4 ms·

Indeed! But getting ~80% back from batteries or pumped hydro looks even better[1]. Or, the ~90% that EnergyVault claims[2]. I'm very much pro-renewables and pr

Indeed! But getting ~80% back from batteries or pumped hydro looks even better[1]. Or, the ~90% that EnergyVault claims[2]. I'm very much pro-renewables and pro-storage, but that doesn't mean that all forms of storage are equal. This one has certain advantages (cheap, simple, tough), but the roundtrip efficiency kinda sucks compared to some other options. And I'm guessing (without checking) that the 30% is for short-term storage (eg, overnight). In which it will drop over time, as the battery inevitably leaks heat into the environment. I'd guess that after 6 months (summer to winter, as you mentioned) it will lose a non-trivial amount of energy so the efficiency will be even lower. [1] - Actual rates from deployed systems from the EIA: https://www.eia.gov/todayinenergy/detail.php?id=46756 https://www.eia.gov/todayinenergy/detail.php?id=46756 [2] - https://www.energyvault.com/newsroom/energy-vault-announces-commercial-availability-of-transformative-utility-scale-energy-storage-technology https://www.energyvault.com/newsroom/energy-vault-announces-...

Schroedingersat 4y ago
Laege scale thermal storage has surprisingly low losses to the point where a 10GWh scale system can disregard them over a year. Cube square helps a lot here. Your heat engine is where you lose all the energy. Batteries are great and getting better. There's a reason most storage uses them, but they're ill suited to multi month storage due to cost. The best contender here is chemical fuels (which have similar efficiency penalties). Energy vault is, at best, a very poor subsitute for batteries. A 1000 tonne block on top of a 500m tall stack can store $20 worth of solar energy and might sell it for $80. If you use it annually, then you get $60/yr. If you use it 1000 times a year it might be viable compared to current generstion batteries, but that will change quickly as there's no real way to make it more economical. Ignore it and tell anyone who suggests using it as anything other than 4hr storage they're probably being scammed. Thermal batteries are the budget option. You can store about 5-20Wh/kg in whatever thing that's lying around that doesn't turn into a gas or become too corrosive. By every other metric they are terrible, but they are, in the most literal sense possible, dirt cheap (although you generally use low grade sand or maybe water because dirt is too valuable). The use case is mostly providing very cheap heat with energy that is not valuable enough to store with other methods. Insofar as using the concept for work (where the efficiency penalty lies), the logic would probably be something like: -- I have free energy noone wants (curtailed wind during midday summer) -- I have free sand noone wants (someone dug a hole and wants to get rid of it), but it costs me a little more to move it and build a silo than I will get paid to dispose of it. -- I have a free steam engine with its exhaust hooked up the district heating noone wants (a coal generator went bankrupt because renewables replaced it). I'll put the heat in sand, then make steam with it in winter (extracting 20-30% of the original energy as work), then pump 50-60% of the original energy into homes as heating (losing about 20% in the pipes and steam engine and some during storage).
- breischl 4y ago
  Interesting point about the economics of Energy Vault. I had not actually done the math before. $80 seems low for current prices but it's about the right order of magnitude. OTOH there is at least one project being built to store 100 MWh, and they have a bunch of investors for millions of dollars. That's a bunch of smart people who surely did at least 5 minutes of math to validate the idea, which makes me wonder what they saw that's missing from this analysis. It's possible it's all just a big scam, but strikes me as unlikely. Also, doesn't thermal storage have the same problem with long-term storage of energy not producing a lot of revenue? Perhaps the capital investment is lower, but the yearly income would be order-of-magnitude the same. I was pretty much agreeing with you about thermal being the cheap option, but thanks for the breakout. Do you know of any examples of 10 GWh thermal storage? Just curious, I haven't really poked into that area too much.
  - Schroedingersat 4y ago
    Energy vault is targeting <4hr. Capex might be cheaper than current batteries, but given that the big project they announced recently is a chemical battery I think those investors are looking to cash out at an IPO rather than see the cocnept succeed. > Do you know of any examples of 10 GWh thermal storage? Just curious, I haven't really poked into that area too much. There are a few pilot projects in the 100s of MWh to GWh range. 100MWh is enough that your surface area is quite low.

Schroedingersat 4y ago

Laege scale thermal storage has surprisingly low losses to the point where a 10GWh scale system can disregard them over a year. Cube square helps a lot here. Your heat engine is where you lose all the energy. Batteries are great and getting better. There's a reason most storage uses them, but they're ill suited to multi month storage due to cost. The best contender here is chemical fuels (which have similar efficiency penalties). Energy vault is, at best, a very poor subsitute for batteries. A 1000 tonne block on top of a 500m tall stack can store $20 worth of solar energy and might sell it for $80. If you use it annually, then you get $60/yr. If you use it 1000 times a year it might be viable compared to current generstion batteries, but that will change quickly as there's no real way to make it more economical. Ignore it and tell anyone who suggests using it as anything other than 4hr storage they're probably being scammed. Thermal batteries are the budget option. You can store about 5-20Wh/kg in whatever thing that's lying around that doesn't turn into a gas or become too corrosive. By every other metric they are terrible, but they are, in the most literal sense possible, dirt cheap (although you generally use low grade sand or maybe water because dirt is too valuable). The use case is mostly providing very cheap heat with energy that is not valuable enough to store with other methods. Insofar as using the concept for work (where the efficiency penalty lies), the logic would probably be something like: -- I have free energy noone wants (curtailed wind during midday summer) -- I have free sand noone wants (someone dug a hole and wants to get rid of it), but it costs me a little more to move it and build a silo than I will get paid to dispose of it. -- I have a free steam engine with its exhaust hooked up the district heating noone wants (a coal generator went bankrupt because renewables replaced it). I'll put the heat in sand, then make steam with it in winter (extracting 20-30% of the original energy as work), then pump 50-60% of the original energy into homes as heating (losing about 20% in the pipes and steam engine and some during storage).

breischl 4y ago
Interesting point about the economics of Energy Vault. I had not actually done the math before. $80 seems low for current prices but it's about the right order of magnitude. OTOH there is at least one project being built to store 100 MWh, and they have a bunch of investors for millions of dollars. That's a bunch of smart people who surely did at least 5 minutes of math to validate the idea, which makes me wonder what they saw that's missing from this analysis. It's possible it's all just a big scam, but strikes me as unlikely. Also, doesn't thermal storage have the same problem with long-term storage of energy not producing a lot of revenue? Perhaps the capital investment is lower, but the yearly income would be order-of-magnitude the same. I was pretty much agreeing with you about thermal being the cheap option, but thanks for the breakout. Do you know of any examples of 10 GWh thermal storage? Just curious, I haven't really poked into that area too much.
- Schroedingersat 4y ago
  Energy vault is targeting <4hr. Capex might be cheaper than current batteries, but given that the big project they announced recently is a chemical battery I think those investors are looking to cash out at an IPO rather than see the cocnept succeed. > Do you know of any examples of 10 GWh thermal storage? Just curious, I haven't really poked into that area too much. There are a few pilot projects in the 100s of MWh to GWh range. 100MWh is enough that your surface area is quite low.