> Data from the U.S. Energy Information Administration (EIA), processed by Ember’s US Electricity Data Explorer, shows that in the 12 months ending April 2025, solar generated 83.1 terawatt-hours (TWh) of electricity, compared to 81.6 TWh from natural gas.
I thought this was going to be the usual headline bait-and-switch where it would be revealed to only be the top source of electricity for like two hours around noon on a single Saturday, but I'm happy to have been wrong here.
California should hit their battery target far earlier than their 2045 goal based on the cost decline curve and manufacturing ramp of stationary storage.
LFP stationary battery storage appears to be at ~$52/kWh in China [1] (compare to $181 in 2018 per Our World in Data). California buys a lot of stationary storage from Tesla though, which is preparing to start production at the new Sparks, NV LFP cell production facility [2]. Ford is building their own facility in Michigan [3], with 35 GWh of capacity.
That $52/kWh price isn't for raw battery modules, it's for a fully packaged bulk storage system. Which means the raw battery module price is significantly less than $52/kWh. Wowsers.
I just bought an ecoflow for around $350 for 1 kWh. It uses LFP, I guess there's a lot of room for it to fall in price.
I wonder if there will be a tipping point where people start defecting from the grid, making it more expensive for people still on the grid, giving them a bigger incentive to defect.
Your $350/kWh for the EcoFlow is actually pretty good - that's lower to mid-range for consumer LFP power stations right now [0]. The best consumer units are still hitting around $300/kWh, but there's a long tail of more expensive options. Still a huge gap from utility-scale ($52/kWh), but the trajectory is steep.
The "death spiral" scenario you mention is interesting - we're probably still a few years out from consumer storage being cheap enough to cause mass grid defection, but the economics are getting compelling fast. The portability premium and all the inverter/BMS/housing costs still add up compared to raw cells, but even those are dropping quickly.
They will also pull forward the economic tipping points for longer duration 8 hour to 10 hour systems needed to shore up ‘Round The Clock’ renewables use cases, which disproportionately stand to benefit.
I never understood the difference between standard systems that deliver the power over a 4 hour interval versus longer duration systems of 8 hours or more. The amount of energy delivered is the same, it is just delivered more slowly. What is the factor that makes delivering over 4 hours more cost effective than 8 hours?
X hour batteries is mostly because battery people think in MWh and grid people talk in MW (and confusingly both call that capacity.)
The hours designation is the MWh / MW i.e. how fast you can empty the battery into the grid at full throttle.
The economics comes in because batteries are expensive and you want to target the highest and lowest prices during the day, charging and discharging totally to maximize revenue via price abitrage. A 1 hour battery is going to be shaving the very peak prices of each day soaking up solar at noon and displacing expensive gas peaker plants in the evening.
As the battery MWh gets bigger, but not the ability to put it all on the grid, that implies you'll be charging and discharging for 4 or 8 hours each day which means you'll be paying and getting paid closer to the average prices in both directions.
It only works economically if the battery is itself cheaper (and/or there's more renewables pushing down prices for longer periods of the day)
Yeah I’ve never understood this for lithium ion systems. Maybe some parallel or series the cells differently to get different total max power outputs? But I don’t expect that would affect cost either way.
With flow batteries there are definitely differences since the power and energy components of the system can each be scaled independently from each other. Ie need more total energy then just expand the amount of liquid electrolyte storage you have.
yesterday, right.....solar goes into the grid, the grid powers any industrial scale desalination.....
silliness aside, there will soon be formats where co generation will be economic, solar to power desalination, the generated brine bieng used for just salt, and then also refined for sodium to make batteries to capture more solar to power plants to turn sand into glass and purified silicone to make more solar and refine aluminum and copper.
agriculture in areas next to the ocean, where there is no fresh water, and essentialy free, barren flat land....fully robotisised greenhouses
the current scale of PV production is unbelievably huge, and growing....
land might be too expensive in California, but in Peru on the coast, there are never any clouds and vast areas that only need water to be productive, also as a desert, there are very few bugs and diseases to deal with, and even a short distance between operations would serve as an effective quarantine
No, that is still onshore wind according to Lazard.
Solar seems set to overtake though both are predicted to continue to drop in price and the cheapest examples of both are currently the same price to build new as running existing nuclear or gas combined cycle plants.
> Data from the U.S. Energy Information Administration (EIA), processed by Ember’s US Electricity Data Explorer, shows that in the 12 months ending April 2025, solar generated 83.1 terawatt-hours (TWh) of electricity, compared to 81.6 TWh from natural gas.
I thought this was going to be the usual headline bait-and-switch where it would be revealed to only be the top source of electricity for like two hours around noon on a single Saturday, but I'm happy to have been wrong here.
California should hit their battery target far earlier than their 2045 goal based on the cost decline curve and manufacturing ramp of stationary storage.
Battery storage growth has been incredible and you can see the gains almost weekly:
https://www.gridstatus.io/records/caiso?record=Maximum%20Bat...
It looks like batteries are now able to displace 100% of imports (which are mostly gas) for a period after sunset, eg here from yesterday:
https://www.gridstatus.io/live/caiso?date=2025-07-08
Even just a few weeks ago, imports would begin as soon as the sun set.
LFP stationary battery storage appears to be at ~$52/kWh in China [1] (compare to $181 in 2018 per Our World in Data). California buys a lot of stationary storage from Tesla though, which is preparing to start production at the new Sparks, NV LFP cell production facility [2]. Ford is building their own facility in Michigan [3], with 35 GWh of capacity.
[1] https://reneweconomy.com.au/watershed-moment-big-battery-sto...
[2] https://electrek.co/2025/07/01/tesla-unveils-lfp-battery-fac...
[3] https://electrek.co/2025/06/25/ford-stands-by-controversial-...
That $52/kWh price isn't for raw battery modules, it's for a fully packaged bulk storage system. Which means the raw battery module price is significantly less than $52/kWh. Wowsers.
I just bought an ecoflow for around $350 for 1 kWh. It uses LFP, I guess there's a lot of room for it to fall in price.
I wonder if there will be a tipping point where people start defecting from the grid, making it more expensive for people still on the grid, giving them a bigger incentive to defect.
Your $350/kWh for the EcoFlow is actually pretty good - that's lower to mid-range for consumer LFP power stations right now [0]. The best consumer units are still hitting around $300/kWh, but there's a long tail of more expensive options. Still a huge gap from utility-scale ($52/kWh), but the trajectory is steep.
The "death spiral" scenario you mention is interesting - we're probably still a few years out from consumer storage being cheap enough to cause mass grid defection, but the economics are getting compelling fast. The portability premium and all the inverter/BMS/housing costs still add up compared to raw cells, but even those are dropping quickly.
[0] https://gearscouts.com/power-stations
Its 35 for cells.
For comparison a project in NZ recently finished cost 550 usd / kwh (which also includes site, etc)
From your first link:
They will also pull forward the economic tipping points for longer duration 8 hour to 10 hour systems needed to shore up ‘Round The Clock’ renewables use cases, which disproportionately stand to benefit.
I never understood the difference between standard systems that deliver the power over a 4 hour interval versus longer duration systems of 8 hours or more. The amount of energy delivered is the same, it is just delivered more slowly. What is the factor that makes delivering over 4 hours more cost effective than 8 hours?
X hour batteries is mostly because battery people think in MWh and grid people talk in MW (and confusingly both call that capacity.)
The hours designation is the MWh / MW i.e. how fast you can empty the battery into the grid at full throttle.
The economics comes in because batteries are expensive and you want to target the highest and lowest prices during the day, charging and discharging totally to maximize revenue via price abitrage. A 1 hour battery is going to be shaving the very peak prices of each day soaking up solar at noon and displacing expensive gas peaker plants in the evening.
As the battery MWh gets bigger, but not the ability to put it all on the grid, that implies you'll be charging and discharging for 4 or 8 hours each day which means you'll be paying and getting paid closer to the average prices in both directions.
It only works economically if the battery is itself cheaper (and/or there's more renewables pushing down prices for longer periods of the day)
Yeah I’ve never understood this for lithium ion systems. Maybe some parallel or series the cells differently to get different total max power outputs? But I don’t expect that would affect cost either way.
With flow batteries there are definitely differences since the power and energy components of the system can each be scaled independently from each other. Ie need more total energy then just expand the amount of liquid electrolyte storage you have.
Is there an expected date we can start doing industrial scale desalination with excess electricity?
These people do it with solar panels…
https://solarwatersolutions.com/
We first need to replace all fossil energy with electricity, so there's a long way to go still.
> We first need to replace all fossil energy with electricity
We really don’t. Desalinated water is certainly more socially useful than a bunch of other uses of power that we don’t question.
[dead]
yesterday, right.....solar goes into the grid, the grid powers any industrial scale desalination..... silliness aside, there will soon be formats where co generation will be economic, solar to power desalination, the generated brine bieng used for just salt, and then also refined for sodium to make batteries to capture more solar to power plants to turn sand into glass and purified silicone to make more solar and refine aluminum and copper. agriculture in areas next to the ocean, where there is no fresh water, and essentialy free, barren flat land....fully robotisised greenhouses the current scale of PV production is unbelievably huge, and growing.... land might be too expensive in California, but in Peru on the coast, there are never any clouds and vast areas that only need water to be productive, also as a desert, there are very few bugs and diseases to deal with, and even a short distance between operations would serve as an effective quarantine
Sadly I think AI is going to push that further out if we ever get to that point.
Yet prices keep going up.
Gahh 8k true up bill on top of paying 1k a month ?
[dead]
Need that anakin meme.
And it’s got the cheapest electricity right. …. The cheapest right?
No, that is still onshore wind according to Lazard.
Solar seems set to overtake though both are predicted to continue to drop in price and the cheapest examples of both are currently the same price to build new as running existing nuclear or gas combined cycle plants.