Nuclear has effectively zero unit cost, it's all fixed cost. Once you build the reactor, you generate at full capacity all day and night and take whatever the market rate is.
That means you get paid less during the day when solar is the cheapest provider, but you still match its price and supply your full generation capacity even then, because you can, because anything is more than nothing and the incremental generation cost is effectively zero. Then you make more at night when solar requires storage, and even more still whenever the day was overcast and energy prices had to rise high enough to suppress demand enough that the depleted storage isn't fully exhausted before sunrise.
The capacity factor is always 100%, it's the price you get at any given time that varies. But price variability in itself is no problem either, as long as the average price is above average cost. And that average includes not only times that it rains for multiple days in a row, but also winter, and that with consideration of people also needing to switch from oil and gas to electric heat.
"as long as the average price is above average cost"
And that's the killer. Add enough solar and wind and the average price craters, even if solar and wind cannot handle everything (without help from dispatchable sources and/or storage).
The problem with that argument is that it proves too much. It applies to everything, because solar and wind are the same as nuclear, their incremental generation cost is effectively zero once the capital is paid, so you take whatever the market price is.
Which means you have a problem.
It's physically possible to build enough solar and wind and storage to even handle extended periods of low sun and low wind. But if you do that then on a normal day with a normal amount of sun and wind, you have a large oversupply and the price falls to zero and everybody goes bankrupt.
That happens less if you add real baseload like nuclear to the mix, because something with a fixed generation capacity doesn't have a periodic shortage requiring you to compensate with an overcapacity which bankrupts everybody on a normal day.
Solar and wind will build out until it is no longer profitable to do so. That means: when they have driven the price of electricity when they are operating, averaged over time, down to the point they can't get a return on investment. This price point is far below where new nuclear needs it to be for nuclear to earn back its construction and operating costs.
What drives all this is that the levelized cost of energy from solar and wind is much less (a factor of 3 or 4) than the levelized cost of energy from new nuclear plants.
Nuclear, or at least many existing nuclear plants, has an additional disadvantage: it cannot cut output rapidly. If power from renewables suddenly floods the market, prices can go negative. Renewables can just stop selling in that situation, but nuclear is forced to continued to run and eat the negative revenue. The low power density of the renewables sources, usually depicted as a negative, is the source of this advantage: sunlight absorbed in PV modules can just be allowed to dissipate as heat there with no negative effects.
(In fairness, I should also mention that there are subsidies in the US that encourage renewables to keep generating even at negative prices. These subsidies will have to go at some point, and perhaps that point is now.)
What will be the final death knell for nuclear will be when short term storage gets cheap enough that the times when that's discharging will also be economic death zones for nuclear plants. I expect few of the existing operating nuclear power plants to survive after that.
> Solar and wind will build out until it is no longer profitable to do so. That means: when they have driven the price of electricity when they are operating, averaged over time, down to the point they can't get a return on investment.
It seems like you're expecting this to be a slope rather than a cliff.
The problem with generation methods with no incremental generation cost is that absent some coordination/collusion, you go straight from a price somewhere above breakeven to basically zero as soon as you have any significant amount of oversupply, because everybody would rather get something than nothing.
So even if the average wholesale price is currently above 2c/kWh and you can bring capacity online that generates at 2c/kWh, you won't, because the act of doing it would create oversupply, cause the average market price to fall to below 2c, and you and everybody else would lose their shirts.
It's basically a market that bankrupts everybody without some coordination, but part of the value of that coordination includes preferring some amount of stable generation capacity to avoid the high cost of supply emergencies when low supply from the unstable generation methods coincide with each other.
> What drives all this is that the levelized cost of energy from solar and wind is much less (a factor of 3 or 4) less than the levelized cost of energy from new nuclear plants.
This isn't accounting for variable supply and demand. The price it costs to generate in the summer sun isn't the real price when the unmet demand is in the winter night, and it isn't going to be economical to shift the demand by six months using energy storage. But if you had enough solar to provide heat in cold climates in winter you would have so much oversupply the rest of the year that you wouldn't make a cent for nine months out of twelve.
> Nuclear has an additional disadvantage: it cannot cut output rapidly. If renewables suddenly flood the market, prices can go negative. Renewables can just stop selling in that situation, but nuclear is forced to continued to run and eat the negative revenue.
If that actually started happening on a regular basis there would be obvious solutions like resistive heaters or on-site energy storage which can be charged during those periods and then sold for a profit when prices are higher. (Thermal storage could work really well considering the reactor generates heat to begin with and they already have existing heat-to-electricity systems on site.)
And that's assuming all of this "smart grid" stuff doesn't ultimately succeed in preventing that from happening by increasing consumption as prices fall so that they don't actually go negative to begin with.
> What will be the final death knell for nuclear will be when short term storage gets cheap enough that the times when that's discharging will also be economic death zones for nuclear plants.
It's only speculation that this will actually happen. And even now people like to use overly optimistic numbers. Storage costs a certain amount if you charge it up every day and then discharge it again every night, but if you want the storage to be able to handle generation undersupply over a period of a week or more, you need a lot more of it which will generally go idle, which requires you to charge higher average prices per kWh.
> I expect few of the existing operating nuclear power plants to survive after that.
The existing nuclear power plants will keep going as long as the market price is above the operating cost. The capital costs are sunk.
That means you get paid less during the day when solar is the cheapest provider, but you still match its price and supply your full generation capacity even then, because you can, because anything is more than nothing and the incremental generation cost is effectively zero. Then you make more at night when solar requires storage, and even more still whenever the day was overcast and energy prices had to rise high enough to suppress demand enough that the depleted storage isn't fully exhausted before sunrise.
The capacity factor is always 100%, it's the price you get at any given time that varies. But price variability in itself is no problem either, as long as the average price is above average cost. And that average includes not only times that it rains for multiple days in a row, but also winter, and that with consideration of people also needing to switch from oil and gas to electric heat.