Nuclear Energy // Deconstructing the Bear Case

[@kevinmorris9101]

I thought this was a good video. I'm a Nuclear Engineer in the industry. Most people get a lot of things wrong. Like most of your videos you are pretty spot on. However, wasn't very bearish ;-) Hit me up if you want for info on future videos if you have questions.

[@thelimitingfactor]

Glad to hear it! Thanks man Yeah, it's a cavernous topic. I had to place a lot of bounds on each video to keep them at a reasonable length 😁 The next two videos are scripted and have been reviewed If, by the end of the series, you see that there is a big hole in my logic, do please let me know!

[@gronkotter]

Yeah he basically dismissed the bear case from a gut feel.

[@akeslx]

Very insightful! It’s rare to see non biased information on controversial topics these days.

[@thelimitingfactor]

Thank man! I try, lol

[@m4miten]

Thank you for the fantastic videos! Wishing you happy holidays!

[@thelimitingfactor]

You're most welcome, and thanks for the support!

[@GeeWit]

Nice! So glad that you're diversifying your channel and analysis to other complex tech topics...

[@klauszinser]

Agreed. A new area Jordan has taken over and handled very well. Also its good to see that he challenges the ideas from Tony Seba who's expertise sometimes had not been taken serious. (For the German Green Party I question if they are capable to understand? Further, as with communism we had to realize sooner or later ideology will break down). Looking forward to the two upcoming videos on this subject.

[@GeeWit]

​@@klauszinser Yes, the root was my war back in the day. With communism, its progeny or incarnations, it's never about capability of understanding or reason - it always leads back to the ideology :: 1970's Cold War Russian linguist in Bayern (and other places).

[@brucec954]

The problem with all the existing nuclear plants in the US (not to mention coal) is that they were built in the 1960's and 70's and will require expensive upgrades to keep them going so the current cost of those plants will not be as low as they are right now going into the future. The current plants are also not fail safe like the proposed new plants because the power industry in the 1960's based them on the Navy submarine reactors which did not have to worry about losing their cooling water which is what happened in Fukushima. There is also the p[roblem of the nuclear waste / transporting it for safe storage which we have just been kicking the can down the road on.

[@thelimitingfactor]

Correct, but none of that cuts against the fundamental argument of this video Those are topics for future videos

[@aliendroneservices6621]

Concern about spent-fuel is an argument *_for_* a rapid and massive *_expansion_* of uranium-fired power.

[@john_in_phoenix]

Yes, almost all have actually been "extended" already beyond their original design service life. I think in a couple more decades, they will become far more dangerous and cheaper to build new plants. The waste problem is actually pretty easy, just costly. France "recycles" 96% of their spent fuel rods, and vitrifies the remainder. The US stopped recycling because of fears of proliferation. It's actually easier to separate fissile plutonium from the spent fuel rods than it is to refine weapons grade uranium. When India did this to produce a nuclear weapon, the USA overreacted and shut down the one small recycling plant in the USA.

[@bobwallace9753]

@@john_in_phoenix Those old plants become more dangerous every day. Exposed to radiation, metals become brittle. US reactors were designed for a forty year use. If we continue using them past their design lifespan we are eating into the safety factor the design engineers included.

[@aliendroneservices6621]

@@john_in_phoenix "India's first nuclear weapon was made from plutonium produced in a Canadian research reactor that was given to India for peaceful purposes" (Google AI) India did not use a commercial power-service reactor. Reactor-grade plutonium is not as easy to build a bomb out of, and such a bomb would not be as easy to maintain and deliver on target.

[@helmutshotthesheriff1942]

Thanks!

[@thelimitingfactor]

Your most welcome!

[@markumbers5362]

OK. Electricity demand may increase however in my case our family owns 3 homes. We have solar on all three and a battery on one plus 2 EVs. Over the year we generate twice as much electricity as we use and the roof area of our houses is only 25% covered. If solar and battery prices drop 50% by 2030 and retail grid electricity prices stay the same or most likely increase then that needs to be taken in to consideration.

[@thelimitingfactor]

I'm not sure what you're trying to say.

[@markumbers5362]

@@thelimitingfactor I guess if solar with battery prices fall by 50% that is way past the tipping point where being off grid is cheaper meaning a large percentage of city homes will not need the grid and pretty well every rural property. In other words even though electricity demand will increase that may not mean grid demand will increase. In fact it may decrease significantly.

[@4literv6]

Residential is tust 18-22% of total energy consumption though.

[@ryuuguu01]

​@@markumbers5362 This has happened in Pakistan already. Pakistan was the third-largest destination for Chinese solar exports2024 but the official (grid) use electric dropped.

[@markumbers5362]

@@ryuuguu01 It's interesting. Jordan of course is right with everything he has presented but I suppose it's not just the cost of generation. It's the personal cost to the consumer which determines whether the grid is even relevant. Another unexpected thing happened to us when we got our first EV too. My parter and I used to drive about 50-50 in our cars but when we got the EV, because it is so cheap to run we now use it 80% of the time. If other people are like this then oil demand will shrink even faster. Fun times ahead.

[@02dpoirier]

Great work on this piece! Excited to watch the series in its entirety.

[@thelimitingfactor]

Thanks man! Hoping it proves to be useful

[@robertosfield]

Nuclear "baseload" isn't helpful to grids, it just means dispatchable sources have to cycle much more to match the actual demand, increasing overall system costs. It isn't at all compatible with renewables as it just forces them to curtrail to compensate for the constant power output from Nuclear, again increasing costs. Really important aspects of modern and future girds not mentioned in video are interconnectors - these help transport excess generation from region to region, the distances can between countries like in we have in Europe or even far enough to handle the disparity of generation and demand across time zones.

[@aliendroneservices6621]

Nothing needs to be compatible with wind-and-solar.

[@thelimitingfactor]

I covered this in the video, but it appears you ignored it to get on your hobby horse

[@rayhardt]

"It isn't at all compatible with renewables as it just forces them to curtrail to compensate for the constant power output from Nuclear, again increasing costs. " Renewables need storage to get a reliable supply - if storage pays off with renewables - it also pays off with nuclear. And transport over huge distances? I am from germany - to get a grid that can handle the expected production peaks we have to invest about 600 to 700 billion euros for germany alone. And you dream of a net to transport excess generation from one region to another. With wind and solar the excess generation also happens when our neighbours do see excess generation from wind and solar and vive versa.

[@RangeMcrangeface]

Regardless of cost competitiveness, some minimum amount of nuclear needs to be maintained from a national security and scientific research perspective. We can’t have nuclear physicists exist only in academia.

[@helmutshotthesheriff1942]

Better than "easy answer", you always have THE RIGHT QUESTIONS. Again a brilliant video ❤️😊👌

[@thelimitingfactor]

🤠✊🏼

[@brentfriess]

Industrial process heat for things like chemical and cement production is one area that most renewable sources are not able to compete with nuclear on. I think this area holds the greatest potential for nuclear energy as well as for heat and power in northern areas that currently rely on diesel.

[@aliendroneservices6621]

*_Industrial process-heat_* is best provided by cheap, reliable *_electricity._*

[@stephenfidler1005]

Thermal industrial storage like Rondo have the advantage of 5 hours of electricity for 24 hours of heat. They will take the cheapest available electricity whether daytime solar evening wind or nighttime nuclear. They can also store excess and regenerate electricity. Also, where grid constrained, can colocate with solar.

[@jamanjeval]

How does the cost analysis work out when you consider not only operating cost but decommissioning cost and the long term storage of waste? What is the cost of upkeep and operating a radioactive waste storage facility for 10,000 years?

[@thelimitingfactor]

As I understand it, nuclear operators pay into a fund for that that's government run Decomissioning was included in the costs presented

[@jamanjeval]

“paying into a fund” to support a 10,000 year facility sounds a bit like “trust me, bro” economics. I’d like to see a deeper dive on what goes into the estimates. What does it take to plan for something like this? That’s about as much time as when Göbekli Tepe was first inhabited to now.

[@sportbikeguy9875]

​@@jamanjevalPretty sure they bury it in a tunnel and eventually just fill the tunnel with concrete to make it inaccessible. Not like they need to operate some facility until the radioactivity reduces

[@dirkohlhausen7671]

@@sportbikeguy9875what about ground water leakage, geological and tectonic shifts? You have to deal with that too, more than „cover it with concrete“….

[@yourcrazybear]

@@dirkohlhausen7671 "@sportbikeguy9875 what about ground water leakage, geological and tectonic shifts? You have to deal with that too, more than „cover it with concrete“…." You bury it in areas without tectonic shifts.

[@gianluigicassin868]

Very interesting and looking forward to the next videos. The only item I felt you did not cover is how many old nuclear plants can be revamped to feed the new demand, in the USA and outside. I feel existing nuclear will keep running also cause of the decommissioning costs. For new plants, I believe NIMBY is the biggest hurdle, even higher than Capex and financing, especially here in Europe

[@thelimitingfactor]

Good point! There's not a whole lot of nuclear reactors that are mothballed Nimby is one of the biggest hurdles But, that's the promise of advanced nuclear reactors

[@AnthahaAnthaha]

Nuclear power capital cost will get cheaper as more is built. The high LCOE cases are based on first of kind construction. If you are going to do nuclear, do it big.

[@Andre-95]

Not true, in the past when France built out it's nuclear it actually got more expensive as they learned more. Nuclear really don't have what it takes to get on an S-curve today especially vs solar.

[@peterfireflylund]

@@Andre-95 Definitely true in South Korea.

[@yourcrazybear]

@@Andre-95 "Not true, in the past when France built out it's nuclear it actually got more expensive as they learned more." Please do explain how more knowledge suddenly makes things more expensive. "Nuclear really don't have what it takes to get on an S-curve today especially vs solar." Solar have to be paired with batteries in order to make sense, but once the battery production is scaled up it will be a great power source in the future. But it still can't be the base power in regions more closer to the poles.

[@svnri]

@@yourcrazybear "But it still can't be the base power in regions more closer to the poles." It isn´t. Thats only Canada, The Nordics and Russia (and Alaska). Most of them use hydroelectric, geothermal and wind as base and have been for years. Iceland is already 100% renewable with geothermal and hydro. Pretty much the only one of these countries that isn´t a frontrunner in renewables already is russia.

[@thelimitingfactor]

Which is notably why included a Size vs FOAK-NOAK graph to tease what's coming

[@MrFoxRobert]

Thank you!

[@tonydeveyra4611]

If the problem with nuclear has been that they can't ramp production up or down, meaning that they end up overproducing energy during periods of kow demand...then batteries should benefit nuclear plants in a similar way that they benefit renewables. Nuclear plant operator charges its batteries during periods of low demand, then discharges batteries during periods of high demand to make more money, right?

[@thelimitingfactor]

Correct. In the past, they did it with pumped hydro 🤠

[@tonydeveyra4611]

@thelimitingfactor I thinknthat batteries are earlier in the cost curve than solar, so the price drop on batteries will probably be steeper than price drop in solar in the years to come. It would be interesting to see the comparison of how much really cheap batteries benefit existing nuclear vs existing older solar/wind that gets curtailed. I drive past the turbines in the San gorgonio pass on the way to palm springs pretty frequently, and the number of times that a small fraction (like

[@Battle_pigeon]

Looking forward to the SMR conversation. Seem well suited for AI facilities near cities.

[@thelimitingfactor]

I agree

[@kennethng8346]

One thing I've always wondered was why nobody combines solar electric and water desalinization. When there is heavy demand generate electricity. When supply exceeds demand, divert some of the energy to water desalinization. In essence you are getting the desalinization "free" because the energy would otherwise be wasted.

[@Haliotro]

Because you can sell the energy to somewhere further away where demand isn't zero... using it for desalination would in effect be throwing away money

[@TMKSilenced]

This is being done, but only a few instances to date because it is difficult to store hydrogen @ large scale. Some unique geologic structures exist to do so and are being developed. Look up Aces Delta (Advanced Clean Energy Storage) which is one of the developments.

[@edgewood99]

Why would you want a plant that can only produce fresh water if its sunny? It doesn't work at night...or in cloudy weather and if the solar panels are damaged. Water supplies need 24/7/365 production. ZPE is the actual choice we need to make: kzbin.info/www/bejne/iJXLoIx_n6dkqrs&pp=ygUNYXNodG9uIGZvcmJlcw%3D%3D

[@seancollins9745]

We need thorium salt reactorssl

[@jamesengland7461]

​@@patreekotime4578 aint nowhere so hot that most of a pond is going to evaporate 😂

[@NO3V]

Do all those prices include worst-worst-case, no-government-aid-whatsoever-no-matter-the-case full-coverage-including-liability insurance?!

[@thelimitingfactor]

It's not included for any of the technologies if that's what you're asking. All the technologies are subsidized out that whazu Oil, Renewables, and Nuclear. All of them

[@dirkohlhausen7671]

@@thelimitingfactorinteresting! In Germany we insure all our Power Plants. But didnt with NPP…. We as tax payers had to pay the „insurance“ in case of a problem…

[@alexisbono24]

@@thelimitingfactor I would also expect that the cost of the water supply for nuclear would not be included, also decommissioning costs, but the main point here must be that you cannot dismiss the excluded but identifiable costs just because they could apply to other forms of generation, as the proper methodology would be to include reasoned estimates for all of those. In Australia, the current proposal is for the Government to totally fund all the predicted (and unpredicted) costs for nuclear power, and to ignore the other costs as these would all be borne by the taxpayer. My expectation is that if those costs were included, the balance would shift well away from nuclear.

[@jonathanmelhuish4530]

@@dirkohlhausen7671 Fukushima has cost $87 billion so far to clean up... and they're not done yet. Insurance companies don't want to take on that much risk. Without strong government support, nuclear would never get built.

[@stephenfidler1005]

Thanks for a thoughtful critique of RethinkX. One technologhy may well have an impact is enhanced Geothermal. It ticks both 24/7 availability and the ability to vary the output. It is much more geaographically diverse than vanilla geothermal and the issue is how fast the price comes down. Anothervplus is reuse of all the oil drilling knowledge. Another factor to consider is how much latent storage will be provided by EVs which will enhance the financials of both variable and nuclear.

[@b0570nk4]

one thing worth mentioning is that as with any other technologies, new tech usually have chance to improve given enough time, my feeling is that there still are a lot of improvements waiting for us within the solar industry and not many people are willing to acomodate this fact while making predictions - an example that we can see if with ev batteries where like you said, its not just the cost of raw materials dropped heavily due to high demand but also the efficiency of the downstream technologies change, which cannot be said about combustion engine motors.. also, what are your thoughts on perovskites? do you think it could potentially replace the conventional solar panels in or at least add new use cases where this tech might be utilized where the standard solar panels would not be acceptable, in relatively near future? thanks again for your video contribution!

[@thelimitingfactor]

Sure thing man! I don't have a view on perovskites. I'm not that deep into solar. My view is that it will continue to follow a Wright's Law Learning curve, and I'll fast forward to the implications in the next video

[@FlorentHenry]

Several elements about nuclear power: - power/production can be adjusted on a half hour basis - not days. See RTE/EDF achievements on the topic. Caveats: design dependent, increases wear - production can be adjusted based on seasonality by shutting down reactors depending on seasons and their needs to either save money or schedule maintenance. Caveat: needs a centralized control of the energy production which is incompatible with the free market (monopolistic state owned energy company in France provided less costly energy than now, after being forced to open up to competition Also Lazard methodology on costs is highly questionnable. For reference, EDF is forced to sell current nuclear power plants electricity at 42€/MWh. It estimates a cost of ~50. New price will be set at 70€

[@thelimitingfactor]

Correct! A whole video could be made on that topic. Good points I focused on what mostly 'is' rather than 'can be' to avoid going down too many rabbit holes.

[@FlorentHenry]

@@thelimitingfactor I understand and agree with your thinking. There is a lot of information to master on the topic of energy to get the full picture. Hard topic to vulgarize with our short attention span

[@Clark-Mills]

Nuclear clean-up costs at reactor EOL and a nice target (dirty bomb) are a couple of things I'd like to throw out there. Solar BIPV & solar farms, both combined with V2G really will round out the weekly energy cycle. The yearly energy cycle will need a bit more thought; Mega-HVDC-grids and the likes of pumped-hydro if you can swing it. Actually, with enough excess solar electrolysis and it's wasteful conversion rate might have a place - especially stored underground without that high pressure compression and cooling. I think the UK is mandating V2G for EVs - Elon may not appreciate that as it eats into Megapack/Powerwall/VPP territory. Merry Xmas from NZ - we have a natural aversion to nuclear [as you'll know]... ;)

[@thelimitingfactor]

Thanks for the support! 1) Decommissioning was included and nuclear operators pay into a waste fund 2) Yes, there are definitely risks to nuclear. But, even with the accidents we've had, the negative externalities of nuclear have proven to be relatively low compared to other technologies 3) I'll start adding that complexity into the next video 💯 Nuclear has a tough path to walk if it's to compete! Thanks! Merry Christmas to you too!

[@mlsasd6494]

@@thelimitingfactorwe are yet to have a war with nuclear power plants as targets. In such a case we would probably be closer to a chernobly than a fukushima, so i would be cautious with risk statements. To this day you are warned to collect mushrooms in some regions in bavaria, Germany because of radiation and chernobyl is a wasteland to this day.

[@frankcoffey]

There has never been a nuclear power plant in the US that was finished on budget or on time. How do you get financing when you can't give an accurate prediction of how much it will cost and revenue doesn't even start until it's finished, and you can't even be sure when that will be?

[@thelimitingfactor]

They keep getting approved and built because either the sunk costs are so great or because people feel the project, whatever it is, is worth the extra money I say "whatever it is" because this isn't unique to Nuclear plants. This is just a common feature (not a bug) of project work.

[@frankcoffey]

@@thelimitingfactor Imagine how much easier it would be to get approval for a solar and battery install than a nuke plant. You could choose any cheap land you want , even in more than one location. Nuke plant could take years before you can even start building.

[@veronicathecow]

I wonder if the graph contains all the energy and costs involved in the decommissioning and the dealing with accidents?

[@thelimitingfactor]

Yup, decommissioning included, which I've already said in half a dozen replies and is also listed on the bottom of the Lazard page. I also addressed the accidents. All of these technologies are either subsidized or outsource some of there costs to the public in the health/ecology aspects

[@bobwallace9753]

The high installed cost of new nuclear makes it hard to justify. Most nuclear reactors, especially those in the US are old and nearing the end of their expected life span. The cost of a mix of wind, solar, and storage is becoming so low that it is hard to justify building new nuclear to replace worn out plants. Many grids will not price in climate change. For those girds we should expect a mix of wind, solar, storage, and CCNG. Combined cycle natural gas can serve as the deep backup for when there is not enough wind and solar to keep batteries charged. When the grid sees a coming shortage due to low expected wind and sunshine, CCNG plants can be brought online to provide additional supply. And for those grids that price in climate change the CCNG plants could be fueled with biogas or synfuel such as ammonia produced by wind and solar electricity.

[@thelimitingfactor]

Agreed! Nuclear has a tough road ahead.

[@RicksPoker]

Very glad to hear about this new series. Looking forward to the remaining videos. I hope you cover the small modular Thorium Molten salt reactors that they are building in China. (The first Thorium Molten salt COMMERCIAL reactor is funded and under construction. It is expected to come online in 2026.) Warm regards, Rick.

[@thelimitingfactor]

Yep! I'll cover more advanced reactors in the series There is a narrow pathway it can walk to success, but it's not guaranteed 🤞

[@aliendroneservices6621]

China is not building any such thing.

[@TMKSilenced]

Jordan, while true a Gas Turbine simple or combined cycle plant can be developed in a few years; i've seen others reference that all the manufacturers of the large frame GT's utilized in power plants have a large backlog which may have contributed to Microsofts choice.

[@kaya051285]

They are really cheap here in the UK we just built a CCGT recently for $450M for a 840MW unit That's just 54 cents per watt

[@thelimitingfactor]

Definitely! There's a backlog on a lot of power grid related things right now

[@rickrys2729]

As an energy analyst and engineer that worked on recent Chinese nuclear projects, I would not count out the RethinkX projections. Utilities will be slow to buy into new nuclear power after the experience of Southern Company on the Vogtle nuke plant (PWR), the enormous cost of the accidents of Chernobyl and Fukushima and the uncertain cost of new designs. China is researching various fuel cycles including thorium, but they have slowed nuclear and are building massive low-cost renewables. If a nuclear renaissance comes it will likely be a decade to scale.

[@aliendroneservices6621]

No one is building RBMKs. AP1000 is *_walkaway safe._*

[@rickrys2729]

@@aliendroneservices6621 AP1000 needs power and pumps to cool the decaying spent fuel in the reactor. NuScale and some other new SMRs have true passive cooling with no power required.

[@rickrys2729]

@@aliendroneservices6621 AP1000 needs power and pumps to cool the decaying spent fuel in the reactor. NuScale and some other new SMRs have true passive thermosiphon cooling with no power required.

[@thelimitingfactor]

I agree 💯 I'll get into the jucier stuff in future videos of the series

[@bluemalibu00]

Well thought out presentation. I wonder if the cost of storing nuclear waste for 1000s of years is included in the cost of producing electricity and how much of the mwh cost is for that. As of now most all of the nuclear waste is stored on site at the existing plants and there is no credible timeline for changing that. Shuttered plants are still storing on site and must pay to maintain the facility. Also is the cost of emissions of a natural gas included in the per MWH price? Solar, Wind, and Gas plants can be shut down and the land repurposed - nuclear can't - which is an inefficient use of resources. Nuclear (existing) plants can have accidents that are dangerous and have the capability of poisoning square miles of land - is this included in the per mwh cost?

[@aliendroneservices6621]

"Nuclear waste" is an argument for rapid and massive *_expansion_* of uranium-fired power-service, because: 1. "Nuclear waste" already exists. 2. The way to solve any existing problem is to make it bigger. 3. Everything is cheaper in larger amounts. Uranium is more land-efficient than any other fuel. There is no reason to ever decommission an entire uranium-fired power plant.

[@thelimitingfactor]

Thanks man! Nuclear operators pay into a waste fund

[@matthiuskoenig3378]

Is the cost of poisoning even more square miles of land included in the cost of solar? After all solar pannels create far more toxic waste per kwh than nuclear. And doesn't have anywhere near the safety requirements for disposal so it actually poisons the soil and water more often.

[@errorwww]

Although rethinkX bare case scenario probably won't happen, it doesn't mean any change in nuclear energy. It's very capital intense, mostly government build them with blank check from tax. Investors don't want to touch it. MS is different, coz Gates fan of the nuclear. But their original cost will skyrocket until they have a running nuclear power plant. Europe shows different examples. France had big difficulties coz nuclear need maintenance. UK and Finland will have a new power plan with a 3x-4x of the original price, not to mentioned the decade delay....

[@thelimitingfactor]

And Germany screwed itself by shutting all its nuclear down too quickly. This is the problem with being a fanboy of once tech over another. We have to think pragmatically what works in each individual situation.

[@addanay]

8 hope you're right. Nuclear should have more capital input. Yes renewable is better than current but not best. If we could combine the 2 or once we can milk it directly from the sun itself idk what's better than nuclear. What do you think?

[@manzourahmed3383]

Why not generate solar on residential and commercial (factories, warehouses. malls and carparks) where it's needed?

[@thelimitingfactor]

Certainly! and that will happen

[@FutureAzA]

A delightfully balance bit of thoroughly researched information.

[@thelimitingfactor]

Thanks man 😊

[@FoamyDave]

Not sure I fully agree with your counter analysis. RethinkX may not be 100% correct but I think they are specifically considering the reduction in rejected energy (65%) and I don't hear that in your analysis. Many reports of increase in energy demand includes the historical 65% of rejected energy. With wind and solar and batteries there is little rejected energy so it goes further (nuclear also has a low rejected energy number) . I do agree that fully depreciated nuclear will remain and probably for a long time becasue it exists (is permitted) and is depreciated. Also, many utility comissions give first priority to nuclear becasue it cannot be ramped so it gets that advantage whereas overproduction from other sources must be cutailed. However, new nuclear is, to date, unproven and despite increased public sentiment there is still long permitting and even longer and expensive construction. Finally, all the generic public sentiment is far from reflecting local accpetance when a utility proposes the actual construction near a community (NIMBY-ism). You certainly have given me something to think about but I don't think new nuclear is going to be a part of the equation in the end.

[@thelimitingfactor]

More to come. 🤠 As I said in this video, I am not anti-nuclear or pro-nuclear Just because I think the bear case is terrible, doesn't mean I think the bull case is great

[@beniaminosani2719]

every grid has a 30% of consumption that is always there. so modern, efficient nuclear would work for that. + modernised reactors have a decent flexibility already today

[@kaya051285]

AI data centres are best placed in cold locations for cooling reasons Nukes actually work best in colder locations too While solar gets less and less economic in colder northern regions

[@thelimitingfactor]

Bingo, latitude matters

[@aliendroneservices6621]

Cooling accounts for low single-digit percentages of total cost of ownership (TCO) of any given data-center. More important is *_proximity to data markets._*

[@fireofenergy]

My main concern is the overall EROEI of solar and battery, wind and battery and just whatever most efficient advanced reactor that could be mass produced efficiently, including the amounts of energy needed to make the nuclear fuel. It seems that the Energy Returned On Energy Invested for the nuclear would be higher, meaning that it would be able to supply the power for more build outs, and for humanity. So, we need to do an energy audit for the mining, refining and manufacturing, and recycling, for all the different parts and go from there. Intrinsically, the one with the least about of energy needed to deliver "x" amount of energy, over say, 30 years, would be the winner. EROEI is the absolute law, physics law! Since energy costs money, it seems that the concern over EROEI is also valid for monetary consideration, only after fossil fuels have been "shut down" by the UN, or whatever that stringently adheres to the UN net zero baloney (I don't believe regulations and lucrative carbon fees are going to help make energy cheaper). Thanks for your in depth videos!

[@thelimitingfactor]

Great perspective! I wish humans thought on longer timeframes. If they did, Nuclear would be the obvious choice. In place of EROEI we used capital and operational costs, which are flawed because it's usually bounded by like a 20-40 year timeframe

[@fireofenergy]

@@thelimitingfactor Thanks for the reply 😀

[@reinhardfuchs5181]

PV makes sense, but windenergy costs with incentives, grid-expansion, shadow plants, redispatch and service over lifetime at least 40Cent/KWh ! with dual-fluid reactors on market windmills are scrap metal !

[@aliendroneservices6621]

Wind and solar are both infinitely-expensive, on a sustained basis. *_No fossil-fuels = no wind-and-solar._*

[@dewibermingham816]

It's interesting that in the case of US electricity, that you are discussing, you barely mention CO2 emissions. Many other countries are driving for low carbon base load which effectively completely excludes gas because of the unknown, but certainly high, cost of removing CO2 from the exhaust stack.

[@thelimitingfactor]

I thought it was enough to say that Nuclear has the lowest CO2 emissions. Adding a page of content about CO2 would have been redundant. This channel isn't about activism.

[@martinv.352]

A great overview of the U.S. electrical energy market. I assume Solar installations gets weakened because of high tariffs. In countries with lots of sunlight and cheap labor costs, solar decreases to 1 cent, even you have to store 1/3 in batteries. Meanwhile storing 1 kWh once a day costs only 1 Cent. In Germany (I am from Germany), there was a political decision to turn off all nuclear plants in the year 2023. The newest plants were 35 years old and were designed for 40 years. But 2028 nuclear had been finished anyway. Shifting the end of old nuclear plans has got a technical limit. But the main reason for turning off the last 6 plants 3-5 years earlier is the problem with negative electricity prices. Last summer, we had one hour with minus 600 Euros on the hourly spot market. Germany has nearly no smart meters in the private housholds so you cannot shut down solar. Instead you have to shut down big plants. Turning wind turbines out of the wind works, but is not sufficient. In France with 65% nuclear energy, the first emergency shutdowns of nuclear power plants were necessary to keep the power grid stable. I am not sure if A.I. really needs so much more energy. It is a race between additional computer power and more efficient computer processors, which have to be specially designed for the new A.I. needs.

[@thelimitingfactor]

All good points worthy of discussion!

[@TheDesertraptor]

I beg to differ on solar/wind being cheap. I live in Adelaide, South Australia. SA is 70% reliant on renewables. We have the highest cost of power in Australia if not the world. I am currently paying 47c per KWh during peak and not much less during off peak

[@thelimitingfactor]

That's why I was very careful to talk about the whole grid picture and the niches of each generation type Solar and wind aren't cheap if that's all you build It takes a combination of generation technologies to make it cheap That's why I so heavily stressed this in the video

[@TheDesertraptor]

@thelimitingfactor SA like many states has an interconnect between states. The other state still has majority brown coal power which helps keep our lights on when our largest gas plant can't fill the gap.

[@SolAce-nw2hf]

Get your own solar panels and a home battery and don't pay the 47c per kWh. I assume these are Australian Dollars, so it's about 0.29 US dollars per kWh. This does not sound too bad if it includes keeping those expensive peak power power plants online to ensure grid stability until energy storage can take over.

[@Mrbfgray]

Amazes me that Three Mile Island is even feasible to reopen. My crude understanding is that Diablo Canyon plant is only running at 1/3 of it's "design capacity", if we had any respectable leadership here in Commifornia--that would be upfront on the table.

[@GruffSillyGoat]

In any multivariate analysis approach, particularly in complex ones like the energy markets, one needs to consider all varying factors on both generation and demand sides. The assumptions on the demand side seem to have had a more stable set of future assumptions, particularly the treatment of the growth curve in data centre/AI demand, in constrast with that of the generation side where more variability was considered or perhaps presented in the video. This lead to a scaling imbalance, or perception of such, on the demand side that produced a potential artefect of the modelling performed that sustained the conclusion that favoured sorucing energy from existing nuclear generation vs the disruptive new generation competition. In particular, the energy demand of data centres as well as the technologies use to perform AI computations and the corresponding levels of energy consumption are all highly active and heavily invested areas of research and commercialisation activities. The demand curve of the data centre used is likely to be much shallower than shown as a result of these endeavours. Examples of activiy that come to mind are: - server and network device consolidation - such as the use on chip/platform photonics for communication (such as Lightspeed's Passage) significantly consolidating processing units onto wafer level components (rather than server level) using low light based communication fabrics speeding up chips and removing the energy expensive (and cost expensive) use of dedicated fibre networking devices. Basically, more bang for your buck via a reduction in the number of servers and external interconnections required for the same energy footprint; - enhanced processor architecture - again either using photonic processors (Lightspeed and others), specialised AI dedicated processor fabrics (Nvidia) or quantum chips optimised for AI (google). As well as many other's either in R&D phase or pre-commercialisation, all three of the ones named have announced comerciallised products that if adopted can reduce the energy demands of AI per interaction. More power for less energy; - progress in AI (and general data centre) software algorithms and processs - AI is maturing and to date has focussed on the what can be done and how, this is shifting into the what-with considerations to optimise processes not just for range/accuracy of results and computation speed but also for processing demand usage. Areas such as quantum AI, predictive forward pruning, early path elimination and contextual clustering can improve outcomes and also reduce the degree of explore everything and then summarise the results brute force used in AI today, reducing energy use per interaction significantly. Do things smarter/differently, more effectively and more efficiently.

[@aliendroneservices6621]

Greater efficiency = greater power-service demand.

[@GruffSillyGoat]

@@aliendroneservices6621 - Is this a comment on use of English, or an observation that with improved efficiency comes more demand?

[@aliendroneservices6621]

@GruffSillyGoat The latter, obviously.

[@GruffSillyGoat]

@@aliendroneservices6621 - agree, demand expands supply, which drives innovation to reduce costs of which energy use has become one of the dominant factors in IT. Although in IT there is a quasi-finite limit when the cost of not doing outweighs the cost of doing something, then the deamnd tends to settle at the _just goodenough_ level, well at least till something new comes along.

[@aliendroneservices6621]

@GruffSillyGoat The limit to power-service demand on Earth is already the thermal limit, which is around 200,000 terawatts (10,000x the current average global thermal power level). You're not going to reduce demand for uranium-fired power plants by making compute more efficient. *_A data-center is a machine which turns electricity into money,_* and there are no limits to that.

[@seancollins9745]

We need thorium salt breeder reactors, we have millions of yrs of thorium and we can decimate our current waste stockpiles into useful isotopes or starting fuels

[@thelimitingfactor]

The issue is extracting that material from the Nuclear waste. No small feat Either way, I'll be covering advanced reactors in future videos

[@seancollins9745]

@@thelimitingfactor Actually it's rather simple chemistry, Kirk Sorenson " flibe energy" covers it in detail. you might want to look him, up and interview him. dude is brilliant !!!! I would love to see more investors behind him and with the new admin coming in hot and heavy for nuclear " required for EV's and AI to be brutally honest along with massive grid upgrades, yeah we need that super bad. So I think 2 places to invest, grid infrastructure supply chain and nuclear base load particularly thorium. We already figured it out in the 1960s

[@YellowRambler]

The main problem with Thorium Molten Salt Reactors is bureaucratic red tape. China does not have that problem with bureaucracy, and will own the Thorium Molten Salt Reactor market unless something is done about the bureaucracy.

[@aliendroneservices6621]

@@YellowRambler There are many known problems with the thorium molten salt power-service reactor. One is the poor *_capacity-factor._*

[@YellowRambler]

@@aliendroneservices6621 Please define “capacity-factor”

[@deanmcmanis9398]

Thanks for covering the topic of nuclear energy technologies. Looking at this as a series of problems that need to be resolved, and demands met (as opposed to which camp you want to support) will better serve everyone's needs. This program was a great beginning, and I look forward to seeing your next program in this series!

[@thelimitingfactor]

You're most welcome!

[@frax7696]

What about nuclear fuel - is there anough to increase the power capacity? To get rid off the nuclear waste isn't cheap and you ignore it for kWh costs!

[@thelimitingfactor]

Nope, that's factored in. And yes, there's plenty of nuclear fuel. The sourcing isn't ideal though

[@aliendroneservices6621]

The Earth's crust contains 75 trillion tonnes of uranium, which is *_10 billion years' worth_* at the current global all-fuels burn-rate of 20 terawatts (TW).

[@testboga5991]

If solar is made smart, it can just be shut down if there is too much energy. Would be interesting if we ever get enough storage capacity to buffer the grid at a large scale.

[@jimurrata6785]

Why, when storage is far, far cheaper than having capital sit idle. Shutting down solar doesn't increase lifetime like it would with a thermal source subject to mechanical wear. If fact the opposite is true, because then *_all_* insolation has to be adsorbed as heat, rather than some 20-30% being converted to electrical energy.

[@frankcoffey]

If you have enough batteries any excess energy can be stored for later use. Those batteries are cheaper than a nuke plant and can be brought online very quickly. They also don't need years to build out and can be expanded at any time. Batteries can even be located in various locations to create a distributed storage system with no single point of failure.

[@jimurrata6785]

@@frankcoffey Yep. And the other thing is that grid tied batteries don't really have to be that energy dense. Because they're stationary they don't care about NMC chemistry, there are far more stable, deep cycling and less costly alternatives that work just fine. All these people that want to change electrons into heat, or hydrogen, or pressure, or.... have absolutely no grasp of thermodynamics or RTE. Battery I/O is 90% or more!

[@frankcoffey]

@@jimurrata6785 With the right management software even used EV battery packs work great. One company has some that will manage down to the cell level. Getting every last bit of service out of old battery packs.

[@dzikdziki2983]

Wind and solar have high operating cost because they loose efficiency fast and needed replacing

[@joeseward3422]

Thank you Jordan for all your hard work at helping to reduce FUD. I appreciate your level handed approach at looking at our ever growing need to explore and develop our energy needs. IMO most people don't understand that the cleaner and cheaper we can make our energy the greater we can potentially increase human flourishing.

[@thelimitingfactor]

Sure thing man! Thanks for the kind words

[@mjoelnir1899]

Regarding your picture or graph regarding capacity factor, I strongly disagree. Here in Iceland geothermal is above 90 % and looking at what France experienced in the last years, nuclear power can not claim 90 % under all circumstances. Hydropower capacity factor is not dependent on an inherent factor but on the design. Is the power station designed to run steadily or designed as a top station. There are hydro power stations designed for a nearly 100 % capacity factor running steadily with little storage and others for 25 % with significant storage in reservoirs, both have their use. For many power station the point is not an inherent capacity factor, but how they are used. That will influence their real use capacity factor. Nuclear has big limitation in regards to be able to follow the up and down of electrical usage, as they are limited detachable and should run at a steady output. Wind and solar have a varying capacity factor. But they are rather cheap to build, have very low running cost and when their is overproduction, they can quickly be shut down or limited. There variability is predictable, as there are good weather predictions even weeks in advance. I agree with nuclear will go out of use. The main point is the cost of new nuclear plants. The investments are very high, so high that low running cost will not compensate for that. A lot of the current nuclear power plants existing today, were build with very high subsidies by the various governments. I do not expect more plants to be build without more subsidies, that dwarf any subsidies used to to start of wind and solar. The cost of power from new nuclear power plants is, according to Lazard, about double the cost of new not subsidized wind and solar. I would claim that a combination of solar and wind with cheap gas turbine power stations, will give the same security to a grid, as nuclear power and gas powered top stations, for a far lower investment and lower electricity prices. With time we have to throw into the mix storage, today mainly batteries, and non variable sources of green energy like geothermal. I do not believe in shutting existing nuclear power plants down before they get aged out, but those days will come and they will go of line one after the other. The comment, that solar and wind need backup power, is a red herring by the nuclear aficionados. Somebody bringen this argument is rather hypocritical, because nuclear does need it too, because it does not follow the ups and downs of the electricity usage. So if you add that extra cost to solar and wind, add it to nuclear too.

[@thelimitingfactor]

There are always exceptions. Overall, the information in the video is accurate in most cases. To tease out every point in detail would take about 4 hours, provide little additional value, and no one would watch it

[@mjoelnir1899]

@@thelimitingfactor I am not talking about teasing out every detail. But one should stop presenting some myths, that are old and long ago refuted. Yes, one should not shut down nuclear plants before they age out, the investment is done and it is an unnecessary waste of resources that can produce cheap and carbon free electrical power. No, there is no sign, also not from the small modular reactor crowd, that nuclear power stations will be able to compete on price without huge subsidies. Yes I know that wind and Solar is subsidized in many regions, but so are fossil fuels, and nuclear to a degree that has not happened again. Today you can build up solar and wind without subsidies and compete on price.

[@jackgoldstein9297]

Great article, but I believe you missed several key considerations. Nothing is mentioned of the overall cost of nuclear, which has in the past been plagued by years of permitting hurdles, years of delays, and still no good means to store waste. There is a good deal of talk about small reactors and how they are better. They will likely still have large permitting hurdles, which are only appropriate considering g safety concerns. Geothermal energy, eg. Fervo Energy has great baseload characteristics, but unlike nuclear energy, it can be easily throttled. Their first plant was a test, only 4 MgW, but they are now building a 40 MgW facility in California. Permitting should be much less cumbersome, and they are using existing tech from the oil and gas industry to make electricity from steam turbines using horizontal dilled wells dilled below 5000 ft. below ground. They were started through funding assistance from Google, precisely for Data Center power. There is no nuclear waste, and it uses common steam turbines for power generation. There is less waste heat, which has been an issue with nuclear plants when cooled by ocean water. Nuclear will be around for a while, but unless there are dramatic improvements in safety and quicker permitting, Think X is likely close to what will play out.

[@thelimitingfactor]

I didn't cover those things because they weren't relevant to this video. That is, they didn't affect the primary point, which is that Nuclear isn't going to all be shut down by 2030. That's five years from now and plants are being restarted rather than shut down, and because of, interstingly, gas rather than solar and batteries. Those points will be covered in the upcoming videos because they're finer and more nuanced points to discuss. As per the introduction, this is a multi-part series.

[@peterjohn5834]

Great logic and considerations. As a comment if existing nuclear is found to be economic and new nuclear is uneconomic, then there is no incentive to produce more nuclear. Especially if the shear scale of the solar and wind integrated by China completely dwarfs nuclear. Further I think the China model will spread as corporations are only interested in their potential returns now and certainly not investing in something that will be even more relatively expensive in 10 to 15 years time.

[@thelimitingfactor]

That's what the next two videos are for 🤠

[@stephenohki7276]

I think a lot of great points are raised. I did not hear the "NIMBY" issues of new nuclear raised as in the US, although the public may generally approve of the concept, permitting, construction, waste etc is a very sensitive area. Would be interesting to hear your thoughts on that concept in future videos.

[@thelimitingfactor]

I'll be mostly focused on the economics in this series. That's because if the economics work, they'll find a way to get them built. I will be getting into the permitting, regulation, and construction side of things a bit though 🤠 There's some interesting points to be made there

[@kaya051285]

Average nuclear power station in the USA has 1.7 reactors if that could be increased to 3 per station that would make space for 70 additional reactors 3 should be possible in France it's 3.1 reactors per power station

[@iandavies4853]

@@kaya051285more fun when one fails too. Can get cascade of failure.

[@thelimitingfactor]

@@kaya051285 Excellent point.

[@Longsnowsm]

Thanks for the video, glad to hear you covering this. Nuclear is a big deal if it can be ramped fast enough to meet the demand. Sounds like you are listening to the people in the know and same group I am following. It looks like they need to sorted out the cost issues. SMR's are supposed to be the answer, but they haven't yet demonstrated the cost savings, and speed to market. So now we wait to see who is going to get projects approved and can actually execute to get it done. Right now there are too many people with their hands in the cookie jar. Nuke doesn't take too long, we take too long. We are watching China executing nuclear and quickly. Nuclear is a good baseload and really does offer a hybrid renewable mixed with nuclear for the biggest cost benefit. I think DOE is covering it right now with that angle. The cost of electricity is climbing fast so the arguments that nuclear is too expensive does not appear to take into account how expensive electricity has already. The next thing they have not taken into account is over production of nuclear energy can be used in desalination and hydrogen production. Those advocating hydrogen could get their lower cost green hydrogen source by tapping that excess production. There are more revenue streams that can tie into nuclear. The AI and electrify everything are driving the huge increase in demand. If big tech is making commitments to nuclear it is a clear indication they expect the demand to skyrocket and are willing to pay up for it.

[@thelimitingfactor]

I'll be getting deeper into this in the next couple of videos 🤠 Good points

[@boatthere867]

for myself i've always look at using nuclear power for the base load and use solar and wind with batteries for the high peaks in enage loads

[@kennethng8346]

The trouble with solar and wind is that their peak production does not match peak usage. What we need is to combine them with grid sized energy storage. One suggestion is to replace some of the ballast wind turbines have with battery storage.

[@JohnboyCollins]

My thoughts on nuclear: it's most interesting when you think about moving nuclear fuel. If you think of energy as a network with latency and bandwidth constraints, nuclear arguably represents the ultimate in high bandwidth, high latency. Grid power is ultra low latency, low bandwidth. Chemical fuels are in between. For example, if you were to use the entire Pacific DC Inertie (largest hvdc line in the US) to charge a Spacex starship, it would take several hours to provide that many BTUs. But with a few LNG trucks you can get the ball rolling. Likewise, the colonial oil pipeline in the US provides about 300x the energy bandwidth per dollar as the Pacific DC Inertie, and about 1/10th the loss per mile. But portable micro-reactors, for examples, could represent a whole new level of energy bandwidth far beyond LNG and diesel trucks, potentially opening up new application like high-energy mining in remote locations. It is a mistake to think of energy as a static network. Highly dynamic allocation and deployment is crucial for a highly productive economy. Nuclear as a "base load" technology is less interesting imo. Long-term power satellites will be hard to beat for base load.

[@aliendroneservices6621]

The problem with micro-reactors is neutron-economy. Once you're dealing with all of the neutrons being produced, you might as well have built a *_large_* reactor.

[@thelimitingfactor]

@@aliendroneservices6621 BINGO!!!

[@JohnboyCollins]

You both seem to have missed my point.

[@larsnystrom6698]

It seems odd that new nuclear power would be much more expensive than old, fully depreciated nuclear power. Long-term economic calculations don't work like that!

[@thelimitingfactor]

I'm not sure what you mean Nuclear was on a cost decline curve, and then it reversed in the '70s due to regulation and catastrophes

[@TheDesertraptor]

Australia has a ban on nuclear yet our opposition has put nuclear on the table for the upcoming election. Most Australians are over paying high prices for power so are getting on board with nuclear but if I am right you are saying our best option is to increase gas generation which is now only used as backup to renewables.

[@SolAce-nw2hf]

In the Netherlands we are building some more gas power plants just too handle the peak loads. We have a lot of renewables and only one small nuclear power plant. Estimates from our national expert are that buidling two new nuclear power plants will cost about 80.000.000.000 euros and will take about 15 years to complete (3200MW each). Building energy storage, more wind turbines or more solar panels is all cheaper and faster to complete, but the current government still wants to go ahead. As for Australia, it has zero experience with nuclear power and much more sunshine and space for solar and wind power. Why bother with nuclear? It's not going to be cheap and certainly not going to be running before 2040.

[@Gjehcyekcgev]

Properly arranged data can help you support any decision. I don't believe you are attempting to arrange the data to achieve a specific outcome. That level headed approach to data Is why I enjoy your channel, well that and the intro music 🙂

[@thelimitingfactor]

Bingo. Thanks man!

[@microsrfr]

Please include insurance (which is provided by US government), security costs (such as US coastguard 24/7 protection of waters near an ocean-cooled plant) and deca- billion dollar decommissioning costs. Also spent-fuel reprocessing or secure ultra long-term secure storage costs.

[@thelimitingfactor]

I already covered that other comments That was taken into consideration That is, it's worthwhile assuming that you're not having an original thought and checking what other people have said

[@ronmorrell9809]

Does the cost of new renewable power include grid expansion? Several projects on Indian Reservations have been approved for Federal stimulus funding. Because the grid can't carry the power, the stimulus budget will expire before Power Purchase Agreements can be signed. Balancing the grid including timing of AC 60-Hz frequency is easier if massive turbines are physically spinning with momentum. TerraPower, is building a nuclear plant in Wyoming. It will replace coal as a power source, but continue using much of the existing turbines, alternators and grid connection. As a bonus, it is variable. It heats a large tank of molten sodium. During peak demand, water flow through the sodium is increased, generating more power by cooling the sodium During low demand, the cooled sodium is re heated. In contrast to wind or solar, nuclear generation sites can be selected close to the site of consumption.

[@thelimitingfactor]

The economics of each site are unique

[@jonathanmelhuish4530]

Do the nuclear cost figures include the cost of long-term storage of nuclear waste? Often that cost, as well as the 'insurance' of a nuclear plant, seems to be born by governments, in additional to the generous subsidies they usually receive for construction and/or high guaranteed prices for energy produced. It therefore wouldn't surprise me if these costs aren't included in the price, as nuclear operators rarely pay them.

[@thelimitingfactor]

1) My understanding is that, although government managed, operators pay into a nuclear waste fund. 2) As for decomissioning, Lazard includes those costs. 3) The 'insurance' and environmental cost is a good point, but that's where next generation nuclear reactors come in for new builds. Beyond that, doesn't the government always step into clean up major accidents by any industry? Oil spills, etc It's just with nuclear it's concentrated rather than distributed compared to the oil industry. And, it's also true of renewables like wind power. I just learned the other week what's happening with all the old wind turbine blades. Yikes. That is, it's easy to target nuclear, but it's an issue that seems to apply to all the technologies in one way or another. Solar might be the exception, but I'd have to do some research on what happens to those at EOL.

[@dirkohlhausen7671]

@@thelimitingfactornever buy something on what they say it can be, by it for what it can do right now. And right now there are very very few to none next gen commercial!! reactors… so i wouldn’t believe one thing they promise (same with Fusion)

[@peterfireflylund]

@@thelimitingfactor another point is that the nuclear industry shouldn't pay for unnecessary worst case costs imposed by the state (Fukushima). The state has a tendency to be extra, overly cautious (to such an extent that the overreaction is much more dangerous than the plant failure). If the state wants to do that for political reasons, the state should pay for that.

[@UlrichHarms-ci1ov]

The supposedly include some money for decomissioning and waste disposal. However in the US the actual cost still have to show. So far in Europe the cost there were well higher (like 2-5 times) than expected and planed. Chances are this also applies to the US with well high costs than the money set aside so far. Insurance and much of the developement costs are government payed so far.

[@jonathanmelhuish4530]

​@@thelimitingfactor Sure, the price of cleaning up oil spills should also be bourne by the oil industry and included in the cost... though as far as I understand it, they do pay for it via their marine insurance? Nuclear really does seem to get special treatment here, because the cost of cleaning up after nuclear disasters (e.g. Fukushima = $87 billion so far) is so enormous that no insurance company would cover it. And really think the government swoops in when a wind turbine falls over or somebody smashes a solar panel? Or that even if they did, the costs would even be in the same ballpark? Sorry but your bias on this topic is really shining through, even when you try to present the "bear case"!

[@jimurrata6785]

Im really interested in how quantum computing innovation/adoption may blunt the increasing energy demand of AI data centers over the coming decade, if these projections are based simply on the flops required. Regardless, i see the premature shutdown of currently certified nuclear as freaking absurd waste of of an existing resource because of 'boogie man' kneejerk reactions. (Yes, im looking at you, California, Germany, New York...

[@thelimitingfactor]

Quantum computing as I understand it is a completely different use case Totally agree with the rest though!

[@jimurrata6785]

@thelimitingfactor I guess that depends what you're asking it to do. Right now there's a bunch of parlor tricks where "AI" is being used as a catchall for generating jabberwocky and deeply flawed nonsense in response to simple query's. That's simply building towards machine thinking and a 'conciousness' in silicon. The LLM's seen today can't be trusted in any case, but where it has potential is in physics, chemistry, biology and deep simulations that use tremendous amounts of supercompute power. Think what's going on at places like LLL & Bluffdale. If you're trying to simulate fusion, fluid dynamics on a global scale to run forward _and_ backward, or tease out connections in infinite data points. Today _these_ situations require tremendously complex algorithms and an incredible amount of power. But right now each of us has the equivalent of yesterday's supercomputer in our damn pocket, able to run on a few ah of battery and deliver results in fractions of a second. Closed systems don't require power for processing, only to create an operating environment.

[@davepermen]

Well, I'll wait till you're over that nuclear phase. I'll come back to watch once you talk about relevant technologies that actually matter.

[@thelimitingfactor]

Looks like you'll be back for the next video then 😂

[@NaokiWatanabe]

Why compare fully depreciated nuclear energy against new solar/wind when there are fully depreciated solar/wind systems on the grid as well, and that number is only growing and accellerating as those projects pay off relatively quickly. By 2030 when solar/wind/battery systems have dropped by 50% you're going to see many times more fully depdeciated solar capacity on the grid as well. It's fine to look at depreciated costs but many nuclear power plants are already into their first or even second life extension and facing increasingly large operational/maintenance costs, they will have to compete with 5-10 year old renewable projects outputting power for next to nothing. Deprecated nuclear won't be competing with new solar in 2030+. The more important note though is I think you're a little too bullish on the AI use-case. Datacenter electricity demand may double by 2030 but will remain a fraction of the nation's demand, perhaps ~9%. For context residental homes account for 38% of demand today. We have to build out infrastructure to support everythihng and it may make sense for datacenter operators to take advantage of those economies of scale rather than trying to operate their own nuclear industry off to the side. Looking at the Microsoft example, lets be clear they are investing ~10x more into renewables than they are into nuclear. Back in May, Microsoft signed a deal with Brookfield Asset Management to invest more than $10 billion to further develop renewable energy capacity and already has something like 34 GW in contracted renewable energy in 24 countries. Google, Meta, Amazon, and others are investing in renewables at similar scales. It's clear from their collective investment strategies where they think this is going. And Microsoft did not announce that they were restarting a nuclear plant, to be clear. They announced a purchase deal with Constellation Energy and there are assumptions and caveats attached to that. 1. Constellation Energy has not guaranteed the plant (specifically, Unit 1) will be up and running by the 2028 goal line. 2. It's currently only permitted to run to 2034 and they need a 20-year extention for this to work. 3. Unit 1 was built in 1974 and would be 54 years old in 2028, 60 in 2034, and could be 80 years old in 2054. Time will tell of course but I think projections for nuclear energy set by the World Nuclear Energy Association and International Atomic Energy Agency are probably accurate. At best, if output is tripled, its share of electricity generation may remain the same as it is today.

[@thelimitingfactor]

Because the vast majority of nuclear is fully depreciated, which is why I said that in the video Because solar and wind are on on an exponential curve, the majority of those are new generation Additionally, the whole point was to say that nuclear isn't going to shut down in the next few years The argument you're bringing forward has no place in that narrative That's a totally separate discussion that will occur over the next few videos That's why I made it a video series, because it's complex topic

[@cottsak]

It might be transparent to folks not living where I do, but here in Australia there’s been a political discussion raging for the past 12+ months about the future of nuclear generation on Australian grids. Putting aside the massive burdens like repealing laws which prevent the use of generation reactors etc, I find most of the assumptions in your video here Jordan to apply to countries with extensive experience in existing grid-scale nuclear generation. Maybe South Korea is another good example which compares to the US. But in places like Australia where there is zero experiences and track record for nuclear generation, and therefore no evidence to support the levelised cost over any timeframe, we can’t assume any kind of competitive economics for a nuclear mix in our energy in the next 20 years. It’s just going to take too long to see these numbers flatten out. This is why I think we won’t see commercial investment into nuclear in AU any time soon. It’s more likely that here in AU, renewables + storage will economically “compete” with nuclear simply because nuclear isn’t an economically viable option in 20 years. What do you think?

[@thelimitingfactor]

I totally agree! One of the things working against nuclear is that humanity just doesn't consume enough power It's only viable in the largest countries

[@peterfireflylund]

You don't have to develop your very own bespoke reactors, you know. South Korean reactors are on the cheaper end so why not buy some of those?

[@yourcrazybear]

@@thelimitingfactor "One of the things working against nuclear is that humanity just doesn't consume enough power " What do you mean here? Enough power for what? A massive build out of nuclear in order to reduce the construction costs?

[@garymenezes6888]

Anastasi In Tech, just dropped a video about new chips in development, that won't need nuclear power stations to power data centres. So the race is on, more efficient chips or more nuclear power stations. kzbin.info/www/bejne/omqslGOGo6-facU

[@aliendroneservices6621]

Greater efficiency = greater power-service demand.

[@thelimitingfactor]

Why not both?

[@garymenezes6888]

@@aliendroneservices6621 Not necessarily, when the UK switched it's lightning to more efficient LED bulbs, the power demands of/from the grid dropped off.

[@sergiotarsiero]

While i think nuclear Is doomed on the long run, running existing reactors for some 20 years After the programmed end would be a Fair enough idea. I'm not convinced about the catastrophic predictions on Energy consumptions, since a lot of patents on industrial activities are popping up that drive ti consistent Energy reduction (i.e. iron and concrete for example) while materials evolve at rapid pace. IA drains a lot of Energy, yes, but actual processors are not rally optimized for IA nor are algorithms.

[@thelimitingfactor]

Could be! I'll get more into that next videos

[@fredkroh6576]

What also can be added to the mix, apart from data centers, are manufacturing processes that can be switched to cheap electricity supplies (solar & nuclear during the middle of the day and wind & nuclear at night). This would include steel production using hydrogen reduction and arc furnaces and fertilizer production using hydrogen produced from electricity rather than natural gas. Optimization comes from utilizing excess capacity, not throwing it away.

[@thelimitingfactor]

Yes!

[@YellowRambler]

Thorium Molten Salt Reactors pretty much answers most of the issues you raised, sadly thanks to decades of bureaucratic red tape China will own this market unless something is done about this bureaucracy. Think about it we had uranium powered pressurised water reactors for about 3/4 Century with very poor uptake, how many major crisis do we need to make this archaic coldwar reactor design desirable?

[@aliendroneservices6621]

TMSR power-service reactors do not exist. They have never existed. China is not building them. China is building its own version if the AP1000 power-service reactor.

[@YellowRambler]

@ try a search for “China 🇨🇳 Thorium Molten Salt Reactors”?

[@thelimitingfactor]

I'll cover that in future videos

[@markgemmell3769]

Fully depreciated nuclear is not as cheap as fully depreciated solar wind and batteries. CAPEX for new nuclear compared to new SWB is insane. Therefore no new nuclear will be built by anyone with a spreadsheet. However, if you want weapons-grade materials... 😂

[@thelimitingfactor]

Correct for the most part. However, new nuclear is still being built, or at least attempted. Either way, the main point of this video is that nuclear plants won't be all shuttered by 2030 I'll get more into the interaction with SWB in the next video

[@TroyFrank-h1k]

AI companies are currently gushing losses on AI investments, and there's no obvious solution to that in sight. Also, latest AI models seem to suggest improvements have stalled, and aren't getting better anymore in spite of now costing ~$500 million per training pass on new models. What I'm getting at here is that AI is a bubble, and one that's likely to burst in the next year or two. It's too early to claim that it's going to drive up future electricity demand, because it might not by much.

[@thelimitingfactor]

1) Internet companies gushed losses for years as well 2) No, the improvements haven't stalled. They're just starting on a new exponential curve in other dimensions, like the ARC benchmark for reasoning. 3) Yes, it's a bubble, but any new economic engine experiences those. It'll pop, and we'll end up with cool new toys when the dust is settled (at low cost) 4) Incorrect, so far the forecasts have proven to undershoot, not overshoot. The forecasts I showed were conservative as well, and I said as much. Stop watching pop-sci channels and the news

[@marcussalinas6335]

I think a general counter point is that not every company that needs a lot of power will be able to spin out a fully depreciated nuclear power plant. Therefore there only option will be to build a new one or build solar. And that solar will produce in excess and will be sold to the market and will drive prices down. Thank you for the charts. I’m pro solar and battery storage but even I was shocked to see the cost per megawatt for a new nuclear plant.

[@thelimitingfactor]

The purpose of this video was to show that Nuclear won't collapse So, what you're saying isn't a counterargument, but rather a separate topic that'll be covered in upcoming videos ✊ And sure thing! Glad there was some useful info in there.

[@jasonproud9535]

So battery storage can help with solar, wind, and nuclear. Battery could store the excess overnight. I didn't realize that they may have to pay when providing too much.

[@heinzbongwasser2715]

Reacting to demand in the grid demand is of course possible for nuclear power. France is doing it since 50 years. Just insert the control rods a little bit. Repeating it all the time does not make it right.😂 I will result in uneven burn of the rods which is no problem at all.😊

[@aliendroneservices6621]

Makes more sense to dump the excess power to resistive shunts.

[@thelimitingfactor]

Yeah, covering the exceptions, like France, wouldn't have affected the message or conclusion of the video, so it wasn't worth the side bar. There's a cost to France's approach - which I think you're alluding to in your comment 😀

[@heinzbongwasser2715]

@@thelimitingfactor You said nuclear power plants cant do load following which is completely wrong. i mean the least at least is not wrong. France is not an exeption because they use the same type of reactor everybody is (PWR). France is just the only one where you can see it because if you only have nuclear you have to adjust do demand which it can no problem.

[@dzikdziki2983]

Nobody but China and Russia is actually building reactors in a reasonable speed. This is all theoretical. Practically we can't build those reactors.

[@charangohabsburg1]

You forgot to account for the limited availability of uranium and the possibility of a steep price increase due to (not only but mainly) China's plans to expand nuclear power generation.

[@aliendroneservices6621]

Uranium is limited to the 75 trillion tonnes in the Earth's crust, which is 10 billion years' worth.

[@thelimitingfactor]

That's because it was irrelevant to this video The point is that the bear case was laid out terribly for nuclear

[@YellowRambler]

There’s more than one type of uranium, there’s very little of right kind of uranium, that why they have breeder reactors to convert unusable uranium to the usable of uranium.

[@erikkovacs3097]

We've probably hit the limit of low prices with solar and battery. We'll never see a 90% decrease again.

[@thelimitingfactor]

That could actually be a good topic for a video. I don't know the answer

[@ryuuguu01]

Battery costs went down 20% this year. I see no reason to think we have seen the limit on the cost decrease. Grid-scale stationary batteries will likely see an even greater cost decline as they are not as mature as EV batteries. I think a 50%~75% cost decrease for stationary batteries over the next 5~10 years is reasonable.

[@Decarbonize11]

I came to the same conclusion in my series on nuclear power. The future of nuclear power isn’t glowing: it will not play a big role in decarbonization kzbin.info/www/bejne/i5bXk6aCdrWXm7c

[@thelimitingfactor]

Cool! I'll take a look

[@raymondleury8334]

You omit the primary energy fallacy in your analysis. We will need more electricity, but less overall energy in an electrified world.

[@thelimitingfactor]

Nope, that's taken into account. We need a metric f*ckton more energy to fully unlock humanity's potential

[@raymondleury8334]

@@thelimitingfactor in advanced economies we will need less overall energy. This is largely because electrified solutions are much more efficient. EVs are 3-4 times more efficient than ICE and heat pumps are >3 times more efficient than the best alternatives. In developing countries the story is different as they currently don't use much energy.

[@aliendroneservices6621]

​@@raymondleury8334Improved efficiency causes increased demand. The world is heading toward *_Kardeshev Type 1,_* whether you like it or not.

[@jamesdubben3687]

Nice discussion, "firming costs" is the new term for me. But I tend to treat battery and solar as tech products that can change (for the better) quickly. Mechanical systems are very stable in development (not much change). Small modular Nuclear? I seems like PowerPoint presentations so far (that venture capital pit).

[@aliendroneservices6621]

Small reactors and/or "paper" reactors are indeed non-starters. They've been around since the 1940s. They were a bad idea then, and they remain a bad idea now. The current *_Large_* Modular Reactors (LMR) might be the best choice for now. AP1000 seems to be the best of those. Factories could be set up to produce the myriad individual modules and submodules which make up each AP1000 reactor unit. There's even a way that the steel-reinforced concrete containment domes could be mass produced in smaller pieces in factories, and then connected together into domes once onsite.

[@CapoRip]

Unfortunately, your title image is of a coal-fired power plant.

[@aliendroneservices6621]

Good catch. Coal-fired power-service *_also isn't going away._*

[@thelimitingfactor]

Ha! Interesting. What's the name of the site called? Based on the domes, I'd assumed reactor vessels.

[@aliendroneservices6621]

@@thelimitingfactor There are no smoke stacks (I found wider versions of the same photo you used). Therefore, it can't be a coal-fired power plant.

[@GruffSillyGoat]

It looks computer generated, perhaps from a stock image library somewhere. A quick image search suggests its adapted for differing needs, and used in a fair few places including: - The IMF use it for their _Nuclear Resurgance_ paper. - It's used on the IPSE GO simulation library produced by SimTech a virtual simulation company. The library is titled _"Advanced Power Plant Library with Gas Turbines,"_ that supports modelling for planning gas turbine sites. - An indentical image, but from a different viewpoint, also appears to be used by used by a company called Silixa, who produce fibre optic linked sensors for hazardous/hostile settings; the different viewpoint image that shows a smoke stack that is obscured by the cooling towers in the one in this video.

[@aliendroneservices6621]

@GruffSillyGoat Found the alternate-angle image with smokestack. Thanks. I doubt it's CGI, because it would be easier to simply take drone photos of a real power-plant.

[@bayjor]

In general, you do a good job of researching the subject of your videos, but this time you completely missed the point on the cost of nuclear power. The cost of nuclear power has hidden cost that are not covered in this video. A 1000 Mega watt reactor uses 27tons of uranium per year. Of this amount only 10 % is used to power the reactor. The other 90 % is not used and in the US is ban from being reprocessed. This requires the nuclear plant to store this spent fuel in a tank of water for about 10,000 years. Pluss the supply of uranium in the world will only last 80 years at the present rate of consumption. The mining, processing, and disposing of the uranium to power the nuclear reactor are just some of the hidden costs in using nuclear energy not discussed in the video.

[@aliendroneservices6621]

A dry-cask is not a tank of water. Reprocessing is easier if you first let the fuel cool for a few decades in a dry-cask. The current supply of uranium is 75 trillion tonnes, which is 10 billion years' worth *_if uranium alone were to replace all current fuels._*

[@thelimitingfactor]

You missed the point of the video, and you didn't read the other comments. I didn't enumerate the those costs because it wouldn't have impacted the conclusion. 1) 1GW uses 27 tonnes of Uranium. So what? That's included in the cost. Generating energy will always involve materials and/or fuel. 2) Yes, obviously, it is generating nuclear waste. But, operators pay into a nuclear waste fund, the other technologies also have ugly environmental side effects, and they're also subsidized in various ways. 3) There's plenty of Uranium, just like there's plenty of lithium. The world is big.

[@testboga5991]

Gas and coal is much more expensive than the numbers in the chart. The future costs of their climate damage is just transferred to the future.

[@thelimitingfactor]

I think that's true with all of these technologies They are all being subsidized in one way or another It's difficult to actually create a levelized cost metric that takes everything into account

[@JohnCorrUK]

Net Zero religion clouding your judgment

[@allangraham970]

I agree with rethink x. Battery tech is going to play a big roll in future power generation. The price per kWh will continue to decline. Seen future sodium batteries predictions of $10 per kWh. New battery technologies that are still to be invented and/or commercialised are likely to play a roll in medium to long term energy needs. As time goes on there will be less need for nuclear fission electrical power generation. Nuclear does not have a good track record of being built even close to being equal or below the price that the nuclear plant feasibility study assumed to justify its construction

[@thelimitingfactor]

The 20 year agreement with MSFT, it it moves forward, completely nukes their thesis. Their thesis is that 70% nuclear will be shut down or throttled by 2030. Not happening. Either way, I'll be covering the bull case for SWB in the next video and explain why Nuclear has a tough road ahead. Too much fanboyism out there. It leads to poor outcomes.

[@yonseienglish]

As a longtime follower of TLF, this is the least well-researched video I’ve seen. Most in the energy transition know that nuclear is a massive black hole of graft and corruption (see Vogtle in GA) linked to nuclear weapons industry. This video DID NOT touch on longterm waste disposal risk, proliferation risk, corruption (see what happened with HB6 in OH), nor Amory Lovins (founder of RMI) insights on nuclear (aka it’s a very bad idea), construction times (10+ years), longer opportunity costs of building instead of renewables, and simple fact that no private company is stupid enough to insure NPPs. Everyone knows SMRs (see many recent bankruptcies) have failed to date and likely will in the future. This video has some good points vis-a-vis the ReThink X report, but I think it misses the forest for the trees. Nuclear is still an expensive way to boil an egg. See ‘The Energy Transition Show with Chris Nelder’ podcast for all their episodes on nuclear power (searchable index with transcripts) for a detailed run down on the rabbit hole. Also, Lazard for nuclear DOES NOT include any of risks above in their pricing, so it’s not actually a fair number.

[@aliendroneservices6621]

@@yonseienglish Citing "waste disposal risk" is an argument to expand uranium-fired power, because: 1. Spent-fuel already exists. 2. The most-effective way to spur development of solutions to any given problem is to expand that problem. 3. Everything is cheaper in larger amounts, including waste-disposal.

[@thelimitingfactor]

As long time creator of TLF, people say this on many of my videos. In most cases, it's because someone gets emotionally attached to a technology and ignore what I actually said in the video because they feel attacked. This seems to be one of those cases. The purpose of the video wasn't to hype Nuclear, which it appears you wanted. Instead, it was specifically to cover the holes in the bear case. In upcoming videos of the series, I'll cover the challenging path ahead for Nuclear. How can I be balanced if I don't cover the flaws in the Nuclear bear case? Or, don't they exist?

[@Andre-95]

1. While there is a lot of nuclear fuel on earth on a cosmic time scale it's really not that great vs how long you could use renewables aka end of earth/sun. I will add that I think it got a lot of potential for off earth applications. 2. Using Microsoft or rather what one single company does to secure energy in the short term is a poor indicator, as one can observe with Toyotas obsessions on hydrogen vehicles. 3. LCOE is in the kindest way I can say it, a joke. It assumes an uptime that especially for fossil fuels that can be far too optimistic and the more renewables gets added the worse they will be off. For solar it will assume that residential roof»commercial roof»grid scale in terms of cost per kWh, but this completely ignores the cost of distribution/transmission both residential and commercial rooftop solar is heading towards a point were even if the cost of producing electricity was free, the cost of distributing that electricity would still be higher than rooftop solar. 4. Close to the end of 2023 it was predicted that solar deployment would be ~400GW for the year, as what always happens that number was adjusted up to ~440GW shortly after the year ended. At this point in time we are looking at ~600GW which is again going to be adjusted up as we get into the new year. And I think you have to be working for the IEA to believe that is going to stagnate anytime soon. 5. The baseload fallacy and peoples absession with it. So I heard recently someone say that once SWB is 60% of generation everything else is a peaker plant. Now I'm not sure I agree on the 60% but it did get me thinking, "what really is baseload?" Well the first thing Google and Wikipedia will tell you is that "It's the minimum level of demand", Well that really seems like a 3rd grade level answer, after all we can already do smaller installations of just solar and batteries and there is certainly no traditional "baseload" there in such a setup. Now baseload is a term invented by the incumbents and if we look at what role it has played it's to produce as much as possible from few plants aka being more efficient and cheaper to operate. To sum that up it's really just "the cheapest majority to produce" and a peaker is just trying to fill the voids it can't fill. So going by that logic once SWB is ~60% of production and also the cheapest to produce, everything else is a peaker plant trying to fill the voids and a peaker nuclear plant sounds incredible expensive. Whether the cheapest majority is variable through the day or flat really has nothing to do with anything other than being a construct. For anything other than SWB + some other renewables like hydro which simply has a spot for water management (but limited in potential) it all really hinges on how low SWB can go in cost and that's honestly not a bet I would take. 6. While the IEA energy outlook 2023 on solar has been proven wrong the 2024 version again is quick to follow a linear prediction that has no place in a world of falling cost curves and I'm sure they will adjust it again next year, because nothing says good data when you're wrong every year. Another case of garbage in garbage out.

[@thelimitingfactor]

1) The energy market doesn't operate on cosmic time scales 2) Many companies are, not just MSFT. Musk, Altman, etc 3) LCOE is fine. People always get their noses in the air on any financial metric. There is no perfect financial metric, that's why we have so many of them. I'll stop there. You're criticism that I've read so far is just that - criticism. And, you aren't adding anything to the conversation.

[@aliendroneservices6621]

3) Then ban wind-and-solar from every grid. Problem solved.

[@Andre-95]

@@thelimitingfactor 1. I was exaggerating a bit with the cosmic time scale but if we heavily pivoted to fission supply would certainly outlast me but just based on current demands of energy we would use up several materials on a scale far less than how long humans have been around, which means it's just kicking the can down the road when we already got permanent solutions. But in the end $/W is king and every year solar and batteries keeps going down (barring fringe events like 2022 for solar demand but even that has been corrected for by now). 2. My point was simply pointing out that just because one company does or say something doesn't mean it's a good indicator since you can like find that for any point. I simply believe that MSFT chose to take advantage of the existing plant, after all it's already there. 3. "There is no perfect financial metric" we agree there but the issue is that it inherently favors incumbent technologies and then use it to compare to the disruptor. And the issue is that the more solar and wind you add to the grid the more wrong LCOE will be as it assumes an utilization rate for the incumbents that will keep going down simply due to falling costs. Well you ignored the rest but I'll add that when Pakistan entered an energy crisis this year they didn't build nuclear but solar at record speeds crushing any predictions for the country.

[@hcjpbluesky9916]

Don't believe Lazard's LCOE. Just look at cost to build, the 80-90% capacity factors and do your math out 60 years. The bondings for these plants can mature near 2060 (GA, SC). They're capital intensive, long-duration and, with storage, should see the elevated capacity factors that add a simple 2-4 cent/KWh O&M to the calculation of total cost (w/interest) / years of service. -You don't get $120-140MWhs this way. You get much less. Not sure, but I think they make the same weak CF assumptions others do, in order to get levelized cost. I'm one of your X-subscibers. If you keep this up, I might re-subscribe, LOL. What's X, anyway? I mean, most of #EnergyTwitter has reconstituted on BlueSky.

[@thelimitingfactor]

Anybody worthwhile in the energy space is still on X If somebody isn't willing to stick around to fight for their ideas, maybe they don't have much faith in their ideas because they know at a deep level, those ideas are flimsy 😉

[@hcjpbluesky9916]

@@thelimitingfactor Pretty sure Elon censors by amping what he wants seen. I'm still on X. I log in, don't waste time toggling off "For You(Elon)" but instead go directly to the several handles I follow most, read them, then I log back off. My "Following" timeline started showing me less from who I'm following (only ~120, not hard to check) and more political ads, or replies from the new blue-check community, LOL. That's cumbersome/inefficient, stifles discussion. So, like others, I've gone looking for signal.

[@JSM-bb80u]

In my opinion nuclear should be defined as renewable. Even though world uranium is limited it's still able to power the world for billions of years woth breeder reactors. And there's thorium too. It's also extremely easy to combine solar with nuclear with plummeting battery costs.

[@Andre-95]

Every renewable energy source can be used until the end of the Earth or the Sun, nuclear can not do this. Unless you are pessimistic about earths short term survival (cosmic scale) getting depended on an energy source you will have to pivot away from in the future seems a bit foolish.

[@JSM-bb80u]

@@Andre-95 Ok. But the uranium isotope (uranium-235) which is used in nuclear powerplants would last 200 years if we solely relied on it. But that isotope is only 0.7% We can use other more abundant uranium isotope (uranium-238) through breeder reactors. Which would last upto 30,000 years. And there's nuclear fusion too.

[@Andre-95]

@@JSM-bb80u Humans have been around for longer than that and baring any fringe events we will be around for a lot longer to, 30k is really just a blink away on the cosmic time scale of millions and billions of years. We have a fusion reactor in the sky we can already capture energy. Both have good applications for off earth though.

[@JSM-bb80u]

@@Andre-95 We need both solar and nuclear. The biggest problem with solar is it takes a lots of land. The theoretical minimum for 1 kW solar panel is 1 m². A 1kW panel would produce 3-4 kWh on average. For industrial processes we need a lots of energy. To smelt 1 kg of aluminum we need 14 kWh of electricity. A 50% renewable, 50% nuclear is the perfect combination for grids.

[@RushWasABand]

Nuclear is at odds with the ever growing bottom of the duck curve. It's a very limited tool in the toolkit

[@thelimitingfactor]

It's more complex than that. I'll be getting deeper into the overall mix in the next video

[@concinnus]

Yeah, I always shake my head when there's always some idiot who says 'but we need nuclear for base load'. No, it's the opposite of what we need.

[@iandavies4853]

Exactly. Other options are more transmission (National grid) or flexible load. And thermal storage for heating / cooling. Future option is Dyson sphere coupled with compute - how close is that to present tech abilities?

[@bobo-cc1xw]

The grid is changing. more electric vehicles has shifted base load. Do we need more of it with current assets but should not rule out for the future. Alot of the peak load has been flattened The power grid is going to be maxed out a response to that is a lot of batteries and shifting which will impact base load and need. Could invent a case that we need batteries and nukes. However nukes are not that hard to change to variable output it is just a conservative industry.

[@johnstubbe3113]

😅 It is disingenuous to analyze nuclear power, without analyzing the cost to refine the fuel

[@Sdvbr]

USA: Price-Anderson Nuclear Industries Indemnity Act. The buck with traditional fission nuclear power plant insurance stops with the tax payer. Try competing with a technology that can't even be sufficiently insured in the private sector, by law. I have rarely seen this as part of the discussion. The, admittely fairly low, risks are sozialized, profits are privatised.

[@thelimitingfactor]

The same is true for oil and gas. The negative externalities are born by the public. It's just that nuclear's side effects are much more apparent when they manifest. shock and awe, lol

[@Sdvbr]

Were thousands of years of nuclear waste storage factored into the 30 cents as well? Shock and awe, they weren't? I will believe that nuclear is lower co2, but never that is low cost.

[@yurona5155]

Just a couple of random additions to the bearish case: -putting aside any safety concerns, there are still some security risks to be aware of, i.e. nuclear facilities make for great sabotage/missile targets and require tight controls on breeding weapons-grade Pu as a non-proliferation measure -nuclear tends to discourage investments into a more decentralized, i.e. resilient, grid architecture and therefore create single points of failure, whereas the renewable competitors are ridiculously easy/cheap to install/maintain/replace -there are deeper reasons to be skeptical about long-term economic viability, i.e. a minimum level of inherent technological complexity combined with a highly specialized supply chain might actually be an insurmountable (relative) cost driver

[@thelimitingfactor]

Why not add to the bull case too? That is, as I said in the video, it's about using the right tool for the job. Fanboyism is bunk

[@yourcrazybear]

"putting aside any safety concerns, there are still some security risks to be aware of, i.e. nuclear facilities make for great sabotage/missile targets and require tight controls on breeding weapons-grade Pu as a non-proliferation measure" On the contrary. Just look at how Russia doesn't send missiles into nuclear power plants in Ukraine. All other types of power plans are however open targets. And it's also not cheap to install the new power lines that have to be connected to all renewable power plants all over the world. Connecting a cable to a centralized place is way cheaper. Nuclear power plants also adds to grid stability.

[@yurona5155]

@@thelimitingfactor Agreed. I simply stuck to making arguments in favor of the bearish case because that's what the video was about. There's no fanboyism involved on my part. When it comes to energy production, I'm about as technology-agnostic as it gets...

[@yurona5155]

@@yourcrazybear Security risks very much depend on the threat model. While at the current stage of the Russo-Ukranian war targeting nuclear facilities doesn't make strategic sense for either side (because of the escalation dynamics involved), for terroristic actors and/or nuclear powers in a peer-level conflict that calculus looks VERY different. The ins and outs of energy market design / grid planning are quite intricate indeed (as a first approximation, they're based on a 'triangle' of production, storage and transmission capacities, where any two of them can compensate for a lack of the third). I.e. there will always be some economic trade-offs to be made as finding the right mix of energy sources is part of a multivariate optimization problem. In the specific case of a nuclear-focused strategy you're basically trading away storage and transmission capacity (i.e. scalability/flexibility and robustness) for - somewhat technology-dependent and possibly temporary - gains in production efficiency. Whether or not any specific energy source aids in providing grid stability depends entirely on the topology of the actual grid. In a certain (purely mathematical) sense, optimizing for stability actually requires a form of decentralization that wide-spread use of nuclear plants tends to disfavor.

[@yourcrazybear]

@@yurona5155 "@yourcrazybear Security risks very much depend on the threat model. While at the current stage of the Russo-Ukranian war targeting nuclear facilities doesn't make strategic sense for either side (because of the escalation dynamics involved), for terroristic actors and/or nuclear powers in a peer-level conflict that calculus looks VERY different." It will always makes more sense to target other power plants that can't leak radiation. "In a certain (purely mathematical) sense, optimizing for stability actually requires a form of decentralization that wide-spread use of nuclear plants tends to disfavor." A form of decentralization? How about stop being so diffuse in your comment? All nuclear powerplants will obviously not be build at the exact same spot in a country. And power plants like hydroelectric and nuclear are quite good at balancing the grid. Windpower and solar panels on their own are not.

[@gronkotter]

I normally really like your videos but you get a LOT wrong here. Baseload isn't a thing that's necessary. Demand is highly variable. Flexible generation is mandatory, but not baseload. Baseload does not fill in when wind and solar is low - flexible assets do. Running at a loss.... is a loss. If the price is zero the reactor is losing money by generating. It does indeed make sense to shut down for extended low demand periods like spring where aggregate price across a month may not be interesting enough, but summer/winter are profitable. This is the first sign a reactor is going to close down, but it is economically rational in the medium term. Yes, old nuclear is cheap to run if fully amortised. But old nuclear often closes when its licence expires. The median closure age is 40, closely followed by 45 years of age. That is why Rethink X believes nuclear generation will drop quickly - old plant will be regulated out of existence. This is common around the world, nuclear generation is dropping globally outside of China. The cost in Lazard assumes no cost overruns. Cost overruns have occured in every single project built in the West this century. 100% probability of double or triple the cost when you actually build something outside a desktop model. "AI is a match for nuclear" - renewables ARE the generation of choice, all the tech companies are investing a hundred times as much in renewables. The nuclear restart example is cute, mainly because they can avoid paying network charges. The growth in demand doesn't mean nuclear will be chosen, it's not "a rising tide floats all boats", but that the winner will win more.

[@thelimitingfactor]

I didn't say it was "mandatory," I said the different generation types were complimentary. I'm not going to read the rest given the first point miscontrued what I put in the video. It's likely that the rest of your points are equally as skewed

[@gronkotter]

@@thelimitingfactor Please do read the rest, that's the least impactful point. And it's not complementary. Flexible complements inflexible. I've got 15 years experience in electricity and you do not. If you are going to get in a huff over criticism you'll lose a subscriber.

[@steverobbins4872]

I'm not ideologically opposed to nuclear, but I think building more nuclear plants would be a huge mistake for the US for several reasons: 1. We are already building wind, solar, and batteries far faster and cheaper than we could ever produce new nuclear plants. So it seems to me that there is simply no point in trying to build more nuclear. 2. We still don't have a solution for the nuclear waste problem. The industry has been kicking the can down the road for decades, and will continue to do so. In short, this problem will never be addressed until there is some huge disaster that causes massive public outcry, and then we'll have some sort of extremely expensive and wasteful response. 3. We currently import about 95% of our nuclear fuel. I want the US to be energy independent.

[@thelimitingfactor]

Good points! I'll address some of them in upcoming videos.

[@yourcrazybear]

"2. We still don't have a solution for the nuclear waste problem. The industry has been kicking the can down the road for decades, and will continue to do so. In short, this problem will never be addressed until there is some huge disaster that causes massive public outcry, and then we'll have some sort of extremely expensive and wasteful response." Considering how long time the waste must be stored adding another 80 years to it is inconsequential. "3. We currently import about 95% of our nuclear fuel. I want the US to be energy independent." So massive tariffs on Chinese solar panels then? Because USA have failed miserable at producing solar panels at great volumes.

[@jasonneugebauer5310]

I think people do not look at energy sources from a operational needs perspective, so I am creating this metaphor to help people appreciate the values of different energy sources. The metaphor: the grid is like a city's commuter rail service. Coal is like steam locomotives, a little expensive to own and operate and mostly dependable but the smoke is a bit dirty. You build a fiew coal powered steem engines with several cars each and keep them running on a set round the clock schedule, while maby running them a bit faster during peak demand and slower at night when most people are sleeping. Natural gas conbined cycle is like a diesel electric train, inexpencive and highly efficient use of hydrocarbons. So you have several of this type engine all pulling several cars each. Again, faster during peak usage and slower at night. Natural gas peaker plants are like trains powered by jet engines powerful but very inefficient and expensive to own and operate. You can run these engine any time you want as fast as you want with however many cars you need at the moment. This is what you use when you needed to move people but ran out of other engine types. Solar energy is like a train engine powered by solar panels. The conductor will tell you this is the least expensive train to ride on. And, while it took a lot of energy to build the train and they did have to modify the tracks, the cost to move a passenger is minimal. When the sun shines these trains just start popping up on the grid and moving passengers when this happens the ticket price to ride any train on the system plummets. Then, at some point it gets cloudy and these solar trains stop, open the doors and tell the passengers to get out and find another train. This shutdown also happens every night and almost all the time during winter. Wind is like a train that runs faster the more the wind blows. The train is expensive to build requires some additional tracks and needs to be overhauled or replaced every 10‐15 years. But, when the wind blows strong, this train runs fast and at very low cost while carrying lots of commuters. Unfortunately this train only runs 35% of the time, mostly in the spring and the fall. Hydo electric with dams is a train that was expensive to build, but almost free to operate so long as there is sufficient water in the dam. This train is incredibly powerful, reliable, and can carry more cars as needed up to its maximum number. Battery power is like taking the tires of a Tesla Model 3 and putting it on the train tracks. This train goes ludicrous speed, but it's expensive to purchase doesn't hold many passengers and will need to be replaced every 10‐15 years, not really practical for moving around the working class people like me. Nuclear energy is like a train built by NASA powered by a perpetual motion machine also designed by NASA. Needless to say building this train costs 10-50 times the cost of the other engines and building the perpetual motion machine is as cheap and easy as NASA building a rocket, but we get the energy for free. This train must run at all times with a full load of passengers because otherwise the innitial cost will have bankrupted the entire city. As the commuter rail service operator, you must have reliable, on time service for all passengers at all times. So, you start with running all your available nuclear powered trains, because yo don't want to bankrupt the city. Next, you add your hydro electric trains because almost free. Then you add your conbined cycle natural gas trains because efficient. Then you run some coal powered trains because they are reliable and you need reliable transportation. At his point you have mostly covered your fixed constant flow of passengers. Now that your base demand is covered you have to deal with peak commuting times when demand doubles or triples. But wait, we have wind and solar out on the tracks picking up customers whenever these trains have power.

[@thelimitingfactor]

I always appreciate a well thought out comment 💯

[@jasonneugebauer5310]

@@thelimitingfactor thanks.

[@suggesttwo]

Well placed windows and shudders are still the KING 👑 of low cost free energy production. 0:28

[@thelimitingfactor]

I really want to build a house

[@rotors_taker_0h]

I fully appreciate continued trashing of RethinkX unhinged visioneering (even if you would never admit you doing that). Would be great to see some taming of their reports in the other domains.

[@thelimitingfactor]

😂 Definitely not the intent. But, I do love that word, visioneering. I'm stealing it.

[@rotors_taker_0h]

@@thelimitingfactor thanks, doubly proud to hear that because english isn't my native language :)

[@yourcrazybear]

And what other domains would that be?

[@rotors_taker_0h]

@@yourcrazybear they hype up precision fermentation as a solution for all food problems and humanoid robots as coming labor replacement and also how all of that would change humanity overall.

[@yourcrazybear]

@@rotors_taker_0h "and humanoid robots as coming labor replacement and also how all of that would change humanity overall." Well humanoid robots will take over quite a lot of jobs and this will have a big positive effect for humanity. So what are you not agreeing to in that area?

[@suggesttwo]

Natural gas is second. Same price but adjustable. Nuclear is by far the most dangerous if something bad happens. No one dies if nothing happens no matter the technology.

[@RicksPoker]

Lots of people believe the media narrative that nuclear is dangerous. It actually has the fewest deaths per terrawatt hour of any power source including solar and hydro. Why has the media hated nuclear for decades? My guess, is that those that own shares in big media, also own shares in fossil fuels. Warm regards, Rick.

[@thelimitingfactor]

If you're strictly looking at how many people are dying from each technology, nuclear is the safest

[@yourcrazybear]

"Nuclear is by far the most dangerous if something bad happens. " It's obviously the safest.

[@suggesttwo]

@@yourcrazybear Nuclear plant provide 10% of the electricity made. 15% of US electricity. Nuclear is safe says the fools who can't see 10 seconds into the future. Chernobyl, Fukushima. When things get old the can succome to multiple failure points. My car has had multiple failure points this year. Battery no start. Wouldn't charge over 11.5v. Replace battery. Power steering line failed on the way home. A week later no start. D connection from ignition switch to starter relay failed. Replaced. 2 weeks later click no crank when raining. Terminal from battery to starter relay corroded. Clean with wire brush on drill.

[@suggesttwo]

@thelimitingfactor You mean have died. First principals reveal that nuclear plants can go supercritical and blow up and contaminate large areas like in Russia in 1986. Maybe there weren't enough redundant sensors at Chernobyl. At Fukushima they were not prepared for a tsunami. The boy scouts teach the scouts: be prepared. The tsunami caused by an earthquake could not be anticipated. 3 reactors melted down. Here's the bottom line. It's a fool who says: "It can never happen." If a lot of them are built it's going to happen. In business you hope for the best and plan for the worst is acceptable. Global Warming Climate Change didn't happen in 2000, 2010, 2013, 2016 and 2020. Stop scaring the kids for nothing. Didn't happen because the plants are eating it. All of the excess CO2. Plants don't care if CO2 came from an animal or an engine.

[@arubaga]

No excuse to delay developing Molten Salt Reactors.

[@thelimitingfactor]

Agreed, they need to move fast

[@bobwallace9753]

There needs to be a careful costing before throwing more billions at the latest nuclear concept. Look at the history of nuclear. Promise a reasonable cost but come in at 3x that promise.

[@arubaga]

@ Just one or two study reactors to work out the kinks again. Should not cost billions. Didn’t China took 10 years to get through their MSR study phase? They are doing only one commercial deployment for their next step.

[@bobwallace9753]

@@arubaga Please explain the cost savings in realistic numbers. No one has done this so far that I've been able to discover. Nuclear reactors are nuge construction jobs that take years and pretty much always overrun cost estimates by huge amounts. Look at the original estimates for Vogtle 3&4 and what their final costs have risen to.

[@arubaga]

@@bobwallace9753 Obviously we have to first construct the study MSR reactors to regain the knowledge that has been partially lost since the 1970’s. MSR should produce much less nuclear waste, but I am sure figuring out potential cost saving will be one of the goals of a study program.