> Solar is cheaper than nuclear the rest of the time, so (due to the way energy markets work) we pay solar producers the cost of nuclear generation, creating strong incentive to build out more solar + battery.

The way energy markets work is that prices change based on supply and demand. So you build a lot of solar and then the price goes down at times when solar generation is high. (There isn't really enough solar in the grid for this to have happened yet in most places.)

Because nuclear has high capital costs but low unit costs, it continues generating all the time even if prices are lower some of the time. It needs to hit a particular average price but will take essentially any instantaneous price. It can't operate with an average price of $0.02/kWh but it can sell for that at noon in July as long as it's getting enough to offset it at midnight in December.

So, you add a bunch of solar to the grid and as a result the price of electricity during daylight hours in the summer falls below the average breakeven cost for solar. Nuclear is still getting more than that at night, when the price won't fall below the daytime price plus the cost of storage, and in the winter, when solar output is lower and heating demand is higher.

Because it's still recovering a lot of its costs on summer nights and even a little bit on summer days, the winter price nuclear needs never gets to be 4x as high, but it goes up some.

Now the solar companies are looking at that price and trying to decide whether it's worth getting. If they add more solar it's going to generate e.g. 4kW in summer, 2kW in spring, 2kW in fall and 1kW in winter. So for every 9kWh they generate, only 1kWh comes in winter, and that's during the day.

Suppose solar needs to average $0.04/kWh to be sustainable but we're already at $0.02/kWh in the spring, summer and fall. Then solar needs the price during the day in winter to be at least $0.20/kWh before wanting to add more. On top of that, if they do add more, then the price during other seasons drops. If those hit $0.01/kWh during the day then solar needs the price during the day in winter to be $0.28/kWh. And, of course, then the price in the other seasons would drop again, until solar would need to have $0.36/kWh in the winter because by then they would only be getting paid in the winter. Whereas nuclear still makes some in the summer at night.

On top of this the demand is higher in the winter at night, but it's not higher in the summer at night. So to use solar with storage, you would also have to amortize the cost of the additional storage during only the winter months. And the winter nights have more hours than the days, don't forget. So for solar to supply all power in the winter it needs the daytime price to be $0.36/kWh and the nighttime price to be more than twice that.

Then solar is intermittent. People still need heat when it's cloudy. Overbuilding solar sufficiently to handle the heating load on a cold cloudy day in winter? Nope. Hopeless. You'd have to use long-term storage in addition to short-term storage and that's a whole new category of expensive when we were already looking at prices nobody was going to like.

So what happens instead? Well, probably the daytime price in the non-winter months gets to something like $0.02/kWh and the daytime price in the winter gets to something like $0.20/kWh and you fall below the breakeven point for building more solar. Then nuclear takes the same $0.02 and $0.20 during the day and something more than that at night and continues to cover its costs.

Really the problem here is not that solar is going to push out nuclear in the winter, it's that natural gas could unless you have a price on CO2. Or more likely, without a carbon price people might not switch to electric heat instead of continuing to heat with natural gas and fuel oil.

So we kind of need a carbon price to do heating. Or to solve it from the other side (which would be even better), we need to get nuclear to be more cost efficient so it can out-compete natural gas since solar won't.

All of the discussions here conveniently ignore the existance of Wind. Which fortunately has higher yield in the months when there is less sun.

So why are the Swedes investing heavily in nuclear energy again, after nixing the nuclear exit they had on the books?

HVDC (and even the grid in general) doesn't transmit all that much power. The largest currently existing line - Changji-Guquan UHVDC link in China - transmits 12GW. It's significantly more than what an average long-range link of current grids transmits, yes. But is it a lot?

Looking from consumption side, my home city of ~1 million people has several coal-fired plants, producing 1.5 GW of electricity and about 5GW of heating. Plus an hydropower station producing 6 GW. Most of that is consumed by an aluminium plant, but nonetheless, it's also part of the city. So that's roughly 12GW on a cold winter day (I suppose we do want to make heating cleaner as well), and probably 6 GW in summer. Heat pumps could be used to reduce power consumed by heating, but even the air-source pumps are not cheap, and they don't provide much efficiency gain in the cold. Ground-source pumps are extremely expensive and reqiure heat replenishment or the ground will freeze - such is the balance here.

So, the world's largest link to power just one city, out of tens of them. It quickly gets prohibitively expensive.

The only realictic answer, it seems, is annual-scale storage. I hope that "dirt pile storage" works well enough and succeeds, batteries are just too expensive and material hungry, hydrogen is problematic to store well either and we don't seem to have good enough scalable direct carbon capture to synthesize methane or propane.

> Sweden is just about the worst case, there's very few countries/people that far north.

Sweden is worse but it's still a significant issue in e.g. New York or Paris or Auckland.

> There's genius invention called "wires". HVDC has transmission losses on the order of 3.5% per 1,000km. You don't have to colocate the solar.

It's more than 1000km from the places that get cold to a part of the world where it isn't winter.

Suppose we ignore that it's winter in the US Northeast and Southeast at the same time and run HVDC 2000+ km to Florida because it gets an extra hour of sunlight. Long distance transmission can't be used to counter seasonal output and regional weather at the same time because one requires the generation to be spread everywhere and the other requires it to be concentrated closer to the equator. If we concentrate the solar in Florida to mitigate winter in New England then we're screwed when Florida is overcast.

Wires and HVDC transmissions are nice, but they have a fairly large downside. They are major infrastructure projects that cost a lot of money and they don't produce any energy. Adding that cost to the solar panels makes them significantly more expensive, and solar/wind farms owners are not exactly willing to bear that cost.

You don't need to colocate the solar, but you need to make sure you can get that power when you actually need it.

During crisis nations are going to restrict exporting electricity and prioritizing their own residents. Electricity that is generated in Germany is not going to warm up Nordic countries if Germany doesn't let it.

Wires are also susceptible to sabotage, especially undersea ones (which are the current major connection points to Europe).

> The cost to build is funded through debt that’s paid off over time. The construction costs aren’t in the past; they’re in the present, and in the future, in the form of debt payments.

That isn't a future cost, it's a past cost with a future payment date. It's like taking out a mortgage on a piece of land to buy some lumber and build a house on it. The past cost is buying the lumber; the hardware store isn't going to give you a refund six months after you already paid them and used the lumber to build the house. What you have now is a house instead of money and separately a mortgage against the house.

What do you think happens if you don't pay the loan? Is the bank going to get a refund from the hardware store? No, they're going to take the house, sell it to someone else for whatever they can get for it and apply the money to the loan. And then the house continues to operate as a house.

The same thing happens with a power plant. If the plant company itself has a bank loan and isn't making enough to pay it, the bank is going to foreclose, sell the plant to someone else, possibly take a partial loss, and the new owner -- who might have gotten the plant for a lower price than it originally cost to build -- is going to continue to operate it as long as its revenue exceeds its operating costs.

And that's assuming the plant was funded with a bank loan. If it had investors then there is no loan payment; the "loan payment" is when the company pays the owners dividends. If they were expecting the plant to pay enough dividends to recover their initial investment plus interest and its operations only generate enough revenue to repay most of the original investment but no interest, then they continue to operate the plant (or sell it to someone who does) because "recover 90% of the original money instead of the expected 200% of the original money" is still significantly more than "recover 0% of the original money by closing the plant".

What happens if the operating company can't pay the debt? The bank repossesses the facility. Now what does the bank do with a solar facility? Does it (A) let it rot, losing massive value, (B) run a bulldozer through it, destroying massive value or (C) find a way to operate it, receiving profit from doing so?

> grid infrastructure that has reach across timezones,

"Night" reaches across more time-zones than you can build your grid across.

Never mind "winter".

> nuclear powerstations are not as reliable as you might think - they often go offline.

Define "often".

They are actually a lot more reliably than you seem to think: the capacity factor of the US fleet, for example, was >90% for the last decade(s). And that <10% offline time includes the planned refueling/inspection/maintenance times.

Nuclear power plants are incredibly reliable.