Hacker News Reader

Comment by coder543 | Hacker News Reader

>> Grids have also repeatedly been shown to handle more renewables than every previous prediction would make, and we haven't hit the limit.

> Yes dispatchable generation may fail to materialize at times, but renewables “fail” to provide consistent power every single day, when the sun stops shining at night or wind stops blowing.

As I said. You're just repeating the old arguments. People thought that small percentages of renewables would destabilize the grid, then that didn't happen, so then they said a slightly larger percentages would do it, and it didn't. This tired theme has been repeated ad nauseam for the last decade or two.

I agree that you need some amount of Base Load, but renewables haven't been the problem yet, and energy storage is the solution, long term, along with Demand Response. Small amounts of grid energy storage have been shown[0] to have disproportionately high effects on improving grid stability. We might need less than you predict.

As it is, since we are still successfully adding more and more renewables to the grid, and renewables aren't being the source of blackouts, SMRs have to compete with renewables on cost, and they simply don't. SMRs also don't compete with Combined Cycle plants in terms of cost either, so which one are utilities going to choose?

Peak demand for the grid is in the late afternoon / early evening, so the amount of battery storage needed to "shift" solar production by a few hours is not as much as you would think.

Wind power produces more power at night than during the day, and it produces more in winter than summer, which is quite convenient given how solar produces more in the summer and during the day.[1] They make quite a complementary pair of power sources.

There are seasonal concerns, which is where some combined cycle plants come into play, even if they don't operate for most of the year, but that's also not SMRs. Combined Cycle is way cheaper than these nuclear SMRs. If you over-build on wind and solar, you can go a long way even in times of year with "less" wind and less sunshine, and with the low cost of wind and solar... lots of people are looking to overbuild as a partial solution that doesn't require batteries.

[0]: https://en.wikipedia.org/wiki/Hornsdale_Power_Reserve#Benefi...

[1]: https://www.osti.gov/servlets/purl/1368867

View on HN

The overwhelming majority of electricity demand is base load. Usually on the order of 70-80% [1]. We don't need "some" base load, almost all our demand is base load.

Electricity storage is nowhere near the scale required to make a dent in the electricity grid. To put this in perspective, the US alone uses about 500 GWh of electricity every hour. Worldwide this figure is about 2,500 GWh per hour. The storage facility you linked to was the biggest facility in the world when it was first constructed, and it stored only 129 MWh of electricity.

At our current rate of battery production it'd take us a century of dedicating 100% of our battery output to grid storage to reach 1 day's worth of storage. Battery production is expected to increase, but it's unclear whether raw material inputs can keep up with manufacturing demands [2].

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

2. https://tradingeconomics.com/commodity/lithium

Hornsdale Power Reserve was nowhere even close to being the biggest energy storage facility, even when it was constructed. At that time, the biggest facility was the Bath County Pumped Storage Station at 24,000 MWh, which has since been surpassed again.

Or just take any hydro electric dam. The Swedish hydro power currently stores 17.7 TWh.

It is not a battery in how it can fluctuate due to natural flow concerns, but some fluctuations are fine.

Dams are geographically limited. You can't just build more of them. A fully decarbonized grid's ability to build renewables is largely determined by the availability of hydroelectricity for dispatchable power. Sweden gets ~40% of its electricity from hydro, another 40 from nuclear, and 20 from intermittent sources.

You can increase their peak power production to deal with the variability.

For example this hydro plant, about the size of a nuclear reactor, maxes out at ~1 GW while the production is on average 250 MW.

In other words, it facilitates renewables.

https://en.wikipedia.org/wiki/Harspr%C3%A5nget_hydroelectric...

A fully decarbonized grid's ability to build renewables is, in fact, not at all determined by the availability of (watershed) hydroelectricity.

In particular, hydro storage can be built in hundreds of times as many places as watershed hydro generation. And, there are numerous other practical storage methods.

Ready combined-cycle gas generation capacity is more important, most places. Those will be incrementally converted to consume imported synthetic ammonia.

> Ready combined-cycle gas generation capacity is more important, most places. Those will be incrementally converted to consume imported synthetic ammonia.

Synthetic ammonia that comes from where? Ammonia is currently produced by steam reformation, which emits CO2. You're talking about electrolyzing water to produce hydrogen, and using renewable energy to carry out the Haber process. Nobody is using this strategy of energy storage, so it's priceless in a very literal sense: there is no way to estimate how much such a system would cost. I'd take an expensive solution over a priceless solution any day.

Synthetic ammonia will come from electric synthetic ammonia production plants. A few production-scale plants are already under construction, but of course hundreds more will be needed. Ammonia will not be, primarily, a storage medium, but a transportation medium and fuel, although it stores well under light pressure. Any tropical country can put up a solar farm and begin exporting ammonia to places less blessed with reliable sunshine -- for fertilizer, at first.

Insisting technologies that people are already spending billions of dollars on building out were not first shown to be viable is a peculiar position to take.

Efficiency of producing Ammonia is 15%? How is this ever going to be practical?

https://caseyhandmer.wordpress.com/2022/07/22/we're-going-to...

When marginal cost of energy is zero, other things than energy efficiency become more important. But where is this 15% figure coming from?

You are assuming a very ideal scenario, which is dangerous.

Even so, do you have another figure? This is a complex molecule, it is naturally inefficient to produce compared to hydrogen, which is below 50%.

To quote the Wikipedia article you cited:

> This demand can be met by unvarying power plants,[2] dispatchable generation,[3] or by a collection of smaller intermittent energy sources,[4] depending on which approach has the best mix of cost, availability and reliability in any particular market.

Renewables are baseload. They mainly compete with other traditional "base load" power plants, i.e. coal and nuclear, all of which rely on dispatch able generation for evening out the peaks and trowths. Your calculation also doesn't make sense, you never need storage to cover the whole electricity generation in all of the US. It's like saying in an all nuclear scenario we need to build plants to cover twice the peak demand because all plants could be under maintenance at the same time. You don't build with 100% redundancy.

> 500 GWh of electricity every hour

GWh/h, or simply 500 GW ;)

Not quite the same thing. Watts are power. Watt-hours per hour (or jules per hour) are average energy consumption during one hour.

Yes, but in practice Watts are almost always an average over some sample period. Generation capacities for power plants are given in Watts, not Wh/h. Wh/h gives you a hint that the sample period was an hour, but that isn't necessarily the case. I don't think Wh/h is the correct way to describe the sample period.

This is highly geographically dependent, but water displacement “batteries” that pump water to an elevated basin when power is on and then let it run through turbines back down to a lowered basin seem like a really simple, effective solution that can work at scale.

It does not scale.

https://dothemath.ucsd.edu/2011/11/pump-up-the-storage/

Good article. I stand corrected.

It is not a good article, at all. See above. It is cleverly written to fool the unwary.

I’ll back out then/leave the debate to those who have the time to sort through all of this. The gist about the enormous amount of space needed in that article seemed accurate/more than I thought.

The article is foolish from top to bottom.

It says exactly one right thing: you can put a reservoir on a hilltop, and the "head" is the height of the hill, not the depth of the reservoir.

You do not need a mountain. You do not need a high valley. It is cheap to build an earthen dike around the top of a hill, leveling off the hilltop peak for material.

The hill does not need to be steep; a shallow slope just means a longer penstock, which costs more.

A 300-meter hill is high enough for practical use. In most cases only a few hours' storage is plenty; you only need enough to make firing up a gas generator an occasional event.

Luckily the author took every opportunity to err on the side of underestimating the amount of water storage needed, to head off such pedantry that wouldn't significantly change the math.

Unfortunately, he got it wrong anyway.

Let me know when you have any evidence of what you're saying.

I don't know what "at scale" means in this context, but they certainly do work [0].

[0] https://en.wikipedia.org/wiki/List_of_pumped-storage_hydroel...

By “at scale” I mean building enough pumped storage facilities to meet current energy demands when the wind isn’t blowing and the sun isn’t shining. It seems much more likely that we could build sufficient pumped storage facilities than the kind of battery farms the poster I was replying to rightly points out are not likely to meet required needs for the foreseeable future, if ever.

That depends on the geography.

Bangladesh, The Netherlands, Kansas, maybe Texas, etc? No, not useful.

Norway, Wyoming, Idaho, California, Switzerland, Nevada, Arizona, etc.? Oh yeah. Very doable.

My personal pet idea: a big ‘tower’ on the seafloor. Yes, costs are immense, but scaling would really work well. (Making the ‘well’ or tower twice as big would not make the cost double).

In ascii art:

   ———|  |——

      |  |                       

      |++|

With ‘| |’ the walls of the well or tower, —- the sealevel, and ++ the water inside the well. Energy would be gained through letting seawater go ‘into’ the well/water tower. (Edit: attempt to make ascii art work)

You don't need a tower. A boat with a winch and a weight does the job cheaply. For extra credit, more weights.

Limited by the displaced water weight. Prone to storm damage. But the general idea could work and you could even do it entirely under water, saving you from having to dig out a mineshaft, just sink a bunch of caissons to create an underwater tower. As long as they stack and can be re-inforced so they don't end up shearing under the pressure of flowing water. Maybe even open up the sides to a lattice to reduce that resistance.

Existing supertankers can displace in the hundreds of thousands of tons. New, they cost under $100M. Scrap, much less.

They are not bothered much by weather, in normal operation. With that much weight hanging well below the surface, they would be very stable. (Racked together side by side, moreso.) Think of it as a very long oil platform. Those endure any weather with no difficulty.

No need for anything attached to the bottom except anchor chains.

When planning ocean operation, make sure almost everything is at or well above the surface, and everything complicated (motor/generator, winch, power conversion) is well protected from exposure. Ocean deployments that expose expensive stuff to wave action or put it underwater fail, reliably. (Expect tidal generation that puts generators underwater to fail spectacularly.)

Offshore wind has the nacelle well encapsulated and everything that moves far above the waves. Nothing is underwater except the pilings, the end of the support post, and a wire.

Yep, I tried to acknowledge that in the original comment when I said it’s geographically dependent.

It still seems like the only solution that currently exists and actually works on the scale required. I’m not sure what realistic solutions are for regions where hydro pumped storage is not a viable option.

A handful of hours storage, 60% curtailment (with most of the surplus used to create hydrogen and such for chemical feedstock) and a small amount of HVDC leaves a remainder small enough that meeting it with biogas and existing hydro is possible, meeting it with hudrogen is not costly, or meeting it with LNG is not a huge problem.

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

The world produces about 3TW of electricity, 2022's battery production was 760GWh and it's growing 30-50% yoy.

Ideally degrowth happens, but while we negotiate that, the |bhattery industry is at the scale required and is rapidly shedding critical mineral requirements (sodium ion is at GWh scale now, and much larger supply chains with all abundant materials come online in june).

> I’m not sure what realistic solutions are for regions where hydro pumped storage is not a viable option.

Pumped thermal.

https://aip.scitation.org/doi/10.1063/1.4994054

It is not either/or.

It is all/and. There will be many kinds of storage.

True, yes. It just seems to me like conventional battery farms are not going to make much sense in a lot of places, and hydro pumped storage will. Another user pointed out that the size inefficiency is severe enough with hydro pumped storage that it is less grid scale suitable than I thought, so that difference between hydro pumped storage and conventional battery farms is less than I thought. Regardless, energy storage is a huge problem that’s going to require lots of deference to practicality and as many simple solutions as possible. Frankly I’m not sure it is possible, which is why I’m a big advocate for nuclear.

Building out enough storage will be a huge task.

We can't use much of it yet because we haven't the renewable generating capacity to charge it from. (Charging it from fossil fuels would be beyond stupid.) Several varieties are getting cheaper very fast, and new, cheaper ones are being invented, so when the time comes it will cost a lot less. Likewise, carbon capture: building that out now would be stupid if it diverted money that could be spent building renewables that displace carbon emission.

It is also why building nukes is stupid: they displace way less carbon emissions, per dollar, than renewables, and first spend a decade displacing none at all. For the price of the coal burned waiting for the nuke to come on line, you could build that much solar, never mind what you are wasting on building the nuke, and it would start displacing carbon emissions almost immediately.

Batteries and pumped hydro are very far from the only practical storage media. But the assertion you read that pumped hydro does not scale is deliberately deceptive.

I would read Tom Murphy's posts from a different light. His message has consistently been 'unbounded exponentials aren't real, why are you pretending they are?'.

In that light 'pumped hydro can't scale' means 'stop trying to use this for Petawatt hours of energy storage you idiot, you'll destroy everything'.

Some of his more recent posts have been worryingly easy to coopt by the 'maek moar fossil fuels and nuclear' crowd, which his previous posts indicate he should be even more strongly opposed to.

The general message of degrowth and steady state economy is positive even if 'pumped hydro can't scale indefinitely' sounds a bit like 'pumped hydro can't decarbonize the current scale of the economy', and he gets the scale required to maintain status quo consumption a bit wrong.

I do wish he'd make this distinction clearer though.

I would be satisfied to see him stick to facts.

OK if by "highly geographically dependent" you mean "not highly geographically dependent".

You need a hill, but there are a very, very large number of hills. Most usually you need to build an earthen dike around the top of the hill, although often a natural feature allows it to be shorter.

The reservoir does not need to be deep because the "head" is from it to the bottom of the hill, or even to the water table well below that.

The reservoir or reservoirs are good places to float solar farms.

I forget what country was trying to build a bunch of these but an article popped up on here about this a while back and they cited lack of suitable locations as one of the main barriers for building more.

The most ideal locations are natural that require minimal land reshaping. Some places in the world have a lot of suitable locations like that. Others don’t.

But despite that it still seems much more feasible to build lots of hydro pump stations like that than it does to build other forms of battery farms.

There was a paper here recently which showed there were a large order more locations s for this than previously thought.

Places with not many hills will of course prefer other storage media.

Places with lots of hills will also prefer other storage media if they turn out to be cheaper. It is far from clear yet how costs will settle out. It is anyway not time yet to build more than just enough storage to shift the few hours from peak generation to peak use.

Kansas and the Netherlands would like a word.

Both have plenty of transmission lines to places with hills.

Exactly. And the hills are not actually that far away.

So it's supposed to be impossible to get the 3 hours or 9TWh (12 years of 2022 production) of battery needed to shift the grid to >99% renewables (including biogas from waste plant matter, planned HVDC, and existing hydro for dispatch) with the much larger than current lithium production that is already pipelined, but scaling Uranium production and the entire reactor and enrichment supply chain 10x with no significant pipelined expansion is trivial?

What happens when the known Uranium resources run out 12 years after that?

1) not that it would happen, but if it did, they won’t just run out.

2) There are vast Uranium deposits in the Southwest and Canada that were identified and either mined a little, or left untouched after WW2 - because we had so much, we didn’t need them.

Early lithium was ‘hard to find’ until it wasn’t, and now there are vast reserves.

Lithium is harder to extract per power unit though, as Uranium is incredibly power dense.

> Lithium is harder to extract per power unit though, as Uranium is incredibly power dense.

Y'all just love making up lies out of whole cloth. This is incredibly wrong and you didn't even consider the idea of checking before deciding it was true. Stanning for nuclear does involve incredible density but it's not power or energy density.

Weight for weight, the amount of lithium you need for a 1kW renewable system with diurnal storage (80-160g/kWh for 8-12kWh to provide for 1kW) and the amount of Uranium you need for 1kW of nuclear reactor (45-60MWd/kg @ 32% thermal efficiency with a 7.4:1 tails/fuel ratio with a 3-6 year fuel cycle) are about the same: Roughly 1kg.

The lithium battery will last 2-4x as long as the Uranium fuel (12-20 years vs 3-6).

At 1-7% the lithium ore is 1-2 orders of magnitude more concentrated than the 0.01-0.7% of most Uranium deposits (Canada's untapped high yield deposits are about the same as good lithium ore but they are deep underground, unique, and only a tiny fraction of what would be needed). The typical ore mined in a mass expansion scenario (0.01-0.03%) would have an energy density between that of coal and crude oil.

Mining for the uranium would involve around 100-1000x the quantity of mined or leached ore and tens to hundreds of times as much leaching chemical.

The Uranium extraction process doesn't end at the mill. Enrichment is just as involved as brine extraction. The end product of the Uranium fuel costs twice as much as an LFP grid battery and has about the same embodied energy per joule delivered (or double if from an underground or deep open pit mine).

Diurnal lithium battery storage is irrelevant where good pumped hydro is available, and is far more than is needed to reach 85-90% VRE (which can be done with 3 hours).

Early uranium was also hard to find, then tens of billions were spent trying to find more. There are fairly reliable methods of estimating how much hasn't been found based on the rate of finding it vs. the effort spent and the answer is there's not a lot undiscovered at concentrations that make fuel affordable. It's also largely irrelevant because any you find after you open your 3-8TW of nuclear plants (which are somehow built in 12 years) is not going to be developed before they all run out.

Thank you for the detailed responce, am I correct in understanding this is for once-through fuel cycle? I was not able to find figures for total burnup once fuel reprocessing is taken into account. I think analysing large-scale transition to nuclear only makes sence with reprosessing.

> The typical ore mined in a mass expansion scenario (0.01-0.03%) would have an energy density between that of coal and crude oil.

This is really interesting, because I have seen a lot of hand-wringing about lithium mining, describing it as physically impossible.

We seem to be mining 8.5 billion tons of coal a year, and 100k tons of lithium a year. Assuming 2% for lithium, that puts us at 4 million tons of ore.

So we need 2000 times less earthmoving equipment to achieve the quantity of lithium we currently consume?

I am assuming here that 'Coal' is the name for stuff that's dug out of the ground, so comparing 'coal' and 'ore' is correct for estimating earthmoving required.

> Diurnal lithium battery storage is irrelevant where good pumped hydro is available

To me the entire point of this scenario is, what do I do if my country doesn't have it. That's going to be the dilemma facing half the world.

> I have seen a lot of hand-wringing about lithium mining, describing it as physically impossible. While cost remains to be seen, it is clear the amount we need to mine is much lower than that of coal.

Very little mining or extraction is actually physically impossible, just as safes are rated not as "impossible to crack" but "takes an expert with best tools 16 hours to open", known measured resources are ranked by economic feasibility .. how much effort will it take to produce the end product (and is that cost worth it).

Lithium | Uranium | Copper | etc .. there are pros, cons, issues and problems all the way through any mineral extraction process - the one solution not proposed nearly enough is for populations to just consume less.

Re: Lithium specifically:

> Let us consider, for example, electric cars. To give an idea of this effect, producing a battery weighing 1,100 pounds emits over 70% more carbon dioxide than producing a conventional car in Germany, according to research by the automotive consultancy Berylls Strategy Advisors.

> Furthermore, lithium mining requires a lot of water. To extract one ton of lithium requires about 500,000 liters of water, and can result in the poisoning of reservoirs and related health problems.

These are our trade offs.

[1] https://phys.org/news/2023-01-paradox-lithium-net-zero-emiss...

No safes I’m aware of are rated to ‘16 hours with tools’, though I imagine some engineer somewhere promised a particular customer that.

Best normal ratings are TL-60 (aka tools for an hour), or TR-60 (tools and a torch for an hour). A big, big difference.

If you lean back and look up there's a chance you might see the point that you missed.

Assume I said "X hours" if that makes you more comfortable.

Also thank you for maintaining civility and expressing genuine curiosity. It's easy to forget that not everyone who is a fan of nuclear is just using it as a tool to attack anything that threatens fossil fuels just hecause those ones are the loudest my distaste towards some of the other commenters caught you in the crossfire and I apologise.

> So we need 2000 times less earthmoving equipment to achieve the quantity of lithium we currently consume?

Both metals have a variety of mining methods including extraction in liquid form, and both need a leaching step in large amounts of chemicals if dug up whole (with the exception of canadian ore which is high purity) but that's the general gist of it. The Lithium requires between 1 and 3 orders of magnitude less space/industry/chemicals and a bit less energy for the same target use. Nickel, manganese, copper and phosphorus also have significant impact but still about the same total impact as the Uranium for a 90-95% grid decarbonization use case up front (but longer lived and recyclable for the battery ingredients). Building a mix of solar and onshore wind to cycle the battery requires a subset of the ingredients of a reactor like an EPR (with maybe a bit more concrete, zinc and steel depending on wind capacity factor, but much less chromium and a number of other higher impact materials).

Offshore wind is about the only renewable technology with significantly larger mining impact. It is better than fossil fuels or delay but does need to come down (Iron Nitride should help here, but there is still a lot of steel and copper -- although an order of magnitude less than the nuclear and fossil fuel interests will tell you).

> Thank you for the detailed responce, am I correct in understanding this is for once-through fuel cycle? I was not able to find figures for total burnup once fuel reprocessing is taken into account. I think analysing large-scale transition to nuclear only makes sence with reprosessing.

Reprocessing does very little without a positive breeding ratio. Unless you create more Pu239 than you consume U235 it's just a small boost in U235 efficiency. You can verify this by looking at the isotope mix of waste for your reactor of choice and the isotope mix and burnup of MOX. An APR has a breeding ratio around 0.6 and consumes 60% of the bred Pu without reprocessing. A PHWR has the advantage of (in principle) being able to extract the last 5-15% of energy without plutonium extraction (which is incredibly polluting and expensive), but it doesn't result in much of an increase in net output, nor does it reduce the amount of Pu240, Pu241, Am242 and Am241 (the very bad high level long lived alpha emitters) that must be dealt with by more than 20%.

Breeders have a host of technical barriers, and a breeder-heavy strategy is actually impaired by spending money and the fairly finite easily accessed startup fissile material on PWRs. The net result of most proposals is about 10x the burnup (as fertile material is notnplannednto be recovered and multiple roundsnof breeding and reprocessing result in problems that don't have proposed solutions), but there is no plutonium separation process that is either affordable or environmentally sustainable if scaled to the TW level.

> To me the entire point of this scenario is, what do I do if my country doesn't have it. That's going to be the dilemma facing half the world.

Sodium ion batteries are a commercial technology now at the 1-5GWh level with a massive scale up being completed in June (100s of GWh/yr), and use all abundant materials (Iron, sodium, carbon, water, aluminium). Pumped hydro is also far less limited. Other chemistries that are in scale up include Zinc Bromide (there is a process that can retrofit existing lead acid production being scaled among others), Iron, and Vanadium (often available as a side product of Uranium in greater quantities at about 1-3hrs of storage per year of uranium fuel), but generally not valuable enough to extract). The main geopolitical danger is access to silver, and building a local industry.

Also buying lithium on the open market has far fewer opportunities for geopolitical domination than systems dependent on fossil fuels or U235 as there are many low quality deposits and the bottleneck is largely extraction. Your country could spend $3-10k per US-citizen-of-primary-energy on imported batteries or $300-1k on imported lithium once then develop domestic recycling and manufacturing and then you are done for decades (maybe replacing 5% of it per year once the kWh per kg of Li stops improving).

Bwaha, 1kg of uranium produces 8.64 x 10^13 joules of energy when fissioned. That's ~ 24 MWH per gram, not per KG. You're off by ~ 3 orders of magnitude.

So what are you talking about? Because your math at the start is so far off, it’s pretty hard to tell here.

You haven't understood his post, he is correct under worst-case scenario for nuclear -> once through fuel cycle achieves 5% burnup, that's 4 million MJ/KG, or roughly 60MWd/kg

Enrichment requires that we throw away most of uranium as depleted uranium, that's what he means by "7.4:1 tails/fuel ratio" That's gives us ~20 years of power at 1KW, but we have are converting to electricity, so it really only gives us 6 years or so.

The main flaw with his argument is, in my understanding, he assumes a once-through fuel cycle, whereas France and others reprocess nuclear fuel. It also discards reactors that work on un-encirched uranium like CANDU.

I am not sure about the ores, but I am greatfull for the detailed writeup

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

Reprocessing doesn't turn fertile material into fissile. It mostly just costs money and pours more fission products into the nearest body of water than Fukushim released. It also extracts the dregs of Pu239 and U235 to get another 5-15% of final energy out. CANDU + reprocessing is a bit better than a PWR, but still significantly under 2x the energy in the original U235.

Closed fuel cycles are a myth so there is no need to specify 'once through'.

Just gonna double down on the 'it's so dense' myth huh?

Do it with a real fuel cycle, real turbine efficiencies and stop trying to conflate the fertile content with the fissile.

Here's a hint to get you started: the reactor fleet uses about 67500 tonnes of raw Uranium and produces about 2650TWh each year. New reactors are about 1.8x the average and most SMR proposals are worse. The world produced 760GWh of batteries from around 75,000 tonnes of Lithium.

Also note that 2650TWh / 760GWh is 3500 which is less than the number of cycles an LFP battery will last.

If you’re going to throw out bullshit numbers, don’t be surprised when you get called on it.

If de-rating, show your math. Which you didn’t.

This whole thing is hilarious anyway, as I’m far from a nuclear advocate. Just pointing out you’re not actually telling the truth while going on your rant.

And it’s all clearly apples to oranges with no sense.

Lithium is not destroyed when used in these battery systems, it can be recycled indefinitely if we cared. It also doesn’t produce any actual power, it’s storage.

Uranium is burned/fissioned, and actually is gone. And actually produces power.

Wow. You got triggered hard by that. My comment was in response to the pearl clutching over the imagined requirement of Lithium mining as a component of a system that generates electricity from wind and solar.

If we take diurnal storage being provided by lithium batteries as a given, then a VRE system producing 1kW >95% of the time using existing commercial technology requires roughly 1kg of lithium in 6-12kWh of batteries which needs to be recycled or replaced every 12-20 years.

A system producing 1kW of electricity from fission 85% of the time using existing commercial technology requires fuel made from 1kg of mined Uranium. The Uranium needs replacing at least every 6 years. You can calculate this easily from burnup (25-60MWd/kg thermal), thermal efficiency (30-38%), and the Uranium required to make a unit of fuel (~8kg per kg of fuel).

Hence the pearl clutching is revealed as disingenuous nonsense and the proposed alternative to the terrors of Lithium mining is revealed to be worse, and why you are triggered.

(2) is already included in the "resources" figure.

> So it's supposed to be impossible to get the 3 hours or 9TWh (12 years of 2022 production) of battery.. but scaling Uranium production and the entire reactor and enrichment supply chain 10x with no significant pipelined expansion is trivial

Yes, that is exactly right!

Energy density of natural, raw, unenriched uranium is 1,000,000 MJ/KG, and energy density of a lithium battery is 0.46 MJ/KG.

1 kilo of uranium gives you 1,000 times more energy than a kilo of lithium will be able to 'process' over the entire 10-year life expectancy of the battery.

You will need 1,000 fewer excavators, dump trucks and people involved in mining if you choose uranium.

If price of uranium increases 3x nobody cares, fuel is like 5% of cost for nuclear. If price of lithium increases 3x it's a disaster, the entire transition to electric vehicles will fail.

We don't even have enough batteries for vehicles, the grid needs another solution, either hydro + hydrogen storage or nuclear.

Now do it again with the electricity output of fuel cycles that actually exist and the amount of lithium in a new battery.

The only density here is that required to mindlessly parrot factoids about the theoretical thermal energy content fertile material as if they were relevant to electricity output of fissile content.

> If price of uranium increases 3x nobody cares, fuel is like 5% of cost for nuclear. If price of lithium increases 3x it's a disaster, the entire transition to electric vehicles will fail.

If the price of Uranium triples, the raw Uranium becomes as expensive as renewables' LCOE in one year rather than 5. If the price of lithium triples, it's still irrelevant because PHES still exists and so do Sodium Ion, Zinc Bromide and Iron batteries.

> We don't even have enough batteries for vehicles, the grid needs another solution, either hydro + hydrogen storage or nuclear.

The amount of batteries required for >90% renewables is tiny compare to the amount needed for overly large EVs everywhere. If you're going to change one of these variables to stop climate change faster, far better to unban light electric vehicles in countries with mandatory monster trucks and spend a few percent of that nuclear reactor money on transit, low speed roads and pedestrianisation.

A near 100% renewable grid doesnt require as much storage as you'd think:

https://reneweconomy.com.au/a-near-100-per-cent-renewables-g...

This model projects that a 98% solar/wind/hydro grid is possible for Australia by building an additional 450GWh of storage. 1.3x snowy 2 (which is 350GWh).

Unfortunately, Snow 2's storage figures are rather misleading. That 350 GWh cannot be used daily, that's its total storage capacity which takes a month and a half to refill. The cyclic capacity of Snowy 2 - as in, the storage capacity that I can pump back into the facility using excess renewable power - is only ~40 GWh.

And again, this is geographically limited storage system: pumped hydro requires just the right geography of an upper and lower reservoir spaced not too far apart. Not too bad if you're a sparsely populated country with huge amounts of land per capita. But it isn't a solution for most countries.

>Not too bad if you're a sparsely populated country with huge amounts of land per capita.

Even a pretty skeptical assessment of Snowy 2 believes it's 240 GWh:

https://majorprojects.planningportal.nsw.gov.au/prweb/PRRest...

"If based on the active storage volume of the ‘lesser reservoir’, Talbingo, the theoretical energy storage capacity is about 240 GWh"

Once again this is just JUST ONE project, already under construction and it already will cover somewhere between 30-40% of the energy storage requirements to get the entire country to reach a 98% solar/wind powered grid.

Dismissing the possibility of this order of magnitude of storage as simply impossible as the OP did when it is already under construction is asinine.

>And again, this is geographically limited storage system: pumped hydro requires just the right geography of an upper and lower reservoir spaced not too far apart.

There have been multiple studies on this. Unlike hydro, the geography for pumped storage is not rare throughout most of the world.

This is the most perplexing talking point against pumped storage, and frankly, reminds me of when people used to pick up on pro nuclear/carbon lobbies sneering at solar/wind for being infeasible because it was < 1% of the grid back in 2014.

It will be used where it is a good solution. Where it is not, others will be used. Use of other methods elsewhere does not detract from its usefulness in places suited to it.

Australia is 1 - Huge, low density, lots of choice where to place renewables, 3 - has some of the best places for solar in the world, 4 - good solar production in winter.

Try Austria, and find that solar production falls 5-10x in winter, there is almost no good location for wind, and the problem is much harder.

Indeed, central and eastern Europe are some of the worst places on Earth for renewable energy. They are "nuclear's last stand". What this means, though, is that in a post-fossil world energy intensive industries will simply move elsewhere. Why build your aluminum smelter in an energy ghetto?

I would add Canada and Alaska, but I think you are right, only Europe has high population and poor conditions for renewables - a terrible combination.

This is not about industry, they need to run their economy, residential heating, etc.

Canada isn't great for many renewables, but hydro supplies most of its power. There's only a couple outlier provinces using coal.

Most of Ontario's power comes from nuclear energy. And Ontario is the biggest economic zone in Canada.

That's wrong. 34% of Ontario's power comes from nuclear energy.

58% as per https://www.ieso.ca//en/Learn/Ontario-Electricity-Grid/Suppl...

> Most of the electricity produced in Ontario is generated at nuclear and hydro plants...

Nuclear: 58%

Hydro: 24%

Like I said, most of Ontario's power comes from nuclear.

You're right. My mistake was reading the capacity numbers.

Austria has just the right geography for pumped storage and is connected to a Europe-wide grid generating plenty of excess solar and wind.

> The overwhelming majority of electricity demand is base load. Usually on the order of 70-80% [1]. We don't need "some" base load, almost all our demand is base load.

The article you linked doesn't really back this up, at least in the way this discussion means it. It shows that with flexible production, you can drastically scale back on "traditional" base load power sources, and that is representing a real power grid in Germany. Nothing about the graph actually says "this is as much renewables as you can pack into this power grid".

If you look at the graph closely, you'll notice that solar is big during the day, and wind is big during the night. With greater installed wind production capacity, the fossil fuel lines would drop drastically in the graph. It's that simple. We would still need to have "peaker plants" available until there is enough grid energy storage capacity, but combined cycle natural plants work fine for that. We can keep pushing down the time they need to on by building more renewables even without batteries.

> The storage facility you linked to was the biggest facility in the world when it was first constructed, and it stored only 129 MWh of electricity.

I specifically linked that one because it talks about how ridiculously profitable it has been, and how much of an impact it has had on the local grid. If it can save money for a traditional grid, then it is a no-brainer for utilities to install bigger and bigger grid batteries. More demand for batteries means more battery production facilities, increasing global production capacity over time.

However, the world is also transitioning to Electric Vehicles, and most EV manufacturers are offering V2G (vehicle to grid) solutions, so millions of EVs can contribute a portion of their battery capacity to the grid in the future, and the grid can compensate them for their contribution.

> Battery production is expected to increase, but it's unclear whether raw material inputs can keep up with manufacturing demands [2].

Lithium is not exactly rare or hard to extract, you can even extract lithium from saltwater, so this argument seems specious. But, various alternative chemistries are being explored which could help in different ways.

> At our current rate of battery production it'd take us a century of dedicating 100% of our battery output to grid storage to reach 1 day's worth of storage.

How did you determine that we need a full day's worth of energy storage? We can drastically decarbonize the grid (and lower electric costs for consumers) with a lot less than that, based on what I've seen, but this is a highly speculative part of the discussion so it's interesting to hear how that number came to be.

We can get a handle on how much storage needed by optimization based on real weather data, minimizing costs based on various assumptions on cost of wind, solar, batteries, and long term storage. This web site lets you do that to obtain "synthetic baseload", the equivalent of what a nuclear plant could provide:

https://model.energy/

If we do this for Germany with 2030 cost assumptions and 2011 weather data, 6 hours of batteries are needed and 289 hours of hydrogen storage. Hydrogen storage is quite cheap, if nowhere near as efficient. It's very useful here, reducing the optimal cost by nearly a factor of 2.

For the US as a whole, the optimum solution uses 6 hours of batteries again, but 106 hours of hydrogen. For just Texas, 2 hours of batteries and 254 hours of hydrogen. California alone is 16 hours of batteries and 70 hours of hydrogen (likely due to wind optimizing to zero under those assumptions.)

I plugged this in with existing storage technologies and existing energy demand for just the USA (500 GW). It turns out we'll only need... 6,000 GWH of battery storage!

https://imgur.com/TGRMOBw

By comparison, the entire world only produces ~400 GWh of batteries each year. So it'd only take a decade and a half of global battery production to satisfy the storage demands of just the USA. The rest of the world would be left with zero EV or electronics production for a decade and a half and no grid storage to show for it.

Thanks for the site: it's a good tool to demonstrate just how unfeasible energy storage really is.

> Thanks for the site: it's a good tool to demonstrate just how unfeasible energy storage really is.

It’s amazing how clearly you can see the trees without realizing there’s a forest.

Battery production capacity has been scaling like crazy and will continue to scale like crazy: https://www.woodmac.com/press-releases/global-lithium-ion-ba...

It doesn’t matter if battery prices increase some, as you have made that the cornerstone of your argument. Many many comments ago, I linked to the Hornsdale battery. Based on the revenue, you can do the math: the batteries could cost a lot more, and it would still have been profitable. Batteries also don’t make up the entire cost of grid scale storage: the inverters, the transformers, even the cabinets and control computers cost money.

As it is, mining is a lagging indicator. Once mining scales up, the cost of the commodities will naturally go back down. While there is profit to be made, too many people will open mines, which will bring the profitability back down to earth. It’s a tale as old as supply chains.

This same inability to imagine how quickly solar and wind would drop in price led to numerous “experts” making absurd claims about solar and wind being economically infeasible at any scale. The battery supply chain is scaling. There will be price volatility, but the volume is growing by leaps and bounds.

It will take years for the world to transition, but it also takes years to build and install the necessary wind and solar. Battery production won’t remain constant, and it won’t even increase slightly. It has and will increase drastically.

Predictions and actual capability are two vastly different things. Outside the realm of predictions, back here in reality, battery costs are actually increasing rather than declining [1].

"Just scale up mining" is easier said than done. Steel is a widely used commodity. If we could "just scale up mining" and exponentially decrease the cost of materials, why haven't we been able to do this with steel? Care you explain why "just scale up mining" will work for lithium when it hasn't for plenty of other commodities?

1. https://about.bnef.com/blog/lithium-ion-battery-pack-prices-...

I said prices would be volatile. Scaling has worked for plenty of other commodities. As I said, I’m out.

> Scaling has worked for plenty of other commodities.

Besides transistors and other electronics, which commodities have seen consistent exponential growth relative to cost?

That’s an irrelevant question. I hate that I’m still bothering to respond when you don’t seem interested in changing your position.

There is a fixed cost to extract a kilogram of lithium using a given method. When demand spikes, lithium suppliers will charge more and reinvest those profits into increasing mining capacity. If they had their way, they’d keep prices high, but anyone can open a lithium mine. Lithium is not hard to find. Those people will undercut the previous miner in order to attract buyers, which will force all miners to lower prices back to reality.

As long as prices are high, more and more mines will open as it is suddenly an attractive resource to mine. This will happen until prices start to drop to some percentage-over-cost where things stop being attractive, and the market reaches equilibrium again.

If lithium was profitable to mine for $X/kg for many years, then that is the price it will naturally return to. It is a commodity. There is no special value for getting your lithium from Corporation X or Corporation Y. The only value is the material.

That’s how commodities work. The price is determined by how hard it is to extract, and whether demand has recently spiked (or subsided). Gold is expensive because it

If this were some rare substance, extraction difficulty would increase noticeably with time, but we’re a long way from a shortage of ways to mine lithium. Some ways might be more expensive, so those mines will only open if the prices stay high, but the prices won’t rise indefinitely this century. That is such an illogical assumption. Perhaps the natural price of high volume lithium production is higher than we have today because those more expensive methods are convenient, but it won’t be enough of a price increase to matter in the grand scheme of things, and this assumption of a higher price is a pretty flimsy one to base predictions on even then.

Eventually, as batteries age, they will be recycled, and the need for lithium mining will likely drop sharply, bankrupting some of these mines until production and (profitable) demand are matched again.

Higher prices would indeed incentivize increased demand. But again, this requires higher prices. Understand that renewable activists are predicting that lithium battery production will not only increase exponentially, it will also exponentially drop in cost per KWh as this happens.

This is not going to happen because, as you point out, lithium mining will only expand if prices climb higher to make otherwise unprofitable reserves profitable. And since raw materials now dominate the cost of batteries, this is going to increase the cost of batteries. There's no having your cake and eating it too: in order to increase lithium mining capacity, battery cost is going to have to grow, not shrink.

You edited your post after I replied, so I'll have to edit in response:

There's no one "difficulty of extraction" factor for each commodity. The reality is that there's a diverse variety of reserves all of which are easier or harder to exploit, even for the same commodity. "The price is determined by how hard it is to extract" is at best a huge simplification.

Higher commodity prices make it viable to extract the more inaccessible reserves, but those reserves will only be profitable so long as prices remain high.

> If lithium was profitable to mine for $X/kg for many years, then that is the price it will naturally return to.

Nope! This is completely wrong. Commodities don't "naturally return" to any price. Prices is a result of supply and demand. It could be we find some other battery chemistry that blows lithium out of the water. In that case the cost of lithium will probably collapse well below $X/kg. Conversely, if countries start to try and provision significant amounts of grid battery storage, the costs will grow even higher as more and more inaccessible reserves need to be exploited to supply market demand.

There is no "natural" price of commodities, whatsoever. If the demand for lithium is going to rise and keep rising, then the cost will rise and keep rising unless some breakthrough makes it way more efficient to mine. Given the fact that we've been mining for centuries and we have huge demand for minerals other than lithium, I'm not optimistic on a 100x improvement in mining efficiency.

It's interesting because coder is very locked into their view on one price.

I don't know lithium at all, but if you follow natural gas / fracking / oil production the question of what reserves are economically recoverable is HIGHLY price dependent (and highly variable). On low end $10-15/barrel cost of production or better. The only reason we have any production in the US is because oil costs more. If you said we would double oil usage prices would spike incredibly.

Same thing with solar net metering, Solar activists demand that soler gen (often during periods with solar curtailments already in effect) be reimbursed based on the same cost as electric rates during the evening (usually peak demand and low solar / wind). Again, depressing how economically nonsensical this all is for a supposedly scientifically grounded effort.

Anyways, rational heads (and economic forces) tend to prevail. I think we will see more TOU rates, and ideally this will at least create some market forces around storage.

The prices have gone up, so mining is expanding. There’s a clear next step to the way commodities work. That next step is not for prices to continue increasing dramatically, things should level off and eventually return to normal, as I explained.

> Understand that renewable activists are predicting that lithium battery production will not only increase exponentially, it will also exponentially drop in cost per KWh as this happens.

Battery prices don’t need to keep dropping for them to take over the world. They’re already cheap. Of course people would love for the prices to drop, and they have historically been dropping slowly, but the basic elements will cost a certain amount, so there is a price floor. I don’t know where that is, but it is lower than we’ve seen, because every step in the battery production chain has been making a profit up to this point, including the miners.

The fact that prices have increased only indicates a mismatch between supply and demand, not that the natural price needs to be this high, or that it needs to continue rising. As I pointed out, everyone was already profitable at a lower battery price than what we have today. If batteries are expensive because of lithium, more lithium mines will open and drop the price. If batteries are expensive because the battery makers are price gouging, someone else will undercut them.

The main concern for something like this is market distortion by patents. If the best way to make batteries is locked behind a patent, that can cause prices to be unnaturally high. This concern does not apply to elemental lithium.

> Higher commodity prices make it viable to extract the more inaccessible reserves, but those reserves will only be profitable so long as prices remain high.

You’re practically quoting my comment to me. I addressed that. It is possible that the price will settle higher than it is today if those methods play a big role, but it won’t matter. You don’t seem to appreciate how insanely cost effective batteries already are today. A modest price increase is fine, but it is still illogical to assume prices will stay high. Lithium is not hard to find. Why would the market settle on expensive extraction methods? It is a strange assumption to start from. We’re not talking about something that’s rare.

Either way, the outcome is unchanged unless batteries increase in price exponentially, as your earlier comments apparently assumed.

For the second time, there is no "natural price". You need to iron this kind of wishful thinking out of your head. Companies want to buy more and more lithium, so unless you've got some breakthrough that makes mining a heck of a lot cheaper the cost will keep going up and up as more and more inaccessible reserves need to be exploited. If I have a reserve that costs $4/kg to operate when the price is $5/kg and another reserve that costs $7/kg, I may open the latter if the price rises to $8/kg. But if the price drops back down to $5/kg it's unprofitable to operate.

Here's a way to articulate this that might better mesh with your mental model. The "natural price" of a commodity isn't static. If demand increases and the only way to meet demand is to use more and more expensive mining operations, then this raises the "natural price". As per the above example, the "natural price" rose to $8. It won't drop back below $8/kg unless either we find a way to make mining cheaper, or demand drops back to levels that can be satisfied by my cheaper-to-operate reserve.

You have too many assumptions about the availability of lithium, and lack of “easy” reserves. That’s the fundamental disconnect here. Lithium is not inaccesible, and it is super common. The “hard” ways to mine lithium aren’t that hard.

We’re not talking about gold or platinum, where your discussion points would be relevant.

I’m not wishing over here. But anyways, I’ve tried to explain the reason why the market disagrees with you, and why people are rapidly seeking to integrate batteries into the grid and into cars, which would not be possible if you were correct about lithium becoming insanely expensive.

Battery production will keep scaling at a huge rate, and the prices will be fine. A modest price increase is irrelevant to anything discussed here today.

Lithium is far from easy to extract. It also needs to be refined on-site because its raw form is too low-density to economically transport. So most lithium mines require large amounts of water to fill brine pools. The number of sites that are actually viable to produce lithium are not nearly so common.

Again, the market disagrees with you not me. Lithium commodity prices are rising, and so are battery prices. People are indeed seeking to integrate batteries into the grid - I'm not disputing that. I'm pointing out that these batteries are actually becoming more expensive [1], not less and it's unlikely this trend will reverse on a dime.

1. https://www.utilitydive.com/news/new-york-battery-storage-co...

Commodity prices are rising… causing more mines to open… causing more supply… which will cause prices to drop. Why do you keep pointing to that? Your argument makes no sense. That’s how commodities scale.

I'd check out something like oil which is a large and high dollar market.

The relationship between production and price is well established, and no, production does not move inversely to price. This is because the marginal cost in almost any of these areas is higher per unit of production as production increases. That also of course makes sense logically.

Gold as well, same thing. You start by grabbing gold nuggets of the ground. The effort to mine more gold has not gotten easier, it's gotten harder over time.

I already addressed all of this with the difficulty of extraction. I even called out gold earlier in the thread, possibly more than once.

The abundance of lithium is nothing like the abundance of gold, and oil is a different subject altogether because we have extracted a truly staggering amount of it, and then burned it. We’ve extracted something like 135 billion tons of oil, and a “mere” 250 thousand tons of gold. So, hundreds of thousands of times more oil than gold has been extracted.

Lithium is about 20ppm (parts per million) in the Earth’s crust. Gold is about 0.001ppm: https://en.wikipedia.org/wiki/Abundance_of_elements_in_Earth...

Lithium is 20 thousand times more abundant than gold, roughly.

Lithium is reusable, analogous to how we handle aluminum. As demand is ramping up, we will need to ramp up extraction, but we’re not throwing it away, and eventually we will get a lot of our lithium from recycling.

> Commodity prices are rising… causing more mines to open…

Yes.

> causing more supply…

Yes

> which will cause prices to drop.

Nope! Remember, the expanded supply is only profitable to operate at higher prices. So the price will drop if demand drops and then supplier will operate the more efficient extraction sites. But if demand stays high, so will prices.

You should be linking to evidence that prices will rise uncontrollably if you want to support your argument that battery production won't keep increasing.

I will remind you that this is your absurd argument:

> By comparison, the entire world only produces ~400 GWh of batteries each year. So it'd only take a decade and a half of global battery production to satisfy the storage demands of just the USA. The rest of the world would be left with zero EV or electronics production for a decade and a half and no grid storage to show for it.

> Thanks for the site: it's a good tool to demonstrate just how unfeasible energy storage really is.

Which is totally different from "battery prices might increase a little", which I have repeatedly indicated is not a problem, and I even pointed out that it is a possibility a long time upthread from here.

Again we're talking about 6,000 GWh of grid storage for just one country. And that's also competing with batteries for EVs, and batteries for electronics. That site estimate's for storage necessity are also pretty optimistic. Most estimates for a completely decarbonized grid - not using peaker gas plants - call for weeks of storage not days: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

And again, battery production is scaling like crazy. That is an undeniable fact for now. You're the one speculating that it will suddenly stop, and you have presented flimsy evidence at best. Your entire argument falls apart if batteries continue scaling at a reasonable price.

> Your entire argument falls apart if batteries continue scaling at a reasonable price.

The only logical choice is to assume that the battery industry will keep doing what it has been doing, unless you present some strong evidence to the contrary, which you have repeatedly failed to do, as has been pointed out by myself and others. The commodities are behaving as expected, which is to say that they're rising in the face of new demand, since supply hasn't had time to catch up.

You're proposing that the sky is falling, but it looks just fine to the rest of us.

Obviously production would have to be scaled up. It's dishonest to present this as some sort of insurmountable barrier, for all potential battery chemistries or other storage modes.

Batteries are not transistors. Their input costs are skyrocketing, and sure enough the end costs of batteries are now starting to rise, too [1]. It's dishonest to pretend that continued exponential growth is guaranteed.

The cost of an automobile shrank from a million dollars inflation adjusted to a hundred thousand over the course of the 1900s. Assembly line manufacturing continued to shrink this down to $10,000 by 1920. Would it be safe to assume that a car would cost $1 by the end of the century given the past rate of a 10x drop in price every two decades?

The reality is that most products are not transistors. They don't get better when you make them smaller. A car will always contain a certain mass of metal, and will not cheaper than the cost of that input. Manufacturing already accounts for under a quarter of a battery's cost. The rest is dominated by cathode and anode material [2]. Battery manufacturing had already become a resource extraction problem.

1. https://about.bnef.com/blog/lithium-ion-battery-pack-prices-...

2. https://www.visualcapitalist.com/breaking-down-the-cost-of-a...

You could not be more wrong about batteries. You made a huge error about the cost percentage that lithium makes up. It’s possible — just possible — that other people have researched batteries too, and maybe they have a better sense of where the industry is going.

It makes no sense to assume that we’ve reached “peak battery”, at all.

Continuing this argument is pointless, and really does seem like you’re choosing to ignore reality. I’m done with this conversation.

> You made a huge error about the cost percentage that lithium makes up.

Care to explain? I'm always intrigued by commenters asserting that an error has been made, yet neglect to explain the error.

I already responded to you elsewhere. Perhaps you ignored it? https://news.ycombinator.com/item?id=34460846

Cathodes are not a giant chunk of elemental lithium, as you believe.

But lithium still dominates the cost of the cathode. Iron is not the main driver of battery expenses.

13% is not 50%. It makes a huge difference in the cost scaling under discussion.

In fact, nickel dominates the cost of the cathode, in case you missed what I had posted. But there are nickel-free chemistries out there.

And for those nickel-free battery chemistries, what inputs are the main cost drivers? And as per my response, nickel is also experiencing shortages and cost spikes.

My core assertion remains true regardless of the fact that different chemistries require different inputs: battery production has become a resource extraction problem, rather than a manufacturing problem. Unless we find a way to somehow make mining exponentially more efficient, we're not going to be seeing exponential growth in battery production.

Your core assertion remains false because (1) lithium is not the only battery chemistry, and (2) batteries are not the only, or even cheapest, storage technology.

CAT-L, the biggest battery maker in the world, is today ramping up sodium battery production capacity. Will you now insist on a sodium supply bottleneck?

Lithium is the only viable battery chemistry we have presently. Lead acid decays after 100-200 cycles. Who knows if sodium will be viable, it'd be a lot better to argue for its efficacy when it's actually on the market.

The other forms of storage have other shortcomings: hydroelectric storage is geographically dependent. Most of the other form of storage people are mentioning in this tread have never been operated outside of prototypes: like hydrogen electrolysis or compressed air. Again, technological feasibility and market viability are two vastly different things. If compressed air storage works, but it's more expensive than nuclear power what's the point?

Lithium is not, in fact, the only viable battery chemistry. CAT-L, the world's largest maker of batteries, is today ramping up production of sodium batteries. Numerous other chemistries are also being fielded, at $billions scale.

Pumped hydro storage is not, in fact, geographically dependent: there are a lot of hills (which you knew). Compressed air is used in production (which you knew). Hydrogen electrolysis is operated in the millions of tons (which you knew).

Relying on falsehoods is a strange way to argue. It depends on your audience remaining ignorant.

I don't care about company hype and marketing. Get back to me when I can actually buy a sodium battery and throw it on a test load to see if it's living up to the promises. It's not available yet, that's the reality. Tech that's 5 years away has a nasty habit of staying 5 years a way for a lot longer than that.

I got a real kick when you insisted hydroelectricity storage is not geographically dependent and immediate followed by saying it needs hills. And it requires more than just hills. It needs a hill, with a reservoir on top, and another reservoir at the base of the hill to collect the water, and another body of water to fill said reservoirs. The conditions for hydroelectric storage are way more specific than "it needs a hill".

I agree, relying on falsehoods are a bad way to argue. That's why it's bad to insist that storage is an easy problem to solve to try and make the case for intermittent sources.

> Get back to me when I can actually buy a sodium battery and throw it on a test load to see if it's living up to the promises.

Sure, right after you get back to to inform us about the approved and operating permanent nuclear waste storage facility in the US with all the waste waiting in temporary holding facilities having been placed there.

We already have it. Yucca mountain was built and could receive waste, but Congress blocked it legislatively.

Yucca mountain is, in fact, not an "approved and operating permanent nuclear waste storage facility in the US with all the waste waiting in temporary holding facilities having been placed there". It is not even "approved and operating".

It's ready to receive waste, it's just political grandstanding withholding approval. It would also be a waste to seal up non-reprocessed fuel since it's a future source of fissile material. Regardless, nuclear waste is a non-issue entirely. We disposable of chemicals far more toxic that never decay all the time, but people get their arms worked up over the word "nuclear".

And a nuke plant needs an unused containment vessel.

Betting that the biggest battery manufacturer in the world will default on billions of dollars in orders for sodium batteries, next quarter, would be a courageous position to take. But I doubt you are shorting CAT-L.

And then there are various thermal storage schemes. Pumped thermal (reversible thermal cycle with hot and cold storage) could have up to a 75% round trip efficiency with cheap materials and no geographical restrictions.

https://aip.scitation.org/doi/10.1063/1.4994054

And has anyone actually built an electrical thermal storage system - not district heating - with 75% round trip efficiency? Or is this just more marketing hype?

> Thanks for the site: it's a good tool to demonstrate just how unfeasible energy storage really is.

Turns out even good tools can give bad results if intentionally used in bad faith.

If you include a small amount of dispatch to represent long distance transmission, biogas or hydro things change dramatically. Prices are for real finished projects, and are going down for wind, solar, and storage very rapidly.

10% Dispatch: https://model.energy/?results=99cee0624b3f3f84ed7fe4f03b968c...

5% Dispatch: https://model.energy/?results=50ab975942dbab37b25ef471759b4d...

2.5% Dispatch: https://model.energy/?results=524af6e91d482ff646b2a9fcbff5a1...

At 2.5% even using gas for dispatch would only add 10g of CO2 per kWh to the total. Well within the error margins of estimates from various low carbon sources and significantly less than taking a year longer to build out low carbon generation.

Restricting scope to subsets of the country:

PNW with no offshore wind because it has the worse renewable resources:

15% dispatch + imports and 9hr storage: https://model.energy/?results=a72cdef1b55cc6a27ce7b9ba9c534e...

West virginia is similar: https://model.energy/?results=adcdcd3d6e54c0c80b0bb39ebe75c5...

California with 5% dispatch: https://model.energy/?results=b3166c36c03318779a65aef9d2f482...

A little off-topic, but has anyone ever developed residential hydrogen production / storage solution? What's the smallest scale hydrogen storage avaliable?

Hydrogen is not very scale-sensitive. Many people would say it is not safe enough for residences.

> However, the world is also transitioning to Electric Vehicles, and most EV manufacturers are offering V2G (vehicle to grid) solutions, so millions of EVs can contribute

V2G is idiotic. It cost £10 for full-charge of Tesla in UK pre-ukraine. Half of that money is distribution cost, not energy cost. What will you earn from buying low, selling high in V2G? Let's be generous, say it's £5.

So the battery lasts 1500 cycles, you are going to make £7500. And the battery costs £17,000 to replace?

No-one will be damaging their expensive car battery to earn fraction of a dollar to earn pennies. For anyone to do this, it has to be 2x above break even.

> Lithium is not exactly rare or hard to extract, you can even extract lithium from saltwater, so this argument seems specious.

You can extract anything from seawater, even gold.

"Using electrochemical methods, comparable to those used in electroplating, gold has actually been extracted from the ocean, but unfortunately the cost of the process is five times the value of the gold obtained."

“It's technically possible to extract lithium from seawater,” Cui says. “But it's all about cost. And currently it's too high.”

There's no need to be so aggressive. V2G would likely operate similar to Net Metering, which the UK (and many other countries) does have experience with: https://uk.renogy.com/blog/net-metering-for-home-solar-in-th...

According to one website that I found, the average residential electricity cost in the UK is $0.482/kWh. Simple math: 1500 cycles * 75kWh * $0.482/kWh = $54,225 in offset electric costs, which would more than cover the hypothetical battery replacement cost. It is well above the 2x break even that you requested.

However, I expect most people do not have electric bills that high, so those people would not allow that much of their battery to be used by the program, unless they were participating in this program for a very long time. They would be more than fairly compensated for whatever cycle life was used by the net metering program. Yes, I recognize that the electricity would have come from the grid at some previous point in time in most cases, so some adjustments would have to be made to the concept of Net Metering to make this a sensible proposal, but there are ways to make it work if you don't just declare that everyone is a bunch of idiots. Electricity is much more valuable to the grid at certain moments in time than others. ""Transmission costs"" are irrelevant when the grid needs more power. Any kind of V2G system would involve exporting power when the grid desperately requests it, not just at a random hours of the day when the grid doesn't even want it. Virtual distributed grid batteries are already being proven.[0] This solution can definitely be scaled up.

Also, that estimate of battery replacement cost is based on the battery magically having zero marginal value at the end of its lifecycle in your car, which is not at all how things work. A battery that people are unwilling to keep using in an EV will still have significant value if it is directly transplanted into a stationary storage application, but oftentimes, it is just a handful of cells that are causing balancing issues, so replacing those for a few dollars can restore a lot of missing capacity. Even after the battery is done in stationary storage, it will still have value in the highly concentrated resources which can easily be recycled into a new battery pack.

So, no, it would not cost that much to replace the battery. Even today, you could throw a used battery pack on eBay as a worst case scenario and get thousands of bucks for it, but in the future as more of these batteries are being cycled through the market, it will be more economically viable to have businesses built around making this as painless as possible.

If you consider V2G idiotic, then you're effectively asserting that any form of grid battery is idiotic because those have to work on very similar economics, and that is a strange position to hold when we have real world examples of grid batteries being profitable even today.

> Half of that money is distribution cost, not energy cost.

I also have no idea where you got the notion that UK electricity costs are 50% transmission fees. It is 3%.[1]

> You can extract anything from seawater, even gold.

I'm not saying seawater is absolutely the best available source of lithium, but some people claim to have proven real success with it.

"The researchers estimate that their system can extract 1 kg of lithium from seawater at a cost of $5 (Energy Environ. Sci. 2021, DOI: 10.1039/d1ee00354b). “Our process is quick, energy efficient, and scalable,” Lai says. “And the system runs continuously and is compact and easy to operate.”" https://cen.acs.org/materials/inorganic-chemistry/Can-seawat...

Using gold in your response the way you did seems intended to be an insult, which is completely unnecessary. I'm not proposing alchemy.

Maybe you know more than these researchers whose literal job is to know about lithium and seawater, and that's fine, but I have never claimed you can extract meaningful amounts of gold from seawater. It is infeasible. Lithium is far more common than gold, though, which makes a huge difference in the feasibility.

[0]: https://electrek.co/2022/09/02/tesla-virtual-power-plant-gro...

[1]: https://www.nationalgrid.com/electricity-transmission/who-we...

> I also have no idea where you got the notion that UK electricity costs are 50% transmission fees. It is 3%.[1]

I expressed myself incorrectly - I meant that the various non-generation costs, like transmission, various middlemen and fees, etc. roughly double the cost of energy between powerplant selling it and it reaching my house.

> Maybe you know more than these researchers whose literal job is to know about lithium and seawater

I am just slightly jaded from seeing many research breakthroughs, especially in batteries, never make it to market, and only really consider something once there has been at least one commercial facility operating. My understanding is that I can't buy lithium from seawater at the moment.

> If you consider V2G idiotic, then you're effectively asserting that any form of grid battery is idiotic

On the contrary, I think if you are buying a used car battery on the cheap, cycling it optimally and you don't need an overpriced Tesla-authorised mechanic to fix it, the results should be much better. Safety is less of a concern too. Or you can buy Lithium iron phosphate.

I think people are very precious about their cars and have many worries about damaging the battery, some of them irrational. There are also logistical challenges, repair is harder, etc.

I think the economics you outlined are the very best-case scenario, with war in Ukraine causing energy crisis and assuming nearly free electricity avaliable to charge.

"Lithium is not exactly rare or hard to extract, you can even extract lithium from saltwater, so this argument seems specious. But, various alternative chemistries are being explored which could help in different ways."

Lithium is abundant as far as its total share in the Earth's crust goes, but it is my impression that mineable concentrations are rather rare, unless you are willing to spend obscene amounts of energy on purification. That is why only a few countries in the world actually produce lithium commercially.

We have some reserves of lithium in Czechia, but they would only be economically mineable if the bulk price rose significantly (as in, ten times or so), plus the mining methods would introduce a lot of poisons into the environment. Mining is usually seriously dirty.

Until recently, there wasn’t much demand for lithium. That’s the main reason for limited mining, in my opinion.

The cost of lithium was 9% of the cost of a lithium ion battery cell according to one analyst last year. Battery packs cost more than just the sum of the cells.

If lithium rises in price, it will affect battery prices, but a 10x increase would “only” double the cost of the battery cells, and the packs would be slightly less affected than that.

Realistically, there should be plenty of lithium for less than that, but it takes time to ramp, and lithium mining does not have to be destructive. The mining can be as simple as evaporation ponds in the desert[0]. (You’ll need refining either way.)

https://www.engadget.com/2019-02-24-the-big-picture-lithium-...

> The mining can be as simple as evaporation ponds in the desert[0]. (You’ll need refining either way.

The OP you are replying to lives in Czechia, they are not blessed with optimal location. Most of the world isn't.

The optimal location is a paradox, it is a desert, At the same time it has to have abundance of water for extraction and evaporative pools. Very few locations have that. Extracting from non-optimal locations costs more.

like 4 people are trying to explain this to you, but you keep repeating the same thing

I'm pretty sure it has almost exclusively been one person that has been explaining, and they have received numerous rebuttals from other people, but okay.

And no, I did not suggest the person I was replying to should do this in Czechia. Every country doesn't need to mine every resource. Thanks for throwing in a strawman.

What is the origin of the claim that lithium makes up 9% of the cost of a battery cell? Anode material is the majority of the cost, according to this: https://www.visualcapitalist.com/breaking-down-the-cost-of-a...

In reality it's 50% of the cost, so a 10X increase in input price would amount to a 5x increase in cost. Battery production has become a resource extraction problem.

https://www.newsweek.com/precious-metal-values-are-raising-b...

> In 2021, lithium comprised 9 percent of the cost of a battery cell. Nickel represented 12 percent of the cost. But for 2022, those numbers have risen to 13 percent and 21 percent, respectively.

I had just glanced at the Google summary, so I only saw the first part. Call it 13%, then.

It is not anywhere close to 50%. The cathode is made of more than just lithium. 50% is an incorrect interpretation of the graphic in the article you linked.

Great, so we have to solve shortages of lithium and nickel. Nickel's cost is also up ~3x over the past few years: https://ycharts.com/indicators/nickel_price#:~:text=Nickel%2....

My point remains: battery production has become a resource extraction problem. Even completely optimizing manufacturing to the point that it costs nothing would only reduce the cost of batteries by a quarter.

Which just shows that you don’t understand commodity pricing. If gold was as common and accessible as lithium, gold would be a lot cheaper.

What don't I understand? Ultimately price is determined by supply and demand. Gold's supply relative to its demand sets its price. If some massive load of goal was dumped on the market, we'd expect a downswing with increased supply and no commensurate increase in demand (although gold being bought by speculators might distort this).

We're seeing a big upswing in lithium demand, mostly for EVs. And supply isn't catching up leading to higher costs, which will in turn lead to higher prices and consequently lower demand. Unless greater supply is secured or demand is reduced, this increase in price is going to persist.

If the predicate is that mining of a commodity cannot ever expand, the how is building new nuclear supposed to work? Where does the Uranium and Gadolinium for fuel rods come from?

And tautologies will persist in being tautologies.

Higher prices will result in investment in increased production, shift of demand to lower-priced alternatives, and investment in alternatives. Increased investment results in cheaper and larger production of everything.

Unobtainable lithium is a strange choice of hill to die on.

Increased investment does indeed incentivize the exploitation of less accessible reserves that would not otherwise be profitable. But those higher prices have to be sustained to keep those additional operations profitable. If prices goes down, those mines close and diminished supply pushes prices back up to equilibrium.

It's not unobtainable lithium I'm worried of. It's difficult supplying 50x as much lithium over the next couple decades with no significantly increase prices.

> But those higher prices have to be sustained to keep those additional operations profitable.

That is not, in fact, the case. A higher price need persist only long enough to pay the capital cost of developing the resource. The millionth ton is routinely much cheaper to extract than the first ton.

And, of course, we don't need 50x lithium production in any case. That is your private fantasy. No, thank you, not buying.

Look at the the lowest point of energy demand. That's base load. How big is it relative to the peak of energy demand? Depends on the season, but it's usually 70-80% of peak demand. So, the vast majority of energy demand is in fact base load. I'm really confused about why renewable proponents talk about base load all the time - it's really not relevant to decarbonization of the grid.

Intermittency of wind and solar aren't just daily: you also have longer-term periods of cloud weather blocking solar and lower wind speed hampering wind power. Actually running a majority renewable grid requires either hydroelectricity, or fossil fuels.

The majority of Germany's electricity comes from fossil fuels [1]. It's not a mostly renewable grid, occasionally supplemented by peaker plants. It's a majority fossil fuel grid supplemented by renewables. By comparison, here's France's electricity production [2]. One of these is a mostly decarbonized grid. The other is a primarily fossil fuel grid, supplemented by renewables here and there.

> However, the world is also transitioning to Electric Vehicles, and most EV manufacturers are offering V2G (vehicle to grid) solutions, so millions of EVs can contribute a portion of their battery capacity to the grid in the future, and the grid can compensate them for their contribution.

This is an idea that no sane grid operator would ever accept. First of all, most of these vehicles actually lose energy in cold weather [3]. And if people leave to go on vacation, then we have blackouts because our energy storage solution drove away for a week? No to mention, plenty of people drive their cars around during the day doing chores or work and charge them at night. Those people are going to be a net drain on the grid.

> Lithium is not exactly rare or hard to extract, you can even extract lithium from saltwater, so this argument seems specious. But, various chemistries are being explored.

The market demonstrates otherwise. At the end of the day, if there's a shortage of lithium and the price goes it up it doesn't really matter what people are writing on tech forums.

1. https://en.wikipedia.org/wiki/Energy_in_Germany#/media/File:...

2. https://en.wikipedia.org/wiki/Energy_in_France#/media/File:F...

3. https://news.ycombinator.com/item?id=34120237

> Look at the the lowest point of energy demand. That's base load. How big is it relative to the peak of energy demand? Depends on the season, but it's usually 70-80% of peak demand. So, the vast majority of energy demand is in fact base load. I'm really confused about why renewable proponents talk about base load all the time - it's really not relevant to decarbonization of the grid.

I think this is just a confusion of terminology. What "renewable proponents" are talking about is baseload power plants. To quote the Wikipedia article that you linked to:

"Power plants that do not change their power output quickly, such as large coal or nuclear plants, are generally called baseload power plants."

If you've been missing that key piece of terminology, I can see the source of the confusion. People just call this "base load" to be short, for various reasons. You can criticize this if you want to, but that is what is happening.

> The majority of Germany's electricity comes from fossil fuels [1]. It's not a mostly renewable grid, occasionally supplemented by peaker plants. It's a majority fossil fuel grid supplemented by renewables. By comparison, here's France's electricity production [2]. One of these is a mostly decarbonized grid. The other is a primarily fossil fuel grid, supplemented by renewables here and there.

This has nothing to do with my previous comment. I have no idea what point you're trying to get at.

> This is an idea that no sane grid operator would ever accept. First of all, most of these vehicles actually lose energy in cold weather [3].

You're really confused about a lot of things in this part of the discussion. EVs have less range in winter because more of the energy is being used to heat the cabin, and that was done with resisitive heating until recently (more vehicles are starting to use heat pumps). The grid operator would not be insane, except from the point of view of a very traditionalist grid operator. The future is dynamic.

> The market demonstrates otherwise.

It really doesn't? The market demonstrates that demand has risen sharply, and it will take time for production to catch up. Every demand spike results in a "shortage".

"base load power plant" is a meaningless term. Base load is a feature of energy demand. Power plants produce energy, the same power plant might serve peak load, base load, or both.

EVs are not a solution to energy storage. Even ignoring the challenge of getting people to hook their cars up to the grid, the battery production figures aren't remotely close to what we need.

People have been assuming that battery cost will continue to decline exponentially. We'll need a century to provision just 18 hours of battery storage at current rates. Renewable activists hand wave this saying that battery production will increase a hundred fold. In reality, prices are rising: https://about.bnef.com/blog/lithium-ion-battery-pack-prices-....

> We'll need a century to provision just 18 hours of battery storage at current rates.

That's a weird way of spelling enough batteries were produced in 2022 alone (about 760GWh) to replace the entire world's nuclear fleet with a renewable mix with minimal overprovision, more flexibility to meet peak demand and a lower forced outage rate.

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

> Renewable activists hand wave this saying that battery production will increase a hundred fold. In reality, prices are rising: https://about.bnef.com/blog/lithium-ion-battery-pack-prices-....

And this has nothing to do with reserves and everything to do with extraction. The price spike of lithium was a combination of unpredicted demand and the supply dip from covid (because brine ponds take about 3 years to process).

Lithium will be irrelevant to grid storage anyway. There are already two GW scale sodium ion factories running and the supply chain CATL is building will completely dwarf them as it is much larger and compatable with the TWh/yr of existing lithium cell production and packaging facilities.

> People have been assuming that battery cost will continue to decline exponentially.

I think the incredible success of the transistor industry has created some very unrealistic expectations about the pace of technological advance more generally. I know this is a fallacy I used to fall for. If a pocket sized computer today can beat out a supercomputer the size of a small building from when I was a kid, anything seems possible. But in reality, the rapid miniaturization of transistors is an extraordinary outlier.

> EVs have less range in winter because more of the energy is being used to heat the cabin, and that was done with resisitive heating until recently (more vehicles are starting to use heat pumps).

I thought it was this PLUS the fact that lithium batteries degrade quicker in freezing temps?

https://www.livescience.com/61334-batteries-die-cold-weather...

I agree cold has some effects on batteries, but the main impact on EV range is related to heating, so linking an article on EVs is not very useful when talking about stationary storage.

Batteries generate heat when charging and discharging, so grid batteries should naturally keep themselves warm, but maybe in extreme climates it would be worth adding some heat pumps to keep them in the optimal temperature range for longevity. It all depends on how the cost calculations work out, but it is not some major obstacle.

> Batteries generate heat when charging and discharging

/scratches head.

Huh? How do you explain why my iPhone drains so much quicker in the cold then, if it's able to heat itself from the discharge?

> but it is not some major obstacle.

Ignoring efficiency of battery consumption is a hilarious oversight. If it costs you energy ("lets just pump heat pumps on them") to keep an energy source from depleting faster, then you're still depleting energy faster because of said energy required to heat the thing. That's not trivial.

A tiny battery generates a very tiny amount of heat. Batteries have internal resistance, so charging and discharging is not perfectly efficient. Larger batteries generate much more heat by moving larger amounts of energy. Secondly, consider the relationship between volume and surface area. Volume grows to the power of 3, and area grows to the power of 2. A small battery naturally has a much lower ratio of volume to surface area than a big battery. With a higher ratio, it takes a lot longer for the heat to be lost to the cold because of how heat transfer works, and that's in addition to the larger batteries generating much more total heat energy. Thirdly, consider that the goal of an iPhone is to use as little power as possible so that your battery lasts as long as possible, which also means generating as little heat as possible. A grid battery does not have that same goal.

It is not a "hilarious oversight". Heat pumps are incredibly efficient these days even in very cold weather and there are ways to mitigate the heat loss, such as insulation. Grid scale battery deployments up to this point are typically already done in small structures, so the batteries aren't just sitting out directly in the elements. Obviously it's better if you don't have to do anything at all, but it is a very legitimate option. Every system will have some losses, no matter what. A very efficient heat pump could result in less losses than relying on the batteries to resistively heat themselves in extreme cold.

Just because it wouldn't work for a phone has nothing to do with how it would work for a grid battery, since scaling things up has a dramatic impact on efficiency for what we're discussing.

For how condescending your comment seemed to be, you seem to be missing a lot of basic points. Please give people the benefit of the doubt. I didn't come here to be a jerk to anyone, I was trying to help people understand things they seemed to be missing, but HN got super toxic over this topic. Some of this stuff is very nuanced and difficult to explain, which means that some people have no patience for it. If I'm wrong about things, or make a mistake in my explanation, I'm always happy to learn, but your comment was not attempting to teach anything.

Why are you talking about grid batteries when I was talking about EV batteries?

"Base load" is just a description of how demand is behaving currently, its now high as we are using fossils unsustainably and keeping intra day price variations artificially low. It can dip much lower and hourly pricing and spot market means the usage adapts to production without blackouts. (this also makes storage profitable to build where needed)

Every fact you are saying may be true, but “fossil fuels are a bigger cause of blackouts than renewables” doesn’t follow from the facts your gave, you’re just assuming it is true. That is the last thing I will say on this.

> you’re just assuming it is true

I'm not assuming it. I'm referring to the very real, major blackouts that have occurred in the US over the past couple of years. These events have plenty of reliable sources that tell exactly what happened.

In the Texas blackouts, the problems were coal and gas plants going offline due to the cold that they weren't winterized against. It had nothing to do with Wind or Solar failing unexpectedly, despite the governor's claims early in the blackouts.

In the recent rolling blackouts in the Southeast, TVA and Duke Energy reported that the cause was their coal and gas plants freezing up. Nothing to do with renewables again.

I cannot recall any major blackouts caused by the fossil fuel plants operating normally while renewables failed to produce on schedule. I would love to see some examples, if they exist.

The Cleantechnica article that I linked to earlier (which you surely didn't read) also provided quotes on this exact topic, and that analysis agrees with my own, FWIW.

I could dig into the specifics of these events and provide more sources, but you don't seem likely to care.

Perhaps unintentionally, you changed your argument from “fossil fuels are responsible for blackouts” to “ I cannot recall any major blackouts caused by the fossil fuel plants operating normally while renewables failed to produce on schedule”. These are different arguments.

I am very much on team renewable, but the only reason those fossil fuel plants are needed in the first place is because of the very nature of renewables. So it’s disingenuous to say that fossil fuels are responsible for blackouts when it’s the dispatchability of renewables that required fossil fuel burning in the first place.

I don’t think the GP meant anything other than this.

> So it’s disingenuous to say that fossil fuels are responsible for blackouts when it’s the dispatchability of renewables that required fossil fuel burning in the first place.

No... as I recall, in the Texas blackouts, the renewables actually generated more power than originally forecasted. If the natural gas and coal plants go offline completely, that has nothing to do with the dispatchability of renewables. The fossil fuel plants would be part of the grid regardless, because it takes decades for those plants to reach the end of their lifecycle and be decommissioned.

The grid relies on every type of power doing what it says it will do. It is possible to predict when solar and wind will deliver power, and how much they will deliver. The stability of the grid relies on that. The fossil fuel plants were the ones that had problems. If they had produced power as they normally do, there would have been no blackout.

> These are different arguments.

Do you still think I changed my argument? I'm fairly sure I didn't, but I can see how you reached that conclusion without the clarifications above.

> the renewables actually generated more power than originally forecasted.

They were forecasted to produce 6% and actually made 8%, vs 41% from the prior week. You can’t blame renewables in this case, but you aren’t presenting facts in an unbiased way.

My primary point is that renewables are working the way that they say they will work. Certain people refuse to admit that fossil fuel plants struggle in adverse conditions, but then rant about renewables being somehow unreliable.

Renewable integration into the grid is heavily dependent on forecasting. You may think it is biased, but I see that statistic as things going better than forecasted. With proper forecasting, you can overbuild and make up for lower production. With proper forecasting, you can employ demand response.

When the fossil fuel plants turn off completely and unexpectedly, there’s nothing you can do, because no one planned for that. The grid was relying on the power plants, and they weren’t there. If the grid had planned for the unreliability of those plants, the grid might’ve had more renewables to make up for it, who knows.

People building renewables plan for the variability. As I’ve mentioned, energy storage is essential for the long term.

> there’s nothing you can do, because no one planned for that

You're comparing a future of renewables to a present of non-renewables. You could also add in additional generation to cover the gap you mentioned in the future as well.

You may not have intended to change your argument, but I think you were assuming evancox100 was arguing against renewables where I thought they were just making a point about your language. I can’t speak for them or you, but I think their comment was fair in the absence of the clarity you’ve now provided.

Energy is fungible, so if both fossil and renewables are both online and producing energy how can you blame one over the other? My understanding is that neither source of energy was winterized to transmit the energy to the grid, not that they were unable to produce. So blaming one or the other is a red herring.

> But unlike utilities under traditional models, they don’t ensure that the resources can deliver power under adverse conditions, they don’t require that generators have secured firm fuel supplies, and they don’t make sure the resources will be ready and available to operate.[0]

The question should squarely be "can renewables create the same base load and peak load required to run the entire demand of the grid"? From there than we can talk about the other topics such as what is required to deliver the energy, which is cheaper to operate, etc.

[0] https://judithcurry.com/2021/02/18/assigning-blame-for-the-b...

> My understanding is that neither source of energy was winterized to transmit the energy to the grid, not that they were unable to produce.

Transmission was not the problem in any of these blackouts that I’m referring to. Powerlines failing affects extremely localized parts of the distribution network, but the coal and gas plants literally stopped producing. It was a production failure.

https://www.texastribune.org/2021/02/16/natural-gas-power-st...

https://www.texastribune.org/2021/02/16/texas-wind-turbines-...

On the other question, that goes back to how we desperately need energy storage and demand response.

FTA:

> "It is possible to “winterize” natural gas power plants, natural gas production and wind turbines, experts said, which prevents such major interruptions in other states with more regular extreme winter weather."

This is really my point. The production sources (and how the elemental source of that energy is acquired) and transmission of energy to residents can be winterized regardless of what they are (wind, gas, etc.). So the point isn't whether wind or gas is better.

> that goes back to how we desperately need energy storage and demand response.

Don't disagree (because this would create a much larger buffer), however if you have fully winterized energy production and transmission, then this point is moot.

I agree with most of what you say, but renewables can’t be counted on, so they can never be blamed. This is tautologically true and yet a meaningless point.

If a hospital loses power to lifesaving equipment, 100% of the time it is due to a failure of backup generators.

> I agree with most of what you say, but renewables can’t be counted on, so they can never be blamed.

Renewables are fairly predictable, so they can be counted on. The production is variable, but not unreliable or unpredictable. It's an important distinction, and it gives more time for the grid to coordinate with Demand Response (or peaker plants) to match load and production.

Obviously the weather models involved are still improving, but for solar especially, it's easy to predict when the sun will go down. Cloud coverage and wind forecasting are active areas of development to make things easier and more predictable for everyone.

> Renewables are fairly predictable, so they can be counted on. The production is variable, but not unreliable or unpredictable. It's an important distinction, and it gives more time for the grid to coordinate with Demand Response (or peaker plants) to match load and production.

Was Germany's months-long period of low wind in 2022 predictable in this sense? Was Germany supposed to stop using electricity for months as Demand Reduction? I buy that you can maybe store enough power for overnight demand if solar generates enough during the day (at some potentially large cost), but wind can just stop blowing for weeks. Storage cannot solve that without outrageous cost, or blackouts.

I can't find any actual accounts of the wind stopping entirely for "weeks" in Germany. It is a theoretical possibility. Periods of "low wind" are fully expected in summer, since there is usually more wind power generated in winter, although I'm not sure when you're referring to specifically. Similarly, solar produces more power in summer and less in winter.

Critically, we have the ability to transmit power over distance. That's the whole point of having a grid, instead of just having each person operating off-grid and only having access to the resources that are within arm's reach of their house. Individual solar panels may be under a cloud or individual wind turbines may experience no wind, but it is tremendously less likely for entire regions to experience this for an extended period of time.

The US has encountered several winter storms that brought a number of our fossil fuel plants to a halt for days at a time, causing blackouts. That's not theoretical.

These are what I'm remembering re: Germany and reduced wind. This was during winter:

https://news.ycombinator.com/item?id=28521533

https://public.substack.com/p/german-emissions-from-electric...

We do have transmission, but it is lossy and expensive, and the places it is cheaper to transmit from are geographically adjacent and are going to have similar solar and wind conditions.

If you're talking about Texas' gas plants, yeah, that was not good. But gas plants are an essential part of preventing blackouts in a wind/solar grid, so I'm not sure this is a point in wind/solar's favor.

Rare "dark calm" backup can be done with gas plants. The gas can be hydrogen. Hydrogen is not cheap, compared to what natural gas typically costs, but for rare event backup it doesn't have to be cheap.

Reply on news.ycombinator.com