[strainer]

> assume that we need 24 hours of storage or roughly 57600MWh. A modern redox flow or lithium ion battery can cost as little as $100 per kWh

I think the flow battery can get even cheaper than this for 24hr storage. Recent price claims[1] were $150/kWh at 4hr, and $100/kWh at 8hr - this puts the cost of charge/discharge rate at $400/kW and the cost of the energy storage at $50/kWh.

1kW @ 4hrs = (400+ 50* 4) /4 = 150$ (50+400/4)

1kW @ 8hrs = (400+ 50* 8) /8 = 100$ (50+400/8)

1kW @24hrs = (400+ 50* 24) /24 = 66$ (50+400/24)

So it really was a bit crazy to estimate 500$/kWh for that storage, 200$kWh would be absolute maximum present price for flow battery, and under $75kWh seems likely.

[1] https://news.ycombinator.com/item?id=18751639

[erikpukinskis]

Is it really sensible to believe any press release about future batteries?

[strainer]

The press release contained current and next years prices - 200$/kWh @ 8hr storage is the current cost of flow storage.

Its not really sensible to price 24hrs of li-on storage considering li-ons charge/discharge rate capability is fixed. There is no use for the capability to charge and discharge 24hours of power within a couple of hours, but you cant avoid buying it with li-on. With flow batteries you price the discharge rate and the capacity separately. This is why flow batteries should work out a fraction of the cost even at todays price.

[epistasis]

I would rate the reliability of a battery press release slightly higher than a signed-off budget and timeline for nuclear construction.

And when it's press from a battery maker, rather than a battery researcher's university press department, the battery press release is far far more reliable. I'd say their cost overruns will be at most 20%-30%, compared to nuclear's usual 200%-300%.