I think[0] we should probably be aiming for a few hours of storage, even with every affordable grid interconnect (which would help a lot with short winter days). We use less power at night — and some of what we do use is incentivised by lower nighttime electricity costs from power plants that don’t scale down well — but I expect we will continue to use some at night forever.
For the sake of Fermi estimation, I assume average electricity use in a developed nation is 1kW/person, and that average all-forms power use in the same is 5kW/person.
Transportation is almost certainly going to be batteries or synthesised fuels like hydrogen, and can only be backed by gravity storage if you have the kind of beamed power that would be banned by international treaty on the grounds of being too easily weaponizable[1].
If you’ll permit me to assume grid interconnects and lower nighttime demand than at present due to different pricing incentives, we might be able to need a mere 3kW-hours/person of electricity storage per night.
Using the lifetime cost estimates in the article, 3kWh/person/night is about $0.51/person/night for gravity storage and $1.10/person/night for LiIon. Neither is bank-breaking, but cheaper is better.
However, the volume required is a different question: 3kWh of batteries has a volume of 4.3 to 12 litres, while 3kWh of gravity storage is 1.1 metric ton kilometres[2].
You can make the distance required smaller by using more mass, but if I assume the average home mass is about 200 tons, you’d still have to raise the entire building 5.5 meters every day to store the same energy as a backpack of batteries. I do not expect construction on this scale to be the optimal solution in general, despite it being a very good idea in some specific cases such as hydro dams and preexisting deep shafts.
[0] Armchair opinion — I’m a software engineer not a civil engineer.
[1] Assuming they’ve read or watched any Larry Niven, the Bobbiverse, The Expanse, Babylon 5, or the news at any point in the decade following the second week of September 2001 — 'A reaction drive's efficiency as a weapon is in direct proportion to its efficiency as a drive.'
Large scale hydro electric dams are already large scale batteries across months. The specific day of the week energy is released is not particularly relevant let alone time of day. 6.6% of US electricity is from hydro, assuming 2/3 flexable your looking at 4.4% of daily power demand or ~1 hour of full grid storage every day is already built.
Excess capacity is required for storage to work and reduces the need for energy storage. But, once you start talking excess capacity, transmission dramatically reduces the need for storage.
Assuming storage costs say 2x as much as wind generation then building excess wind capacity is worth it until ~1/2 of a wind turbines output is wasted. Thus if we are looking at 3kWh of storage that can be banked at any time of the day we are looking at a lot of excess capacity, yet somehow still supposed to have a 3kWh per night deficit.
For the sake of Fermi estimation, I assume average electricity use in a developed nation is 1kW/person, and that average all-forms power use in the same is 5kW/person.
Transportation is almost certainly going to be batteries or synthesised fuels like hydrogen, and can only be backed by gravity storage if you have the kind of beamed power that would be banned by international treaty on the grounds of being too easily weaponizable[1].
If you’ll permit me to assume grid interconnects and lower nighttime demand than at present due to different pricing incentives, we might be able to need a mere 3kW-hours/person of electricity storage per night.
Using the lifetime cost estimates in the article, 3kWh/person/night is about $0.51/person/night for gravity storage and $1.10/person/night for LiIon. Neither is bank-breaking, but cheaper is better.
However, the volume required is a different question: 3kWh of batteries has a volume of 4.3 to 12 litres, while 3kWh of gravity storage is 1.1 metric ton kilometres[2].
You can make the distance required smaller by using more mass, but if I assume the average home mass is about 200 tons, you’d still have to raise the entire building 5.5 meters every day to store the same energy as a backpack of batteries. I do not expect construction on this scale to be the optimal solution in general, despite it being a very good idea in some specific cases such as hydro dams and preexisting deep shafts.
[0] Armchair opinion — I’m a software engineer not a civil engineer.
[1] Assuming they’ve read or watched any Larry Niven, the Bobbiverse, The Expanse, Babylon 5, or the news at any point in the decade following the second week of September 2001 — 'A reaction drive's efficiency as a weapon is in direct proportion to its efficiency as a drive.'
[2] http://www.wolframalpha.com/input/?i=3kWh%2F%289.8m%2Fs%2Fs%...