Comment by toomuchtodo | Hacker News Reader

Comment by toomuchtodo | Hacker News Reader

TLDR battery storage can replace grid inertia (“ancillary services”) historically provided by thermal generators (coal and various forms of fossil gas). This is a distinct service/capability from energy arbitrage and has its own revenue stream.

South Australia (where the Hornsdale Power Reserve was installed) runs a minimum of 80MW of fossil gas generation to provide these services (in concert with synchronous condensers), and this year the grid operator AEMO plans on lowering that minimum further, approaching a point where they’ll allow running on renewables and batteries alone around 2025-2026 (once a new interconnector with adjacent New South Wales is complete).

https://reneweconomy.com.au/remarkable-south-australia-may-s...

(also of note are new capabilities in battery and solar power controls that allow them to form grids independently vs simply following existing grid reference)

https://reneweconomy.com.au/aemo-to-fast-track-grid-forming-...

https://www.sciencedirect.com/science/article/pii/S235246772...

South Australia is building brand new gas peaking plants. Seems like fossil fuels continue to be a core pillar of their energy strategy.

Irrational decisions still occur occasionally.

https://www.energy-storage.news/no-longer-economically-ratio...

https://www.energy-storage.news/battery-storage-30-cheaper-t...

https://www.cleanenergycouncil.org.au/resources/resources-hu...

¯\_(ツ)_/¯

The cost however remains remarkably high. A combination of renewables and giant piles of lithium is more expensive than nuclear, by quite a bit. [1]

Anyways, the future is a mix of all sorts of things and this is a strict improvement so it's hard to be anything other than supportive.

[1] https://spectrum.ieee.org/what-energy-storage-would-have-to-...

You’re assuming long duration lithium storage vs excess renewables with curtailment. Australia is going 100% renewables at a reasonable cost with solar, wind, hydro, transmission, and some batteries. No nuclear is required to decarbonize.

https://aemo.com.au/newsroom/media-release/engineering-frame...

https://aemo.com.au/initiatives/major-programs/engineering-f...

I'm not actually making that assumption, I linked an article with the numbers as [1] in an edit to my initial reply.

Nuclear is renewable :) [2]

But yes, adjusting demand is far more efficient than adjusting supply. However it seems there's limited appetite.

> No nuclear is required to decarbonize.

No but it's less expensive, and better from a space utilization and environmental impact perspective. It's the best energy source we have. Safest, cleanest and lowest carbon. And most reliable. [1] Not perfect, but better than anything else we've got.

[1] https://sustainablereview.com/nuclear-energy-is-better-than-...

[2] https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

I’m not against nuclear but it is a long way from being cheap and easy. Look at the recent andventures with Hinkley C. And that’s with an existing industrial base for nuclear power. 25 bn pounds is 45 billion AUD which is what previous governments paid to wire the whole country up with a patchwork of broadband (inflation unadjusted). And don’t say they small modular will step into the breach because it is not a proven technology and until the first 10 are deployed may never be.

Australia has no nuclear knowledge base and would require a nationwide debate about getting into the nuclear power game so we would have at a minimum a 20 year lead time to build something.

We could be totally decarbonised by that point and we wouldn’t even have a reactor online.

As to your other points, have you seen how much space we have down here? Cramming a plant into a specific area is not a net benefit. We’ve got a geographically diverse population with a lot of empty space, it would be better to have more plants in more diversified locations.

We will end up passing nuclear by, the economics and time lag just don’t stack up for us

> We will end up passing nuclear by, the economics and time lag just don’t stack up for us

Again the economics are better than what is being done. However.

> I’m not against nuclear but it is a long way from being cheap and easy.

Ah yes just like JFK said about going to space. "We choose to decarbonize the grid with solar not because it is the best option, not because the cheapest option or even because it is hard - but because it is easy and politically expedient."

You haven’t managed to make any strong assertions, just a smattering of pseudo-intellectualism. Thanks and good luck

You can politically thank The Greens for no nuclear power plants in Australia (apart from Lucas Heights, medical).

Ironically, their 100% outright, non negotiable no nuclear policy has pushed heavy carbon emitting coal power plants for the last 30 years until the recent policy shift of winding them down in lieu of solar.

Is this true?

Greens sometimes have influence in balance of power, but ultimatly they are a smaller party with limited influence or policy making decisions.

Logically that view seems to look at one side of the position only, if they had the power to stop nuclear you'd have to think they also have the power to significantly reduce/replace coal, which hasn't been the case.

Generally doesn't seem fair statement to me but open to reading information that shows otherwise.

No. It's not true. The Greens as a party didn't exist when the Australian Labor Party resolved to ban uranium mining and join the Nuclear Non-Proliferation Pact in the mid 70's. The various Australian anti-nuclear movements (plural) largely arose and thrived during the late 70s and 80s. Public opposition to nuclear energy started to wane in the 2000s, accelerated by concerns relating to climate change.

I think generally it wasn't the 'Green Party' but rather the wider environmental movement that then had a wide influence on the left in particular. The right meanwhile generally is just happy with fossil fuels and does whatever is cheaper.

That was always the problem with nuclear, it didn't have a political ally.

In one of the few cases where it did have political allies, it was in France when the realized how much control over oil is in the middle east. So French elites turned on fossil and invested in nuclear. And within 20 years they solved the problem. Its pretty instructive.

It is not true; the main impediment to nuclear in Australia has been some combination of the Australian Labor party and the people who vote for them. They've been staunchly anti-nuclear for a long time.

"Labor will ... [p]rohibit the establishment of nuclear power plants and all other stages of the nuclear fuel cycle in Australia;" [0]

[0] https://www.alp.org.au/2021nationalplatform

> recent andventures with Hinkley C

Building individual prestige one-off projects will always be a bad idea. This is the kind of things government do if they just need to hand out pieces of the pie to everybody.

A better example would be what South Korea has done in the UAE. Building 4 reactors 1 after the other in a country with no experience and no labor force. They did it pretty quickly and the unit price and unit speed goes up with every one.

In the UAE, construction started 2012, commissioned 2020. But importantly, after that they finish every year.

In reality, nuclear is not actually that slow if a government is really willing to do it and build a plant that has been built before.

RE "...Australia is going 100% renewables at a reasonable cost with solar, wind, hydro, transmission, and some batteries....." The level of network supply reliability ( as in outages ) and the cost will be higher than the predicated "reasonable " in my opinion. Id like to know how several days of "overcast weather over extended areas " will be dealt with. Batteries are ok for short term storage .

Hydro. Snowy 2 and Tasmania. Worst case, we’re still burning some small amount of fossil gas, or perhaps ammonia or hydrogen generated from renewables that would’ve otherwise been curtailed.

Consider that enough energy falls on the Earth every ~60 minutes to power humanity for a year. We just need to continue to drive down renewables and storage costs. The world is already fusion capable, at a distance.

It doesn't matter that nuclear is cheaper if you can't build a nuclear plant.

The difference is that a lithium battery facility can be installed and functional in 12-24 months. A nuclear plant takes 12-24 years.

Much like World War I strategies, a solution that nobody has the political will to implement isn't a solution.

> The difference is that a lithium battery facility can be installed and functional in 12-24 months.

These things aren't comparable without putting it together as a system.

> A nuclear plant takes 12-24 years.

You're thinking about wild outliers there. It's around 6.5 years median [1].

[1] https://hannahritchie.substack.com/p/nuclear-construction-ti...

Funny article. It talks about time to build nuclear reactors but never mentions Russia.

Why would it?

Rosatom builds faster than South Korea, which takes 9 years to build a reactor [0]. For example, the reactor in Belarus was built in 8 years[1]. The first reactor at the Akkuyu site in Turkey is going to be built in 6 years [2].

[0] https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...

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

[2] https://www.world-nuclear-news.org/Articles/Major-components...

> Funny article. It talks about time to build nuclear reactors but never mentions Russia. > ... > For example, the reactor in Belarus was built in 8 years[1].

Sorry to break it to you but Belarus is still not russia, yet.

If you look again at the article you'll see lots of countries that have nuclear reactors are not specifically mentioned, however in the one section where countries are specifically mentioned a group for all former USSR countries is called out that includes russia and other countries. This will be due to how reactor data was recorded prior to the dissolution of the USSR.

Slava Ukraini

Just like UAE isn't South Korea but the reactors there are designed and built by a South Korean company. What's your point?

Sorry to break it to you, but the USSR dissolved over 30 years ago.

You are confused. The article talks about reactor construction time in countries.

E.g. Hinkley Point C is being built in the UK, that means it gets recorded in data about construction in the UK, even though it is designed by a French company, and owned by a French and Chinese company.

The same applies for reactors designed by Rosatom and built around the world. A VVER-1200 built in China may be a russian design, but the data is recorded as construction in China for obvious reasons.

> Sorry to break it to you, but the USSR dissolved over 30 years ago.

Yes, you're the one that was confused why Belarusian projects did not count as russian, and you ignored the Former USSR information in the article. If you would like to know why she aggregated the data that way and didn't break it out into before and after USSR dissolution, you should go and post a comment asking on her Substack.

Firstly, it's meaningless -- the construction time depends on who is building, not where.

Secondly, it doesn't matter -- reactors in Russia are built by Rosatom just a little slower and still on par with South Korea.

I have zero interest in asking her about anything.

> the reactor in Belarus was built in 8 years

That’s slower than average [1]. Striking, considering how much where you build influences timelines as much as who’s building. (I doubt Minsk subjects Rosatom to community input, labor requirements or environmental reviews.)

[1] https://hannahritchie.substack.com/p/nuclear-construction-ti...

Yes, lots of the reactors are built by China which takes 5-6 years to do the job. All other countries take more time.

"community input, ... environmental reviews"

It all happens before the start of construction, doesn't it?

[0] https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...

> lots of the reactors are built by China which takes 5-6 years to do the job. All other countries take more time

This is incorrect: Russia builds new nuclear power plants slower than the average plant built by Japan and Korea in the last decades. The slowpokes are Europe and the U.S.

> happens before the start of construction, doesn't it

No. It’s an ongoing process in developed countries. (Including, increasingly, China.)

Why don't you support your statements with sources?

"the average plant built by Japan"

Japan stopped building reactors after Fukushima disaster. If you cut corners, build a reactor in five years and later it blows up, that's not what you want, isn't it?

"It’s an ongoing process"

What kind of effect does it have on the construction? If none, it's just ongoing show.

That's from start of "building" not "planning". In the US, getting the permits and handling the planning far exceeds the actual construction.

In addition, there have only been 7 reactors constructed in North America since 1990 and none from 1995 to 2005.

I'm actually a fan of nuclear reactors. However, in the US, we simply don't have the political will to build them.

Wind and solar also have many years (and decades) of planning involved. And many also fail to get past the planning phase.

> However, in the US, we simply don't have the political will to build them.

Fair enough, though this article is about Australia. Here in the US we do have the TVA that is looking to go ahead with new nuclear. But we will likely lag other countries such as Canada (OPG and SaskPower are proceeding with new nuclear builds).

Chicken and egg thing. If you listen to the DOE Loan Program head, he says it takes so long and costs so much because each plant is a bespoke hand-crafted special construction project. They can be made much cheaper and much faster if you standardize the design and copy-paste them to where it's a good fit.

This isn't an issue inherent to nuclear power at all, but instead in how we approach it.

This isn't speculation either, China is building 150 new plants by 2035. [1]

> Much like World War I strategies, a solution that nobody has the political will to implement isn't a solution.

That's just a tautology.

[1] https://thediplomat.com/2022/09/powering-chinas-nuclear-ambi...

Again, you’re saying that it’s cheap, but then responding to criticism they’re bespoke by presenting an untested technology that has further lead time delays and may even be vapourwear. You still can’t spin up an industrial base in less than 5-10 years, and there sure as hell isn’t the political will to ask the Chinese to build it for us

I didn't say it was cheap. I said current nuclear reactor technology is cheaper than a mix of renewables and lithium batteries. They're both quite expensive, and headline renewable 'low costs' exclude storage. But that's ok, I think the fact it's expensive isn't a reason to avoid either of them.

I then said that we can also make it significantly cheaper than it is to construct nuclear plants, and offered a path presented by someone very knowledgeable about the space. Jigar Shah from DOE. [1]

> ... but then responding to criticism they’re bespoke by presenting an untested technology that has further lead time delays and may even be vapourwear.

What untested technology lol, nuclear has been around since 1955. I'm talking about layouts.

> You still can’t spin up an industrial base in less than 5-10 years, and there sure as hell isn’t the political will to ask the Chinese to build it for us

That's ok just get the Canadians to do it. Check out the CANDU family. [2] And you know all those solar panels are being made in China right? 85% of all panels - 45% of polysilicon comes from Xinjiang. So that's exactly what is being done. [3]

[1] https://www.youtube.com/watch?v=rXRmlcsqHtk

[2] https://en.wikipedia.org/wiki/CANDU_reactor

[3] https://www.utilitydive.com/news/china-made-solar-pv-created...

They are actually quite cheap. I did some basic calculations on what it would have cost Germany if they in 2000 had started a the 'Gründewende' with nuclear. And it would have been massively cheaper then what they have been doing.

Modern CANDU reactors can be really great. There is some new Thorium/Uranium based fuel that will revolutionize CANDU operations. One of the biggest issues with CANDU is that it high labor cost to change the fuel bundles. Some new fuels will allow the reduction of those operations by like 4x, saving on labor cost.

It takes 12-24 years if its a lazy one off with little political support.

We also shouldn't measure time for a single plant, but for many.

Fact is, France built like 50 in 20 years. And if you look at the UAE, its much more instructive. A country with no experience.

Handed out a bid in 2009, first plant active 2020 and then plants going online 2021, 2022, 2023. And if you just keep going like that eventually you can finish 2 or more per year. France was finishing like 5-6 plants a year.

If you look at Germany in 2000 for example. Had they done what France did in the 70/80s, Germany would have carbon free right now and they would have spent far less money then the 'Green Revolution' that they are engaged with now (and all the massive cost they will be paying the next 20 years). If you just do the math, based even on the prices that South Korea charged ULA, nuclear would be a no brain. And nobody can tell me Germany wouldn't have the engineering skill to do this.

That article is 4 years old the prices it assumes are more than what they actually are already. With large scale Sodium Ion battery production expected to start this year in China it will be interesting to see what profilates faster grid scale battery storage or home battery packs.

I mean, that points to a Joule article that says: “If other sources meet demand 5% of the time, electricity costs fall and the energy capacity cost target rises to $150/kWh.” (Ie from $20/kWh). In other words, the grid can easily go to 95% variable renewable using storage costs the same as what they are right now. I should point out even nuclear power tends to use fossil power for backup, and synthesizing hydrogen for industrial uses could also be used to provide that last 5% affordably.

That’s why when you add hydrogen season storage (to lithium daily storage), renewables are very competitive with nuclear even at 100% bAsElOaD. See: https://model.energy for an illustration of this.

But seriously, getting to even just 95% clean energy would be a massive win, and the most important thing is getting to 95% clean within 20 years, not 100% in 40. (Or actually, getting to 80% in 10 is even more important, and to do it in a way that doesn’t massively increase energy costs, because we also have to eliminate natural gas for heating and petroleum for transportation.)

(And I’m a big proponent of nuclear, but you can just build renewable WAAAAY faster in 2023. And that’s what matters… well that and not prematurely decommissioning existing nuclear!)

Reply on news.ycombinator.com