4 ms·

> they are still orders of magnitude too expensive This isn't right as of today, stored energy charged by solar is about $0.20/kWh ($200/MWh) which is within s

> they are still orders of magnitude too expensive This isn't right as of today, stored energy charged by solar is about $0.20/kWh ($200/MWh) which is within spitting distance of the cost of new nuclear. And when you average this cost with the cost of solar/wind delivered directly, it's a huge cost winner. https://www.lazard.com/media/451566/lazards-levelized-cost-of-storage-version-60-vf2.pdf https://www.lazard.com/media/451566/lazards-levelized-cost-o... This is just the cost today, batteries are dropping precipitously in cost. If storage dropped in cost multiple orders of magnitude, and there's reason to believe that it will drop at least an order of magnitude, then thermal cycle electricity will completely obsolete.

jeppesen-io 5y ago
Thank you for saying this. I find in technical circles, like HN, it's often underappreciated how dramatically solar, wind and stored energy are becoming cheaper year by year while production capacity continues to grow
- FabHK 5y ago
  Question is whether batteries are getting better fast enough. Cost per energy stored might be coming down relatively quickly, but energy density (energy per volume) or specific energy (energy per mass) are improving only very slowly. As a consequence, I think long-range electric passenger jets, for example, are many decades off.
  - epistasis 5y ago
    In the context of grid electricity and nuclear, this parameter doesn't matter. For long-range passenger jets, batteries will likely never be feasible. Aviation will have other solutions.
  - jillesvangurp 5y ago
    By "very slowly" you mean double the capacity per kg in the last decade at a tenth of the cost? Because that is more or less what happened in the last decade. If you don't believe me, the Nissan Leaf originally had 21KWH of capacity when it launched in 2010. The smallest variant of the latest model has 41 KWH. You can also get a 60KWH version. And you can actually install new batteries into an original Leaf and double the range while slightly lowering the weight. And of course those batteries are now a fraction of the cost that they were in 2010. Back then replacing the battery would have cost tens of thousands of dollars. People are getting that done for around 5K now for a battery that is literally twice the capacity. The discarded batteries typically end up being part of some grid storage solution. Another doubling in the coming decade is likely. As is further reductions in price per KWH. Maybe not 10x. But probably more than 2-3x. It's profitable now to buy expensive storage solutions for grid providers. Ten years from now that will be a lot more attractive.
- epistasis 5y ago
  It's funny because the technical crowd should be best situated to understand the incredible tech curves of solar, wind, and storage, as they have experienced the advancement of computer tech. However, for other technical fields, such as civil engineering, they are not used to dramatic cost drops. Nuclear has a surprising negative learning curve. This is obvious when comparing different reactor designs over time, with ever increasing costs. But the more surprising finding is that building the same reactor multiple times get more expensive, not less. This is true even during France's successful build in the 1970s, so regulatory changes can't be the sole reason for increasing costs: https://www.sciencedirect.com/science/article/abs/pii/S0301421510003526 https://www.sciencedirect.com/science/article/abs/pii/S03014...
jillesvangurp 5y ago
Yes, it gets better when you consider that homeowners that are putting solar on their roof are essentially providing large amounts of power to energy companies without these companies having to actually spend on infrastructure. It's home owners who bear the financial burden for that as well as the risk. It gets even better if you consider many people are also installing batteries or plugging cars with vehicle to grid functionality (like the new Ford pickup truck is capable of as well as several other EVs). A lot of houses with solar and batteries enables the creation of virtual power plants that can supply many GW of power to the grid during peak hours. So, when demand is highest, they can sell power at the most lucrative rates straight from batteries that were charged for free during the day. And they don't even have to buy the batteries or the panels. All they need to do is give a cut of the profits to the home owners. Basically, the way the EV market is developing, there are going to be many millions of large capacity batteries on wheels plugged into the grid at any time. All power companies have to do is find a way to tap into all that power. Even discharging them a little bit collectively creates a lot of power. E.g. supplying half a KWH over the course of an evening barely moves the needle on a car. Most EVs have something like 40-50KWH of battery. So we're talking a percent or so. They'd be trickling out electricity at a low rate of something like e.g. 500W or even less. Times a two million plugged in cars is about 1 GW of capacity. Do that for an hour and you get 1GWH of power at the cost of very slightly draining the batteries on a few million cars. That's just car batteries. Many home owners also have batteries installed into their houses, and solar on the roof to charge them. And many of those setups are net producers for large parts of the year. A gas plant costs about 1000$/MWH to operate. Possibly a bit more lately due to the high gas prices lately. So, 1GWH of power is about 1M $ in cost. That's worth paying something to home owners. Most of them are happy to just have the KWH slashed from their monthly bills. The rest is basically pure profit. Any GWH of gas they don't have to burn adds to their profits. Power companies are spending billions on expensive grid batteries to lower that cost.

jeppesen-io 5y ago

Thank you for saying this. I find in technical circles, like HN, it's often underappreciated how dramatically solar, wind and stored energy are becoming cheaper year by year while production capacity continues to grow

FabHK 5y ago
Question is whether batteries are getting better fast enough. Cost per energy stored might be coming down relatively quickly, but energy density (energy per volume) or specific energy (energy per mass) are improving only very slowly. As a consequence, I think long-range electric passenger jets, for example, are many decades off.
- epistasis 5y ago
  In the context of grid electricity and nuclear, this parameter doesn't matter. For long-range passenger jets, batteries will likely never be feasible. Aviation will have other solutions.
- jillesvangurp 5y ago
  By "very slowly" you mean double the capacity per kg in the last decade at a tenth of the cost? Because that is more or less what happened in the last decade. If you don't believe me, the Nissan Leaf originally had 21KWH of capacity when it launched in 2010. The smallest variant of the latest model has 41 KWH. You can also get a 60KWH version. And you can actually install new batteries into an original Leaf and double the range while slightly lowering the weight. And of course those batteries are now a fraction of the cost that they were in 2010. Back then replacing the battery would have cost tens of thousands of dollars. People are getting that done for around 5K now for a battery that is literally twice the capacity. The discarded batteries typically end up being part of some grid storage solution. Another doubling in the coming decade is likely. As is further reductions in price per KWH. Maybe not 10x. But probably more than 2-3x. It's profitable now to buy expensive storage solutions for grid providers. Ten years from now that will be a lot more attractive.
epistasis 5y ago
It's funny because the technical crowd should be best situated to understand the incredible tech curves of solar, wind, and storage, as they have experienced the advancement of computer tech. However, for other technical fields, such as civil engineering, they are not used to dramatic cost drops. Nuclear has a surprising negative learning curve. This is obvious when comparing different reactor designs over time, with ever increasing costs. But the more surprising finding is that building the same reactor multiple times get more expensive, not less. This is true even during France's successful build in the 1970s, so regulatory changes can't be the sole reason for increasing costs: https://www.sciencedirect.com/science/article/abs/pii/S0301421510003526 https://www.sciencedirect.com/science/article/abs/pii/S03014...

jillesvangurp 5y ago

Yes, it gets better when you consider that homeowners that are putting solar on their roof are essentially providing large amounts of power to energy companies without these companies having to actually spend on infrastructure. It's home owners who bear the financial burden for that as well as the risk. It gets even better if you consider many people are also installing batteries or plugging cars with vehicle to grid functionality (like the new Ford pickup truck is capable of as well as several other EVs). A lot of houses with solar and batteries enables the creation of virtual power plants that can supply many GW of power to the grid during peak hours. So, when demand is highest, they can sell power at the most lucrative rates straight from batteries that were charged for free during the day. And they don't even have to buy the batteries or the panels. All they need to do is give a cut of the profits to the home owners. Basically, the way the EV market is developing, there are going to be many millions of large capacity batteries on wheels plugged into the grid at any time. All power companies have to do is find a way to tap into all that power. Even discharging them a little bit collectively creates a lot of power. E.g. supplying half a KWH over the course of an evening barely moves the needle on a car. Most EVs have something like 40-50KWH of battery. So we're talking a percent or so. They'd be trickling out electricity at a low rate of something like e.g. 500W or even less. Times a two million plugged in cars is about 1 GW of capacity. Do that for an hour and you get 1GWH of power at the cost of very slightly draining the batteries on a few million cars. That's just car batteries. Many home owners also have batteries installed into their houses, and solar on the roof to charge them. And many of those setups are net producers for large parts of the year. A gas plant costs about 1000$/MWH to operate. Possibly a bit more lately due to the high gas prices lately. So, 1GWH of power is about 1M $ in cost. That's worth paying something to home owners. Most of them are happy to just have the KWH slashed from their monthly bills. The rest is basically pure profit. Any GWH of gas they don't have to burn adds to their profits. Power companies are spending billions on expensive grid batteries to lower that cost.