I had exact the same discussion here 5 and 10 years ago (it will be ready next year!), I'm willing to bet that storage isn't going to cut it in the next 10 years.
There's at least two of orders of magnitude missing with the current storage gen and unless a new tech revolution happens, that's not going to work.
The supposedly massive storage which is built in your link doesn't even cover half a day of winter load.
Anything below 200GWh is a proof of concept at best.
I remember those discussions. 5-10 years ago people were summing all electric cars, including ICE cars starting batteries, to prove the scale was irrelevant.
Now the goalpost is shifted to "not even a single winter day without any other input of electricity". Which is a high 90s percent decarbonized grid. Not fully decarbonized, but almost.
In California storage is now timeshifting 50 GWh daily. An expansion that has come in the last few years.
Battery prices are down to $50/kWh when not using extremely expensive western batteries. Which means in the near future 50-200 kWh systems attached to houses. Excluding the BEV providing demand response to also help shape grid demand.
I think you should update your priors to 2026 data. We're in the point of the S-curve where batteries goes from nowhere to everywhere in the blink of an eye.
Just like solar was almost insignificant in 2020 adding a mere 140 GW over the year while in the first 6 months of 2025 we added 380 GW of solar.
Storage is time-shifting a lot of energy but it can't do long periods of time in large volumes - still a significant problem. What you are referring to is small incremental changes (like peak shaving or demand response programs in single hour increments). And it isn't only a co-ordination problem (which many companies are solving) its a physical/physics problem.
I am all for energy storage and solar - I've worked extensively in both. Their continued growth is a huge asset for humanity. That said they aren't a panacea and doesn't cover the full spectrum of energy needs even with continued cost reductions - they have constraints due to the physical reality of the world and how power is produced.
California has huge power cuts and has the benefit of having a very mild winter on top of that (can we even call a 10 degrees celcius minimum as winter? Personally I wouldn't).I wouldn't use them as a successful example of anything.
> In California storage is now timeshifting 50 GWh daily. An expansion that has come in the last few years.
I don't think you realize the scale of the problem, France alone consume 90GWh per day in winter, yes one day. And that isn't going to be any better with all petrol consumption switching to electric.
50 GWh shifting is just a proof of concept at best.
And yeah sure, see you in 2027, for sure it will be the year of storage this time.
The storage and grid upgrades have essentially removed all power outages in California? Again you operate on old information.
On the other hand California do have an absolutely massive air conditioning load in the summer.
> And that isn't going to be any better with all petrol consumption switching to electric.
Electrifying transportation is expected to add 15-25% extra load. A load that is extremely flexible in when it runs and thus perfectly match renewable intermittency.
In 2025 alone China added 168 GWh of storage.
I think you don’t realize how much even 50 GWh of storage causes the entire Californian grid to transform.
> On the other hand California do have an absolutely massive air conditioning load in the summer.
Well good for them but the vast majority of the western world has the opposite load, reduced load in summer where the panels operate full capacity and massive consumption when they produce close to nothing.
> In 2025 alone China added 168 GWh of storage.
Great, and they use over 1400 GWh per day.
In 2025 alone, it means they added an astonishing 3h of electricity storage (I'm rounding it up for you as a bonus)
> In 2025 alone, it means they added an astonishing 3h of electricity storage (I'm rounding it up for you as a bonus)
That’s really good isn’t it?
It would be unusual for solar to produce zero during the day, and the night is presumably going to be around 12 hours (in terms of solar generation). Energy usage is presumably less at night too.
The storage is already meaningful, with 3 of the 12 hours of zero generation covered (assuming usage is flat over a 24 hour period, which it isn’t), and if they keep adding at that pace it’ll be very significant.
Yes, the missing piece here is most of the demand is in winter but most of the solar production in summer.
Daily load shift is a solved problem since the 70s with dams anyways, it's not the issue with solar. The issue is season load shift which is still science fiction as we speak.
Most of the time yes, but most of the time isn't a good answer when we talk about a grid. See the nuclear issue in France which had an even worse wind generation issue compounding the problem.
Not to mention the variability which is 10x worse than solar.
That is the expected variability? On-shore wind has capacity factors between 25-40% depending on location and size of wind turbine. That it reduces to 11% is expected.
Given that this happened once it is also quickly pushed higher by storage.
How would add nuclear power to this grid mix? Yes, that is over 100% of demand being generated by rooftop solar.
"Expected variability" is nonsense, you can't just say "yeah it doesn't work, that's normal"
> How would add nuclear power to this grid mix? Yes, that is over 100% of demand being generated by rooftop solar.
Absolute nonsense again, Australia has one of the dirtiest electricity generation of the developed world with 65% fossil. Nowhere near the totality of the demand is covered by solar, even if again they have almost no real winter.
> "Expected variability" is nonsense, you can't just say "yeah it doesn't work, that's normal"
Are you saying that hydro electric dams are nonsense because they store an intermittent energy source for later usage?
With renewables lowering the price floor it means that if you can utilize them you have a competetive advantage.
> Absolute nonsense again, Australia has one of the dirtiest electricity generation of the developed world with 65% fossil. Nowhere near the totality of the demand is covered by solar, even if again they have almost no real winter.
This is an australian state. Which often has 100% of its demand either covered by rooftop solar or wind power.
Without trying to brush the example aside, how would you add an inflexible new built nuclear power plant to the mix? How will you force everyone to buy expensive electricity coming from it?
Which storage can solve. The problem is that "base load" as we know it is dead. It was only ever an economic construct, never a physical one.
Distributed renewables are unraveling the grid monopoly, meaning you can't just foist enormous nuclear subsidies on the tax payers anymore. They will vote with their wallets.
Storage can not solve firming in most of the world, maybe South Aus being exception, maybe. It can be solved by gas peakers or gas peakers combined with bess if you go fossils way, or hydro if you have it or nuclear. In Germany it'll be gas+bess but mostly gas, according to their Fraunhofer ISE org. In UK it'll be mostly gas too. In DK it'll be imports from hydro rich nordics.
It's interesting you mentioned nuclear subsidies when Germany poured on it's EEG renewables scheme alone more than double the cost of entire french nuclear fleet, both adjusted in today money
Who cares if there's a tiny portion of gas turbines left (optimized for low CAPEX) when we've solved high 90s% of the problem?
Its like complaining about the 3% fossil gas usage in France today when we still need to decarbonize shipping, agriculture, aviation, construction etc.
It is trivial to run gas turbines on carbon neutral fuel when the time comes if we determine they are still needed.
> It's interesting you mentioned nuclear subsidies when Germany poured on it's EEG renewables scheme alone more than double the cost of entire french nuclear fleet, both adjusted in today money
This is a backwards looking metric, we need to look forward based on the costs today. Are we paying 2011 solar prices or 2026 solar prices when building renewables in 2026? We pay 2026 prices.
Look at the proposed French EPR2 program. 11 cents/kWh 40 year CFD and interest free loans with the first reactor coming online in 2038.
Just an absolutely insanely large handout from tax money to force new built nuclear power into existence.
While the competition in renewables and storage are built on massive scale without subsides.
It's not tiny portion - it's a fully parallel grid. Germany as example needs to have 80GW+ of gas per Fraunhofer.
French epr2 nuclear will have smaller subsidies than german biomass. The handout of tax money for epr will be the equivalent of about 1.5-2y of german eeg now or even less in the future since it's projected to grow due to ren self cannibalization.
It's not that trivial to run gas firming on carbon neutral fuel aka biogas. First you don't have enough fuel, second- their opex will get so high due to low CF that you'll need a separate market for that and owners will be sure to ask a lot of $ for this firming to get profit and compensate no demand periods.
And we are talking only about direct subsidies. Germany will start subsidizing transmission this year tpo because their household prices are highest in EU, about 6bn/y. Most of this transmission is due to distributed ren expansion and need to avoid curtailment
That is a question for the 2030s to answer. Maybe demand response, storage and similar is enough? Either way, locking in trillions in nuclear handouts when the entire energy system will be reshaped before they are even online is by far the most stupid thing we can do today.
> French epr2 nuclear will have smaller subsidies than german biomass. The handout of tax money for epr will be the equivalent of about 1.5-2y of german eeg now or even less in the future since it's projected to grow due to ren self cannibalization.
Now you're trying to compare with the worst, because you know how outrageous the comaprison becomes when comparing with solar, wind and storage.
You do know that the EEG payments have been quickly reducing due to not needing subsidies anymore? And €20B in subsidies per reactor, which you did try to hide in "1.5-2y of eeg" is just a horrific waste of money.
> It's not that trivial to run gas firming on carbon neutral fuel aka biogas. First you don't have enough fuel, second- their opex will get so high due to low CF that you'll need a separate market for that and owners will be sure to ask a lot of $ for this firming to get profit and compensate no demand periods.
Or hydrogen, or hydrogen derivatives. Just pick whatever the maritime industry and aviation settles on as they decarbonize.
Yes. That is called "capacity markets". They already exist all around the world. Generally very cheap to run.
> And we are talking only about direct subsidies. Germany will start subsidizing transmission this year tpo because their household prices are highest in EU, about 6bn/y. Most of this transmission is due to distributed ren expansion and need to avoid curtailment
You do know that an electrified society requires 2 - 3x the grid size right? No matter the path we take we will need to massively expand the grid.
The only reason for curtailment is because Germany haven't divided the country into more markets because they expect to resolve the transmission bottlenecks in a few years.
France wants to spend less than 20bn/unit, afaik more like 12bn/unit and half of all the sum will be 0% govt loans, about 35bn (hence 1.5-2y of EEG), rest covered by edf.
EEG in Germany is projected to rise per EWI because it's paid more frequently
Hydrogen is insanely expensive. To think it'll be more economical than nuclear is strange. Germany isn't building transmission just for electrification, but mostly due to distributed generation and the need to avoid curtailment, like sudlink.
And I want the cost of renewables and storage to drop by 80%. That doesn't make it correct.
If you actually look at the EEG data over the years it is vastly down from peaking in ~2020-21.
Today yes. But that is what aviation and the maritime industry is looking and. In in the case of the maritime industry mostly derivatives, likely ammonia due to voluemetric constriants.
So like I said. Just pick whatever they settle on in the 2030s. No need to rush out trillions in handouts from tax money for new built nuclear power today because hydrogen is not cheap.
The absolut worst thing we can do today is lock in trillions in handouts to new built nuclear power just as the energy grid is fundamentally being transformed.
That's like betting on the steam locomotive when the age of diesel had already arrived. Would that be reasonable?
Since that incident storage has been scaling massively. How does a nuclear plant compete with zero marginal cost renewables?
https://oilprice.com/Energy/Energy-General/The-Quiet-Unravel...