[sofixa]

> There are good reasons to reject nuclear power on purely historic analyses of the projected vs actual costs of nuclear projects

That is if you forget to factor in the lifespan of nuclear projects, which is easily 2-3 times longer than solar and wind, and doesn't require associated (not yet existing) massive storage.

(Not saying this for Australia specifically, there is no nuclear industry there whatsoever so any new project will have significant human resource obstacles on top of all others; just as a general point which is so often forgotten).

[Retric]

If people had been responsibly building nuclear 20 years ago the world would be better off today, but today ramping up storage and renewables seems like a better use of R&D investments and subsidies.

Nuclear and solar can each scale to ~40% of the annual supply for most grids without storage, but for different reasons they both need increasing amounts of storage as you ramp them past that point.

Solar because the sun doesn’t shine at night and peak consumption is mornings and evening, but Nuclear because demand varies though the day and season while the costs per kWh increase the more its capacity factor drops. France both had lower capacity factors and exchanged a great deal of power with its largely non nuclear neighbors. Exchanging power with less nuclear countries doesn’t scale to a worldwide increase in nuclear.

However nuclear also costs more per kWh as a baseline and runs into similar problems as the percentage of solar energy increases. Without storage, a 20% solar 30% nuclear grid is less profitable for nuclear than a 10% solar 30% nuclear grid. Given the long lifespans of nuclear power plants nobody wants to invest in nuclear if it’s expected to be unprofitable 20+ years from now.

[Ajedi32]

> Nuclear and solar can each scale to ~40% of the annual supply for most grids without storage, but for different reasons they both need increasing amounts of storage as you ramp them past that point.

Is that actually true for nuclear? I did a brief search and it seems like in France at least, many reactors can adjust their power output at a rate of about 1% per minute[1], with some even as high as 5% per minute, which seems like plenty to me. You'll probably need some storage, sure, but a heck of lot less when you only need <10 minutes of backup power before the reactors can kick back on (compared to a grid based entirely on unreliable energy sources like solar and wind, for which you could have occasional dry periods of low generation lasting days or weeks).

[1]: https://www.oecd-nea.org/upload/docs/application/pdf/2021-12...

[Retric]

The important bit here is capacity factor as someone is losing money when a nuclear reactor is sitting around not generating 100% power.

Globally the majority of nuclear reactors have a capacity factor of around 90-92% which means the vast majority of the time they’re getting paid something for generating electricity. France varied quite a bit but was generally below 80%.

Now that doesn’t sound that bad as their cost per kWh only went up by ~15%, but they where also exporting nuclear energy at a loss at night and the weekend to countries that didn’t use much nuclear. If every country tried to go 50+% nuclear then everyone would have a surplus on nights and weekends driving those capacity factors down even further and thus cost per kWh even higher.

France massively subsidized consumer’s electricity prices using taxpayer money, so it wasn’t that obvious to the consumer how expensive it was. However, it’s hard to justify such expenses when there are cheaper alternatives.

[Ajedi32]

> someone is losing money when a nuclear reactor is sitting around not generating 100% power

Isn't that true of literally _every_ possible means of power generation though? Sure, there's an opportunity cost to not running your equipment at 100% 24/7, but only if there's enough demand to actually use that excess energy. You could argue solar and wind have it worse, since they don't even have the option to reduce output when it isn't needed (or ramp it up when it is, though it's a bit weird to think of cloudy days as an "opportunity cost").

Really though, we shouldn't _need_ to argue about which option is cheaper all things considered. Just remove as many regulatory barriers as possible and let the market sort it out. The only issue is that, as it stands, the regulatory barriers to nuclear are way higher than the regulatory barriers for wind and solar. (Governments are bending over backwards to accommodate the later, while in many cases effectively banning the former.) I'm just advocating for equal treatment.

[Retric]

On a level playing field without any regulation or subsidies, nuclear gets crushed.

> Isn’t that true of literally _every_ possible means of power generation though?

No, some peaking power plants get paid to sit around not generating power so they will be there for extreme events. They are paid not for power but for the possibility to generate power. This is viable because their operating costs when off are very low. Nobody can afford to employ 1000 people at a nuclear reactor so that someday the grid might want them to turn on for a few hours a year from now.

The operation vs standby costs of various types of energy generation vary wildly. It’s actually profitable for a natural gas turbine operator to install solar panels that only get used 1/2 the time simply to offset their natural gas fuel costs.

Actually, seeing solar panels installed at a fossil fuel power plant is however seriously trippy.

[hathchip]

France also built out tons of hydro at the same time as developing nukes.

Installed capacity of 1 part hydro and 2-3 parts nuclear is a viable energy source at any scale, nuclear on its own is not.

The problem is that building both hydro and nuclear is even slower and more difficult than just building nuclear.

[pyrale]

France was more like 80% nuke, 15% hydro (some of it being baseload) and 4-5% other at peak nuclear.

[Retric]

Don’t confuse consumption with production, they consumed closer to 60% nuclear electricity though this varied by year.

It’s telling when they recently shut 26 of 56 nuclear plants for weeks and still had enough electricity.

[hathchip]

In terms of energy produced, you are correct. In terms of installed capacity the ratio is closer to what I said.

[pyrale]

I'm not sure how you see hydro being necessary for nuclear, then. Are you talking about reversible hydro, or just any hydro plant?

[hathchip]

France built this hydro because it could not have moved from fossil fuels to 95% nuclear.

You can't realistically bring nuclear up and down in line with a daily demand curve. Hydro provides the peaking. Pumped hydro is better for this, since it provides both positive and negative balancing, but any hydro will do if there is sufficient storage.

If you have a good balance of nuclear and hydro, the nuclear will run close to full capacity and the hydro at less than half capacity, averaged over a day or longer. This is by design.

[pyrale]

> but for different reasons they both need increasing amounts of storage as you ramp them past that point.

Curious to know why nuclear would require storage. Having storage can make any production means more profitable, but there is no reason for nuclear to be non-viable without storage.

[Retric]

They are both technically possible, we could pay ~50c/kWh 24/7 for a 100% nuclear grid without storage in the same way we could build a global electric grid with power cables under the baring sea to use solar 24/7 without batteries. But realistically neither are viable without storage.

However, given the choice charging batteries via wind and solar just costs a lot less, thus why so few nuclear power plants are coming online each year.

[pyrale]

> But realistically neither are viable without storage.

What do you call viable? Because there's a huge difference between the consequences for nuclear not having storage or peakers (a moderate price bump) and wind/solar not having it (basically, hours/days of blackout).

If I reversed the situation and said that wind turbines have as much of a waste problem as nuclear, you'd be outraged, because having to handle radioactive materials isn't comparable to polymers not being recyclable.

[Retric]

>What do you call viable?

Something that might happen Aka Reasonably competitive with alternatives.

It’s physically possible to build infrastructure allowing 100’s of GW of solar electricity to move from Africa to South America across a single grid. But, there’s noway that is actually going to happen as it would be a horrific waste of resources vs local storage even ignoring political problems etc.

The same thing is true of Nuclear. There’s no way you’re going to see anything close to a 100% nuclear grid when adding hydro, batteries, wind, and or solar would drastically lower costs.

However, once you accept people aren’t going to do something that stupid you need to consider what mix of generation and storage is cheapest and how far from that we’re willing to go. That same logic is why nobody is every going to build days worth of battery storage, it’s simply cheaper to have extra capacity that mostly sits unused than extra storage.

[pyrale]

> adding hydro, batteries, wind, and or solar would drastically lower costs.

What does "drastically" means? Fine tuning of the grid doesn't require hydro specifically, batteries are not actually used in any meaningful way in countries currently using nuclear, and the projected savings of having such tech aren't transformative: the intraday difference between peak use and low use in a typical winter day in France are about 20-25% [1]. Sure, shaving 25% off your bill is great, but typically it's significantly less than the difference in price between different European countries.

As for wind and solar, it doesn't really lower costs of nuclear as there's no correlation (wind) or a negative correlation (solar) with peak winter hours.

[1]:

[hathchip]

> and doesn't require associated (not yet existing) massive storage.

That's not quite true. Nuclear generation is more flexible than wind and solar but much less flexible than thermal fuels, particularly gas. Although you can choose when you turn a nuclear power station on and off, you can't do so quickly, and your choices are somewhat constrained (regular downtime is required).

Nuclear worked well in the energy mix when it supplied base load (ie generate roughly the minimum daily load at all times) while coal supplied predictable peak load and gas could supply unexpected peaking. As this mix changes, availability patterns of nuclear start to be more of a burden. In a system without fossil fuels, you would have to have (some) energy storage, regardless of the split between nuclear, solar and wind.

The exact number for how much storage, how much overcapacity, or how much natural gas generation you would need vary with the proportions of solar, wind and nuclear of course, but it's an oversimplification to suggest that they are trivial if you use mainly nuclear.

[MagnumOpus]

To simplify without oversimplifying: Modern nuclear plants can change between 30% and 100% of nameplate capacity within half an hour, and that's how they operate in France for instance. While coal plants might rev up from 75% to 100% in a twice-daily shift change, but are clearly not set up to run at minimal loads.

https://en.wikipedia.org/wiki/Load-following_power_plant#Nuc...

[ZeroGravitas]

The lifespan is included in these calculations already.

This actually favours nuclear since they generally assume that a nuclear plant will be able to sell all its produced energy over many decades, even as solar is predicted to be cheaper than just their running costs for much of that time period.

[sligor]

> lifespan of nuclear projects, which is easily 2-3 times longer than solar and wind

Not sure it is a really solid argument. For wind maybe yes, but life expectancy of solar panels is about 40 years (warranty are 25-30 years but are pessimistic with margins). And with very low maintenance during the lifespan and easy replacement. Nuclear reactors have 60 years lifespan yes, but with massive continuous maintenance during the life cycle.

[invalidname]

As far Storage has become very cheap at grid scale e.g. https://www.abc.net.au/news/2019-09-12/is-renewable-power-ch...

This is an older article. As far as I understand this is even more in favor of green energy now.

[imtringued]

I'm sorry but the lifespan of the power plant is irrelevant.

LCoE is measured in dollars per MWh. If your powerplant is projected to go from $55/MWh to $110/MWh no amount of lifespan doubling is going to change that.

[sofixa]

It's not irrelevant. The cost to construct the nuclear power plant is very high, and is the main part of the price, and if it's amortised over 20 years, the cost of MWh produced is higher than if it's amortised over a more realistic 50 or 70 years.

[Gwypaas]

Locking up capital for 70 to 80 years? Do you know about the time value of money? [1] What you are proposing is FDR and Churchill sitting at the Casablanca Conference at the height of the second world war in 1943 [2] planning out economic investments that will be returned in 2023. Does that sound reasonable?

Likewise, a renewable plant that makes the money back in ~25 years allows the investor to reconsider the investment decision and better optimize the capital allocation two times while the ones building nuclear are still paying for the nuclear plant. The renewable owner could after 50 years decide to invest in nuclear. I wonder which is more efficient?

[1]: https://www.investopedia.com/articles/03/082703.asp

[2]: https://en.wikipedia.org/wiki/Casablanca_Conference