Tesla Switches on Giant Battery To Shore Up Australia's Grid (reuters.com)

Tesla switched on the world's biggest lithium ion battery on Friday in time to feed Australia's shaky power grid for the first day of summer, meeting a promise by Elon Musk to build it in 100 days or give it free. From a report: "South Australia is now leading the world in dispatchable renewable energy," state Premier Jay Weatherill said at the official launch at the Hornsdale wind farm, owned by private French firm Neoen. Tesla won a bid in July to build the 129-megawatt hour battery for South Australia, which expanded in wind power far quicker than the rest of the country, but has suffered a string of blackouts over the past 18 months. In a politically charged debate, opponents of the state's renewables push have argued that the battery is a "Hollywood solution" in a country that still relies on fossil fuels, mainly coal, for two-thirds of its electricity.

Comments

Special Solution for a Special Problem

[careysub]

South Australia (and Australia generally) is a special case for renewable energy since it is a small continent, and sparsely inhabited.

This is a fix for a remote corner in Australia, the edge of the 5th largest population center (Adelaide*) separated from it by 100 miles and isolated by hundreds of miles of emptiness from anywhere else. There is little redundant/backup infrastructure, or all that many people.

More generally battery facilities shouldn't be needed in larger, more populous continents (North America, Eurasia).

The solution to issues of variable power production is to connect the entire continent together with high voltage DC power lines (a nearly century old technology) which can ship power from one coast of North America to the other with losses of under 5%. You build enough excess solar and wind capacity that even under the worst conditions you still have enough for the entire continent (Canada and Mexico should be part of this grid also).

This also allows using the sun out west to power the evening peak back east, and so forth, leveling out production/consumption mismatches.

Pumped storage can service the entire grid since power can be transported long distances. The U.S. currently has enough pumped storage on-line to provide 2.2% of US grid capacity (and about twice this much more has been licensed), so it can be sited where ever geography makes it most convenient.

We need some national-level vision to help bring this about (good luck with that at present), but mostly this can be done by private investment.

*The greater metropolitan area of Adelaide has a population of 1,317,000 which is 77% of the entire population of South Australia (which is 50% larger than Texas). Things get really sparse really fast out past Adelaide's metro area.

Re:Special Solution for a Special Problem

[tlhIngan]

Why HVDC? DC is a bitch to convert between voltage and current, and it (generally) is more dangerous at any given voltage, though to be fair at transmission voltage levels it makes little difference if you're a DC or AC flavored charcoal lump.

If your grid is disparate, then getting them into synchronicity can be a pain. With modern semiconductors, it is however possible to rectify and invert DC into AC quite painlessly.

This is used for grids that have historically never been tied together, as well as new grids which never were synchronized. In Texas, there's a grid intertie that connects the three major US and Canadian grids together so power imbalances can be dealt with. But trying to synchronize the grids is a next to impossible problem, so the intertie uses HVDC internally so it's able to move power between the grids as necessary.

I believe China has a HVDC distribution network for the same reason - too many little grids to synchronize up.

HVDC systems do have lots of advantages over traditional AC systems.

Re:Special Solution for a Special Problem

[Mr D from 63]

South Australia (and Australia generally) is a special case for renewable energy since it is a small continent, and sparsely inhabited.

This is a fix for a remote corner in Australia, the edge of the 5th largest population center (Adelaide*) separated from it by 100 miles and isolated by hundreds of miles of emptiness from anywhere else. There is little redundant/backup infrastructure, or all that many people.

More generally battery facilities shouldn't be needed in larger, more populous continents (North America, Eurasia).

Many folks may not fully appreciate the primary function of these batteries. It is not to levelize renewables, but rather to provide fast response to prevent overloads and voltage/frequency support when there is a sudden event on the system. That is because, as you say, they rely on a small number of lines and therefore don't have the networked/redundancy to maintain reliability. Adding new transmission lines for these long distances is expensive.

With the batteries, if there is a transient event on the grid the hope is they provide voltage and frequency support to ride it through without some overload on a major line. Now that it is operational, it will be interesting to see how well that works and how often that support is needed.

One important factor to note, when batteries need to be available for this type of support they must retain a certain percentage of capacity. They can also use them for renewable levelization or peaking support, but they don't want to discharge them too much or they may not be able to supply adequate voltage/frequency support when called upon. Full discharge/recharge cycles will likely not happen often.

In other words

[Anonymous Coward]

Australia was so quick to go green with renewables, they didn't realize that it's called intermittent generation for a reason. Once they retired their reliable fossil generation, they had blackouts because the wind is only consistent on average, and now they need an even more expensive solution to shore up the wind energy output.

Re: In other words

[Rei]

Indeed. Look at the stats on the battery: 100MW, yet only 129MWh. This isn't storing power for long periods of time; it's providing a power-plant level of power until power plant outputs can be ramped up elsewhere. Hence it physically cannot be used as a "storage for when the wind doesn't blow" solution (unless the wind is only stopped for an hour or so). Tesla does make longer-term storage solutions too (it's done that for solar in a number of places), but that's not what this battery is.

Re:Elon Musk

[Rei]

Sure he did - he made a bunch of Slashdotters (and "experts") look silly ;)

The proposed system would contain 129,000 kilowatt-hours of capacity, meaning the project's cost would start at around $42 million. The head of Tesla's battery division has quoted a cost of about $65 million in the past. Other experts say a system of that size is likely to cost somewhere between $60 and $120 million.

(It cost $50M. And judging from Semi battery prices, if they were to do it again late next year, it'd be a small fraction of that much)

Hope they have a site ready for him, leveling and concreting a section of land can take years. Yeah, yeah, I know Aussie is flat, but not that flat.

Link

Getting them there is only a small part of the problem. The real issue is that Australia can throw lots of roadblocks in Elon's way, from customs to building permits. And there is a hell of a difference between delivering enough batteries in the stated time and building up a power system to use them. I think Musk's ego go the better of him here and he shot off his mouth too fast. Betting that you can do something in 100 days or it is free against the very people who can block you at every move isn't the smartest thing to do.

That's the rub. He hasn't even clearly defined what problem he will solve. The recent articles describe the symptoms of a more complex problem. Musk is proposing to alleviate the local symptoms possibly, not necessarily the core systemic problem which is continuously evolving. And of course he needs to tell them how much it will cost to 'solve' the problem.

Australia clearly won't fall for his rhetoric.

A container ship can cross the pacific in 2-4 weeks so that's not a big deal. Lead time would be a serious problem though for his 100 day boast. Presume it takes 20 days to transport the batteries and maybe another 30-40 to build them all (probably optimistic), they would be left with maybe a month to design, install and test the whole thing. Not saying it would be impossible but it would be a tight squeeze most likely unless he has already built the batteries and designed the system. He could probably get it up and running quickly but perhaps not at full capacity.

Reminder: it was not only done 99 days from the bet, but only 55 days from the contract signing ;)

It's not just the batteries that are the issue, unless Elon is only saying that the batteries will be there within 100 days. Utility scale electrical equipment is generally bespoke and/or custom manufactured and has long lead times. On top of the batteries you need to have utility scale inverters. You need switchgear to direct the produced power, you need large transformers to boost it up to utility voltages. In the organization I work with, we just finished a major electrical upgrade where we purchased 12 pad-mount transformers. While they were out of the catalogue, the lead time before shipping was still 8 weeks, and that was from a major manufacturer. Really big transformers, those capable of working with megawatts of power, take months to manufacture.

When Vancouver lost one of the two large transformers that supply the downtown part of the city, it was found that it would take 18 months to get a replacement manufactured.

I'm quite sure getting all the permits takes longer than 100 days.
Where can I sign the contract for a free 100MW battery storage system.

Given that it took 1.125 days per MWh for Los Angeles, I wo