The SA gov investment is $15M and there is an additional $8M from the federal government[1].

The $150M savings is a little unclear - [1] gives is at $40M/year for 2 years.

Those savings are to consumers and businesses in the state, not to the government itself and there are a number of additional clean energy projects underway in SA, so the maths on the appropriate amount to invest in each one is difficult to calculate.

In particular, batteries save money by removing the (very high) peaks. They are important because they reduce the effectiveness of market manipulation which SA was suffering from. But increased diversity in power generation within SA and an upcoming second interstate interconnector[2] help with this too.

> what other car company has market trading software as part of its energy division?

I think every single energy division has energy trading software? It would be astounding if they didn't - I worked for a power company last century for a little while and they had energy trading software back then.

[1] https://www.abc.net.au/news/2019-11-19/sa-big-battery-set-to...

[2]

This is good news but expansion of only 50% is underwhelming and $15m is small potatoes compared to other energy projects. Ideally they'd be significantly expanding this or building whole new plants.

They're building one in Ventura, CA that is about 3 times bigger than the one in south australia. It will be 100 MW/400 MWh whereas the Hornsdale power reserve is 100MW/129MWh currently and will be 150MW / 129MWh after this update.

Then there is the absolutely monstrous PG&E Tesla battery at Moss Landing (also in California) which will start as 182.5MW / 730MWh, but is forecast to eventually grow to 1.1GWh of energy.

A contract for a "smaller" one with Hawaii Electric that works out to 135MW / 185MWh was just signed in March with Tesla.

etc.

Perfect. Glad to hear they are legitimately expanding and trying out the model in the US.

Maybe diminishing returns?

This is likely it, though I don't know the particulars of this grid. In most grids, there's a small amount of super profitable frequency regulation to be performed. However, there's very limited capacity needed for that. Doing energy arbitrage over time is much more difficult to make profitable, usually. A huge amount of new solar and wind in the US will be built with batteries attached in the next five years, because it's easier to make energy arbitrage work when you're an energy producer (and also make the tax credits for storage work, which typically require that the batter is only charged from zero carbon electricity).

I remember there was an analysis somewhere that suggested we need batteries at 20$/kwh to achieve 100% renewable production. But we only need 150 $/kwh to achieve 80% renewable. So there are probably low hanging fruit to get right now. Hopefully all this extra production will drive battery prices down.

Nah, the state’s broke

What other car company has a legitimate energy division at all?

Mitsubishi?

Fair. Mitsubishi Heavy Industries develops just about everything.