Hacker News Reader

Comment by nine_k | Hacker News Reader

So, both sides are right. Nuclear is too slow, but renewables are not fast enough / sufficient either.

Another thing is that politicians tend to plan that things will continue to go about as well as they are going at the moment, and the only major crisis ahead is the one being discussed at the moment. Nobody has the nerve to think that e.g. France will have trouble with several nuclear reactors at once, and a major war erupts in Europe at the same time. Ah, and a global pandemic right before that. Who in their right mind would vote for an energy budget based on such paranoid view of a possible future?

View on HN

A big problem with projects like these (nuclear and even renewable, but to a bit lesser of a degree) is that these projects outlast leadership. It then becomes easy to support something and it never get done. You make a big speech, but a few gears into motion, but more pressing stuff takes priority because a voter's attention is less than an election cycle. Truth is that if they work hard, get everything in motion you have 1 of 2 outcomes: 1) the next person gets all the praise because it is finished under their leadership or 2) the project goes into major delay as the next leadership doesn't prioritize it (often because it was the previous party's platform).

Honestly I think we could get a lot more done if people were able to disentangle when policies were achieved from when they were implemented. Governments are fantastic for long term investments, but politics works on shorter timeframes than many corporations. Though I think changing the cultural mentality to provide that selective pressure is an unrealistic pipedream. That said, it does also really show the constraints that we operate under and why existential issues like climate change are extremely difficult to solve.

How could this be achieved? I have one model in mind that might enable this type of long-term planning, but it would take quite a bit of work to make the US like China.

Policy Prediction Markets.

The reason China can do it is because they are an autocracy. Voters don't matter and their elections aren't as meaningful, meaning they can have projects longer than an election cycle.

If you want to get the West on a path this way, you have to get the people to have a change in how they evaluate their leaders. But first, you need to have the people trust in the leaders, which they don't right now for good reason. It is a tough nut to crack.

I'm not saying this in approval of authoritarianism btw. I think the results of that are worse, but it does have advantages (which is why it is attractive to some).

I wouldn't put nuclear and renewables as opposed sides. The only reasonable future will be built for a minimum of 80%-90% of primary energy generated from renewbles, and it may be that 10%-20% of nuclear is cheaper than to do it all by renewables.

I don't understand why you say renewables will be too slow. I've only done napkin math, but simple projection of continued growth of existing production capacity through, say, 2035, completely replaces all current electrical generation, as well as other energy uses. And when you account for the much much higher efficient of electrification, meaning that we get 2-3x as much work out of the same amount of energy compared to all the waste with fossil fuels, we can fulfill the developing world's needs for growth too.

But this is just my napkin math from taking production trends from top Google hits, it's not a sophisticated analysis. I would be interested to see one that is realistic, meaning that it doesn't assume that solar deployments plateau tomorrow.

Renewables and nuclear are of course not opposing each other; they are opposing fossil fuels.

According to the article, the UK now has less electrical generation than it would like / planned to have. Partly because the nuclear power plants failed to go online at the planned time. Partly because the amount of renewable generation that has been built in the meantime is not sufficient ether, because renewables need to be seriously overpeovisioned, due to their varying output.

Hence "both were too slow". Were they not, there'd be no energy deficit, I suppose.

They are opposing each other because of the combination of two factors:

* Peak energy demand happens when intermittent sources are at their lowest production [1].

* Nuclear power is as cheap to run at 100% capacity as it is to run at 50% or 25% capacity.

If you build enough nuclear power to fulfill the peak demand when renewables aren't producing, then you've already built enough nuclear to power the grid all day. And since nuclear is just as cheap to run at 100% capacity there's no point in provisioning renewables. Renewables work well when paired with a dispatchable energy source like gas or hydroelectricity. The former emits carbon and the latter is geographically dependent. In areas without hydroelectric potential, nuclear is the only existing option for full decarbonization.

1. https://en.wikipedia.org/wiki/Duck_curve

This is right, and again, I'm pretty pro-nuclear.

The upside of solar generation is its potential for really rapid deployment, and for piecemeal deployment. I can't afford to have a personal nuclear reactor, but I likely would be able to adorn my house with some solar panels, some batteries, and maybe a heat pump to create ice during the day's heat peak and use it after dusk. When I'd be able to afford a house, that is %)

Renewables don’t really oppose fossil fuels. Not to the same extent that nuclear does anyway. Nuclear can 100% replace all fossil fuels as a drop in. Renewables on their own can’t and need significant battery capacity as well to even out the load. Additionally, there are industrial use cases that need high temperature (like 300+ degrees C if my memory is correct). Nuclear power plants can generate this no problem because it’s a massive furnace already. Coal too. Solar and wind cannot.

I strongly believe who think renewables can replace all carbon energy sources are not realistic about what is needed and don’t appreciate the nuance and just think of it as “energy source” rather than as part of a complete energy production / distribution problem.

A small nit, it _is possible_ to generate high temperatures reasonably efficiently using solar concentrators. In fact, it is used as a means for power generation already[1].

Oh, and if you are in a tropical area, you may also use solar concentrators for free-ish hot water (with an insulating tank), and get 24 hours hot water. However, if it was cloudy the day before, it will not be a great experience, and you may need to use amn electric heater.

[1] https://www.scientificamerican.com/article/new-concentrating...

80-90% of energy generated from intermittent source would require immense amounts of storage to even out intermittency. If a silver bullet like hydrogen or compressed air solves storage, this may be feasible. But that's a big if. If fusion becomes viable could it be a viable primary energy source? Sure. But societies aren't going to bet the future of the energy industry on an if.

Every model I have seen shows 80%-90% renewables with almost no storage. Where are you getting your estimates?

Fusion is not a realistic power source for decades, and there's no reason to think it will be cheaper than fission, or that it could ever be cheaper than massive amounts of storage.

Demand for storage varies greatly depending on overproduction, geography, and willingness to continue using fossil fuels. Plans for an 80% renewable grid in the he US are estimated at 12 hours of storage and "just" 3 weeks of storage. https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

The problem is, even 12 hours of storage is a prohibitively large amount. The US uses 12,000 GWh of electricity per day. 12 hours of storage is 6,000 GWh. The entire world produces 400 GWh of batteries each year. Most of it is going to EVs and electronics, not grid storage. Even if we stopped building EVs entirely and directed the world's supply of battery production to just America's grid storage needs were talking about a decade of production. And worse yet these batteries are good for a couple thousand cycles, and will have to be replaced after a few years of use. And even with this 6 terawatts of stotage, in the end we'll still have a fifth of our grid still dependent on fossil fuels.

Again, this is why plans for a renewable grid assume some other form of energy storage will be orders of magnitude better than lithium batteries. That'd be great if hydrogen, molten sand, compressed air, or something else delivers. But that's a big if on which to gamble the energy grid, especially when alternative paths to decarbonization that we have more than half a century operating are within our reach.

80-90% of electricity from renewables is achievable in many regions with no storage other than increasing the intermittency of existing hydro, existing interconnects, and adding dispatchable loads to soak up the curtailed energy which is effectively free.

https://www.nature.com/articles/s41467-021-26355-z

An example is heating which is a huge portion of emissions and is also a dispatchable load.

That's great for regions with hydroelectricity, but the significant majority of the world is not lucky enough to have that geography. Also, the places that do have good geography for hydroelectric storage also usually generate their electricity from dams anyway and thus don't benefit from storage.

Well when you find a region witout hydro or interconnects or sufficient renewable resource that neither are necessary to exceed 75% penetration just by adding variable loads then let us know and we can put the chemical batteries there first.

Most of the USA, for one. Some states like mine are fortunate enough to have 70%+ hydroelectric generation. But in aggregate, the country creates 6% of it's electricity through hydropower. If our plan is to use hydro when wind and solar aren't producing we're very limited in terms of renewable production. The US uses about 12,000 GWh of electricity per day. The entire world produces about 500 GWh of batteries per year and this price is climbing as lithium shortages impact the supply chain. Batteries aren't going to cut it. That's why plans for a renewable grid either comes with either plans for extensive blackouts or some silver bullet solving storage.

Again, "variable loads" is a pretty euphemism for "shut off the power". If we're going to be shutting off power to industries the cost of reduced economic output needs to be counted.

So a country with a lot of existing interconnects and hydro where a complete renewable transition would cost less than nuclear and very barely more than the cheapest fossil fuels is your example?

https://www.sciencedirect.com/sdfe/reader/pii/S2542435121002...

And pretending adding off peak EV charging to carparks or building hybrid renewable+electrolyser stations that provide all their own energy needs and export during high demand is the same as shutting off existing critical infrastructure is laughably dishonest.

>80-90% of energy generated from intermittent source would require immense amounts of storage to even out intermittency. If a silver bullet like hydrogen or compressed air solves storage, this may be feasible.

The silver bullets are called demand shaping, pumped storage and batteries and theyre already here.

The fact that renewables are 5x cheaper than nuclear means that any storage that is, overall, less than 4x the cost of generation makes dispatchable solar/wind energy cheaper than nuclear energy when the sun isnt shining and the wind isnt blowing.

Almost every form of storage is cheaper than that. There is no need for an if. The if is here.

If we need to resort to demand shaping (read: turning off the power) then the cost of that needs to be accounted for. If you only run your smelters, chemical industries, etc. for part of the year then the cost of those commodities, and by extension every product using those commodities, gets more expensive. Increasing the cost of steel by 30% because we can't power our arc furnaces all year round is a massive cost that isn't captured when you just look at levelized cost of generation.

What form of storage are you referring to? The only popular forms of storage are batteries and hydroelectricity. The former isn't produced in nearly enough quantities, and the latte ris geographically restricted.

>If we need to resort to demand shaping (read: turning off the power) then the cost of that needs to be accounted for.

Of course, but the amount of low hanging fruit in this regard is A) ridiculously high B) zero, according to the nuclear and carbon industry models (they think demand cannot be shaped and only expensive batteries can deal with intermittency).

>If you only run your smelters, chemical industries, etc. for part of the year then the cost of those commodities, and by extension every product using those commodities, gets more expensive.

If you produce 2x as much as normal aluminum on a Tuesday when it's sunny and 1/5th as much on a Wednesday you've created more aluminum.

The solar driven smelter can produce more at a much lower cost because electricity is such a large % of the overall product.

There are plenty more applications like that where 5x cheaper electricity that "decides" when you use it is way better than 5x more expensive electricity that you can use any time.

Your straw man is still too strong for you to push over.

Only a small fraction of the arc furnace's power needs to be provided as electricity. A large fraction of it is a reducing agent like carbon or hydrogen which -- shockingly enough -- can be stored as carbon or hydrogen. Your imagined 30% increase due to idle capital is on top of a 50-80% reduction due to lower energy cost.

In industries where energy is the dominant cost, then minor shifts in production methods to do high labour tasks during low energy periods and vice versa will happen regardless of whether nuclear energy is available, because building solar to reduce costs and ignoring the nuclear reactor will still be a cheaper option than buying all of your energy from it. Alternatively the plant does buy the nuclear energy, but only when the sun is not shining because building their own array to use in DC mode costs $10/MWh and the VOM costs of nuclear are $15/MWh.

The only out is mandatory payment to a utility or funding it entirely with public money to everyone's detriment.

Hardly, the UK completely replaced coal and reduced it’s dependence on natural gas with renewables from 2010 to 2020. Meanwhile that wasn’t enough time to bring any new nuclear online in the current environment. https://upload.wikimedia.org/wikipedia/commons/2/2e/Electric...

Looks like renewables where simply the better, cheaper, and faster option. Which is shocking considering how little hydro and solar they have.

Plenty of wind though! Wind farms in the sea are a great idea; UK has a lot of sea at every side.

Well sure. When you effectively ban nuclear through regulatory hurdles, why is it a surprise that you don’t builds any and why is that a conclusion you can draw about the efficacy of nuclear itself? Somehow China is building out quite a bit of nuclear capacity while the west is sitting around dithering.

Reply on news.ycombinator.com