another big selling point of nuclear is it’s easy to drop in at old coal plants that already have grid connections ready to go.

pretending today’s limitations are permanent is a strange way to plan tomorrow’s energy system.

how long are you willing to wait for perfect energy storage?

obviously battery tech will improve, but the point is we cannot afford to assume that problem will be solved 100%, meanwhile experiencing more and more irreversible damage to our environment. We need to put our eggs into all clean energy baskets to get off CO2 as soon as possible. Not whenever battery tech happens to be ready to handle everything.

Nuclear is simply incompatible with renewables, if the price tag is of any concern to you. Nuclear reactors take long and cost a lot to build, still aren’t cheap to operate if you include the infrastructure cost of getting, handling and storing the fuel, can’t respond to chaning demand quickly and are centralized.

The fact is that the price tag is not the only factor.
It is true that a 1000 MW nuclear plant cost about 10 times the equivalent solar field (about 6-9 billions for the nuclear plant Vs about 800 millions for panels) but at this point you use about 15 square Km to just match the output of the nuclear plant, with no storage, and only in the best case scenario.

In a real case scenario the price tag of building a solar field that could replace a nuclear plant is probably not that different anymore.
To produce the same amount of power in every condition you could not install the exact amout of panels, you need way more (double ? triple ?) with the associated costs and space requirements.
And then you need batteries to be able to offer energy also at night and to store energy for the days when the sun is not enough to produce the energy required.

I would say that to match a nucler plant a solar field would cost about half of it for the same power generation but with the additional problem that you need way more space, which itself is a cost.

Nuclear power plants need subsidies over their whole life, mainly to handle the risk inherent with nuclear fuel and the finincial risk of having to handle spent fuel for centuries to come.

We have nuclear power plants that uses less problematic fuel which is easier to handle and less risky.

Nuclear power plants need to run at full power, which costs money every time renewables are pushing the energy price into the negatives.

True, but when the renewables are not producing enough ?

Nuclear power plants need a centralized power grid.

Like a solar farm.

Renewables work in exactly the opposite way. They are cheap to build and maintain, they are variable and decentralized. So to use renewables effectively, you need to build a decentralized grid and ideall, on-premise energy storage.

No, they work in the exactly same way. What is different is that renewables can be made decentralized. Solar panels on the roof produce only for the house, not for the power grid (even if you can connect them to the grid to fill in the surplus).

I think you are comparing a nuclear plant with the small panels you install on your roof but you cannot make such comparison.

So the energy grid of the future will be purpose-built to accomodate renewable energy.

As far as I remember, what was told me when I worked in the field was that power grids does not care about the source, only the power: a power grid do not care if the 10 Gw power it transfer comes from a nuclear plant, a solar farm or from a squirrel on a wheel.

And the grid is already ready, we are already able (ok at least here) to send to the grid the surplus generated by the solars panels on the roof. And adding the batteries will not require anything done on the grid itself (at least this is what my collegues said).

Trying to force nuclear energy generation into that mix only makes sense, if you have a clear, non-monetary reason to do so, like enriching uranium for bombs. For strictly commercial use, nuclear power is dead.

What about being sure that a 5 days of bad weather cannot stop a society ?

There will be diversity in the future grid, but that diversity will be in the forms of energy storage used. I think the grid of the future is almost entirely solar but with different energy storage systems optimized for different time scales. For storage on the order of seconds, you use fly wheels to replace the spinning mass once provided by gas turbines. For daily swings, vast warehouses full of old electric car batteries work well. For longer term storage, you can use chemical energy storage or flow batteries. Think making vast tanks of ammonia or methane. You use excess electricity in the summer, store it as chemical energy in vast tank farms, and then burn it or run it through a fuel cell in the winter.

Although, for seasonal variation, the most practical solution might be just to spam solar panels. It just depends on the price ratio of solar panel cost to storage cost. You could have a grid that just had so much solar deployed that it is able to cover energy demand even on cloudy winter days. You just flat out build that many panels. Maybe then in the summertime you're capable of producing 10x your normal power needs.

Seems wasteful, but this would just encourage a lot of very energy-intensive industries to become more seasonal. We have a farming season, why not an steal smelting season? Why not an AI model training season? Maybe for datacenters, the big runs are done during the summer when power is dirt cheap. Then the companies spend the winter planning next year's runs, etc. Or the extra power could be used to make synthetic fuels from CO2 pulled from the air. Maybe we never figure out how to make batteries good enough to power a trans-oceanic jumbo jet. But maybe we don't have to. You offer dirt cheap energy on a seasonal basis, and people will find uses for it eventually.