Hacker News Reader

Comment by aeleos | Hacker News Reader

I like the idea, but I am wondering how big these will need to be. If one of their prototype towers can store 20 MwH, at each one requires a circle 300 feet in diameter, than it will require a lot of land to do this.

I wonder how this compares to lithium on a power store per square foot metric.

View on HN

They can store (not generate) 20 MWh at, according to the article, a cost of $2M. The article does not discuss how much power they can produce using one unit. That means it would cost roughly a trillion dollars to build enough storage to power the entire US for a day [1]. That's a lot, but you wouldn't need nearly that much, since daily consumption is still roughly balanced to production. You could think of that as an upper bound on the cost of backing the entire grid with this tech, assuming their numbers scale perfectly and are entirely accurate.

Almost everywhere has a lot of unused land, even in densely packed countries like Japan. If it's not good for farming, forestry, or leisure, it would be suitable for this sort of use.

[1]

- 3.8e12 kWh/y electrical use in the US.

- 1e10 kWh/d

- 1e10kWh / 2e4 kWh/unit = 5e5 units

- $2e6 / unit * 5e5 units = $1e12

Thanks, yea I meant store. Fixed.

They spent $2M on the demonstration plant, which is ~1/10th the size. It doesn't store the full 20 MWh. (Or is that 35 MWh? The article seems inconsistent on that.)

The bad news is, the capacity is lower, but the good news is the $2M is the cost to build the first prototype, and prototypes can be extremely expensive compared to the real manufacturing costs.

It seems like the only information available about the price is their estimate that they can hit $150/kWh.

Reply on news.ycombinator.com