $1.2 million.
There's a downside to this. The EPC never has the incentive to hire good people unless they get into real trouble with time and one of the few constraints on EPC are time clauses. I see this in mining because when they build a new mine there's a date on which the mine has to start production because they will have already signed delivery contracts. EPCs handle this by simply hiring more people and at those times they they also hire better people to avoid time penalties.
Trust ne the EPC always wins. In fact you really have to try hard to not make money on a cost plus contract.
EPCs are also incentivised to buy the most expensive equipment for a project. If you have a choice between a $100 million item and a $120 million alternative you buy that alternative irrespective because 20% of $20 million is $4 million an you just made your boss and extra $4 million with that decision.
Also the best thing an EPC ever hears is "we are changing the........(whatever)" because it means you can hire more people, buy more stuff and charge heaps of money to do whatever the change is. This is why the first AP 1000 plant in South Carolina at the Summer plant blew out so horrendously.
You might think that Smart customers would hire smart contract managers to make sure they aren't getting ripped off *BUT YOU'D BE WRONG!* As James explained their are ways the customer dumps that back onto the state. In the mining industry where I have been involved those costs are dumped back onto the rest of the population through tax right-offs and basically the state doesn't earn the money they might think they should earn from having this company in their state.
Here in Australia we had the Gorgon Gas plant blow out from $55 Billion to over $70 Billion. Do you think Chevron paid that or do you think Chevron wrote that off against taxes? Yes they dumped all that onto the people of Australia, just like other companies do in other places.
15:21 I mean no harm... While at this point started to count amount of "You know" used... We know.😅
Thanks for hosting this, Chris! I’m new to your podcast, and it’s great to see such a platform for discussion of these topics in a long form.
If I could offer one idea for improvement, for the more technical stuff, please encourage guests to avoid acronyms, or to at least state what they are the first time they use them!
Thanks again! Looking forward to going through your other vids.
I really like this podcast, I watched Vogtle episodes, SMRs episodes, uranium mining episodes and all of them are so interesting. Thanks for making this podcasts
A suggestion for another episode: What are the building blocks that we need to put in place to be able to permit and construct energy infrastructure projects in a timely manner; focusing on reforming permitting processes to better balance local and national interests.
Regardless of your favourite technology, the time required to get formal permission to build any electrical infrastructure project is simply too long, and too easily derailed by special interest groups who have no local interests, who either insert themselves directly like Greenpeace, FOE, etc. or subvert local interests by propaganda and misinformation to intervene in support of some else's agenda. If we have any hope of achieving energy transition in a timely manner and at a price we can afford, we need clear, competent, and transparent permitting processes that run to clearly understood timelines.
You have put your finger on the key problem: the ability of special interest groups to derail projects by objections timed to maximize the delays and disruptions of projects they do not want to proceed. These groups know that timely completion of high-capital-cost projects is critical to their success and that late-coming objections that delay the project are effective weapons, irrespective of the merits of the objections. The economist Larry Summers once memorably characterized this problem as "the promiscuous distribution of the power to hold things up."
There isn't much to debate about because this isn't an either-or question. We need more nuclear energy and need to get away from oil. Doesn't matter small or large just build, build, build.
The "Does Reactor Size Matter?" teaser title made me think of nuclear fusion. It's often not appreciated just how *big* fusion reactors will be compared to fission reactors of the same power output.
Given the recent PR about fusion, it would be nice if you could interview someone from the fission side to comment on this. Lawrence Lidsky at MIT has been dead for some time now, but back in the 1980s he made the point of just how large the discrepancy had to be.
Since we don't know if we will ever figure out a sane fusion process, we also have no idea what size of a facility that might entail.
I remember sitting cross legged in the gym as a grade 4 student listening to kids who had won provinicial public speaking awards give their speeches.... One bright kid did a talk about fission verses fusion in powering our future I remember she said the holy grail of fusion was likely 10 years away. That talk was given in 1969. more than 50 years ago. As such, I don't think Fusion will happen commercially in my life time.
@@alisonmcgillivray8008 In my opinion, Helion is the least dubious of the commercial fusion efforts, because direct conversion would allow the no-go power density argument to possibly be evaded, as it wouldn't be functioning as a heat source to drive turbines.
It could also be a great source of fuel for conventional fission reactors, since by using DD fusion it has large amounts of excess neutrons. This also presents a proliferation risk, of course.
@@danadurnfordkevinblanchdebunk There are generic arguments that DT reactors must be large. This is because the neutrons must radiate through the boundary of the reactor vessel, and the blanket hast to be at least a certain thickness dictated by nuclear cross sections. Limits on power/area at the wall therefore imply a limit on power/volume in the reactor, and this gets worse as the power of the reactor increases (square-cube law). Contrast this to a fission reactor, where coolant flows through the reactor itself, and area for heat transfer is large. Lidsky's conclusion was DT reactors had to be at least 10x larger than fission reactors of the same power output. Today, ARC would be 1/40th the volumetric power density (and ITER 1/400th) of a commercial PWR (taking as volume that of the reactor vessel), which is consistent with Lidsky's analysis.
Controlled fusion has been pursued since the late 1950s. In the mid-1970s, I was an undergraduate in a nuclear engineering program. A representative from ERDA, the predecessor of the U.S. Dept of Energy, came and gave a presentation on the U.S. fusion program. It involved something like achieving sustained energy breakeven by the late 1980s and having a demonstration electricity-generation plant operating by the year 2000. A couple of the professors pointed out some of the key technical challenges for which solutions did not then exist, including maintaining a stable plasma and developing a reliable first-wall design, and termed the program schedule unrealistic. The presenter pushed back against what he characterized as "can't do" thinking and insisted the program was credible.
Today, at the cusp of 2025, fusion has yet to demonstrate sustained net thermal energy gain for a time remotely compatible with electricity production, and the first-wall problem is still with us. Significant money has been, and continues to be, spent on fusion research and development and some progress has been made. However, nobody should be counting on fusion becoming commercially viable and deployable for electricity production in the next generation or two. If several major technical and economic breakthroughs happen before then, be happy and enjoy the unexpected boon, but in the meantime, make economic plans on the assumption fusion is not a viable option for the economically relevant future.
Great discussion, thank you gentlemen. My response to the question is, 'all of the above'. We need large standardized GW class reactors that can be rolled out on a serial basis allowing construction resources to go from one build to the next over time. We also need to create an environment in which SMR's and advanced reactors can be developed and rolled out on a serial basis. The use case for big units is for markets and grids that can tolerate large single generating units, assuming that the installed cost in $/kW are on target. That said there are many, many grids in the world that simply cannot tolerate large single unit generators. Alberta (AB) for example, the Most Severe Single Contingency (MSSC) is 466 MW. In simple terms, AB cannot connect a 1,000+ MW generator, because when it tripped off-line the AB grid would collapse. There are many grids in the world that also have this challenge.
For AB, assuming no changes on the grid side, 300 - 400 MWe Reactors would be ideal and 300 - 400 MW does not create too big a disturbance in the real-time energy-only electricity market there. For me, 300 MW is about as small as I think we should go. Nuclear technology is amazing, but the cost per unit of production gets high really fast as your total annual production number gets smaller. While the technology is amazing the overheads and one off project costs are eye-wateringly high; having fewer MWh/year in the denominator makes the overall cost of production very high on a per MWh basis.
@@lindsaydempsey5683 Co-located data-centers can soak up any power the grid can't handle.
I was also surprised nobody brought up MSSC issues with regard to 100% all large reactors. I believe this is a point in favor of SMRs, because there is no point to maximizing cost/unit generation if that is not the constraint.
@factnotfiction5915 Diesel peaker backup.
Loved this conversation. Thanks to all three of you!
IMO, nuclear plants don't need to worry about load following at all, just run all out all the time. Grids of the future won't be trying to match supply to load, instead, it's going to work the other way around. Loads will follow supply (eg VPPs managing EV charging, thermostats, and home batteries all to match available supply).
My guess is within a decade or two, a large percentage of customer premises will have several hours of BTM battery installed. They'll charge when supply is plentiful and grid utilization is lower and discharge when supply is constrained or grid infrastructure is more stressed. Locating storage close to end users has a triple benefit: 1) outage protection, 2) can optimize for highest average T&D utilization, 3) can optimize supply utilization.
This change will be great for nuclear heavy grids because plants can just run all out and not have to worry about the headaches and economics of trying to follow load. If nuclear can get its cost low enough, there will be no limit to the percent of generation nuclear can satisfy on a grid.
1. Build out at the maximum possible rate.
2. Match capacity with resistive shunts.
3. Outlaw metering.
Customers will figure out how to use the wasted power.
Instead of batteries, molten salt storage will likely be the best pairing with nuclear energy (like Terra Power's Natrium).
Indeed it is a waste of time to reduce output from Nuclear Power, and cause more wear on the hardware too, without much saving on cheap fuel. Many countries need heating- and hot water can be stored cheaply and in huge quantities compared to batteries. Hydrogen could be next, but it is more pricey than heat to make it, store it and use it.
"Right" vs "Y'know"
I'm not sure if I know
We need to build 10s of thousands of reactors.
We need SML reactors.
Small, Medium & Large reactors!
We know how to build large ones today, lets start building those large ones, and the small and medium ones will follow along when they're ready. In 10, 20, 30 years, we'll naturally land up with all sorts of funky reactors... As soon as we get a thriving nuclear industry and supply train up to speed.
Climate change isn't gonna wait for us to deliberate.
The sooner we've built enough reactors, the sooner we can actually turn off the coal, gas, oil power stations and industrial heat generators.
We also know how to build small ones, we have been building small ones for the military for many decades.
@danadurnfordkevinblanchdebunk small in size, but ginormous in cost, and astronomical in fuel cost.
Naval reactors are just not good for generating electricity for a grid. They're just so different, so specialised to their job.
@@RichHornerCheeseAndJamSandwich False. They are built for millions not billions, in two years or less.
@@danadurnfordkevinblanchdebunk
$/GW?
What fuel do they use? What enrichment?
There's a LOT about their design that is tailored to being able to ramp up from nothing, to everything, and from everything back down to nothing, in mere seconds (probably).
They're also keen on being pretty much surrounded by the biggest heatsink on the planet, i.e. the ocean.
And don't forget that their designs are highly classified, too. Unless you know the secrets?
They're the wrong tool for the job.
It sounds cool if we could use them, but they're just not what we need for grid/industrial electricity & heat demands.
@@RichHornerCheeseAndJamSandwich Sounds like an excellent design for load following. Some commercial floating nuclear reactors are powering coastal towns in the Arctic circle.
The old adage; "A bird in the hand beats two in the bush." You have only two options that have been built and are deployable today, CANDU 6 (Monarch is not ready) or the AP1000. Everything else will be a fight for regulatory approval, FOAK build, financing the a new FOAK, unproven Capacity Factors, etc.
For what it is worth, I think that one of the most constructive climate initiatives that the federal Canadian government could have done is to pay SNC Lavalin to update the CANDU 6 design to EC6. Then license that design through CNSC under the nuclear safety assessment part for use anywhere in Canada. Provide free technology use licensing to any province that want to build them and kick off a serial construction program across Canada. Some federal enabling legislation would also be a help, something to say any site that meets minimum national standards for environmental impact is deemed to be permitted by default as part of Canada's climate change initiatives as a streamlining of the normally tortuous EA process.
You forget the Korean APR-1400 (Barakah) and the Russian VVER (sure, a terrible partner, led by a war criminal, but they do provide some amazing financing).
@@factnotfiction5915 I don’t think that VVER would meet the‘Reliable Partner’ test, but they do seem to know how to build NPP’s. The Koreans on the other hand definitely know how to build, but also make better partners. I think that they have real potential.
Unless Canada develops the ability to enrich its own fuel, building an AP-1000 would be a terrible mistake. America is showing the world that they are an unreliable trading parent and cant be trusted, being dependent on them for fuel would be a stupid move.
Do not forget about Greenland comments.
Agree, but you could also partner with the French on fuel supplies as an alternative.
The U.S. does not come close to supplying its own demand for uranium enrichment, let alone supplying foreign demand. U.S. enrichment capacity supplies about 40% of U.S. enrichment demand, with the remainder coming from abroad. If Canada adopted enriched fuel, all its demand for enrichment services would have to be from non-U.S. sources. Further, the U.S. depends on Canada for natural uranium. Canada is the largest foreign supplier of uranium to the U.S., supplying about 30% of U.S. needs, so the U.S. is much more dependent on Canada than Canada is dependent on the U.S. when it comes to nuclear fuel.
@lindsaydempsey5683 yeah I think we could have a supply agreement with France, where we supply the uranium and they enrich it
Uranium enrichment would be a good industry for Canada
The best take away from this is that we should declare the debate over. (Well-scheduled conversation, Chris!) We need both of these present paths to come to successful fruition. Not in the usual "all-of-the-above, nod-nod to the wind turbines" sense, but because we are looking at two unique each-wonderful toolboxes for the future. Tech typified by the AP-1000 is mature and ready to give us the Terawatts (just a few random nickels from the gutter, please). Let's figure out the political economics, etc., and go forward. Meanwhile, advanced tech has so much to offer in terms of potential simplification of manufacture and new and unique applications. Let's support that, too, and go forward. It's too bad that this has even been ossified into a debate. The anti-water-it's-bad-old-school-and-it-can-explode meme should be buried right along side the it's-only-a-paper-reactor-you-clown-you-don't-know-what-you-are-doing thread. Let's get real, and work together!
After what Ted Nordhaus asked, I wonder. Do you guys know ?
I love confident predictions on schedule and cost for something that has never been built at scale. Still it is all you have as the stuff built at scale hasn't gone so smoothly
The great debate continues. I just want to put my cards on the table and say I fully agree with both sides.
32:40 "The *_Loan Programs Office (LPO)_* provides loans and loan guarantees available to help deploy innovative clean energy, advanced transportation, and Tribal energy projects in the United States. Over the past decade, LPO has closed more than $30 billion of deals across a variety of energy sectors." (US Department of Energy)
Thank you, if I would offer one criticism of this episode, it contained may too many unexplained acronyms
You know this man needs a speech coach.
Or, at least, regular strength-training and meditation.
"You know" preceded too many sentences. But YOU KNOW what? Overlooking that manerism, it's good information.
@acwojtkowiak Understood a simple joke was beyond the pale..
Well wanted to listen but after about 100 abbreviations I had to turn it off! WTF!
“This was great.” AMEN
Privatization of Nuclear Energy in US baffles me. Rosatom is a state corporation in Russia. You shouldn't expect private entities to take upon themselves these enormous multi-billion infrastructure projects, that take a minimum of 5 years to build if you're lucky, and it's more like a minimum of 10-12 years to build in the West in the atmosphere of hostility from the general public and general legal uncertainty from administration to administration.
Appreciate the detailed breakdown! Just a quick off-topic question: I have a SafePal wallet with USDT, and I have the seed phrase. (alarm fetch churn bridge exercise tape speak race clerk couch crater letter). Could you explain how to move them to Binance?
11:37 Hinkley Point? You mean the most expensive construction project in human history? at potentially 20 billion dollars per GW, if it ever actually gets built.
8:59 It's not even about building large. Russian RITM-200 SMR are 55MW. But you need the kind of regulatory environment and the kind of serious state company environment with international political/diplomatic connections to have a market and a will to make these things. Those were icebreaker power units. Old soviet technology. First they put it on a floating platform to power a town in the Far East of the country, for testing. Now they got a large order from Azerbaijan (and a few others) and they're beginning to "mass-produce" them. You can't expect a startup or even your "above average billionaire" to be able to do this on their own (at least not at their current levels of wealth which may sound large in terms of market capitalisation, but doesn't necessarily translate into that substantial of an RnD and manufacturing capability -- at least not compared to the task). Free market is not magic. It could be short-sighted. Someone needs to provide a long-term vision.
Oops. Clean industrial process heat will require 2-3 times more energy generation, and high temperatures reactors will be the only scalable source.
Electricity is a simpler way to provide process heat.
@@aliendroneservices6621 And 2/3rds of industrial heat energy demand is below 300 C and can be provided more efficiently with industrial heat pumps.
@@pauldietz1325*_"Efficient"_* always needs to be defined. Otherwise, statements containing the adverbial phrase *_"more efficiently"_* are meaningless.
Some 2/3 of industrial heat energy used in the US is below 300 C and could be provided by industrial heat pumps. Above that temperature, heat is quite storable (for example in SiC or nickel-doped chromia, which is its own resistive heater, or even sand at 1200 C), so electric resistive heat could exploit peaks in renewable output when prices are low. Former nuclear vendor Babcock & Wilcox has been working on a system for storing heat in sand at 1200 C using very compact fluidized bed heat exchangers.
@@aliendroneservices6621 > Electricity is a simpler way to provide process heat.
Heat is the simplest, and most efficient, method of supplying process heat.
Converting high-grade electricity to low-grade heat is inefficient, and doubly inefficient if heat was used to generate the electricity in the first place.
Chris I am curious as to WHY have you kept Canada's CANDU heavy water reactor option out of this large scale light water or small scale non-light water debate? I am especially curious because the Candu refurbishment teams are delivering on time and on budget?
I suspect the reason is that CANDU is more expensive up front than light water and because they are A, American so not likely to work with foreign designs and B, because the biggest constraint they are trying to argue around is initial financing.
@@jonmce1 is that true ? is a new build candu delivered on time and on-budget more expensive than a time overrun, cost overrun alternative? Or it it like the Avro Arrow? Clearly Superior but it sticks-in-the-craw being better smarter non-american technology?
@@alisonmcgillivray8008Paper airplanes *_with no engines, no armaments, and no mission_* don't stick in anyone's craw.
I think CANDU tech is banned in the U.S. because it technically has a 'negative void coefficient' of reactivity.
@RandyTWester Opposite. Negative is good. Positive is bad. CANDU is positive. PWR is negative.
The mission timespan directly leading to given quantity of iterations over some duration of time is a damn good argument. If anyone was winning it was Krellenstein although maybe Ted is right that everyone is underestimating the political question but i cant judge that. Unfortunately we've attacked and sued our reactor operators for decades. The damage of this may never be able to be undone and so a technological paradigm shift may be necessary for that reason even though ironically the technology gets possibly worse.
Nuclear power is finite. Uranium supply would be consumed within decades if it was producing equivalent power to Oil(cars, planes, trains, boats, tractors, factories, ...)
If you remove the false assumption that the transition to nuclear needs to be rapid, then the solution is obvious.
How cheap is the renewable/storage competition going to be if the putative nuclear transition is extended? Nuclear has to be sprinting down an experience curve if it's going to have a chance.
@@pauldietz1325 Renewable/storage is a fantasy. It doesn't exist.
The competition is low-pressure/high-temperature nuclear.
@@pauldietz1325 It is a short time preference which requires justification by the answer to that question, not a long one.
How can you remove that assumption when we are sprinting toward an extinction event…?
@@moqo There is no extinction event, I know that's disappointing for most Doomers, but you'll get over it.
You know, you know, you know,
I thought this a weird debate. It seems illogical to be talking either/or when it is logical to do both depending on location and capacity requirements. The debate seems to debating whether something that is already in work should be done. In Ontario the first BWRX 300 is underconstruction with 2 more planned. They are being constructed at the Darlington nuclear site so both environmental decisions and depth of experience exist already. It will be a model for other provinces to gain experience in such technology. More expansive use is expected to be used at old coal generating sites where environment and grid issuea are already established. A new large unit hopefully CANDU is planed to built at another existing power generating location.
We need to build new and established reactor design. Gen 2 gen 3 and Four. Both paths must be built out. We need more electricity and clean high temperature heat.
At this point, CANDU increasingly looks like the "safe bet" of a medium sized reactor that can always be built. Oddly, the shortfall in know how in the nuclear industry seems to be control systems these days, not so much the nuclear physics. But, regardless of technology, the stupefying amount of paperwork required to build and operate a reactor needs to come down as does the endless circus of continuous audits. There are way too many overheads. These overheads are prohibitive for any SMR design and they can't just be wished away easily.
Actually, I don't know.
Chris I know I could just Google this, but comments grow your channel. So could you tell me how much water does a Candu reactor use?
Approximately 80,000 liters of heavy water (D2O) for a CANDU 6 reactor.
canteach.candu.org/content%20library/candu6_technicalsummary-s.pdf
Ask an AI, make that dc work harder. Another billion from the tech bros :)
Water isn't *consumed* in any reactor
I understand that, but the water for cooling the spent fuel.
@@marblackCanada No water is consumed for the spent fuel. The spent fuel is kept under water for a period of time. Then in dry casket
Only low-pressure/high-temperature nukes can compete with combined-cycle natural gas, and the NRC will never let that happen. So, let's stick with gas until we can replace the NRC with a regulator that places equal value on safety and affordability.
Nothing can compete with CCGTs
One was built recently in the UK 2022 at a cost of $0.50 / KW and it has a very high efficiency of ~63%
Vs the reactors under construction in the UK at ~$20/KW
Thats 40 x cheaper
And they need ~15x fewer staff
Wind solar nuclear can't compete with that, so we either build CCGTs and accept the carbon emissions or subsiside and guarantee Wind/Solar or Nuclear
There is no shame or failure in nuclear highlighting it needs state guarantee and support because wind and solar do too
I'd like a link. AFAIK, CCGTs are more like $1/W rather than $0.50/W (which is what you meant). The latter is more like the cost of simple cycle gas plants. Did you leave off something from the cost, maybe the cost of the grid connection?
"That's 40 x cheaper"
Then Singapore (95% natural-gas-fired) should have the cheapest electricity. Let's check:
Singapore: $0.240/kWh
Taiwan: $0.090/kWh
Taiwan's electricity is:
Coal: 43%
Natural-gas: 40%
Uranium: 6%
@aliendroneservices6621 I was talking about the construction cost of UK nuclear vs UK CCGTs and it is about 40x cheaper to build 1 watt of CCGT in the UK you can look up both costs
Google
Keadby 2 CCGT cost
Hinckley Point C cost
@@kaya051285I was unable to find solid numbers for Keadby 2 anywhere. All I could find was SSE playing its usual game of *_"hide the numbers"._*
From the pdf "Powering Progress" (October 2020):
"Keadby 2 is one of the largest infrastructure projects to be built in the North Lincolnshire region. The investment by SSE Thermal associated with the project, from development through construction and to the end of its operational life, is expected to total £1.2bn"
How much was the total project cost? No answer.
*_£1,200M / 893MW = £1.34/W_*
Y'know.
So many y'knows.
Wow great show
1% per second above 60% is some pretty insane numbers
58:16 58:26 58:28
I would not be surprised if the reactor control system can make small, incremental reactor-power adjustments at that rate for the purposes of holding reactor power stable around a desired power level, but I would be surprised if the power conversion cycle (i.e., the main turbine generator, feedwater, and feedwater heating systems) can tolerate a sustained rate of power change that large for more than a few seconds.
@@thomasgreene5750 The French have been exercising their NPP fleet (including generators and feedwater) across large % changes every day for decades. This is basically old tech that probably deserves upgrades as improvements in control and prediction are made.
What we really need is large thermal storage/off-take to balance it out with less % changes on the electricity generation side.
@@factnotfiction5915 I'm aware that the existing generation of plants can load follow over a wide range and that they can make small changes rapidly. What I am finding surprising is the implication that they could run from 60% to 100% at 1% per second. That implies a ramp time of 40 seconds from 60% to 100% or from 100% to 60%. I find myself wondering how long the pellet-cladding interface can tolerate up-ramps that fast, and in my experience, down-ramps that large in power, but slower rates, say on the order of 3 or 4 minutes instead of 40 seconds, have run into feedwater heater water-level issues and feedwater pump NPSH issues.
@@thomasgreene5750 I see where you are coming from now.
> I find myself wondering how long the pellet-cladding interface can tolerate up-ramps that fast
There probably isn't much that can be engineered there, given one wants to use existing fuel formulations so I would presume that was not the limiting constraint.
> down-ramps that large in power, but slower rates, say on the order of 3 or 4 minutes instead of 40 seconds, have run into feedwater heater water-level issues and feedwater pump NPSH issues.'
But it may be that the new designs and new pumps can handle this feedwater constraint.
> the implication that they could run from 60% to 100% at 1% per second. That implies a ramp time of 40 seconds from 60% to 100% or from 100% to 60%.
We probably only have half the story here. Sure, between 60% and 100% power, the instantaneous ramp rate is 1% per second (which we need to remember is on the order of 10 MW for a 1000 MW reactor, so not insignificant).
There are probably secondary, understated, constraints, among which is that the instantaneous rate and the sustained rate are somewhat different.
Is there a map that shows which regions provide the required demand for nuclear power generation?
The demand comes after they are built. Provide the high-quality, reliable, cheap power-service first.
Never X! Xitter, maybe? 🤣
If decarbonization is the goal, then industrial
At what battery prices will solar and batteries enable an affordable load following profile?
So AP1000 is 1.1GW and roughly 8.8TWh annually
What would be the cost of a solar farm that can output 1.1 GW for 14h a day followed by ~0.8 GW for 10h a day. This would be better than baseload
Somehting like
1.1 Grid connection
4 Gw solar
12 GWh batteries
Would allow solar load following power station
Worst months Dec/Jan would be ~60% capacity factor (at the grid interface)
And then 8-10 months it can do the ideal load following of 1.1gW for 14h followed bt 0.8GW for 10h
If this load following solar plant can be built cjeaper than Ap1000s then all sunny regions will do the above
This is also super modular
Can build it 1000x smaller than an ap1000
1.1MW grid connections
4MW solar panel
12MWh battery
No good for non sunny locations but probably future for sunny regions
No. All sunny regions will install coal-fired and natural-gas-fired power plants, which is what they are doing now.
Y'know, um, ya know, there's too many you knows in this video ... I literally can't watch it any further.
Was just about to make the same comment.
The AP1000 will be the death of nuclear if that's the "best" option.
Why .. it's the most used
The first two AP-1000 units in China had a cost per kWe of $3154.
@@Fuad_ So, you're going to build them in the US at Chinese wage rates?
It might cost a little more in the USA but I dont think it will be anywhere near $16000/kWe
@@Fuad_ The wage ratio between China and the US is about 3.5, so ballpark it would be $10000/kWe.
what a useless yapfest. How many times is the bald guy going to say that nuclear has a 92% capacity factor, even when computer programmers work during the day and solar works during the day. There was nothing resolved here, just a bunch of yapping, no facts, no figures, The guy who likes big nukes didn't explain how they would be cheaper, and the little nukes didn't explain how it would be cheaper. Maybe we should just admit that the future is Natural Gas. Since you started this podcast, the entire world has buillt a natural gas and refridge and defridge terminal, and we sent both natural gas and other liquid fuels around the world.
Data-centers don't sleep.