[dignick]

There are several problems with fission (probably fusion too) as I understand it:

1. Cost per MW compared to renewables (~$150 vs ~$40 and falling). Here in the UK the government is promising to subsidise this to make it viable.

2. Construction time - average is 10 years, we don’t have that long to wait.

3. Decommissioning is expensive and a long way in the future. Is that cost built into the cost per MW? How can we be sure the money will be protected, and will be enough to cover it?

4. Spent fuel. The project you mentioned isn’t complete yet, but even then it’s a huge liability to leave for future generations to manage indefinitely.

Meanwhile, renewables don’t have these problems and are available immediately. We should be building huge factories to produce wind and solar en masse.

Source for the figures: https://www.reuters.com/article/us-energy-nuclearpower-idUSK...

[formerly_proven]

> 1. Cost per MW compared to renewables (~$150 vs ~$40 and falling). Here in the UK the government is promising to subsidise this to make it viable.

You can't directly compare cost per generating capacity, because nuclear, gas, coal etc. are available according to schedule, while most renewables aren't. Adding storage around renewables to make them schedulable raises costs.

[Schroedingersat]

Nuclear in the UK has a capacity factor of around 60%. Availability is in the 70-80% range.

Yeah it's (usually) planned, but it's a decently long time in which you need those gas plants.

Why not just build solar instead and fuel those same gas plants with hydrogen or methane you plucked from the air with your $20-30/MWh unscheduled electricity?

Plus, you can get solar and storage as an off the shelf item today as a retail customer for less per watt than recent reactors in UK/France or even USA. 8kW nameplate solar and 16kWh storage capacity is about $10k which matches 1kW of net from eg UK projects of around 2.5GW net for 26 billion pounds fairly closely.

Yeah if you live far north or have a long cloudy month in winter you'll be relying on that gas plant, but so does the nuclear reactor. Plus you'll be dumping 10-20kWh/day into the grid on the good days. Provides a decent incentive to figure out how to store it, and even if you're only getting 5c/kWh for it, it'll pay for replacement in 7-10 years or so when prices have dropped another 50-80% without sacrificing your kilowatt.

[neilwilson]

Solar requires land area. Storage requires land area. Britain isn't that sunny - particularly not in winter.

Nuclear has a very small footprint on a crowded island.

Plus we have Rolls Royce SMRs who have been building nuclear reactors for a while.

[Schroedingersat]

That is the admitted cost of Hinkley C and lower bound on the cost of Sizewell (it will go up, they always do). Sizewell is a rolls royce smr. Matching end user retail cost of solar. Right now. By the time sizewell comes online it'll be a fraction. It's also calculated with a 12.5% capacity factor which is winter in the UK. Add in overnight costs and it's extremely one sided.

You could add as much net capacity as the UK has in nuclear in just above the space used for parking cars.

You could add twice to four times that again just on detached house rooftops.

Even as a commercial installation with no other purpose, a 4km square is hardly an insurmountable barrier.

The initial capital budget of sizewell and hinkley alone could provide 30-80GW of nameplate solar or a rooftop system on every building in the country.

If there are trillions in the pot, by all means go ham with fission, but when low carbon sources are fighting for the scraps left over after subsidizing fossil fuels we have to do the thing that is effective first.

[neilwilson]

"If there are trillions in the pot"

There are always trillions in the pot, because the UK is a sovereign country with its own currency. Money is never the issue.

Therefore it's only overnight costs that matter in a build (and hitting the deadline). The real issue is one of manpower and stuff. We don't make solar panels in the UK. We will make SMRs. Therefore we're not reliant on Chinese manufacture, or the whims of export markets to fund them. A problem we're currently having with gas and oil.

To have security of energy supply over time you have to be as decoupled from world markets as possible. We don't want to be in the situation where we're relying on China for replacement advanced manufactures to keep the lights on.

Solar has no reliable capacity in winter in the UK unfortunately. You wouldn't want to rely on solar with several weeks of grey miserable UK winter weather even with storage. The same with wind, which is still suffering from a degradation in capacity due to the as yet unexplained overall reduction in wind speeds - which may itself be a result of climate change.

[Schroedingersat]

Commenting about fiat currency is a pointless distraction when the purpose of using it is as a proxy for labour and materials. 'Trillions in the pot' is just a proxy for a certain amount of access to raw materials, trading ability and labour power, all of which are finite and don't really increase if you sink your country into hyperinflation.

> Therefore it's only overnight costs that matter in a build (and hitting the deadline). The real issue is one of manpower and stuff. We don't make solar panels in the UK. We will make SMRs. Therefore we're not reliant on Chinese manufacture, or the whims of export markets to fund them. A problem we're currently having with gas and oil.

> To have security of energy supply over time you have to be as decoupled from world markets as possible. We don't want to be in the situation where we're relying on China for replacement advanced manufactures to keep the lights on.

Spending an amount on computers and steel and exotic alloys and uranium ore and then also spending 10x as much on labour is no better than spending that first amount on foreign solar panels. Far better to overpay for solar panels by developing a local industry, or overpay for solar thermal systems (which are still vastly cheaper than fission).

> Solar has no reliable capacity in winter in the UK unfortunately. You wouldn't want to rely on solar with several weeks of grey miserable UK winter weather even with storage. The same with wind, which is still suffering from a degradation in capacity due to the as yet unexplained overall reduction in wind speeds - which may itself be a result of climate change.

Renewable-derived hydrogen is already at cost-parity with fossil-fuel-methane derived hydrogen in some markets. Renewable-derived methane isn't much further off and is one of many ways of solving the storage issue. When a gas plant (which you need anyway) and the renewables to provide enough net power in winter and a massive overprovision during summer cost a fraction of nuclear there's no point.

Plus your argument about energy security completely precludes nuclear as an option for over 50% of the world as they're not allowed to make their own fuel. There are also about as many countries with a credible manufacturing base for solar panels than countries with viable uranium reserves, and there is more than one chemistry that you can make solar cells with.

Finally being entirely beholden to one of four or five corporations worldwide is no better than being entirely beholden to one of four or five countries with the cheapest solar.

[pfdietz]

The land area required for most forms of storage is inconsequential.

[dfjklseldklk]

This is oft repeated, but once the cost of renewables is low enough it would have (usually un-modelled) second order effects on end users which may make this less important.

Some industries literally can't turn off production without damaging plant, but if some can then where energy cost is ~30% of fossil/nuclear than it might make sense to over-provision industrial capacity and only run it when the sun is shining.

[legulere]

Scheduling also raises costs though. Costs don’t go down much for nuclear power plants if you let them run at reduced output. Nuclear also isn’t as reliable as it seems as it can be seen in France.

Renewables even have the advantage that they produce electricity mostly when it’s needed. Photovoltaics during the day and wind during winter when heating is most needed.

[fredestine]

macron removed investment and focus on nuclear in his last mandate before realising that being woke made france broke (luckily not like germany broke - you know like in a way that when there is no sun or wind they need to call Putin to send some energy) now he reverted his thinking because money and energy is more important than beliefs when you needs them. just like usa or germany reactivated coal and biden is selling fracked gas to europe at gold price.

[Krasnol]

> macron removed investment and focus on nuclear in his last mandate before realising that being woke made france broke

None of those problems OP described have anything to do with that overblown statement. Investment may have been removed for FUTURE projects but not for current operation which is highly subsidised by the French taxpayer guaranteeing the fixed price the Government decides on.

Also Germany reactivated those dirty plants also to help France out. The whole European grid is helping the nuclear nation out and will continue to do so until France diversifies its power generation infrastructure.

[fredestine]

the investments included maintenance of existing infrastructure. france has more than half its reactor being off right now. guess what? invest more 5 years before and we wouldnt be there today. so yes. it is not overblown it is factual

[Krasnol]

Show me some useful source which says that there have been investments missed 5 years ago and which is why that lead to the current problem. French is ok too.

[dignick]

I’m a firm believer in distributed generation and storage, i.e. solar on the roof and battery on premises. This has the added benefits of reducing load on the grid and increasing resiliency. It should be required for all new buildings in regulations.

[dfjklseldklk]

In fact, solar on roof often loads the local grid in unhelpful ways (they were designed for asymmetric loads).

I'm in favour of renewables, but done in the way that actually makes most economic and environmental sense. In the Uk we threw money at rich people with £15k to install 4kwh solar rigs on their houses. That money would have been much better spent subsidising large industrial installations.

[bakuninsbart]

You can however price this in, and I doubt it accounts for 110$/MW. Furthermore, nuclear energy specifically only runs according to schedule. Reducing or raising output is expensive and slow.

[belorn]

For simplicity, lets use the cost of for every $ that a KG of green hydrogen costs, this mean that the cost per MW will be 30x of that. So if green hydrogen cost $1/KG you the cost in term of MW will be $30.

The current cost of green hydrogen is somewhere between $2 -> $12. That is the production cost. The market price for green hydrogen sits around $4-$20, since there are multiple industries that demands hydrogen.

For 110 to break even the hydrogen need to cost $3.5/kg, and in order to really displace natural gas, it is estimated that it need to reach $1/kg.

Now I noticed that those $150/MW is not a range, so I took a look. Projected nuclear LCOE costs for plants built 2020-2025 places nuclear around $27/MW to $147/MW depending on financing and country (source: OECD Nuclear Energy Agency’s (NEA's) calculation). Russia has the lowest cost and Slovakia or Japan (depending on financing method) has the highest.

So in summery, it can definitively cost more than $110/MW to produce viable green storage solution, especially in northern countries where low duration lithium batteries is not a working solution for long winter periods with low wind production and the sun is only up for a max few hours per day. Nuclear can also be much cheaper depending on where it is built and how it is financed.

[Schroedingersat]

Why would opex and amortised capital scale with fuel price?

[belorn]

The report is likely this one: https://www.oecd-nea.org/upload/docs/application/pdf/2019-12...

The $147 figure is basically the worst case scenario.

[illiac786]

EDF has published a paper stating they can scale 80% down and then up again, every day, within 30min.

In practice, they have done something like 20% within an hour. It was early 2019, there was such a crazy wind that they had to reduce nuclear production also (after already reducing coal, gas etc. to the min).

I think what they really can do is somewhere in between.

sources, it was on l’energeek, but in French.

[jules]

> You can however price this in, and I doubt it accounts for 110$/MW.

What do you base this on?

[ncmncm]

Raises cost, but it is still way below cost of nukes.

[bratbag]

Cost factor isn't as relevant as you think, as this is an apple's to oranges comparison.

Renewables are too unreliable to act as baseline generation for a country.

In the UK last year for example we had very little wind, so we had to ramp up our gas power output to make up for our shortages in renewables. We burned through much of our gas reserves before the Ukraine war started, because of Renewable power unreliability.

Fission is the replacement for that baseline role that hydrocarbons currently fill, not the unpredictable-but-clean role that renewables fill.

The ideal future has both, with renewables producing as much power as possible and fission running on low capacity and ready to ramp up when renewables fall short.

[cycomanic]

Just repeating the baseline myth does not make it true. Nuclear does not compete with gas it competes with coal and renewables. It is often technically difficult but more importantly economically prohibitive to run nuclear as on demand sources. So for both renewables and nuclear you need some sort of storage or peakers.

Moreover nuclear is not the beacon of reliability, Frances nuclear plants were running to only 60% capacity due to maintanance and weather (when it gets hot nuclear plants have to shut down or reduce output significantly). Guess who was picking up the shortfall... German renewables and gas.

Finally, cost is absolutely the main measure: if the cost of nuclear is 3x wind/solar (and the cost of solar is falling exponentially) and you want to replace fossil fuels as quickly as possible the obvious way is to build renewables, you can overbuild 300% at the same cost. At that point you're close to being able to run your grid if you are sufficiently geographically distributed (even without batteries). Moreover in 10 years when your nuclear plant is finished building the price differential is like >5x due to the cost decreases.

[ascar]

> So for both renewables and nuclear you need some sort of storage or peakers.

One thing I don't understand here is the problem with overproduction. If we actually have excess electricity (as in not needed as electricity later) can't we dynamically use that for active carbon capturing? The efficiency of that process isn't even that important then as the main goal is to remove carbon from the atmosphere with carbon free energy.

Having carbon free overproduction sounds like a good thing to me. It's the occasional underproduction that's hard to handle.

[tuatoru]

See Casey Handmer[1]. We are underproducing solar PV by at least 4.8 TW (nameplate) per year: we're only producing about 4% of what we need.

There is no such thing as overproduction, there are only manufacturing bottlenecks in batteries, electrolyzers, and reverse osmosis water plants.

1. https://caseyhandmer.wordpress.com/2022/07/22/were-going-to-...

[toxik]

Calling it a myth doesn’t make it a myth. Power companies have been saying exactly this for years: they need PLANNABLE power generation. Building 3x solar or wind plants means 3x volatility.

[ncmncm]

Building 3x renewables in widely distributed places radically reduces volatility.

Wind is always blowing somewhere. Sun is always out somewhere. Storage is transportable.

[ascar]

> Building 3x renewables in widely distributed places radically reduces volatility.

Is that actually true? Serious question. That sounds like a claim that seems so obvious, but won't hold up to the degree you might think in reality. Just one scenario I'm thinking of are giant storms that have clouds spanning multiple countries. And in that storm scenario even wind power shuts down to prevent damage.

[cinntaile]

https://doi.org/10.1016/j.rser.2015.12.318 has tried to analyze this for the EU. I'm not convinced that daily data provides the necessary granularity though, but more detailed data for the mentioned time span probably doesn't exist. I would try to find some more articles and check if there is a consensus.

[ncmncm]

Such giant storms are rare, and short-lived. A few days' storage outlasts them.

[Galaxeblaffer]

Is storage really transportable ? Like how much energy in any form could you realistically transport for any meaningful distance without using too much of the energy that you are transporting ? Since you made the claim I'd like you to paint any kind of realistic scenario.

[tuatoru]

Hydrocarbons, especially medium-chain liquid hydrocarbons, can easily and safely be transported 10_000 kilometres and further.

Doing exactly that is presently about a quarter of total global international trade by value.

Their advantages of high energy density, safety, and undemanding environmental and handling requirements (distribution can be performed in temperatures from -40 to +40 celsius by almost untrained teenagers), and effectively unlimited storage duration and volume, far outweigh the energy inefficiency of producing them from atmospheric carbon. Especially once PV gets cheap enough.

Edit: I notice I didn't answer your question. For liquid hydrocarbons, I believe the answer is in the single digit percents, perhaps five percent. For LNG, the energy cost is much higher, perhaps as much as a third of the total energy value.

[_ph_]

High voltage DC lines are quite practical over 1000 kilometers and more - Germany already operates an 1.4GW line to Norway, using the Norwegian grid as a storage for electricity.

[ncmncm]

There will be a very great deal of ammonia synthesis, worldwide, just because ammonia is so useful for so many things, ultimately billions of tons annually. Ammonia is very transportable.

Even liquified hydrogen is about as transportable as LNG, which is shipped all over.

[snovv_crash]

It reduces volatility, it doesn't eliminate it. There will still be days when the sun and wind aren't out in a large enough fraction of places that there will be a shortage. It is less likely, but it will still happen. Factories can't just shut down, people can't just choose not to charge their cars or boil their kettles if there's a shortage.

[ncmncm]

When generation flags and local storage looks likely to be depleted, utilities will order a shipment of ammonia from any of numerous solar farms in the tropics.

Most of the time a utility will prefer cheaper local generation, local storage, or transmission-line power before spending on shipped-in synthetic fuel.

[Schroedingersat]

If the goal is 1 unit net, then min(3x, mean(x)) has massively lower variance than x.

The extra power when 3x > mean(x) is just incentive for developing flexible loads and arbitrage.

[toxik]

You’re not listening. They are literally calling for plannable power. I don’t think they’re stupid people either. You cover the base consumption using plannable sources, then use gas/hydro turbines for short term variation. The problem now is that when the wind/solar vary, you can only compensate so much with stored hydro before you empty the reservoir. Then electricity becomes expensive because you burn gas etc, or buy from elsewhere - which is EXACTLY what has happened.

[Schroedingersat]

When solar is 1/8th the price of current fission projects for the same net capacity and falling by double digit percentage per year you have a hell of a lot of money left over for moving the energy around.

Fission is around $10/watt with a mostly-plannable capacity factor of 60-80%. Fission is a cakewalk compared to fusion.

Solar is around $0.5/watt with a capacity factor of around 25% and falling rapidly.

The operating and capital costs of a gas plant are around the same with a plannable capacity factor in the high 90s.

So by spending $6 on solar, and $2 on a gas plant. You have $7/watt left over to figure out how to turn free electricity into hydrogen or methane at 50% efficiency and store it for a year.

We already have electrolyzers that work for $0.5 to $1.5 watt at around 50% efficiency.

Hydrogen storage is hard, but that left over $5 per 4kWh/yr should take care of it. If it doesn't, sabatier rractors are getting cheaper too.

The only thing we have to do for people to start using them is stop the coal and gas subsidies.

This is also just one of many options. Salt cavern CAES is similarly viable

[tuatoru]

> It is often technically difficult but more importantly economically prohibitive to run nuclear as on demand sources.

I'd just like to point out that the US Navy has an excellent track record running nuclear reactors that ramp up to full and down to zero rapidly, in submarines.

The US Navy does not have quite the same financial constraints as commercial land-based power, but constraints still exist.

I fully agree that solar PV and wind, especially PV, are much more atttractive to investors because you can be earning cashflow from your first MW of capacity while you're installing the second (which takes weeks (or days!) instead of years), and you can iterate and scale this all the way to 10 TW or more of capacity, as the demand requires.

[Schroedingersat]

> Frances nuclear plants were running to only 60% capacity due to maintanance and weather (when it gets hot nuclear plants have to shut down or reduce output significantly).

It's important to use the real deal-breaking flaws when pointing these things out. The relevant figure is availability because capacity also includes energy that could be produced but was not due to having nowhere to put it.

Availability is 70-80% in France and UK and 80-90% in the US.

But due to the long timeline of refuelling cycles you still need a full baseload backup. So nuclear needs long term storage or other uncorrelated backup more than renewables if anything.

The upside is it's easy to have two uncorrelated nuclear plants, so overprovisioning by 30% is sufficient.

That makes something that is already more expensive than solar with the same overprovisioned net capacity factor, and a green hydrogen plant with capacity sufficient to cover, and full gas backup infrastructure even more expensive though. Probably not enough to cover the costs of hydrogen storage yet or someone would be doing it (ignoring massive nuclear subsidies), but prices of batteries, solar panels, and electrolyzers are dropping rapidly. Hydrogen storage or green methane production only needs to become marginally cheaper to make it start happening even sans subsidies.

[jamesrom]

> Construction time - average is 10 years, we don’t have that long to wait.

This is a fallacy in two ways:

1. Scaling up nuclear projects will decrease construction time and cost. Efficiencies are found by with scale.

2. The opportunity cost of not starting nuclear projects now will surely be worse than attempting 100% renewables. The point is that we can invest in both.

[cycomanic]

> > Construction time - average is 10 years, we don’t have that long to wait.

> This is a fallacy in two ways:

> 1. Scaling up nuclear projects will decrease construction time and cost. Efficiencies are found by with scale.

More than half of a nuclear plant is essentially the same as any large scale power plant (goal, gas...). The opportunity for reducing cost through economies of scales is low. Economies of scales work for things build in factories, much less so for construction projects. That is true in general, not just for power plants.

> 2. The opportunity cost of not starting nuclear projects now will surely be worse than attempting 100% renewables. The point is that we can invest in both.

Why? It's the other way around, the actual cost of building nuclear instead of much cheaper and faster renewables causes an opportunity cost, because we can replace fossil fuels much faster building up renewables.

[ascar]

> It's the other way around, the actual cost of building nuclear instead of much cheaper and faster renewables causes an opportunity cost

That's under the assumption the available money, hardware and labor of ramping up solar and building nuclear plants directly competes with each other. That's a pretty strong assumption and I highly doubt there is a strong enough link between any of those three for your argument to have significant impact.

E.g. We should be able to drive rapid solar expansion with government money and subsidies while incentivcing big energy carriers to build nuclear plants.

[cycomanic]

No that's under the assumption that we have limited funding.

[ascar]

That's a non-answer to my comment. Limited funding and a available money is the same thing. The point is the funding isn't so limited that we couldn't do both as we run in other bottlenecks.

[ciphol]

> More than half of a nuclear plant is essentially the same as any large scale power plant (goal, gas...).

Presumably that's the cheap half of the nuclear plant, because coal/gas plants are pretty cheap to build.

[pfdietz]

Coal plants are actually not that cheap to build.

Combined cycle gas plants are cheap because 2/3rds of the power output is from the combustion turbine, which needs no heat exchangers. The steam bottoming part needs two: the boiler running off the exhaust from the combustion turbine, and the condenser to transfer heat to the environment.

[troorl]

> Cost per MW compared to renewables (~$150 vs ~$40 and falling).

Do you count in all the subsidies the renewables get from governments, including the production of the solar panels/wind mills, land ownership, utilities and all kinds of tax cuts and preferential treatment? In my country billionaires own massive solar farms and make tons of money at the expense of everyone else.

[legulere]

Generally speaking the costs in those comparisons are usually without subsidies. However some subsidies are difficult to disentangle from the costs. For instance it’s difficult for nuclear power plants to get insurance, so often states take that responsibility.

Can you tell where you are from?

[raverbashing]

Oil has way more subsidies than renewables. By far

I'm rooting for billionaires to own more and more solar farms please. Let them buy more newspapers and "think-tanks"

Because I'd really like to live on that world where oil has no subsidies, occupies no land and it gets magically transported throught the country

[nixass]

> 2. Construction time - average is 10 years, we don’t have that long to wait

You've been saying that for last 20 years. It's pathetic by this point. Best time to start doing things (anything) was 30 years ago. Second best time is now

[rmbyrro]

> Here in the UK the government is promising to subsidise this to make it viable.

Doesn't make it cheaper, only hides the cost. Subsidies are many times perversive. Prices are communicating something. When Gov messes with it, people and organizations tend to make bad decisions for themselves, society, environment or everything.

[7952]

The UK government system is more nuanced than that. The operator bids for a strike price and communicate something with the bid they offer. That price is then locked in. We will never pay less than that, but we will never pay more either. In successive rounds the strike price is lower. They are setting the price up front to give stability. And that makes sense when most of the cost is upfront. Otherwise renewable prices would just track oil prices.