4 ms·

EIA disagrees with you. Batteries are increasingly being used to respond to price fluctuations, which means storage. https://www.eia.gov/todayinenergy/detail.p

EIA disagrees with you. Batteries are increasingly being used to respond to price fluctuations, which means storage. https://www.eia.gov/todayinenergy/detail.php?id=53199 https://www.eia.gov/todayinenergy/detail.php?id=53199

Eisenstein 3y ago
From your link: > [batteries] can balance electricity supply and demand for periods ranging from a few seconds to a few hours
- Gwypaas 3y ago
  The point being, the "few hours" can be at any time from when the solar panels stop filling the outbound power grid connection, or a trickle all night if that is what the market needs.
  - Eisenstein 3y ago
    You want to run the grid on a few hours leeway? I think you have a profound misunderstanding of the scope of this.
    - Gwypaas 3y ago
      Why are you putting words in my mouth? You have to differentiate the general case with the specific case. If the market solution is not satisfactory add something like Karlshamnsverket in Sweden. Oil power plant getting a tiny subsidy to be in standby during the winter months. Convert it to synthetic fuels when the paltry emissions a few of hours of runtime produces becomes significant in the grand scheme of things. https://www.uniper.energy/sweden/power-plants-sweden/karlshamn https://www.uniper.energy/sweden/power-plants-sweden/karlsha...
- jillesvangurp 3y ago
  Days or weeks easily; it just depends on how much battery you have available. Batteries are currently being produced in the hundreds of gwh per year. That will soon cross over into twh per year. Some numbers to boggle your mind: 1 smallish EV is 50kwh of battery. 100 kwh is 2 EVs 20 EVs is 1 mwh 20,000 EVs is 1 gwh 200,000 EVs is 10 gwh 2,000,000 EVs is 100 gwh Tesla is producing around 2 million EVs/year right now. That's about 100 gwh of battery. Of course Teslas have larger batteries (60-90ish) so it's probably closer to 150 gwh of battery. That's aside from their mega packs and domestic storage of course, which is on track to becoming at least as big of a business as their car business currently is. Lets call it at around 200-250 gwh. Right now. Also, they plan to grow to 10 million vehicles produced per year. So, just Tesla should be producing well over a twh of battery per year pretty soon. That's just Tesla. They are quite big of course but other companies are catching up fast and Tesla actually buys a lot of battery from external suppliers. These batteries have thousands of cycles and last quite long. So, the cumulative battery capacity of this planet is going to range in the tens to hundreds of twh pretty soon. We'll soon have lots of trucks, buses, and other vehicles switching to batteries as well. Millions of them. Very large batteries. And of course grids are deploying batteries by the tens of mwh already. That's a lot of battery production. All that battery production adds up to some pretty significant capacity. Growing by the hundreds of ghw right now every year and thousands of gwh pretty soon. Relative to the total electricity production, that's getting close to potentially keeping up with that if you assume ~365 cycles per year for each battery (on the high side of course). The total electricity production globally is around 25 phw (1000 twh). 25000/365 ~= 70 twh per day. So having a few tens of twh of battery in the market is going to provide plenty of buffer for whatever severe electricity production dips might happen locally. Yes, 70 twh of battery wouldn't last long if you relied on that exclusively. But it won't have to because we can interconnect places with cables and there's always some wind somewhere and the sun comes out pretty reliably every day somewhere and it's not going to be cloudy everywhere at the same time. On average renewable production fluctuates a lot locally but not regionally or globally. We can plan for this with overcapacity, storage, and cables. The bottom line is that we'll have way more battery than we actually need pretty soon. And that's before we start considering long term storage that is also being developed. To be fair, using lithium ion batteries for that would be pretty inefficient from a cost point of view. But it seems we might have tens of twh of that deployed pretty soon anyway. My expectation is that most of that battery capacity will be just sitting there fully charged most of the time.
pyrale 3y ago
> EIA disagrees with you. Batteries are increasingly being used to respond to price fluctuations, which means storage. Your link states that respondents may provide multiple answers. That means that someone buying batteries for frequency regulation may also decide, should the opportunity arise, to do some arbitrage. Because of that, it's pretty hard to claim that storage is viable for a single use case.

Eisenstein 3y ago

From your link: > [batteries] can balance electricity supply and demand for periods ranging from a few seconds to a few hours

Gwypaas 3y ago
The point being, the "few hours" can be at any time from when the solar panels stop filling the outbound power grid connection, or a trickle all night if that is what the market needs.
- Eisenstein 3y ago
  You want to run the grid on a few hours leeway? I think you have a profound misunderstanding of the scope of this.
  - Gwypaas 3y ago
    Why are you putting words in my mouth? You have to differentiate the general case with the specific case. If the market solution is not satisfactory add something like Karlshamnsverket in Sweden. Oil power plant getting a tiny subsidy to be in standby during the winter months. Convert it to synthetic fuels when the paltry emissions a few of hours of runtime produces becomes significant in the grand scheme of things. https://www.uniper.energy/sweden/power-plants-sweden/karlshamn https://www.uniper.energy/sweden/power-plants-sweden/karlsha...
jillesvangurp 3y ago
Days or weeks easily; it just depends on how much battery you have available. Batteries are currently being produced in the hundreds of gwh per year. That will soon cross over into twh per year. Some numbers to boggle your mind: 1 smallish EV is 50kwh of battery. 100 kwh is 2 EVs 20 EVs is 1 mwh 20,000 EVs is 1 gwh 200,000 EVs is 10 gwh 2,000,000 EVs is 100 gwh Tesla is producing around 2 million EVs/year right now. That's about 100 gwh of battery. Of course Teslas have larger batteries (60-90ish) so it's probably closer to 150 gwh of battery. That's aside from their mega packs and domestic storage of course, which is on track to becoming at least as big of a business as their car business currently is. Lets call it at around 200-250 gwh. Right now. Also, they plan to grow to 10 million vehicles produced per year. So, just Tesla should be producing well over a twh of battery per year pretty soon. That's just Tesla. They are quite big of course but other companies are catching up fast and Tesla actually buys a lot of battery from external suppliers. These batteries have thousands of cycles and last quite long. So, the cumulative battery capacity of this planet is going to range in the tens to hundreds of twh pretty soon. We'll soon have lots of trucks, buses, and other vehicles switching to batteries as well. Millions of them. Very large batteries. And of course grids are deploying batteries by the tens of mwh already. That's a lot of battery production. All that battery production adds up to some pretty significant capacity. Growing by the hundreds of ghw right now every year and thousands of gwh pretty soon. Relative to the total electricity production, that's getting close to potentially keeping up with that if you assume ~365 cycles per year for each battery (on the high side of course). The total electricity production globally is around 25 phw (1000 twh). 25000/365 ~= 70 twh per day. So having a few tens of twh of battery in the market is going to provide plenty of buffer for whatever severe electricity production dips might happen locally. Yes, 70 twh of battery wouldn't last long if you relied on that exclusively. But it won't have to because we can interconnect places with cables and there's always some wind somewhere and the sun comes out pretty reliably every day somewhere and it's not going to be cloudy everywhere at the same time. On average renewable production fluctuates a lot locally but not regionally or globally. We can plan for this with overcapacity, storage, and cables. The bottom line is that we'll have way more battery than we actually need pretty soon. And that's before we start considering long term storage that is also being developed. To be fair, using lithium ion batteries for that would be pretty inefficient from a cost point of view. But it seems we might have tens of twh of that deployed pretty soon anyway. My expectation is that most of that battery capacity will be just sitting there fully charged most of the time.

pyrale 3y ago

> EIA disagrees with you. Batteries are increasingly being used to respond to price fluctuations, which means storage. Your link states that respondents may provide multiple answers. That means that someone buying batteries for frequency regulation may also decide, should the opportunity arise, to do some arbitrage. Because of that, it's pretty hard to claim that storage is viable for a single use case.