6 ms·

You can use batteries more than once! If you cycle them once per day for ten years to store solar power until night, your numbers given mean they add $30-100 pe

You can use batteries more than once! If you cycle them once per day for ten years to store solar power until night, your numbers given mean they add $30-100 per MWh to the cost of that stored power (but not daytime use). Not cheap, but in no way a "fantasy". Certainly not in a world with sane carbon pricing.

VBprogrammer 6y ago
So the cost of storage would exceed the cost of generating the same power via a CCGT plant?
- reitzensteinm 6y ago
  I just took parent's numbers at face value to show how they were misinterpreting them. I haven't seen much in the way of true apples to apples comparisons. Most solar plants with storage don't have much storage. For instance I saw a link this week on HN for $30 + $15/MWh for solar + storage, but the storage was only one hour of the nameplate capacity. Based on the falling prices of batteries, I don't think it will be long before solar + storage providing power evenly over 24h (an unrealistic worst case) becomes economical. At a 30% capacity factor, assuming the batteries are charging for 8 hours and draining for 16, that plant would need 5x as much storage for a cost of $30 + $75/MWh. More realistically you'd only need 3-4x as much storage as you get small amounts of sunlight later and earlier in the day, and demand curves are not flat. Halve the price of batteries, which already happened in the 2010s, and that's looking economical even with a zero price on carbon. But most countries are pretty far from saturating solar potential, so you can still build solar without storage for a while. Once you need the battery backing, it's probably going to be cheap enough.
  - emteycz 6y ago
    Are you sure it was $15/MWh? That seems more than 1000x cheaper than state of the art batteries.
    - reitzensteinm 6y ago
      That's how much the storage added to the price of power in the bid, not the capital cost of the batteries.
      - emteycz 6y ago
        Oh! Thank you.
    - teraku 6y ago
      I just checked some prices for small scale PV farms you can plant to your house. It's 16,83 euro-cent for the PV per KWH and 23,74 euro-cent for the storage. Production is 10KWH, storage is 11KWH. So if we (naively) multiply by 1000 we get 23,74€/MWh. And that is for the home market. I know battery storage doesn't scale linearly, but also if you were to build this as a highly-scaled farm, I'm sure prices would also drop as well.
      - bedosh 6y ago
        23 euro cents * 1000 is 230 Euros, not 23. But that is still a factor of 1000 lower then the prices mentioned in https://www.nrel.gov/docs/fy19osti/73222.pdf https://www.nrel.gov/docs/fy19osti/73222.pdf and other web publications. Are you sure about those numbers and what kind of storage do they refer to?
  - macspoofing 6y ago
    > solar + storage providing power evenly over 24h (an unrealistic worst case) becomes economical. That's not the the worst case. Solar output varies by season. So you have to have enough storage for multiple weeks. This also means you have to greatly over-provison solar panels so that they not only provide power, but also enough to charge batteries.
    - epistasis 6y ago
      No, you don't need storage for multiple weeks, that would be silly. Solar is so cheap now that already panel generation capacity is frequently getting oversized compared to the potential inverter output, to maximize costs. Future solar farm design will optimize seasonal generation capacity similarly; size your panels for the seasonal minimum, attach batteries, and you can have year round output of a firm amount with only hours worth of batteries.
      - macspoofing 6y ago
        >No, you don't need storage for multiple weeks You have seasonal, and inter-seasonal variability in a solar and wind output. You need to account for that, in an economy that requires orders of magnitude more energy then being provided by renewables and is still electrifing. What do you mean it's silly? I've seen a few articles now where a baseline required battery storage is on the order of weeks. What do you mean a few hours is all that's required? What if you have a multi-day cloudy, wind-still weather? What do you do then????? >size your panels for the seasonal minimum, attach batteries No. You need to over-provison even for minimum seasonal output, because even in seasonal minimum you still need to charge your batteries. Winter is very long. > only hours worth of batteries If you don't mind constant blanks during periods of prolonged reduced wind/solar output...then sure. I don't what 'few hours' means - there's a day-night cycle every single day. Multi-day and highly variable multi-weeks where solar and wind output is minimal or none are relatively common. Here's what should give you pause and maybe make you reexamine your confidence - nobody is actually building out a solar/wind/battery infrastructure - nobody. We're essentially expanding natural gas use and complementing it with solar/wind (there's a reason why Germany is singing multi-decade contracts to ship Russian natural gas, and why every natural gas company is pushing renewables). Natural gas is not a transitional technology to renewables. It's the end-state for renewables.
Manfredo_1 6y ago
No, that's not this works. Batteries currently cost ~$150 to $250 per KWh. As in, if you want to store one kilowatt hour, you need to spend $150 to $250 dollars on batteries. If you want to store one megawatt hour, then that's $150,000 to $250,000 worth of batteries. Yes, you can cycle this battery multiple times. I'm not sure how you think that this somehow makes storage cheaper. The fact that you can charge a 1 KWh battery and discharge it 1,000 times doesn't magically make it a 1 MWh battery. Battery cost is measured in terms of cost of capacity, in watt hours. It absolutely is a fantasy to use this for grid storage. To put this in perspective, the US uses 11.5 TWh of electricity each day. The entire world produces 300 GWh of batteries every year. In order to achieve the 12 hours of storage for 80% renewables we'd consume global battery production for several years. To reach the 3 weeks storage for 100% renewable we'd need 805 years of the entire world's battery production at current levels. Sure, battery production is set to double over the next decade. But we're still talking centuries worth of global battery production to fulfill the storage demands of just the United States. And we're ignoring the fact that batteries wear out after a few thousand cycles. Again, the idea that batteries are going to make wind and solar feasible is a fantasy.
- reitzensteinm 6y ago
  Parent compared the price of one MWh of power from a natural gas plant with the capital cost of building batteries that can store one MWh. This is a type error.
  - Manfredo_1 6y ago
    The enormous capital costs demonstrates how expensive it makes solar and wind in simple terms. Even if we measure in terms of cost per watt hour stored and drawn over the lifetime of the batter it's still hugely expensive. Most batteries can support 300 to 1,000 charge and discharge cycles. So in terms of overall cost of storage, that's still in the range of $200 to $700 per MWh of energy stored and drawn over the lifetime of the battery. This point is largely moot because we don't even have the manufacturing capacity to deliver the amount of batteries required. We only build 300 GWh of batteries globally each year. To reach 12 hours of storage, the US would need 6 TWh - 20 years of global production at current production rates.

VBprogrammer 6y ago

So the cost of storage would exceed the cost of generating the same power via a CCGT plant?

reitzensteinm 6y ago
I just took parent's numbers at face value to show how they were misinterpreting them. I haven't seen much in the way of true apples to apples comparisons. Most solar plants with storage don't have much storage. For instance I saw a link this week on HN for $30 + $15/MWh for solar + storage, but the storage was only one hour of the nameplate capacity. Based on the falling prices of batteries, I don't think it will be long before solar + storage providing power evenly over 24h (an unrealistic worst case) becomes economical. At a 30% capacity factor, assuming the batteries are charging for 8 hours and draining for 16, that plant would need 5x as much storage for a cost of $30 + $75/MWh. More realistically you'd only need 3-4x as much storage as you get small amounts of sunlight later and earlier in the day, and demand curves are not flat. Halve the price of batteries, which already happened in the 2010s, and that's looking economical even with a zero price on carbon. But most countries are pretty far from saturating solar potential, so you can still build solar without storage for a while. Once you need the battery backing, it's probably going to be cheap enough.
- emteycz 6y ago
  Are you sure it was $15/MWh? That seems more than 1000x cheaper than state of the art batteries.
  - reitzensteinm 6y ago
    That's how much the storage added to the price of power in the bid, not the capital cost of the batteries.
    - emteycz 6y ago
      Oh! Thank you.
  - teraku 6y ago
    I just checked some prices for small scale PV farms you can plant to your house. It's 16,83 euro-cent for the PV per KWH and 23,74 euro-cent for the storage. Production is 10KWH, storage is 11KWH. So if we (naively) multiply by 1000 we get 23,74€/MWh. And that is for the home market. I know battery storage doesn't scale linearly, but also if you were to build this as a highly-scaled farm, I'm sure prices would also drop as well.
    - bedosh 6y ago
      23 euro cents * 1000 is 230 Euros, not 23. But that is still a factor of 1000 lower then the prices mentioned in https://www.nrel.gov/docs/fy19osti/73222.pdf https://www.nrel.gov/docs/fy19osti/73222.pdf and other web publications. Are you sure about those numbers and what kind of storage do they refer to?
- macspoofing 6y ago
  > solar + storage providing power evenly over 24h (an unrealistic worst case) becomes economical. That's not the the worst case. Solar output varies by season. So you have to have enough storage for multiple weeks. This also means you have to greatly over-provison solar panels so that they not only provide power, but also enough to charge batteries.
  - epistasis 6y ago
    No, you don't need storage for multiple weeks, that would be silly. Solar is so cheap now that already panel generation capacity is frequently getting oversized compared to the potential inverter output, to maximize costs. Future solar farm design will optimize seasonal generation capacity similarly; size your panels for the seasonal minimum, attach batteries, and you can have year round output of a firm amount with only hours worth of batteries.
    - macspoofing 6y ago
      >No, you don't need storage for multiple weeks You have seasonal, and inter-seasonal variability in a solar and wind output. You need to account for that, in an economy that requires orders of magnitude more energy then being provided by renewables and is still electrifing. What do you mean it's silly? I've seen a few articles now where a baseline required battery storage is on the order of weeks. What do you mean a few hours is all that's required? What if you have a multi-day cloudy, wind-still weather? What do you do then????? >size your panels for the seasonal minimum, attach batteries No. You need to over-provison even for minimum seasonal output, because even in seasonal minimum you still need to charge your batteries. Winter is very long. > only hours worth of batteries If you don't mind constant blanks during periods of prolonged reduced wind/solar output...then sure. I don't what 'few hours' means - there's a day-night cycle every single day. Multi-day and highly variable multi-weeks where solar and wind output is minimal or none are relatively common. Here's what should give you pause and maybe make you reexamine your confidence - nobody is actually building out a solar/wind/battery infrastructure - nobody. We're essentially expanding natural gas use and complementing it with solar/wind (there's a reason why Germany is singing multi-decade contracts to ship Russian natural gas, and why every natural gas company is pushing renewables). Natural gas is not a transitional technology to renewables. It's the end-state for renewables.

Manfredo_1 6y ago

No, that's not this works. Batteries currently cost ~$150 to $250 per KWh. As in, if you want to store one kilowatt hour, you need to spend $150 to $250 dollars on batteries. If you want to store one megawatt hour, then that's $150,000 to $250,000 worth of batteries. Yes, you can cycle this battery multiple times. I'm not sure how you think that this somehow makes storage cheaper. The fact that you can charge a 1 KWh battery and discharge it 1,000 times doesn't magically make it a 1 MWh battery. Battery cost is measured in terms of cost of capacity, in watt hours. It absolutely is a fantasy to use this for grid storage. To put this in perspective, the US uses 11.5 TWh of electricity each day. The entire world produces 300 GWh of batteries every year. In order to achieve the 12 hours of storage for 80% renewables we'd consume global battery production for several years. To reach the 3 weeks storage for 100% renewable we'd need 805 years of the entire world's battery production at current levels. Sure, battery production is set to double over the next decade. But we're still talking centuries worth of global battery production to fulfill the storage demands of just the United States. And we're ignoring the fact that batteries wear out after a few thousand cycles. Again, the idea that batteries are going to make wind and solar feasible is a fantasy.

reitzensteinm 6y ago
Parent compared the price of one MWh of power from a natural gas plant with the capital cost of building batteries that can store one MWh. This is a type error.
- Manfredo_1 6y ago
  The enormous capital costs demonstrates how expensive it makes solar and wind in simple terms. Even if we measure in terms of cost per watt hour stored and drawn over the lifetime of the batter it's still hugely expensive. Most batteries can support 300 to 1,000 charge and discharge cycles. So in terms of overall cost of storage, that's still in the range of $200 to $700 per MWh of energy stored and drawn over the lifetime of the battery. This point is largely moot because we don't even have the manufacturing capacity to deliver the amount of batteries required. We only build 300 GWh of batteries globally each year. To reach 12 hours of storage, the US would need 6 TWh - 20 years of global production at current production rates.