In Australia, the real value of these big batteries is not necessarily to provide emergency power, but more to respond in milliseconds to variations in demand and to smooth out the 5 minute spot market.
Having a sink and a source for spikes aids grid stability and somewhat helps to reduce the need to build more infrastructure to handle transmitting power from the generators during those peaks.
It also helps avoid the viability of gas powered generators which are very expensive and carbon intensive.
I don't think it's a coincidence that China has developed batteries that neuter the advantage that gas had over coal, but I can't stop wondering if we'd already have solved long-term storage if China hated coal as much as oil and gas.
China can't hate coal (yet). In case of an armed conflict with the West, they will be cut off completely from oil and gas. Renewables don't cover all their energy needs. So they are building a lot of coal plants as backup.
China is the world's 4th largest oil & gas producer. They're dependent on imports because they're the 2nd largest importer. However, they are simultaneously increasing production and decreasing consumption. Those two lines will cross some day. It might be a dangerous day for the rest of the world when China stops becoming dependent on imported oil.
> It might be a dangerous day for the rest of the world when China stops becoming dependent on imported oil.
Why? Would China just try to start a global war? If it's about Taiwan, I'm not sure that USA would care like 5 years ago. USA will probably do the same with South America
China built strategic petroleum reserves that can operate for over a year so that if the US attempted to blockade oil supplies from the middle east during a conflict, China could press on when they decide to obtain Taiwan.
China coal use has dropped below 50% [1], and they currently curtail renewable generation roughly total annual generation of Spain and the Philippines combined every six months [2] (~401 terawatt-hours of power). They are rapidly deploying battery storage to accelerate electrification and making use of energy that would otherwise be curtailed, which should continue to drive down coal use (their electricity markets are currently reconfiguring to use market signals and incentives to encourage this outcome).
They will, in the next five years (at current renewables and battery storage deployment rates [60% of their energy currently comes from renewables, and this increases 10% YoY]), become a low carbon (hard to say net zero yet) electrostate [3] that will continue to scale manufacturing clean tech for the world [4] with minimal carbon footprint.
- "As the power system integrates more renewables, the true value of batteries to China’s energy transition is becoming undeniable," Yang said. China added a record amount of new battery storage facilities in 2025, ending the year with more than half of global capacity. Growth has continued this year, with installed lithium-ion battery storage reaching almost 150 gigawatts by the end of the first quarter, according to Ember. The country is targeting 300 gigawatts of battery storage capacity by 2030, up from 136 at the end of 2025, in addition to expanding its fleet of longer-duration pumped hydro storage to around 160 gigawatts from 66.
Can they procure their own uranium?
Coal still has the advantage of being more local, much simpler, faster to get up and running. Smaller plants also means smaller distribution radius etc. so more infrastructure decentralization. I imagine that's part of the thought process too.
In the case of armed conflict with the West all Chinese inverters will be weaponized to burn homes in the West. The shops will be empty in the West and West will be in very ugly situation outsourcing all the manufacturing. Better it doesn’t happen.
To put things into perspective, China gets half its energy from coal, and about 10% from natgas (increasing to 15% by 2030) and about 20% from oil. China is rapidly increasing natgas use, primarily by building out new pipelines to get more natgas from Russia.
China is also increasing oil consumption as well as exploration, particularly starting with the 14th 5 year plan as well as a special seven year action plan to increase oil production, doing deep water drilling, unconventional (e.g. fracking) drilling as well as a massive shale oil boom. Over the course of that 5 year plan, China discovered more than 225 oil fields and oil production increased from 3.8 million barrels per day to 4.6 mbpd as a result of a ~500 billion CNY investment to increase production. Still China imports about 10 mbpd, so it has developed a massive 1.5 billion barrel strategic reserve, which allows China to weather disruptions in global oil supply even as it races to increase domestic production.
And China is also investing in battery capacity, with installed capacity now at 140 GW or 5% of their energy usage.
The statement that somehow "batteries neuter the advantage that gas has over coal" is not something I understand. Both coal and natgas are reliable forms of energy and are unaffected by battery capacity; you need batteries for unreliable energy such as wind or solar so that you can turn the variable output into something stable that a grid can consume.
> The statement that somehow "batteries neuter the advantage that gas has over coal" is not something I understand.
In thermal electricity plants it takes a lot longer to go from cold & dark to generating electricity with coal than with either Combined-Cycle gas or especially the Open-Cycle "peaker" gas plants. Both types of gas plant also AIUI have a lot better ability to "throttle" their power output.
This means if you have a bunch of CCGT like Britain, you can use those directly to "top-up" the power from your fleet of solar PV and wind turbines, whether that's 1GW of top-up or 20GW of top-up. The change from weather (sun goes down, wind picks up, clouds appear, etc.) is relatively slow compared to what a gas plant can do, whereas it's too fast for coal to keep up as I understand it.
Coal by it’s nature is much less amenable (from a physics/chemistry) perspective to automation and high resolution control. Less pure/refined, granular instead of a gas, harder to start and burns longer once started, etc.
Contaminants in the coal (due to it being a solid mineral) also means it burns with a lot often toxic ash that is hard to deal with.
On the plus side, you can often literally use an excavator to dig it up and put it in the back of a train and then burn it with minimal other capital equipment.
That is ancillary services. And where all battery implementations started before saturating those marketss.
We've now past that and batteries are actively killing the business case for fossil gas peakers while renewables are doing the same for legacy thermal producers.
I am still a fan of centrifugal systems. The lifecycle costs of a spinning mass seem more manageable than the complexities of chemical batteries. The mass of a block of concrete and steal isn't going to degrade over time. And bearings/motors are such mature tech that they can last decades, even a century without replacement.
Genuine question - what is the thought process behind this comment?
Because we both know that gas generators also require a production step... arguably one that's quite a bit more complex than a battery, at ~$1,000,000/MW of installation costs. And that installation of that complex is going to be very carbon intensive (even the concrete foundation alone...).
And then on top the plant actively burns hydrocarbons for the entirety of its existence.
So fill me in here - what is the exact argument you are trying to make with this?
The good 'ole "Lets start with the premise that all energy in our calculation comes from coal transported by truck, and then work out how "green" these technologies are"
that's a one time cost over many years that's quickly amortized. lets assume an upfront cost of ~100kg CO2 per kWh of battery storage and 250g of CO2 per kWh generated by a gas turbine. this means the carbon ROI is there after 400 kWh.
The extraction cost of minerals is only paid once. And a few big batteries will be much easier to recycle than a billion small ones in consumer electronics.
I'm glad they're reporting it as the energy storage capacity and not the power output capacity. Too many news sites like to report the power output capacity in stories like these, which is horribly misleading, and then often they also frame it as though this is equivalent to adding actual power generation to the grid, which is even more misleading.
One reason they report in MW instead of MWh, is that grid operators primarily spec systems by their continuous MW, and only secondarily mention their cumulative MWh. So that's the way they talk about them, and reporters just mimic that. Both are important, but the top-line number is MW, for sure.
One reason for this is that for a lot of grid-support functions, instantaneous or short-term power is vastly more important than long-term energy capacity. The other reason is that traditional generation is only sized by MW; you just keep pouring fuel into it forever, so the notion of MWh doesn't even enter the picture. They're not even thinking in those terms.
You and I, as battery natives, think about it the other way around. But we're not grid operators.
Yup! Some breaker trips, and a 500MW plant goes offline. How much battery storage do you need to compensate? 500MW. The fact that it stores 500MWh and therefore only lasts an hour comes afterwards, that is plenty of time to ask all the other fossil power plants to scale up a bit.
It's probably also a view that is shifting as batteries move from primarily providing short-term grid stability (where power available is probably the more important metric) to longer term storage.
Having worked in the industry for a bit, the reason people talk about "MW of batteries" is because they're usually designed to deliver the capacity for a fixed number of hours. Earlier batteries would do 1 hour (so a battery would typically be 1MW/1MWh) and newer sites coming online are usually 2 hours (so 1MW/2MWh). As battery deployment grows the money is increasingly just in energy arbitrage, so there's a lot of pressure for longer duration storage. If you see a headline MW figure it's a safe bet that it's between 1-2x that in MWh.
I can see why one might intuitively think this, but actually this is just a bad take. For most practical purposes, the main value added by these batteries is not really the size of their storage capacity (afterall, their capacity is tiny relative to the grid's needs), but their ability to quickly sink or supply massive amounts of power.
If these batteries had 10x the storage capacity, but 1/10th the power output, they'd miss out on their biggest advantage, which is capturing mid-day solar oversupply, and providing virtual grid inertia.
Long duration, lower-power batteries will have an important role to play once grids get further along in renewable adoption, but for now, short duration power-supply is the main value they add, and it makes perfect sense to quote battery numbers in terms of power. (not that I'd complain about providing capacity numbers too)
Politically force datacenters to consume only solar and battery. Instead of total collapse, we now have tremendous amounts of cheap energy when the bubble implodes. That seems like the only way to hedge the bet on AI.
Makes sense. With battery technology improving and being the cheapest it's ever been, why not? This is probably the direction most countries will go in the future.
One thing that would really help batteries is permitting reform. The BAAJA reform bill would really help not just the batteries themselves but the transmission lines which are far more difficult to construct thank to an insane system and lots of lawsuits.
Obviously there are a variety of sizes and configurations, but if you assumed 3.5MWh per 20ft container, that's 5771 containers, and if you assumed a PCS container per two battery container, that's 8656 containers, just for an idea of the scale.
I've always wondered why they don't stack them in deployment -- the containers are obviously fine with it! -- but I guess land is cheap enough not to bother.
Horizontally they need plenty of room for airflow, as each container has some HVAC-or-whatever module on the side. But vertically...
So if ALL power generation of the US would stop right now, EVERYTHING would still have power for 25 minutes. (ignoring all safeties tripping, and perfect load balancing)
No. Because neither storage nor generation is evenly spread. In some places you would have nothing, in some you could have potentially hours.
In practical terms, you would most probably get nothing, because situation like this would lead to complete grid breakdown and interruption of service.
Technical detail question: if all of those batteries were on connecting allowing feeding back to the grid, would they be able to synchronise the frequency/phase? What's the threshold/requirement for that to happen?
It can be done, but it would requires additional equipment. Batteries are traditionally grid-following, so they'll just follow whatever frequency the grid is already running on. The frequency itself is then provided by the rotational inertia of "spinning metal" power plants.
But there's no reason that has to be your frequency source. Any single battery installation on its own is perfectly capable of making up some frequency, which is why home power banks with island-mode AC output are a thing - or even just a traditional UPS. Doing the same with thousands of individual battery installations across the grid? You'll need some way to get everyone to agree exactly on frequency and phase, which in practice is probably going to mean getting them to agree on time to within a few microseconds, which could be done (albeit not very sabotage-proof) using GPS.
> The frequency itself is then provided by the rotational inertia of "spinning metal" power plants.
This might or might not be true. I used to live near a large plant that took HVDC from a power plant several states away and transformed it into AC. They used electronics to provide the grid frequency, not spinning metal. All the other local peak power generation was run in grid following mode.
> What's the threshold/requirement for that to happen?
The RoCoF (rate of change of frequency) must be low enough that the inverters can form the grid without letting out the magical smoke. Having inertia is what slows this rate down.
Considering the accelerating rate of installation and the ever decreasing costs of manufacturing, both problems should be alleviated in 10 years. Instead of 25 minutes, most likely 4-6 hours. Instead of extremely uneven distribution, likely all the major population centers will have at least some non zero coverage.
Only if that battery was connected to 500GW of inverter capacity (and the grid was a spherical chicken -- enough intertia-replacement devices to keep timing working, enough capacity to link supply with demand etc)
Grid batteries was one of the few areas that escaped Trump's gutting of the IRA and was included in other Infrastructure bills.
And a side effect of the EV market getting hobbled also redirected battery factory capacity from that market.
I'm not sure what's online yet but there is a pipeline.
edit an estimate from earlier this year:
> By the end of 2025, 20 gigawatt-hours of dedicated storage cell lines had opened, and the industry is on pace to hit 96 gigawatt-hours by the end of this year.
> I was snobby and looked up the numbers because I didn't think this is relevant. But it is.
Kudos to you for bein' "that guy". So many folks sit and comment while having the world's biggest library of information on just about every topic right there at their fingertips and still can't be bothered to take the 30 seconds to look up something they don't know about on their favorite search engine. I appreciate you for havin' done "the right thing" here. :)
- The utility-scale market continued underpin growth, with 18 GWh installed, largely supported by seven gigawatt scale projects. Arizona installed over 6.2 GWh of utility scale projects, the largest quarterly deployment for any state in history.
- In the BTM space, residential storage added 657 MWh and commercial and industrial (C&I) added 1.8 GWh, largely on the back of demand from data centers.
I wonder how much medium- to long-term storage has been installed this year. Batteries are fantastic, but we need better ways to store excess wind and solar for dark winter days when there's no sun or wind.
It's nice to see more storage being added, but we have keep thr order of magnitude in mind: the peak electricity demand for the US is 750GW, so the added storage for the past 3 month corresponds to just 96 seconds of peak consumption.
But we generally don't need to satisfy the peak with battery - peak US demand occurs while the sun is still up, in July, so you can satisfy a large chunk of that with solar still. The hard part is those ~4 hours after solar stops producing enough and it's still hot and people are still up. So what we need is several hours of about half the gap between {nuclear, wind, hydro} and peak. Remembering that we also use even non-pumped hydro as a form of storage, by drawing on it more in the periods just before and just after sunrise/sunset.
(Using that b/c I run the bot and it's data I'm very familiar with. You can see similar patterns in CAISO but they've already got so much solar it's a little more messy/duck-curved by all of the behind-the-meter solar.)
You can see hydro coming online from roughly 16:00 - 23:00. There's still a big gas peak that we'd need a lot of storage to shave, and you're right that it's a long way to go before we do that, but it's not quite as bad as "we need enough battery to satisfy 100% peak demand".
There's likely some spare solar capacity that will help. Currently the peak of solar is almost counterproductive, but converting that to storage is good for both the consumers and the power plants. (Less feed in during the day, lower evening peak)
As far as I know, we have yet to figure out how to capture power from a lightning strike. There is a lot of energy there, but it's static electricity which is completely different from what any grid runs on.
That means very high peak power, making your whole capture infrastructure extremely expensive, but even capturing 100% of an average lightning bolt is worth <$100 in electricity.
yea, i think the low efficiency (plus how to predict where lightning would hit!). Might as well just put up a wind turbine, since wind is much more predictable, not to mention easier to harvest.
The dam only runs about 33% of the time so 8 hours a day and the batteries run nearly every day so just 3 months of additions is nearing half a grand coulee dam of electricity output. Or two Coulee dams a year in grid battery production.
The existing batteries in Texas and California have already delivered a coulee-dam-year (21TWh) so far in 2026.
Building out batteries and solar farms. Balancing the need for peak load demands against the steady state of steam generators. Racing to get nuclear going before running out of space for solar panels.
Is this real or are we living in a Factorio livestream?
Having a sink and a source for spikes aids grid stability and somewhat helps to reduce the need to build more infrastructure to handle transmitting power from the generators during those peaks.
It also helps avoid the viability of gas powered generators which are very expensive and carbon intensive.
China is the world's 4th largest oil & gas producer. They're dependent on imports because they're the 2nd largest importer. However, they are simultaneously increasing production and decreasing consumption. Those two lines will cross some day. It might be a dangerous day for the rest of the world when China stops becoming dependent on imported oil.
Why? Would China just try to start a global war? If it's about Taiwan, I'm not sure that USA would care like 5 years ago. USA will probably do the same with South America
They will, in the next five years (at current renewables and battery storage deployment rates [60% of their energy currently comes from renewables, and this increases 10% YoY]), become a low carbon (hard to say net zero yet) electrostate [3] that will continue to scale manufacturing clean tech for the world [4] with minimal carbon footprint.
It cannot be overstated how fast they are going.
[1] China Crosses Energy Milestone as Coal Falls Below 50% of Power - https://news.ycombinator.com/item?id=49110187 - July 2026
[2] https://news.ycombinator.com/item?id=50132343 (citations)
- "As the power system integrates more renewables, the true value of batteries to China’s energy transition is becoming undeniable," Yang said. China added a record amount of new battery storage facilities in 2025, ending the year with more than half of global capacity. Growth has continued this year, with installed lithium-ion battery storage reaching almost 150 gigawatts by the end of the first quarter, according to Ember. The country is targeting 300 gigawatts of battery storage capacity by 2030, up from 136 at the end of 2025, in addition to expanding its fleet of longer-duration pumped hydro storage to around 160 gigawatts from 66.
[3] https://hn.algolia.com/?q=china+electrostate
[4] https://ember-energy.org/data/china-cleantech-exports-data-e...
To put things into perspective, China gets half its energy from coal, and about 10% from natgas (increasing to 15% by 2030) and about 20% from oil. China is rapidly increasing natgas use, primarily by building out new pipelines to get more natgas from Russia.
China is also increasing oil consumption as well as exploration, particularly starting with the 14th 5 year plan as well as a special seven year action plan to increase oil production, doing deep water drilling, unconventional (e.g. fracking) drilling as well as a massive shale oil boom. Over the course of that 5 year plan, China discovered more than 225 oil fields and oil production increased from 3.8 million barrels per day to 4.6 mbpd as a result of a ~500 billion CNY investment to increase production. Still China imports about 10 mbpd, so it has developed a massive 1.5 billion barrel strategic reserve, which allows China to weather disruptions in global oil supply even as it races to increase domestic production.
And China is also investing in battery capacity, with installed capacity now at 140 GW or 5% of their energy usage.
The statement that somehow "batteries neuter the advantage that gas has over coal" is not something I understand. Both coal and natgas are reliable forms of energy and are unaffected by battery capacity; you need batteries for unreliable energy such as wind or solar so that you can turn the variable output into something stable that a grid can consume.
In thermal electricity plants it takes a lot longer to go from cold & dark to generating electricity with coal than with either Combined-Cycle gas or especially the Open-Cycle "peaker" gas plants. Both types of gas plant also AIUI have a lot better ability to "throttle" their power output.
This means if you have a bunch of CCGT like Britain, you can use those directly to "top-up" the power from your fleet of solar PV and wind turbines, whether that's 1GW of top-up or 20GW of top-up. The change from weather (sun goes down, wind picks up, clouds appear, etc.) is relatively slow compared to what a gas plant can do, whereas it's too fast for coal to keep up as I understand it.
Contaminants in the coal (due to it being a solid mineral) also means it burns with a lot often toxic ash that is hard to deal with.
On the plus side, you can often literally use an excavator to dig it up and put it in the back of a train and then burn it with minimal other capital equipment.
It all makes sense.
We've now past that and batteries are actively killing the business case for fossil gas peakers while renewables are doing the same for legacy thermal producers.
(From a guy who got burned buying stock in a centrifugal system as my first "wicked smaht" investment as a 22yo).
Because we both know that gas generators also require a production step... arguably one that's quite a bit more complex than a battery, at ~$1,000,000/MW of installation costs. And that installation of that complex is going to be very carbon intensive (even the concrete foundation alone...).
And then on top the plant actively burns hydrocarbons for the entirety of its existence.
So fill me in here - what is the exact argument you are trying to make with this?
One reason for this is that for a lot of grid-support functions, instantaneous or short-term power is vastly more important than long-term energy capacity. The other reason is that traditional generation is only sized by MW; you just keep pouring fuel into it forever, so the notion of MWh doesn't even enter the picture. They're not even thinking in those terms.
You and I, as battery natives, think about it the other way around. But we're not grid operators.
If these batteries had 10x the storage capacity, but 1/10th the power output, they'd miss out on their biggest advantage, which is capturing mid-day solar oversupply, and providing virtual grid inertia.
Long duration, lower-power batteries will have an important role to play once grids get further along in renewable adoption, but for now, short duration power-supply is the main value they add, and it makes perfect sense to quote battery numbers in terms of power. (not that I'd complain about providing capacity numbers too)
Diesel is the future.
Horizontally they need plenty of room for airflow, as each container has some HVAC-or-whatever module on the side. But vertically...
Land is cheap relative to a container full of cells.
That would be a lot harder if you stack them since heat likes to rise.
But no clue if that is the actual reason.
It'd just be impressive to see a giant wall of battery. A bunch of little batteries camping in a field doesn't play the same visually. Ah well.
So, this adds over two minutes of average usage.
https://seia.org/research-resources/energy-storage-market-ou... says total storage now is about 200 GWh, or about 25 minutes of average demand.
That sounds so impressive.
In practical terms, you would most probably get nothing, because situation like this would lead to complete grid breakdown and interruption of service.
But there's no reason that has to be your frequency source. Any single battery installation on its own is perfectly capable of making up some frequency, which is why home power banks with island-mode AC output are a thing - or even just a traditional UPS. Doing the same with thousands of individual battery installations across the grid? You'll need some way to get everyone to agree exactly on frequency and phase, which in practice is probably going to mean getting them to agree on time to within a few microseconds, which could be done (albeit not very sabotage-proof) using GPS.
This might or might not be true. I used to live near a large plant that took HVDC from a power plant several states away and transformed it into AC. They used electronics to provide the grid frequency, not spinning metal. All the other local peak power generation was run in grid following mode.
The RoCoF (rate of change of frequency) must be low enough that the inverters can form the grid without letting out the magical smoke. Having inertia is what slows this rate down.
"How much is needed during a no sun and no wind period" could be reasonable comparison.
There are big factories in both Michigan and Kentucky Ford operates it just took over from Blueoval SK.
They retooled Ford plants are now making grid batteries instead of car batteries is my understanding.
Tesla also has battery factories in the USA as well as a lithium refinery.
Batteries are complex but less complex than semiconductors. They are rapidly dropping in cost too.
And a side effect of the EV market getting hobbled also redirected battery factory capacity from that market.
I'm not sure what's online yet but there is a pipeline.
edit an estimate from earlier this year:
> By the end of 2025, 20 gigawatt-hours of dedicated storage cell lines had opened, and the industry is on pace to hit 96 gigawatt-hours by the end of this year.
Space-Tesla installed 3 GWh in 2026 alone.
Kudos to you for bein' "that guy". So many folks sit and comment while having the world's biggest library of information on just about every topic right there at their fingertips and still can't be bothered to take the 30 seconds to look up something they don't know about on their favorite search engine. I appreciate you for havin' done "the right thing" here. :)
- The utility-scale market continued underpin growth, with 18 GWh installed, largely supported by seven gigawatt scale projects. Arizona installed over 6.2 GWh of utility scale projects, the largest quarterly deployment for any state in history.
- In the BTM space, residential storage added 657 MWh and commercial and industrial (C&I) added 1.8 GWh, largely on the back of demand from data centers.
⇒ it includes data centers
Everything using batteries will benefit from it.
https://cleantechnica.com/2026/07/16/chinas-batteries-move-b...
There's still a very long way to go !
For example, if you look at the energy production across PJM from two days ago: https://cdn.bsky.app/img/feed_fullsize/plain/did:plc:hiovtnl...
(Using that b/c I run the bot and it's data I'm very familiar with. You can see similar patterns in CAISO but they've already got so much solar it's a little more messy/duck-curved by all of the behind-the-meter solar.)
You can see hydro coming online from roughly 16:00 - 23:00. There's still a big gas peak that we'd need a lot of storage to shave, and you're right that it's a long way to go before we do that, but it's not quite as bad as "we need enough battery to satisfy 100% peak demand".
But, like with UPS, 96 seconds is ample time to let delicate systems glide to a standstill or to start the diesel generator.
That means very high peak power, making your whole capture infrastructure extremely expensive, but even capturing 100% of an average lightning bolt is worth <$100 in electricity.
Which is why nobody even attempts this.
https://en.wikipedia.org/wiki/Grand_Coulee_Dam
The existing batteries in Texas and California have already delivered a coulee-dam-year (21TWh) so far in 2026.
Is this real or are we living in a Factorio livestream?
But a more typical size would be around 200MW, which could run over a week on 20GWh.