Hacker News Reader

Comment by atoav | Hacker News Reader

They will go like...

storing energy is an unsolvable problem

Yeah sure. But if you have an abundance of electrical power you can literally lift up concrete blocks with a crane and generate electricity by letting them down. You can pump water up and use turbines. If you live in a flat area you can do the same in mine shafts. A German guy developed a way of cutting huge granite cylinders out of solid rock that you can lift up and let down producing high pressure waterflows that can be used to store energy. You can split water into hydrogen and oxygen and run turbines off of it, you can charge batteries. You can warm up sand in insulated containers and use the heat to power turbines. There is so many ways to store energy it is hard to enumerate them.

And if you have much more wind and solar power than you need you don't even need to worry that much about the efficiency of the storage system. Sure, you're gonna select the one that you can finance, maintain and find the space for, but a lot of that stuff could be built literally everywhere.

Storage is no argument any serious engineer would pull out. Serious engineers would say that our electrical grids are not capable to deliver that much energy if we try to replace every form of energy with electrical energy — and they would be right. But this is again not an impossible problem to solve. We know how to transmit electricity pretty well. We just need to build the infrastructure.

View on HN

There are numerous ways that you can store some amount of power - batteries, various potential energy storage options, hydrogen and other synthetic fuels, even stuff like flywheels. The unsolved problem is storing power on the vast scale actually needed to make an 100% renewable grid possible. In particular I think the numbers on storing energy by lifting and lowering concrete blocks work out really bad and a long way from being useful as a form of energy storage.

Exactly.

You’d probably be better off assuming no new grid connections tbh. You can effective force a distributed energy network. It just wouldn’t work for cities.

I don’t think a lot of people outside of homesteaders or people working in energy understand.

I can spend $25k for the solar I need for my house, but I’ll be spending another $50k for energy storage. And I’m in a more southern region… if you live somewhere in Europe, particularly the north, you’d need $50k solar and $50k storage.

Btw that $50k only covers batteries for 36-48 hrs of usage. Have 10 days of heavily cloud cover, gotta pull out the diesel engine.

I’ve looked at alternatives energy stores and nothing really has capacity we need. There’s ways to cut energy usage, there’s other forms of energy capture, but battery storage is basically it for single homes.

Fly-wheels and what not might work to stabilize grids for 24-36hrs. But I highly doubt on a week of energy, for instance. So you still have the same issues as single homes. You need the ability to start up engines (so to speak).

Of course there are alternative energy stores which have the capacity we need. What do you think we are using for energy storage now? Now we have huge piles of coal, natural gas stored in huge caverns, massive tank farms storing petroleum products. So the question is can we convert the electrical energy when wind and solar are at maximum output into gases, liquids, solids. The answer is yes. One obvious gas is hydrogen. And if the liquids and solids include carbon then to be carbon neutral we need to pull as much carbon out of the atmosphere as we will put back in (carbon capture technologies).

So it can be done. The problem is the higher cost than fossil fuels.

“Can be done” typically means “can be practically done”

I can fly to the moon, if I had a hundred billion dollars. That doesn’t mean “I” can fly to the moon. These theoretical cases mean nothing.

Funny that you mention a hundred billion dollars:

> Australia's growing hydrogen sector has an investment pipeline of A$133–A$185 billion ($92-$128 billion), equating to 35% of the country's total non-renewable energy and mineral resource investment, state body Geoscience Australia said July 19.

The answer is no, actually. Pulling carbon out of the atmosphere is way harder than you think is is. The carbon in the atmosphere is at a very low concentration. The Sabatier process is presently only done with concentrated carbon dioxide that comes from industrial byproducts. Pulling it from the atmosphere is not efficient enough to be viable.

Even producing the hydrogen is a challenge. Almost all hydrogen is produced via steam reformation, which emits carbon dioxide. Producing it through electrolysis remains difficult because electrodes corrode and lead to high maintenance costs.

If we go the carbon-capture route, it'll almost certainly be via bioengineering, since lifeforms are able to scale themselves up. Something like giant GMO algae farms in the desert.

The big problem with energy storage technology is figuring out which of the many different options will win the competition.

No. It's fine that different technologies are used in different situations. The big problem is finding technologies which are cheaper than fossil fuels.

Not at all. Fossil fuels have to be phased out, so CO2 charges and tariffs will be increased until they are. What renewables + storage have to be cheaper than is the other non-fossil alternatives.

We didn't wait until low air pollution technologies were cheaper than just dumping soot and SOx into the air, we acted to stop the pollution. Because negative externalities are a shining example of where government action benefits society. CO2 also presents large negative externalities. This is why you see climate deniers try to argue there are no such externalities from CO2 emission, because they understand the logic for regulation becomes inescapable otherwise.

No, the problem is that the one that wins the competition is still not viable at grid scale. The competition isn't against other storage systems, the competition is against energy systems that don't require storage. Nuclear was build at around $1-2 billion per GW when it was built at scale during the 1960s and 70s. It's hard for any renewable + storage system to compete against that.

I've explained over and over again to you that you're wrong. Studies such as this one (and others) confirm what I say. I have wonder why, in the face of repeated explanation, you are so resistant to understanding.

Because your "explanations" are incorrect. You just assume storage systems will become multiple orders of magnitude cheaper than they are today, and offer no reason to justify this assumption.

In fact, read what TFA actually says about storage;

> Key pillars of this new energy system are solar and wind energy, energy storage, sector coupling, and electrification of all energy and industry sectors implying power-to-X and hydrogen-to-X solutions, complemented by upcoming carbon dioxide removal.

I chuckled when I read "Power-to-X". It's a pretty funny way of saying "we no idea how to deliver storage at this scale".

No, I do not assume storage systems become multiple orders of magnitude cheaper than they are today. Even the storage systems available at today's price points would allow renewables to be leveled at a price that could be afforded. Rather modest reductions in cost would be preferable, of course, and we can expect those to occur.

Power-to-X is using technologies that are mostly off the shelf. The particularly important one that wasn't, low cost electrolyzers, has seen tremendous progress. The Chinese are now selling these for < $300/kW, a fraction of the price assumed in the 2030 assumptions in that modeling site I like to point to.

Except storage systems today don't scale and become more expensive when deployed at anything close to grid scale. Google told me that lithium ion batteries are $130-150 per KWh. But it turns out when you try to buy loads of them they become scarce and the price goes up. New York ended up paying over $500 for it's storage: https://www.utilitydive.com/news/new-york-battery-storage-co...

Keep stanning "Power-to-X" solutions. But until you actually have a solution for X that works at grid scale, there's no real solution. Just empty promises.

You edited your post after I replied:

> Power-to-X is using technologies that are mostly off the shelf. The particularly important one that wasn't, low cost electrolyzers, has seen tremendous progress. The Chinese are now selling these for < $300/kW, a fraction of the price assumed in the 2030 assumptions in that modeling site I like to point to.

Except none of those systems scale. If they did, the article would have actually specified the storage system. But then they'd have to stand up to scrutiny over whether that storage system can actually deliver at the promised scale. Because it can't, they use ambiguous language like "power to X".

You are repeating the same tired bullshit here. When supply is constrained, greatly increased demand can drive up prices. That doesn't mean storage doesn't scale, it just means we haven't built enough factories yet.

I'm sure you can remember many episodes in computing where the price of something (DRAMs, GPUs, whatever) temporarily went up because of supply constraints. It would have been deeply foolish to proclaim this mean price declines were over.

A similar thing happened with PV back around 2009 or so. The long term decline trend appeared to stall, because of poly-Si shortages. Dishonest skeptics claimed PV cost declines were over. What happened then? The prices provided incentive for more manufacturing capacity to be added, and the decline tracked back to the long term downward trend line.

Not just factories, also the mines that are needed to feed input materials to those factories: https://www.canarymedia.com/articles/batteries/chart-lithium...

You know what we call it when factories and raw materials can't keep up with the demand and lead to skyrocketing costs? A system that doesn't scale.

DRAM and GPUs aren't bottlenecked by raw materials the way energy storage is. It was a shortage of electrical components, not a shortage of copper. It is a very wrong assumption to think that resource extractions behaves like semiconductor manufacturing.

Yes, and all of those things can be scaled up.

I will further say that for your argument to be correct, it must be correct not just for Li-ion batteries, but for all possible storage technologies. You have not, and indeed cannot, make that argument with any level of honesty, since you do not know the bounds of such technologies. This does not stop you from making it though, which is very sad.

No, you can't just "scale up" mining by a factor of 1000. And yes, that's the kind of increase we'll need to make storage systems viable. I don't think you fully appreciate just how massive an undertaking it is to provide the storage required to make intermittent sources viable. It's estimated it'll take 3 weeks of storage to make intermittent sources viable: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

By comparison, the world uses 60TWh of electricity daily.

There is no realistic plan to build this amount of storage. You have not, and indeed cannot, honestly make the argument that it's within our capacity to build storage at such a scale. Hence why you point to untested storage systems like compressed air or hydrogen electrolysis. We don't know that they will scale, since we have not attempted to build them at scale. You take this absence of evidence as proof that they will scale instead of recognizing it for what it is: and absence of evidence since hardly any of these storage systems have been built. This is your pattern of commenting: you concede that the storage systems that we have built at scale don't work. Then you point to the storage systems that we haven't built at scale and assume they will work at scales dwarfing anything we've built so far.

We do not have the capacity to build storage to make intermittent sources viable. You just assume that every new and untested storage system will be super cheap even when it's built by the terawatt hour. And then you call people dishonest when the point out that you're the one being dishonest about our experience with energy storage.

Your cited link, is also cited in the paper.

They said that with only solar, wind, and HVDC but no batteries you could get 80% renewable.

Or, with 12hr of battery but no HDVC you could get 80% renewable.

And, if you used only batteries, you could get the last 20%.

Which, is pretty good? They're not recommending you ignore other technologies, they're just trying to establish rough costs for simplified models.

They seem to think so back in 2018 anyway:

> “The fact that we could get 80 percent of our power from wind and solar alone is really encouraging,” he said. “Five years ago, many people doubted that these resources could account for more than 20 or 30 percent.”

>But beyond the 80 percent mark...Options could include nuclear and hydroelectric power generation, as well as managing demand.”

Here's where it gets cited in the current work (footnote 275):

> Critics of 100% RE systems like to contrast solar and wind with ’firm’ energy sources like nuclear and fossil fuels (often combined with CCS) that bring their own storage. This is the key point made in some already mentioned reactions, such as those by Clack et al. [225], Trainer [226], Heard et al. [227] Jenkins et al. [228], and Caldeira et al. [275], [276]. However, while it is true that keeping a system with variable sources stable is more complex, a range of strategies can be employed that are often ignored or underutilized in critical studies: oversizing solar and wind capacities; strengthening interconnections [68], [82], [132], [143], [277], [278]; demand response [279], [172], e.g. smart electric vehicles charging using delayed charging or delivering energy back to the electricity grid via vehicle-to-grid [181], [280]–[281][282]; storage [40]–[41][42][43], [46], [83], [140], [142], such as stationary batteries; sector coupling [16], [39], [90]–[91][92], [97], [132], [216], e.g. optimizing the interaction between electricity, heat, transport, and industry; power-to-X [39], [106], [134], [176], e.g. producing hydrogen at moments when there is abundant energy; et cetera. Using all these strategies effectively to mitigate variability is where much of the cutting-edge development of 100% RE scenarios takes place.

> With every iteration in the research and with every technological breakthrough in these areas, 100% RE systems become increasingly viable. Even former critics must admit that adding e-fuels through PtX makes 100% RE possible at costs similar to fossil fuels. These critics are still questioning whether 100% RE is the cheapest solution but no longer claim it would be unfeasible or prohibitively expensive.

> Even former critics must admit that adding e-fuels through PtX makes 100% RE possible at costs similar to fossil fuels

E-fuels and "ptx" are ambiguous terms for energy storage. That plan is making the same allusions to storage systems that do not yet exist. What is "X"? Can it actually be built as cheaply as people promise? Until that's answered, this is just wishful thinking.

If you are assuring us storage won't do it, the ambiguity doesn't matter, as it is on your plate to refute every last possibility. So step up and do you job and refute all possible interpretations of those terms. It's your responsibility to do so. You also have to refute all possible combinations of all the elements in the message you are responding to. But I assume you've done that, otherwise your assurance that 100% renewables can't work would be a lie.

No, the burden of proof is one the one saying that X is a viable approach. You're asking people to prove a negative.

Here's a thought: we don't need fission nor renewables nor storage. Fusion will give us unlimited cheap energy! Now you have to refute every possible implementation of fusion energy - even methods we haven't built or tested - otherwise you're lying. Step up and do your job refuting all possible implementations of fusion!

This is the kind of logic you're making with energy storage, and it's faulty logic. If someone wants to claim the viability of fusion, they actually have to deliver a working fusion reactor and at a viable cost. It's not the burden of other people to disprove every possible implementation of fusion.

> Modern 100% RE scenarios often make wide use of power-to-X (PtX) technologies, in particular, power-to-heat [22] and power-to-hydrogen [23]–[24][25][26]. Where direct hydrogen cannot yet be used, such as in the chemical industry or for long-distance marine and aviation transportation, hydrogen can be further converted to synthetic electricity-based fuels (e-fuels) as chemically bound RE and such as e-methane [27], [28], Fischer-Tropsch fuels [29], [30], e-ammonia [31], [32], and e-methanol [33], [34].

This is hydrogen electrolysis and the Sabatier process. There have been attempts to do this, but difficulties continue to prevent it from being deployed at scale. Hydrogen electrolysis is less efficient and maintaining the electeolysis systems expensive.

The Sabatier process (aka power to gas) requires hydrogen as an input so it shares all of the above. It also requires a source of carbon dioxide to fix into hydrocarbons. Usually this is only done opportunistically when industrial processes produce carbon dioxide as a byproduct. The carbon dioxide in the atmosphere is at too low a concentration to be useful. This could be delivered through biofuels but there's not much energy density in them.

If they can actually deliver these systems at viable costs, great. But no one has accomplished this so far.

The main thing that prevents it from being applied at scale is the availability of cheap natural gas unburdened by CO2 charges.

But right now, renewable electrolytic hydrogen would be far cheaper than natural gas in Europe, where NG prices have reached 8x those in the US.

If this were true, then Europe would be building lots of hydrogen electrolysis plants. Get back to me when they're actually built.

Undecided with Matt Ferrell did a video with a mention of costs for some modern techniques a few days ago [1].

Levelized Cost of Storage (LCOS) per MWh:

- CO₂ battery: $50* - Lithium-ion: $131 - $232 - LAES (Liquid Air Energy Storage): $300 - PHS (Pumped Hydro-Energy Storage): $186

*Estimate by Energy Dome.

[1]: https://www.youtube.com/watch?v=GSzh8D8Of0k

First of all, these are almost certainly costs per Kilowatt hour, not megawatt hour.

These costs would skyrocket if the world actually tried to deploy them at grid scale. The entire world only puts out 300 GWh of lithium ion batteries. By comparison, one hour of electricity storage is 2,500 GWh. You'd have to direct all battery Production to storage for 8 years, and pause the production of all EVs and electronics that use batteries. This drove the cost up above $500/KWh in New York for example: https://www.utilitydive.com/news/new-york-battery-storage-co...

Similar story with pumped hydroelectricity. You need the right geography to build pumped hydro. Most of it is in remote inaccessible places (Tibet has huge pumped hydroelectricity potential). The only economically viable sites are the ones that are close to transportation infrastructure. Once those are developed, the cost climbs as more and more remote sites need to be developed.

Liquid air was promising, but the pilot plant took $600/KWh just to build. And who knows what the maintenance costs will be and if it'll live up to it's promised lifespan.

And here I was thinking to have found a promising new YouTube channel. Thanks for the corrections.

> But if you have an abundance of electrical power you can literally lift up concrete blocks with a crane and generate electricity by letting them down.

That doesn't seem like the best idea if you think about it: https://youtu.be/iGGOjD_OtAM

Yeah it’s a terrible idea.

Pumped hydro or compressed air is way better on every front.

Isn't the round trip efficiency of compressed air terrible? You lose a lot of energy as heat when you compress air which you then have to resupply when expanding the compressed air to avoid things freezing over.

There is adiabatic compressed air, where the heat of compression is transfered to a thermal store, cooling the compressed air before it's injected into the storage cavern. Then, on discharge, the heat is recovered from the thermal store and used to reheat the air.

This illustrates, btw, that compressed air doesn't actually storage energy! The energy content of a gas is the kinetic energy of its molecules (and some small contribution from the potential energy of molecules at the moments they bounce off each other, but that's typically minor). The kinetic energy of the molecules is a function of temperature, but not of pressure.

What the air is doing is acting as a store of reduced entropy, which can later be exploited to convert that separate heat back to work at high efficiency. However, one could also do this by PTES, Pumped Thermal Energy Storage, where a reversible thermal cycle produces heat and cold, which are stored separately, then brought back together to recover the work. This is like adiabatic CAES, but instead of storing the cooled compressed air, it is expanded through a turbine and the cold of the resulting expanded gas is stored (say in a tank of mild cryogenic hexane).

Their point isn't that that's the best idea, but that we have lots of things that would actually work, and between them there was never any reason to worry too much about it.

Like you might not know what your going to eat next Friday, you don't know if the cafeteria might sell out of chicken, or if they'll be doing a special on your favourite fish. But that's different from "I will starve next Friday" or "I am allergic to nuts, and there will be nothing but nuts to eat" and it certainly doesn't mean you should put poison in your food today.

It's not physics that the limit, it's economics that suggests green hydrogen will probably be the winner, since as well as storage it has other uses. Similarly for batteries, the dual use tips the balance economically.

> Their point isn't that that's the best idea, but that we have lots of things that would actually work, and between them there was never any reason to worry too much about it.

And the above commenter's point is that it doesn't actually work.

The fact that so many support and energy policy that amounts to "whatever, we'll figure out storage eventually" is astounding. If that's the mentality you're going to take, just let fusion solve all our problems.

I don't like that video, but yes, I don't think that design is the solution for the future. Using water als gravitational storage is quite well developed and has additional benefits (like storing precious water for droughts).

But that is the nice thing going forward: there are many different ways of storing electricity and many of them will be tried. I am sure there will be a lot of new feasible ways of doing so in the market soon.

Energy storage is still hard, unfortunately. I don't think any gravity storage other than pumped hydro (where nature allows for it) is viable.

The question I've been asking myself in case we were to go for battery energy storage: would we be better off using decentralized system with solar + battery on every home / small community or just make massive installations and keep using existing power grid to move energy to homes?

Decentralized doesn't work so well for places that require a high ratio of electrical power to solar radiance - like refineries, factories, medium and high density housing, electric vehical fast chargers, low density housing in places that get cold in winter, and more. So a large fraction of society will need to import some energy from somewhere (either from the grid or a fuel like hydrogen).

Fully decentralised solar + battery might be the best option for rural / semi-rural settings, but even low-density suburbia somewhere sunny could likely benefit from having all those batteries and panels working together via the existing grid.

> Decentralized doesn't work so well for places that require a high ratio of electrical power to solar radiance

Yeah, another example that comes to mind are flats - the roof is a bit small to cover all residents' energy needs.

As always in life it seems that "it depends" but I think you are right about remote places. It is of course always better to have electrical grid since you shift the maintenance to someone else (as long as the power delivered is reliable and cost effective).

The one issue I can think of is current grid may not be able to handle many small energy producers - they can be cut off when peak power is a problem. In those cases individual prosumers would benefit from having a local battery storage system that fit their needs, even when connected to the grid.

Remember the goal to electrify everything. For the home owner this includes space heating. So if you live in the snowbelt you need to have the battery capacity to handle a two week Polar Vortex with temperatures below zero degrees Fahrenheit with very little sun with your fully decentralized solar system. This will require a very expensive battery system!

Even if in theory local storage us cheaper, most people won't want to invest in replacing their system every 10 years when it wears out.

I think we will see mostly utility scale storage. Most people will have an electric car that they can plug into, but that will only be done to keep the fridge cold when the utility goes out.

10 years seems like a short time, what if we make it 30 years - same as solar? You would need to update it roughly once during your lifetime.

The other idea is having small communities that manage their own energy needs. Meaning 1MWh energy bank for 100 homes instead of 1GWh for 100 000 homes.

Battery life doesn't always work the way you want it to.

I don't know what the limits of different battery chemistry is, but I suspect it won't work. Though 10 years was just a number I made up, I think it is reasonable for discussion.

Yeah, I think you might be right although AFAIK they still hold on to about 80% of their initial charge after that time (10y), which is quite a lot. 30 years was just a random thought that seemed like a sensible minimum to be somewhat low-maintenance solution.

I am just trying to think about different possibilities, they pros, cons and what might be net win in this case. Your remark on people not wanting to service yet another thing is spot on, grid always works for them.

I think it will be a mix of both. Local storage at home is great for some grid independance during blackouts and usually you save some money when you have you own solar roof.

But local storage usually will be more expensive per storage capacity vs. larger storage facilities. This means, a large amount of storage will be there, because it is cheaper. Also the power company can control it better than home systems.

And of course there the electric cars which could partially contribute to the grid storage.

On land you just can’t make it tall enough, you get roughly 2.5kWh of potential energy per km of delta-h per ton of mass. I think if you store energy in a gravity battery out in the ocean where the depths are 10km, you may end up with a system competitive with pumped hydro.

The incremental cost of a ton of mass and a ton of buoyancy are probably gonna be a factor of 2 to 10 less than lithium batteries. For any storage project: if it needs $1B of infrastructure costs for wires, dynamos, robots etc. just to start then you must be talking about building at least 10GWh before it is practical. so probably millions of tons for an oceanic gravity battery

Storage via gravitational potential energy is extraordinarily inefficient.

Let's say you have a medium city, like Columbus, OH. This city purchases about 890 GWh of electricity per year [1] for costumers and street lights. Let's say they need to store 100 GWh. A kg of matter raised 1 meter stores about 10 joules, or watt-seconds. We need to store 100 times 3600 billion of watt-seconds, which is 360 trillion. You can trade mass for height. For example you can raise 36 trillion kg to a height of 1 meter. That's 36 billion tons or gigatons (GT). Or you can raise 3.6 GT to a height of 10 meters, just 36 MT to a 1km hight.

Let's say you want to use blocks of stone. The total US production of blocks of stone in 2021 was 2.3 million tons [2]. You need 15 times as much to only cover Columbus, OH, and only assuming a 100% efficiency in the conversion of the energy from electricity to gravitational potential energy and back.

The only substance that can be useful for this type of storage is water. But even for water, you need the same huge quantities. 36 million tons of water is a lake with a surface area of 1 square kilometer and a depth of 36 meters. If you have some mountains around you, then you can do it. But Columbus, OH is out of luck. It does not have mountains, or even hills around it. Building an artificial hill that's 1000 meters tall, that has a 1 km2 lake at the top with a depth of 36 meters is a monumental undertaking, that was never attempted.

Of course, you can try to play around with the tradeoffs. Maybe you build a lower hill with a larger footprint. But how long is it going to take to offset the emissions from such a huge construction project?

Bottom line: pumped hydro makes sense, if geography allows you. If not, it is uneconomical by many orders of magnitude.

[1] https://www.columbus.gov/greenpower/ [2] https://pubs.usgs.gov/periodicals/mcs2022/mcs2022-stone-dime...

> Storage via gravitational potential energy is extraordinarily inefficient.

I think the point is that what we do right now and what has been considered good practice is extremely inefficient, engineering-wise and economically.

This city purchases about 890 GWh of electricity per year [1] for costumers and street lights. Let's say they need to store 100 GWh.

Why do they need to store over 1/9th of their yearly energy usage?

If you switch to renewables, you need to store energy.

I’m questioning the amount (which is what the whole argument about this being impractical is based on).

Do you really need to have a store of 11% of the city’s entire yearly energy usage? Like, 5 (almost 6!) entire weeks of energy? That seems ludicrously high to me.

Ok, let's say it's one week. Is that still ludicrously high? The scale of the challenge is the same. Storage via gravitational potential energy is not realistic, except if geography is on your side.

I think an entire week’s worth also seems high.

But, yeah, on the whole you’ve convinced me. This does seem like a ludicrously inefficient form of energy storage.

That was my point, that gravitational storage is surprisingly inefficient.

But since you challenged me on the 11%, I did some research, and now I think that storage may not even be needed (aside from the day-to-night storage via batteries, which is already economical nowadays).

The EIA gives [1] the US electricity generation by month and source for 2020 and 2021. Solar and wind are small nowadays, but if you were to scale to cover the entire annual needs of the US only with solar and wind, you'd get a profile where on some months you get more than you need and on some months you get less. The longest stretch is in the Summer months (presumably because of AC). During those months, for 3 or 4 months in a row you produce less than you consume. You need about 6% or 7% of total annual consumption in storage to make up for the deficit. If we include conversion losses, and an additional buffer, then my original 11% sounds about right (although I just pulled it out of thin air).

However, if your just use natural gas for those 6-7%, then you don't need storage at all. Cutting the fossil fuel generation from the current 60% to 6% means reducing emissions by more than 90% (lost of current generation is from coal, which emits about double the CO2 than the equivalent gas generation).

You can also overbuild solar and wind a bit, let's say by 20%, and the deficit gets reduced from 6-7% to less than 3%. You can then balance the load using the hydropower that we already have.

So, I think the whole storage problem may not be such a big problem after all.

[1] https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...

Lifting heavy things up is cool but expensive. Batteries are way more compact but don't last too long.

It's a hard problem. Hopefully one of those magic battery tech from the past will be ready at some point.

On the contrary, the difficulty of switching to full electricity is very country dependant: some countries rely heavily on gas, some use electricity for everything. I didn't have gas in any of my houses for the last 10+ years.

On the automotive front, you already have the market and the high price of EV to make sure EV adoption will be gradual. I'm sure over time we can adapt to the increased demand with incremental improvements.

Do you know what's up with molten salt batteries? I've heard GE was big on it but then after investing a ton of money they just closed it down.

Is the technology not viable? Seemed to me like a green solution that is working ok and can be used now.

This is a useful chart (though it would be nice if efficiency was included).

https://en.wikipedia.org/wiki/Hydrogen_storage#/media/File:A...

The world uses 60 TWh of electricity daily. About twice that in terms of total energy use. Even just one hour of energy storage is a massive undertaking. Estimates to reach a 100% renewable grid call for 3 weeks of storage: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...

No solution we have can deliver even remotely close to that amount of storage. Sure, you can build giant flywheels or pulleys or what have you. But can you build them at the required scale? What is the rate at which you can build them?

> Sure, you're gonna select the one that you can finance, maintain and find the space for

My point is that this is way, way harder than you seem to think.

I wonder what the FUD line will become once solar plants start coming with 2 or 3 hours of storage because it optimized the ROI of the inverting and transforming equipment. (This shouldn't take too long, since the battery costs are almost there already.)

Already happening in the US:

https://www.rechargenews.com/energy-transition/us-energy-sto...

> The research house expects 9.4GW of energy storage capacity to come online this year and a further 15.6GW in 2023 ,with nearly 60% to be co-located with other – chiefly solar – power plants.

> Developers added 1.33GW of energy storage capacity in the fourth quarter of 2021, exceeding the previous two years combined, and energised more than 3GW last year.

Reply on news.ycombinator.com