I would be nice to see it be a nighttime base-load provider, but I think it has been kneecapped for so long in so many different ways, that it is effectively a dead technology (in the US at least).
It's a little less than renewables and a little more than coal.
> I would be nice to see it be a nighttime base-load provider
The bigger issue with Solar is the dramatic difference in generation between the summer and winter. So not only do you need something for "nighttime," you also need something for "winter."
A proper multiply connected grid goes some way to solving the winter problem by allowing southern solar farms to sell excess production to the north, similarly for the east and west to reduce the effective length of the night.
But such a grid would require many billions of dollars in high voltage DC lines to be constructed to move that power such distances. We simply don't spend that much money on infrastructure to make that feasible.
Not OP - but dead as in "not financially viable to add new grid capacity". Had we gone all in on nuclear 40 or 50 years ago, that might be a different story but alas. You can buy so much more solar + storage than nuclear capacity now, and the former has definite timelines and almost zero construction risk. The latter bankrupts whichever companies touch it.
Vogtle 3/4 saw negative learning rates (the second was more expensive to build than the first) and came in a decade behind schedule at over $20,000/KW. Enormous solar + storage projects are coming online at more like $2,500/KW. Their levelized cost is less than half the comparable cost to a nuclear plant without all the risk.
At current prices, to power the UK for 36 hours off battery storage, it would cost (according to Gemini) £150bn to £400bn for the batteries alone - that doesn't include purchasing land, wiring them up and so on, nor the cost of the wind farms and solar panels to feed the batteries. We don't/can't/won't do hydro pumped storage in the UK.
Sizewell C, our massively overpriced nuclear power station, comes in at about £100bn.
At current prices, to power the UK for 36 hours off battery storage, it would cost (according to Gemini) £150bn to £400bn for the batteries alone - that doesn't include purchasing land, wiring them up and so on, nor the cost of the wind farms and solar panels to feed the batteries. We don't/can't/won't do hydro pumped storage in the UK.
Sizewell C, our massively overpriced nuclear power station, comes in at about £100bn.
It comes in at £100bn for 3GW - that's about 7% of total UK consumption. A huge amount to be sure, but the choice isn't £100bn for Sizewell or £400bn for batteries.
Rolls Royce SMR for Czechia at 0.5GW is about £3bn
To be clear - this doesn't exist. They "aim" to start construction in 3 years and it's a FOAK project so will surely be delayed. The initial target was a cost of £2bn - the paper cost has already increased by £1b without a shovel turning over a single grain of sand.
It's a huge shame and problem across the 'West' - but we can't build megaprojects for anything resembling a reasonable cost any more.
Unfortunately for British nuclear advocates - their plants have substantially worse economics than US ones. Hinckley C was granted its license in 2012 and authorized in 2013 with a strike price of £89.50 per MWh when the projected cost was £7 billion -- it's now scheduled to be completed in 2030 at a cost nearing £50 billion. There is literally no amount of baseload generation worth that cost.
The wholesale price coming out of Hinckley will be higher than the retail price in most of the US.
Have you ever asked yourself why this is? There are two basic types of reasons why the UK nuclear industry does so poorly. The first reason are scientifically illiterate "activists" that delay and block construction. But this is only part of it. The other and possibly bigger part is that the UK doesn't build stock nuclear plants. They "Wallace and Grommet" things far too often. In this case, that means not building proven designs but instead building experimental ones. Specifically building high temp reactors which haven't been built before. Normally (at least in the US) we build pilot plants (think 1/10th the normal size, so 100MW or so) and then once we get the kinks worked out, we built the full size plant. In the UK they just built full sized experimental reactors. This is why the UK has so many problems with (the cost of) their nuclear plants. Had they built a stock design (like the French) they wouldn't have these problems.
What made the cost to balloon so much, and the timeline to set so huge a delay? Technical problems? Regulatory hurdles? Locals resisting? Running out of money?
Most of the Hinkley Point C and Sizewell C costs come from interests being paid in the high risk debt raised to build it.
Unsurprisingly it turns out financing such a long term project on a 10% rate debt is not a good idea. (Keep in mind that this debt has been borrowed during ZIRP, where the baseline interest rates where near 0, this rate is just the premium that private actors are taking for the risk to see such a project being canceled for political reason). Of course the sensible way to do it would have been to finance it on long term public debt guaranteed by the British govt, who would then have born the burden of an hypothetical cancelation of the projects (and the cost per MWh would have been 4 times lower IIRC), but public is a forbidden word for the Tories.
10-11 Sizewell C's, assuming the UK 2025 average demand of ~33 GW.
A couple more to deal with outages. Plus some battery capacity, which you want because nuclear doesn't load follow well. At £38 billion per, you need about £500 billion to power the whole country. About 40% of one annual UK budget.
It's possible to imagine building Sizewell C's at considerably lower cost, as is done it other nations that exist, and manage to build such things without the huge multipliers, but that's another matter.
There are nuclear designs that follow load just fine. The UK just didn't build them. Also, the designs that load follow are usually liquid fueled reactors, not solid fueled ones which are normally built. PWRs generally have the lowest utility costs (the actual cost you pay, not the cost to build which is normally quoted) of any power source we have tried. Even the fast-sodium reactors the Soviets built can't match their price. That's part of why building stock nuclear designs works so well (eg the French) but trying to build experimental reactors doesn't. And when you try a new design, you have to build a pilot plant first to work out the kinks. The British always skip this step and this shows in the results.
Why are you pretending that there's no wind in the UK? Or Germany? In fact, all of Europe is basically surrounded by a coastline with huge amounts of wind resources.
Or just gas peak already.
But look, if data centers want to fund a whole bunch of advanced nuclear and wait 10 years for it to get built and will foot the bill for the cost overruns, okay I'm not going to stop them.
But let's face it all. This nuclear stuff which is going to be for data centers is all an externalized cost that's going to be dumped on the public.
I’m not pretending anything. I simply posted some easy to double check numbers. I’m pointing out that just this week the wind didn’t blow for 3 days, and that happens frequently during in European winter. So much that the Germans coined a phrase for it.
The military is given extra leeway in its decision making. There is no NIMBY for military projects. But, as for civilian power plants, we don't really build them in large numbers like we used to. In the 1950s, it was thought that every town would have its own nuclear powerplant, but that clearly didn't come to pass.
It's a little less than renewables and a little more than coal.
> I would be nice to see it be a nighttime base-load provider
The bigger issue with Solar is the dramatic difference in generation between the summer and winter. So not only do you need something for "nighttime," you also need something for "winter."
Vogtle 3/4 saw negative learning rates (the second was more expensive to build than the first) and came in a decade behind schedule at over $20,000/KW. Enormous solar + storage projects are coming online at more like $2,500/KW. Their levelized cost is less than half the comparable cost to a nuclear plant without all the risk.
At current prices, to power the UK for 36 hours off battery storage, it would cost (according to Gemini) £150bn to £400bn for the batteries alone - that doesn't include purchasing land, wiring them up and so on, nor the cost of the wind farms and solar panels to feed the batteries. We don't/can't/won't do hydro pumped storage in the UK.
Sizewell C, our massively overpriced nuclear power station, comes in at about £100bn.
Rolls Royce SMR for Czechia at 0.5GW is about £3bn - https://en.wikipedia.org/wiki/Rolls-Royce_SMR#Cost
The current dunkelflaute with no wind, and little sun, in Northern Europe has been 3 days, or 72 hours. This is a regular occurence.
The longest dunkelflaute was the winter of 1996/97, for 55 days.
Sizewell C, our massively overpriced nuclear power station, comes in at about £100bn.
It comes in at £100bn for 3GW - that's about 7% of total UK consumption. A huge amount to be sure, but the choice isn't £100bn for Sizewell or £400bn for batteries.
Rolls Royce SMR for Czechia at 0.5GW is about £3bn
To be clear - this doesn't exist. They "aim" to start construction in 3 years and it's a FOAK project so will surely be delayed. The initial target was a cost of £2bn - the paper cost has already increased by £1b without a shovel turning over a single grain of sand.
It's a huge shame and problem across the 'West' - but we can't build megaprojects for anything resembling a reasonable cost any more.
The wholesale price coming out of Hinckley will be higher than the retail price in most of the US.
https://www.theguardian.com/environment/2021/apr/28/sizewell...
Unsurprisingly it turns out financing such a long term project on a 10% rate debt is not a good idea. (Keep in mind that this debt has been borrowed during ZIRP, where the baseline interest rates where near 0, this rate is just the premium that private actors are taking for the risk to see such a project being canceled for political reason). Of course the sensible way to do it would have been to finance it on long term public debt guaranteed by the British govt, who would then have born the burden of an hypothetical cancelation of the projects (and the cost per MWh would have been 4 times lower IIRC), but public is a forbidden word for the Tories.
A couple more to deal with outages. Plus some battery capacity, which you want because nuclear doesn't load follow well. At £38 billion per, you need about £500 billion to power the whole country. About 40% of one annual UK budget.
It's possible to imagine building Sizewell C's at considerably lower cost, as is done it other nations that exist, and manage to build such things without the huge multipliers, but that's another matter.
Or just gas peak already.
But look, if data centers want to fund a whole bunch of advanced nuclear and wait 10 years for it to get built and will foot the bill for the cost overruns, okay I'm not going to stop them.
But let's face it all. This nuclear stuff which is going to be for data centers is all an externalized cost that's going to be dumped on the public.
https://gridwatch.co.uk/
I find that sarcastic comments are very hard to pull off on Hacker News; I tend to avoid making them unless I'm sure most readers will get the joke.