Hacker News Reader

Comment by mrpopo | Hacker News Reader

The failure of EDF in France is not due to nuclear power inherent limitations, but to massive underinvestment from the French government. The Energiewende has cost the Germans more than 150 billion €, all that with lenient EU regulations.

Whereas in France, the new EPR in Flamanville, for all it being talked about as a failed investment, has cost 1/10th of that. With 10x the money and economies of scale, we could have had a crazy nuclear capacity with very strict, modern standards.

Whether France will build nuclear or renewables, the problem is they had to do it 10 years ago.

View on HN

Nuclear power has high upfront investments, always. And pretty much all nuclear projects are finished behind schedule and wayyy above budget. After a couple of decades of this, reasonable people may come to the conclusion that these investments and risks are quite "inherent".

Your comparison is meaningless, because the 150 billion € investment is not comparable to a single nuclear plant.

And realizing ten times the current building capacity for nuclear plants with the same (or stricter) regulation and standards? That is almost delusionally optimistic.

> Nuclear power has high upfront investments, always. And pretty much all nuclear projects are finished behind schedule and wayyy above budget.

Which grid-scale solar projects have gone into production on schedule and under budget? Are there any?

Yes, it's actually pretty common. See comparisons across generating technologies in "An international comparative assessment of construction cost overruns for electricity infrastructure" [1].

Here's what the researchers found for solar (section 3.4):

Thirty-nine solar PV and CSP power plants, representing 2374 MW of installed capacity and $16.5 billion worth of investment, were analyzed. (These are utility-scale infrastructure projects, thus they do not include the familiar rooftop solar collectors on buildings or residential configurations.) These projects as a class actually came in $4.2 million less than budgeted, or $200,000 less than expected per project. These solar systems had the lowest average cost escalation per reference class (1.3 percent), the least time overruns (an average of 2 months ahead of schedule), and the lowest standard deviation for amount of overruns and time overruns. They also had the largest total amount of cost underruns, with the entire class of projects costing $163.9 million less than budgeted, though some of this may be explained by dramatic reductions in cost over the past 4 years. As Table 5 indicates, only 3 facilities had overruns greater than $50 million and only 3 facilities had cost overruns greater than 30%, all of which were CSP facilities rather than large-scale solar PV apparatuses.

Here are a few solar projects that outperformed in recent years:

O'Brien solar farm in Wisconsin, completed in 2021 on time and under budget:

https://www.boldt.com/project/edf-renewables-obrien-solar-fa...

Mount Signal 3 in California, completed in 2018 ahead of schedule:

https://www.solarpowerworldonline.com/2018/12/8minutenergy-c...

Sun Streams 2 in Arizona, completed under budget and ahead of schedule in 2021:

https://www.enr.com/articles/55112-best-project-energy-indus...

Hillston solar farm in Australia, reached full commercial operation in 2022 ahead of schedule:

https://www.solarquotes.com.au/blog/hillston-solar-farm-mb24...

Kuala Ketil solar farm in Malaysia, completed in 2019 ahead of schedule:

https://www.pv-magazine.com/2019/09/12/50-mw-solar-park-come...

[1] https://www.qualenergia.it/sites/default/files/articolo-doc/...

1) None of these are grid scale. 2) All of them are heavily subsidized.

The closest I see in that list is Mount Signal 3, which comes in at about 1/3 the capacity of a typical grid-scale generating station (1/6 if you account for the fact that it's producing zero power half the time, on average). It also chews up nearly 2,000 acres of land.

Hillston is 120 megawatts. Sun Streams 2 is 200 megawatts. Mount Signal 3 is 328 megawatts. The coal fired generating plants that retired in the US in the past decade were often smaller than that.

https://www.eia.gov/todayinenergy/detail.php?id=7290

The table above shows that the coal generators that retired between 2009 and 2011 had an average size of 59 megawatts (MW). By contrast, the average size of a coal-fired plant planned for retirement between 2012 and 2015 is 154 MW, more than twice the average size of the units retired during the 2009-2011 period.

In 2022, coal fired generators had a capacity factor of ~48% in the United States:

https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...

Photovoltaic solar farms attained ~25% capacity factor in 2022:

https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...

As a rough guide you can estimate that an American solar farm will generate about as much electricity annually as an American coal plant with half the nameplate capacity. Mount Signal 3 would roughly match a 164 megawatt coal fired generator for annual generation.

We were comparing nuclear and solar, not coal and solar.

164 megawatts (328 / 2, to account for the fact that solar plants are offline a minimum of half the time) is not anywhere in the same league as nuclear power plant, which are typically in the 2-3 gigawatt range, with modern large-scale plants producing 6-8 gigawatts.

The Mount Signal 3 plant also used up 2,000 acres (800 hectares of Imperial Valley farm land.

Also, the decommissioning costs are really high, are usually not factored into the investment and often have to be absorbed by the state.

Current calculations from German decommissioning projects point at 10-20 billion per plant.

> And pretty much all nuclear projects are finished behind schedule and wayyy above budget.

All projects do, including wind and solar.

--- start quote ---

Ernst & Young (EY) has found that an average power and utility megaproject is delivered 35% over budget and two years behind schedule

Of the megaprojects surveyed, 64% were delayed and 57% were over budge. Almost three-quarters of hydropower, water, coal and nuclear infrastructure projects were over budget by 49% on average,

--- end quote ---

https://www.offshorewind.biz/2016/12/02/offshore-wind-projec...

> And realizing ten times the current building capacity for nuclear plants

How much do we need to overbuild wind and solar?

Nuclear power projects are not just over-budget, they start off with a budget of a couple of billions, whereas individual solar or wind power projects are way smaller and therefor carry less of a risk. Nuclear power requires a huge financial backing from a single entity (even it is "fed" by multiple sources). Almost always, the risk is carried by society, the profit, if any is gobbled up by some corporation.

There is no need to "overbuild" wind or solar. There has never been any illusion those sources will produce at 100% capacity all the time. There will eventually be energy storage to level out the peaks, but most countries still have years to build up to the point where this is necessary.

Let's say this number is correct: Flamanville has cost a 1/10 of that and has produced 0 kWh, and won't produce any at least 20 years since the project began. Germany produces around 250 TWh worth of electricity per year from renewables, and has done so to an increasing extent in the last 20 years, especially in the last 10. Flamanville Unit 3, I haven't found the estimated production, but judging from the other two units existing it should be in the ballpark of 10TWh per year.

Nuclear has failed to scale, it's a well documented issue. If we want to decarbonize quickly, I cannot imagine how it can make sense to invest a 1/10 for a 1/25 of the production, and then wait decades longer for it. If decarbonization is the goal, projects like Flamanville are already a failure: even if finished it will take decades and decades to make up for the construction time during which it did not produce a single kWh and did not displace a single molecule of co2. Germany was over 50% coal 20 years ago, it would still be at this level if it had gone that route. It can't be the solution now.

> Nuclear has failed to scale

This reminds me of the joke "Fusion is 30 years away and always will be".

The reason is that a panel of scientist predicted various timelines of viable fusion between 30 years and infinity years based on effective funding. Ever since the funding for nuclear research was significantly below the level of the infinity years[0].

I suspect nuclear fission followed a similar path in the last decades.

[0] https://astralcodexten.substack.com/p/your-book-review-the-f...

> Nuclear has failed to scale, it's a well documented issue.

It hasn't, and it's not a "well-documented issue". What's documented is decades of FUD, fear-mongering, underinvestment etc.

> If decarbonization is the goal, projects like Flamanville are already a failure

Why not look at projects like Fuqing Nuclear Power Plant instead?

> will take decades and decades to make up for the construction time

Ho long will it take to overbuild renewables and all the required power storage for them?

E.g. right now, during the day, Germany's renewables are generating:

- wind: 20% of installed capacity

- solar: 34% of installed capacity

- pumped hydro storage: sucking up 8% of total power generation for recharging

> Germany was over 50% coal 20 years ago, it would still be at this level if it had gone that route.

Which route? E.g. France's carbon output from electricity generation is ~56g/kWh. Germany's is ~340g/kWh. Care to guess why?

It's well documented even by people in favor of nuclear. For example: https://www.construction-physics.com/p/why-are-nuclear-power...

> Why not look at projects like Fuqing Nuclear Power Plant instead

Because the person I replied to talked about Flamanville... Do you even read what is discussed or just frantically go from one post to the other?

I also posted how long it took to reach 50% renewables - 20 years, during which time the EDF/Areva failed to complete almost all of the projects they were involved with. If Germany had gone that route it would still be pumping out 800g/kWh like Poland does and it would be doing that for the last 20 years. Thank God it didn't make that mistake.

I talked about Flamanville because the article is about France and I mentioned France's underinvestment. If you want to talk about well funded projects then yes the Chinese have done a much better job. But you're not really talking in earnest.

I'll repeat again. Nuclear power projects in France took 20 years and were not completed because of underinvestment; about 10% of what the Germans paid for the Energiewende. If the French invested 10x more money, the industry could have been rebuilt. Instead they just perfuse enough to keep it half-alive.

So we should talk about that one that maybe was not a total disaster, but we should not talk about the three others that have been. I notice a lot of no true Scotsman rhetoric by nuclear people in general: France is proof we can build nuclear fast, but they didn't do it right by not maintaining properly. Germany is proof we can run nuclear plants efficiently, but they are incompetent for not being able to make it work. Etc. etc. If only the French had done this or that. But in the end it does not work.

You don't have to repeat anything, I heard you the first time. Flamanville cost a 1/10 of the Energiewende and its output is 0. Its theoretical output is 1/25 of the actual output of German renewables. That's not a good deal. If the goal is to decarbonize, Flamanville is already a failure, for reasons I already posted.

You say you heard me but didn't seem to understand it. That's okay, I notice a lot of misunderstanding of things by anti-nuclear people in general :)

(I'm copying your rhetoric in case you don't notice)

Thank you

Except my rhetoric is based on what you said (coulda would shoulda and it would all be peaches), not on made up impressions of other people. Living in a fact free world is also typical of pro-nuclear people. Take care. I will stop now, since some people got banned for flame wars already, and rightfully so.

The running costs seem to be quite high too. The operator of Hinkley Point C in the UK for example receives a fixed "strike price" of 106.12£/MWh (currently) + inflation adjustment.

https://www.theguardian.com/news/2017/dec/21/hinkley-point-c...

https://www.lowcarboncontracts.uk/cfds/hinkley-point-c

And it seems likely that the French taxpayer (via EDF) will be taking a loss even at that price.

> we could have had a crazy nuclear capacity with very strict, modern standards.

Would that help though, if drought periods hit Western Europe, like last year, and water levels in the rivers are as low as they were?

In France last summer, some plants had indeed to be shut down because of the drought but:

* a minority of plants were involved, it was only an issue because it happened on top of other issues (planned maintenances delayed due to covid, corrosion issues)

* the problem isn't actually the drought, it was the heat. The plants could keep operating, but they would have rejected water too hot, in breach of environmental regulations.

Besides, new plants can be built close to the sea instead of rivers to account for that.

> Besides, new plants can be built close to the sea instead of rivers to account for that.

I think that's a no-go due to the salt in the water that will corrode pipes etc. Not an expert on that, though, obviously.

All the other points don't sound too good either: Corrosion issues, let's fuck up nature with hot water, not very good at all.

It's possible to build nuclear plants by the sea, and it's actually commonly built. There are some in France, China, Korea, etc.

I'm sure, they can figure that out with something similar to irrigation channels and/or solar panels along river/channels to prevent evaporation. Would they do this? That's another question

0.18% of nuclear power generation was lost due to water levels and temperatures.

Also nothing stops you from building nuclear plants in other locations

The inherent limitations are largely cost. Large scale underinvestment plus the financial failure of EDF very much demonstrate that.

The financial failure of EDF are mainly due to european regulations that forced price caps, making france sell at loss price when others needed energy last year (before the mess with plants breaking down everywhere)

Either way it's true that they let the situation degrade far too much while resting on aging infrastructure; it'll take a bit of time to get back on track whatever the direction is.

And those price caps are what kept French retail prices reasonable. Either somebody has to shoulder the high cost or maintenance must be neglected.

In France's case, we had both.

Meanwhile because cost cutting on maintenance backed up wholesale prices next winter are a blowout - far worse than Germany's:

https://www.archyde.com/french-electricity-prices-soar-for-w...

Reply on news.ycombinator.com