Grid Batteries, From Hornsdale in South Australia to California's Evening Peak

The bet was made on Twitter in March 2017, which is part of what made it memorable. South Australia had suffered a statewide blackout the previous September, when violent storms toppled transmission towers and a cascade of failures left the entire state dark. Politicians in Canberra blamed the state's heavy reliance on wind power. A software billionaire in Sydney, Mike Cannon-Brookes, asked Elon Musk publicly how serious he was about fixing the problem with batteries. Musk replied that Tesla would install a large battery system within 100 days of a signed contract, or it would be free. The state government took him up on it. That December, beside a wind farm near the town of Jamestown, a few hours' drive north of Adelaide, the Hornsdale Power Reserve began operating. It was rated at 100 megawatts, then the largest lithium-ion battery on any grid in the world. Its owner was the French renewable energy company Neoen, Tesla supplied the batteries, and it was finished inside the deadline. The stunt got headlines. What the battery did over the next few years, quietly, mattered more.

In short: Hornsdale proved that a big battery could earn its keep by stabilizing the grid in fractions of a second and cutting the cost of backup services. California then showed the bigger use: soaking up midday solar and releasing it in the evening, when demand peaks and the sun is gone. Batteries now reshape the daily operation of grids, but they cover hours, not seasons, and they face fire, supply chain and policy risks.

What a battery does that a power plant cannot

A power grid has to balance supply and demand every instant. When a big generator trips offline, the frequency of the grid starts to fall within a second or two. Traditional plants with heavy spinning turbines resist that change naturally, and some are paid to ramp up quickly to fill the gap. But even the fastest gas plant takes many seconds or minutes to respond fully.

A battery responds in milliseconds. Its power electronics can inject or absorb electricity almost instantly, and it can do so with great precision. That makes it extraordinarily good at the services grid operators call frequency control: keeping the grid steady through the small, constant wobbles of supply and demand and catching it during sudden failures.

Hornsdale demonstrated this early. When a coal unit in another state tripped offline unexpectedly within weeks of the battery's opening, the battery responded faster than the conventional plants contracted for that job. In its first years, the market operators and independent analysts found that the battery helped cut the cost of frequency control services in South Australia significantly, partly by competing with gas generators that had been charging high prices when they were the only option. The project was later expanded, adding capacity, and was used to test a feature called synthetic inertia, which lets batteries imitate the stabilizing effect of spinning machines.

South Australia, by then running on a very high share of wind and solar, used Hornsdale and the batteries that followed it to show that a grid could operate with less fossil generation than engineers had thought prudent.

The duck that ate California's evenings

On the other side of the Pacific, California's grid operator was worried about a chart. Around 2013, its planners published a projection of the net load on a typical spring day, meaning total demand minus what wind and solar produce. As more solar came online, the net load sagged deeply in the middle of the day and then climbed steeply after sunset, when people come home, turn on lights and appliances, and the panels stop producing. Drawn out, the curve looked like the profile of a duck, with a fat belly at midday and a long neck rising into the evening. The name stuck.

The duck curve described two problems. At midday, California often had more solar than it could use, so it curtailed panels or exported power at low or even negative prices. In the evening, it needed a huge amount of power to come online very quickly, which meant keeping gas plants ready. Without some way to move energy from noon to seven in the evening, every new solar farm made the evening ramp steeper.

Batteries are almost perfectly suited to that job. The standard grid battery that California built in large numbers holds about four hours of energy at full power. It charges in the middle of the day when solar is abundant and cheap, and it discharges into the evening peak.

The buildout

California's battery fleet was close to nonexistent in the late 2010s. State regulators ordered utilities to procure storage, first through a 2010 law and a 2013 mandate, then far more aggressively after the August 2020 heat wave, when the state ordered rotating outages for the first time in nearly two decades as evening demand outran supply. In response, the state utility commission ordered large purchases of new resources, and most of what got built was solar paired with batteries or standalone batteries.

The growth since then has been steep. By the mid-2020s California had many gigawatts of batteries on its grid, and on many evenings batteries became the single largest source of power for a stretch of hours after sunset, discharging energy captured from solar earlier in the day. The evening ramp that once depended mainly on gas plants now depends heavily on batteries. Texas, too, built batteries quickly, mostly for the high prices its market pays during peaks and emergencies, and its batteries helped carry the grid through record summer demand.

The federal climate law of 2022 helped by allowing standalone storage to claim the investment tax credit for the first time, which had previously been available mainly for storage paired with solar. Falling battery prices, driven largely by the electric car industry, did the rest.

How a battery pays for itself

A grid battery is a business, and understanding how it earns money explains a lot about where batteries get built.

The first income stream is the simplest: buy low, sell high. A battery charges when wholesale electricity is cheap, often in the middle of a sunny day, and discharges when it is expensive, usually in the evening. Traders call this arbitrage. The larger the gap between midday and evening prices, the more a battery earns, which is why batteries crowd into markets with lots of solar.

The second stream is grid services. Grid operators pay resources to stand ready to stabilize frequency, to ramp quickly, or to hold power in reserve. Hornsdale's early profits came largely from this kind of work. These markets are relatively small, so the first batteries to enter them can do very well, and later arrivals find the prices competed down.

The third is capacity. In California, utilities and other electricity sellers must show they have contracted enough resources to meet peak demand plus a margin. Batteries can count toward that obligation, and many California batteries were built under long-term contracts for exactly that purpose. Those contracts give developers steady revenue that banks will lend against.

The mix matters for policy. A battery that lives on volatile arbitrage profits is a riskier loan than one with a long contract. States that want storage built at scale tend to create the long contracts, through procurement orders or capacity rules, rather than hoping market prices alone will do it.

Batteries in the garage

Not every battery sits in a fenced lot beside a substation. Home batteries, usually paired with rooftop solar, have spread quickly, and utilities have started to treat them as a grid resource.

Vermont offers an early example. Its largest utility, Green Mountain Power, began offering customers home batteries several years ago in exchange for permission to draw on them during peak hours, when power is most expensive. When thousands of batteries discharge together, they act like a power plant, a setup often called a virtual power plant. The utility avoids buying expensive peak power, and the homeowner gets backup power during outages, which in rural Vermont after a storm can mean days.

California and other states have run similar programs, paying households to let their batteries discharge during heat waves. The appeal is obvious: storage that is already built and paid for, spread across neighborhoods, can be called on when the grid is strained. The fairness question is just as obvious. Households that can afford solar and batteries get the backup power and the payments. Programs aimed at low-income households, community centers that double as cooling and warming shelters, and multifamily buildings are how the benefits reach people who need them most.

What batteries cannot do

It is worth being clear about the limits, because the enthusiasm sometimes runs ahead of the physics.

Hours, not seasons. A four-hour battery moves energy from afternoon to evening. It does not store summer sun for a cold windless week in January. For that kind of gap, grids still need other resources: hydropower, long-distance transmission to places with different weather, firm low-carbon generation, or long-duration storage technologies, many of which are still early.

Fire risk. In January 2025, a fire broke out at the large battery facility at Moss Landing on California's Monterey Bay, one of the biggest in the world, inside a building that housed lithium-ion modules. Nearby residents were told to evacuate, and the fire smoldered for days. Newer projects mostly use outdoor container designs with spacing and fire suppression meant to stop a fire from spreading, and many use lithium iron phosphate chemistry, which is less prone to runaway heating. But the Moss Landing fire made local opposition to battery projects stronger in several places, and it is reasonable for communities to ask hard questions about siting, monitoring and emergency plans.

Supply chains. China dominates the manufacturing of battery cells and many of the materials that go into them. U.S. policy has tried, through tariffs and tax-credit rules, to shift production to American and allied factories. That is a defensible goal, but it raises costs in the short term, and its success depends on policy staying stable long enough for factories to be built and run.

Market design. Batteries make money by buying low and selling high. As more batteries enter a market, they compete away the very price spreads that pay for them. That is good for consumers and part of how storage lowers costs, but it means investment depends on market rules that reward the services batteries provide.

The political lesson of a bet

The Hornsdale story is often told as a tech billionaire's triumph. It is more interesting as a story about what a state government did. South Australia's government, facing ridicule after the blackout, chose to back a new technology rather than retreat to what it knew. It signed the contract, accepted the risk and let the results speak. When those results came in, other Australian states and grids around the world followed.

California's story is similar. Its regulators set procurement targets before batteries were cheap and kept raising them as prices fell. Utilities and community energy programs signed contracts. The 2020 blackouts were embarrassing, and the response was to build, quickly, rather than only to study.

Australia, for its part, kept going. After Hornsdale, its states and the national government backed many more big batteries, and the country's grid operator began planning for periods when wind, solar and batteries alone could run large parts of the system for hours at a time. A battery that started as a dare became, within a few years, a routine piece of infrastructure, the kind of thing nobody tweets about anymore. That ordinariness is the real measure of success.

There is a lesson here that cuts against some progressive habits. Batteries got built at speed in California partly because many projects went up on land already used for power plants or industry, with connections already in place, and partly because the state treated reliability as an emergency. Where projects have stalled, it has often been in long local permitting fights and grid interconnection queues. Building clean infrastructure at scale requires willingness to say yes, with safety rules that are clear and enforced.

It is also a reminder of who the grid serves. The 2020 outages and the evening peaks they exposed fall hardest on people without air conditioning, on outdoor workers who come home to hot apartments, and on older residents for whom heat is dangerous. A battery that keeps an evening from going dark during a heat wave is a public health tool as much as a climate tool.

What you can do

Look at how your state's power is changing. The site's climate data page shows the U.S. power mix and clean power by state, including how much comes from solar and wind, the resources storage makes more useful.

If a battery project is proposed near you, take the questions seriously and ask them early: what chemistry, what fire suppression design, what plan the local fire department has reviewed, and what monitoring will be reported publicly. Thoughtful local questions lead to safer projects; reflexive opposition leads to projects being built where people have less power to object.

When federal agencies take public comment on energy rules, from grid interconnection to tax-credit guidance, comment. The list of open federal comment periods shows what is closing soonest, grouped by issue.

Tell your state legislators and utility commissioners that storage targets and clear safety standards belong together. You can find your state and federal representatives by address. The climate policy page follows storage, transmission and the rest of the grid fight.