Hacker News Reader

Comment by manfredo | Hacker News Reader

Yes, nuclear power is less cost effective than fossil fuels. But nuclear, when built at scale in serialized production, can be cost effective. France generates it's electricity at half the cost of Germany, while simultaneously emitting less carbon. The total cost of the nuclear transition program was 400 billion 1993 francs which is ~150 billion USD according to my conversions: https://www.world-nuclear.org/information-library/country-pr...

One of the crucial things to account for is that nuclear's cost is overwhelming overhead cost. So prematurely closing plants vastly inflates the price of nuclear power. This is why nuclear energy costs are so much more expensive in the US, Japan, and Germany. These countries prematurely closed nuclear plants, thus increasing $/MHW.

View on HN

There are nuclear reactors closing in the US not because of regulation, but because they cannot compete with much cheaper wind and fracked gas sources. There is little difference in operational cost between running a nuclear plant at 50% and running it at 100%. So when they start losing key customers to cheaper power, pretty soon they're operating the plant in the red.

The real interesting part of this is that modern power is so much cheaper that it's less expensive to eat the capital losses and shut down nuclear plants early than it is to continue to operate them.

> There are nuclear reactors closing in the US not because of regulation, but because they cannot compete with much cheaper wind and fracked gas sources

The former is intermittent, and the latter emits carbon.

> There is little difference in operational cost between running a nuclear plant at 50% and running it at 100%. So when they start losing key customers to cheaper power, pretty soon they're operating the plant in the red.

Yes, and we're replacing nuclear plants with fossil fuel plants because of it. When California shut down its second to last nuclear plant, it's carbon emissions from electricity generation rose by 35%. And now we're poised to shut down the Diablo canyon plant with solar during the day and natural gas during the evening.

If we cared about reducing carbon, we'd keep the nuclear plant operating and shut down a fossil fuel plant with the surplus energy.

IMO, the most interesting power source mid term is tracking solar power. https://www.nrel.gov/docs/fy19osti/72399.pdf

At grid scale 1 access tracking is ~7% higher per watt which works out to about 2.14c/kWh in an ideal location. https://pv-magazine-usa.com/2019/06/28/los-angeles-seeks-rec...

However, these systems produces ~30% more power outside of the normal range for solar systems. This effectively squashes much of the duck curve and with the right mix and wide geographic distribution work well with nuclear power and modest energy storage systems.

Technically, this makes the duck curve worse, because the actual "problem" the duck curve presents is solar ramplong down in the evening faster than gas can ramp up. But trackers (or west facing panels) make that more like a square wave so it lasts longer but then falls faster than a static south facing panel.

But, the duck curve isn't a real problem anyway. Worst case scenario, some solar gets curtailed , creating an opportunity for storage solutions.

You’re thinking solar is all in the same location. Across a few locations in even a single time zone the ramp up and down is going to be fairly steady. Also, with nuclear in the mix you can be charging grid storage twice a day assuming a daytime and night time surplus. This possibly cuts the cost per kWh of some energy storage in half.

Ramp up/down is not much of an issue, as long as you can predict it. Yes tracking solar looks more like a square wave, but only in an extent a duck looks more like a horse, than a worm.

Market forces do not care if the power source emits carbon or not. We have to deal with that, and I don't like it either. Despite my surface appearance of being anti-nuclear because I point out the irrational, emotional nature of pro-nuclear arguments, I would greatly prefer to keep existing nuclear plants running until end-of-life, rather than replacing them with gas.

But market forces disagree with me.

So what's the alternative? Crushing carbon taxes? That would eliminate the "intermittent" wind and solar solutions that are clean, safe, and cheap, because at the moment they need gas to balance the load. And it would effectively be doubling the cost of energy production in order to subsidize nuclear. And it still doesn't give us an exit ramp, so we'd have to build more nuclear and continue to de facto prohibit solar indefinitely. That's absurd.

What appears most practical to me is taking advantage of gas in the short run to load-balance wind/solar, and eventually moving to energy storage for the balancing, encouraged via increasing carbon tax.

Yes, if we want to eliminate carbon dioxide emissions we should tax or even ban carbon. Why is building nuclear instead of wind and solar "absurd"? France generates over 70% of its electricity from nuclear and it costs them half as much as Germany.

Your wind and solar + gas solution is not a solution. We're still going to be emitting carbon of we are burning gas during the evenings. Energy storage at the necessary scale has not been demonstrated. Nuclear power is known technology. By comparison, solving the energy storage at this scale amounts to saying, "well... we'll figure it out eventually but until then we'll keep emitting carbon."

France is no longer building nuclear reactors. Nobody is, really, certainly not at "do something about greenhouse gasses" scale. The problem is economic. The vast bulk of the cost of a reactor is up-front construction, amortized over 30-50 year lifespans. That requires predictable energy prices, in a market where solar prices dropped over 80% in a decade. There are three different major energy sources all running half the cost of nuclear right now.

This rather impacts the risk of the up-front capital investment. Which affects the cost of the capital upwards. Which in turn makes building new reactors that much more expensive.

We should be building renewables because they're cheaper, safer (please, don't waste breath with nonsense arguments), and far more politically palatable. "Energy storage at the necessary scale has not been demonstrated". Of course it's been demonstrated. It's a problem with a thousand solutions. It's just a matter of getting costs down and production up (which in turn drives down cost). Batteries, thermal, compression, gravity storage... there are so many ways to store and release energy.

I see the "can't store at scale" argument everywhere, and it's utter nonsense, and it never comes with supporting math or facts.

Why would France need to keep building reactors? They're not building more of them because reactors last close to a century.

> That requires predictable energy prices, in a market where solar prices dropped over 80% in a decade. There are three different major energy sources all running half the cost of nuclear right now

And how many of those sources emit no carbon, and deliver power all around the clock? Geothermal and Hydroelectric can, but those are geographically limited.

Solar thermal is zero carbon and 100% uptime, but it's still expensive. Costs are dropping rapidly, though. And as stated earlier, solar + storage or wind + storage also meet your requirements, and are technically viable. It's just a question of cost. Cost of PV solar has dropped so much in the past decade that it could become the winner, even factoring in storage costs. Wind has far less storage costs, because the "What if the wind stops blowing?" theory that sounds so clever isn't really held up by decades of actual data.

Solar thermal is 2-2.5 times the cost of nuclear: https://www.energy.gov/sites/prod/files/2015/08/f25/LCOE.pdf

We could wait for solar thermal to get cheaper than nuclear (and assume that the cost of nuclear also remains static). Or we could just build nuclear power. The latter has the advantage of having consistent generation regardless of weather and time of year, and consuming a fraction of the amount of land.

Wind power has consistent output over long periods of time. But we still need to make the power grid resilient to fluctuations, which would require immense amounts of energy storage. To put it in perspective how infeasible energy storage is, take a look at California's latest energy projects. The current largest storage plant has 183MWh of capacity, and a planned one has a predicted 300Mwh of capacity[1]. By comparison, the Diablo Canyon plant generats 2,2000MWh of energy every hour [2]. These two energy storage plants can only store 5 minutes and 9 minutes worth of power generated by the Diablo Canyon plant respectively.

1. https://www.weforum.org/agenda/2018/11/california-will-repla...

2. https://en.wikipedia.org/wiki/Diablo_Canyon_Power_Plant

Is cost a factor, or is cost not a factor? Pick one. You asked if there was clean, consistent renewable, I gave you solar thermal, and acknowledged it's still cost-prohibitive. But I also expect its cost to plummet as experiments turn to production and lessons are learned.

First of all solar thermal is still intermittent and subject to weather and seasonal fluctuations. It has a built in thermal battery so it's consistent on a 24 hour basis, but it's still a variable source of energy.

Cost is a factor but not the only factor. Intermittency is a factor. Geographic limitation is a factor. Land consumption is a factor.

Cheap intermittent energy is an okay supplement, but cannot reliably deliver when it is needed. Hydroelectric and geothermal are great, non-intermittent clean energy but are impossible to build without the right geography. Fossil fuels are cheap, and deliver power anywhere but emit carbon. Nuclear power isn't as cheap as fossil fuels or intermittent sources. But it's the cheapest non-intermittent source that isn't geographically dependent.

If our goal is to fully replace fossil fuels, then nuclear is the best option (besides building geothermal and hydro where we can). Sure, solar thermal can deliver clean energy without the need for additional energy storage. But we could build twice as much capacity with nuclear and use a fraction of the land, and avoid having to build larger plants in the north and south, and avoid seasonal output fluctuations.

Ignoring clouds, the daily average insolation for the Earth is approximately 6 kWh/m2. [1] So you get about 0.25 KWe averaged daily per m2. Solar has serious land use issues. It is estimated that 1% of the UK would need to be given over to solar to deliver the current power needs. [2]

Noor II CSP delivers 0.66 TWh and is 6.8km2 (200 MW - peak?)

Ringhals Nuclear Power Plant delivers 23 TWh annually (3955 MWe)

A solar plant using the tech of Noor II and the power delivered of Ringhals would be 237km2 - twice the size of Paris.

[1] https://en.wikipedia.org/wiki/Solar_irradiance

[2] https://www.solarpowerportal.co.uk/news/if_solar_covered_one...

> It's a problem with a thousand solutions. It's just a matter of getting costs down and production up

Can you source this? I heard of molten salt, gravity storage etc... but I have not read any paper stating we can scale those solution to 50% or even 20% of our current grid usage

Have you read a paper saying we can't scale those solutions? Can you think of a rational reason they can't be scaled?

The problem isn't "can't". Storing energy is trivial. The problem is cost. How much storage is required, and how much will it cost to build it, and how much will that make the total cost of a new energy system?

All this "can't" stuff is, frankly, reactionary bullshit by some very emotional people who are rather in love with the idea of nuclear energy. That's why you never see hard numbers attached to it.

Storing energy is not at all trivial. You want hard numbers? Here are the hard numbers. California's largest energy storage facility, still under construction, is set to have 300MWh of capacity. By comparison, the Diablo Canyon nuclear plant generates 300MWh of electricity every ten minutes. The largest energy storage facility in the world, still under construction in Utah, is set to store between 1-2 GWh of energy. This is still less than what the plant generates every hour.

Electricity to gas conversion has terrible efficiency. 30-40% for the electrolysis and Sabatier process, and then ~50% efficient for the gas combustion engine. Net efficiency is in the 20-25% range. Hydroelectric storage is geographically dependent. Most of the US is in flat terrain.

California said they would do solar and wind plus storage. Then they realized storage was not possible, and they used fossil fuels instead. Similarly, Germany closed down their nuclear plants, saying they'll build intermittent renewables plus storage. And then they ended up building fossil fuel plants when they realized storage could not fulfill the base load they lost from closing nuclear plants.

Why not store hydrogen? Estimated round-trip efficiency is approximately 40%. Storage in underground caverns / mines.

If most of the energy consumption is supplied directly, the storage part need not be overly efficient as long as it can cover the slack. 40% seems good enough. Even 20% is workable, if need be, just need to over-provision enough PV / wind and the over-provisioning itself significantly reduces needed storage.

If combined with capacious long-distance electrical grid (e.g. HV DC), load scheduling, high-uptime offshore wind, some PV in deserts, maybe thermo-electric solar, some batteries for rapid load following, etc, we really can supply enough power even with 0 coal, gas and nuclear power plants, and it wouldn't even bankrupt us. All we need is will. The technology is already sufficient and with improvements it won't be even that hard.

Trying to rely only on intermittent power sources has huge storage requirements due to weather along with daily/seasonal variation. If grid energy storage was a simple problem it would have been done decades ago.

For example, one estimate is that for Germany to rely on solar and wind would require about 6,000 pumped storage plants which is 183 times their current capacity:

>...Based on German hourly feed-in and consumption data for electric power, this paper studies the storage and buffering needs resulting from the volatility of wind and solar energy. It shows that joint buffers for wind and solar energy require less storage capacity than would be necessary to buffer wind or solar energy alone. The storage requirement of over 6,000 pumped storage plants, which is 183 times Germany’s current capacity, would nevertheless be huge.

https://www.econstor.eu/bitstream/10419/144985/1/cesifo1_wp5...

Land use becomes a real problem with some solutions, so yes it can be a “can’t”

You don't need to use solutions that don't work or scale. You only need to have those that do, and there are.

They just cost more than market rates.

> France is no longer building nuclear reactors. Nobody is, really

The Wikipedia page https://en.wikipedia.org/wiki/List_of_nuclear_reactors believes the following countries are building nuclear reactors: Argentina, Bangladesh, Belarus, Brazil, China, France, India, Japan, North Korea, Pakistan, Russia, South Korea, Turkey, Ukraine, the United Arab Emirates, the United Kingdom, and the United States.

Let’s not forget the military. Regardless of how much the public hates it, the military loves nuclear and would switch all ships over if they could.

The sole running nuclear plant-building project in France is Flamanville... and a disaster.

https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...

It can't be true that keeping a nuclear power plant running is more costly than replacing it with a massive capital expenditure. Can it?

The only possible way is if the alternative is massively funded by debt, and the "cost" is based on historically low interest rates and a repayment timeframe in the decades.

Even then, given the ongoing cost of nuclear is relatively small, how is it possible that the capital expenditure of solar + natural gas is cheaper than the running cots of nuclear? Can someone show me that maths?

When something is true, saying "it can't be true" doesn't make it not true. This is happening to existing nuclear plants in the US. Not plans for plants, not plants in construction.

This caught my eye when a nuclear plant in Iowa (near friends, including one who retired from working there) got slated for shutdown, over a decade before end-of-life. They'd lost a key industrial customer that consumed 30% of the plant's output to much cheaper wind/gas. At that point, operating costs went into the red. The plant was no longer generating the revenue to pay off its own debt. It was a purely economic decision.

And yes, the wind and gas that ate its lunch are also capital expenditures, amortized over time. But they're still much cheaper.

The market is easily distorted by government.

Want nuclear to win? Interest free loans. Want wind/solar to win? Subsidies.

Put a thumb on the scale and make it say whatever you want.

>So what's the alternative?

Nationalization of all energy generation?

That's interesting. It may then matter that French plants are state-owned. It probably makes them cheaper on the long term

The price of power is probably also regulated. You can't really compare France to the US, the French state steps in and regulates pretty much anything.

> the French state steps in and regulates pretty much anything.

That’s a nice way of putting it.

Germany shows that even if your cost of building a reactor was zero, you could still have the reactor shut down due to other issues.

"Following the March 2011 Fukushima nuclear disaster, Germany has permanently shut down eight of its 17 reactors and pledged to close the rest by the end of 2022." - Wikipedia

Edit: This might just be because Germany is as close to the Ukraine as California is to Texas. Also removed dumb sentence.

That's a problem caused by politicians. The same could be said of any energy source.

Reply on news.ycombinator.com