Inside the Tesla big battery: How it made money and cut prices | RenewEconomy

Inside the Tesla big battery: How it made money and cut prices

AEMO reports detail how Tesla big battery helped keep the lights on last summer, smashed prices in specialised markets, and made money for itself from price arbitrage.


The Australian Energy Market Operator has provided further detail on how the Tesla big battery in South Australia pitched in to help to cut prices in crucial electricity markets and deliver an incident-free summer for the National Electricity Market.

The report from AEMO, discussed in its special review of the last summer, and in its latest quarterly market report, delivers the most detailed description to date of how the world’s biggest lithium-ion battery has performed since its opening last December.

It also provides unique insight into the battery’s own performance, its ability to slash prices by breaking open the gas cartel that controlled prices in specialised markets such as grid services, and how it helped keep the lights on and reshape thinking about grid operations.

AEMO is clearly impressed, and is already looking forward to more batteries joining the grid before next summer. Another three batteries – one in South Australia, and two in Victoria – are scheduled to join by then.

“We were watching the Tesla battery with interest, particularly over the high demand periods, and we were quite pleased with what we saw,” says AEMO head of operations Damien Sanford.

He said he was impressed with its ability to “flex” and provide services when needed. “You know that the reserve is there, and that it can inject it very quickly.”

Some Tesla big battery services, such as its super fast response to system faults, are not being remunerated because the market does not value such speed and accuracy – and won’t until rules are changed.

But the battery’s other services, such as arbitrage, are making money.

This graph below shows in great clarity, for the first time, how the Tesla battery has managed to make money through arbitrage, charging when prices are low and selling when prices are high.

According to AEMO, Neoen’s Hornsdale Power Reserve, as the battery is officially known, was dispatched as a load in the SA market for 38 per cent of the December 2017 to March 2018 quarter, for a total of 11GWh .

It was dispatched as a generator (discharging) for 32 per cent of the quarter, discharging a total of 8.9GWh – mostly in the late afternoon, to correspond with higher energy prices.

The graph above shows the average daily dispatch, showing that the load, or charging, is clearly aggregated around cheaper prices in the morning, and the discharging, or generation, is clearly gathered around the afternoon and evening peaks.

As you can see in the graph below, the comparison of the average charge and discharge price reveals an average price arbitrage of $90.56/MWh.

AEMO says this spread was “in a large part due to three days of price volatility in South Australia” – January 18 and 19, and February 7 – during which time the price settled above $5,000/MWh for nine trading intervals.

Most attention up until now has been on the battery’s impact on so-called ancillary markets, which provide essential grid services such as frequency control, and are known in the industry as FCAS.

Earlier this month, we brought you a story about the stunning impacts of the battery in assessments done by McKinsey & Co analyst Godart van Gendt.

The AEMO reports confirm that assessment, even if some of the numbers are slightly different, and help to put paid to Coalition jibes that Tesla’s big battery is too small to do anything of any substance in Australia’s grid.

Conservatives have sought to demonise the Tesla big battery. Senior ministers such as Treasurer Scott Morrison said it was about as useless as the “Big Banana” and Resources minister Matt Canavan has described it as “Kim Kardashian” of the energy industry.

But as AEMO’s Quarterly Energy Dynamics report notes, the introduction of the battery and demand-side response has worked to drive down frequency control ancillary services markets by 57 per cent below than the final quarter of 2017.

“In Q1 2018, FCAS costs were $25 million, representing a $32.7 million (57%) decrease on Q4 2017 levels,” the report says, referring to the table on the left (above).

“AEMO’s FCAS requirements were steady compared to Q4 2017, so decreases in FCAS prices were the main driver of lower costs.”

And driving those lower FCAS prices, the report explains, was the additional supply from two new participants: Hornsdale Power Reserve and demand response from EnerNOC.

“These participants provided FCAS supply from a large-scale battery and aggregated demand response, representing an Australian first for these technologies,” the report says.

“Noticeably, during the first quarter of 2018, these new technologies captured a larger share of FCAS markets … displacing higher-priced supply from existing technologies (largely coal), as shown in figure 17.

“In addition, increased competition through two new FCAS providers has coincided with a reduction in the price of offers from some existing providers.”

As AEMO explains it, the HPR managed to provide regulation and contingency FCAS for 71 per cent and 99 per cent of the time, respectively.

It also provided regulation FCAS to South Australia during the activation of a special constraints that require a minimum of 35MW to be made available on January 14 and March 8.

Regulation FCAS prices in SA during periods of constraint have typically exceeded $9,000/MWh – due to the limited number of suppliers of these services, as we have documented on numerous occasions.

But, with the big battery in the mix and breaking the gas market cartel, average Raise and Lower Regulation prices were just $248/MWh during the January 14 event.

In that instance, AEMO estimates that the big battery reduced the cost of regulation services by about $3.5 million during the five-hour period in which the constraint bound.

The Tesla big battery is also being financially supported through a contract from the state government to provide grid security services in case of an emergency. Those services are yet to be invoked.

The battery is also being used by AEMO as a front-line defence against any catastrophic outcome should the interconnector that links South Australia with Victoria suddenly fail.

Batteries are likely to need revenue from several different serves – including the ability to create micro-grids – to deliver a return on investment.

Mostly they will require the development of new rules within the markets that recognise and value their superior speed, accuracy and flexibility.

In a statement accompanying the two new reports, AEMO chief Audrey Zibelman said the data illustrated the important role considered planning and collaboration across governments and industry played in delivering a secure, reliable and cost-efficient grid.

“While the hotter than usual 2017-18 summer posed significant challenges from increased demand and risks of failure of generation and transmission assets, the power system held up well,” she said.

“We recognise that there are still concerns about pricing but our analysis on the energy markets shows a downward trend on wholesale electricity prices and FCAS costs on average,” she said.

“We are pleased to have created a framework for the use of new technologies that ensures power system security and simultaneously encourages competition.

“In this instance, the entrance of two new FCAS providers introduced competition and put downward pressure on ancillary services prices, which ultimately benefits the consumer,” she said.

Print Friendly, PDF & Email

  1. Electric Boogaloo 3 years ago

    So it made about $800,000 from arbitrage and FCAS prices are almost back to where they were before the closure of Pt Augusta’s power station?

    A waste of money, in other words.

    • Ren Stimpy 3 years ago

      Much needed competition, sending a message to the oversized, centralised, public money built generators which then get privatised and act as monopolies and cartels – get lost soon and don’t come back.

      • john 3 years ago

        As more and more wind, solar PHES and battery storage is built the wholesale price must go down which will further eat into the slim pickings for the 20 odd coal fired stations.
        One after the other they will close.

    • john 3 years ago

      The article states
      Regulation FCAS prices in SA during periods of constraint have typically exceeded $9,000/MWh – due to the limited number of suppliers of these services, as we have documented on numerous occasions.

      But, with the big battery in the mix and breaking the gas market cartel, average Raise and Lower Regulation prices were just $248/MWh during the January 14 event.
      It would appear the outcome is positive for the end user.

      • Electric Boogaloo 3 years ago

        The SA government dropped way more money on this than what it could have on assisting the Pt Augusta power station to remain operational. Add the cost of increased electricity prices on and it’s even worse.

        While the battery may be doing its job, the whole situation is a public policy disaster.

        • Ren Stimpy 3 years ago

          By Pt Augusta station, do you mean the old decrepit maintenance-starved privatised coal station that, like all the others in our nation, is soon due to retire anyway?

          • Rod 3 years ago

            That would be the one. The fact that they only had about 3 years worth of relatively easy to get to coal left also seems to get lost along with the remediation costs.
            He is suggesting the SA Govt should have done what the Coal ash forum are suggesting.

            Waste taxpayer money on a decrepit Coal Clunker. Thankfully, Jay said “no thanks”.

          • Electric Boogaloo 3 years ago

            Three years? More like at least a decade worth.

            You don’t have to take my word for it though as Alinta stated that in the Leigh Creek Mine Closure Plan it submitted to the SA government.

          • Rod 3 years ago

            This forum allows links. Feel free to provide one.

          • Electric Boogaloo 3 years ago

            Because it’s so hard to search for “Leigh Creek Mine Closure Plan”…

            The main body of the report is this 32MB PDF:

            Since you seem to want me to do all the work for you, refer to Table 13 on Page 96: 43.1MT reserves in mid-2014, from a mine that was producing around 3MT/yr.

          • Barri Mundee 3 years ago

            One of the trolls who pops up under various names from time to time and whose only aim is to shitcan anything renewable.

          • Ren Stimpy 3 years ago

            I’m for free speech – trolls, nut jobs, anybody. Come as you are!

          • Barri Mundee 3 years ago

            Sure and I am too but they do not deserve to be taken seriously.

          • Ren Stimpy 3 years ago

            The populist parties count these people as fodder and these folk lap up the PHONey message because WE don’t take the time to engage with them across an education barrier. Let them have their say on these forums I say. We need to engage with them without patronising them. WE might learn something!

          • Electric Boogaloo 3 years ago

            I don’t use other pseudonyms and saying something that you disagree with does not make one a troll.

          • Electric Boogaloo 3 years ago

            Decrepit? No, the power station that was built in the 1980s and was just over half-way through its lifespan with at least a decade of coal reserves available.

            The very same one that was closed because its operator said that it was undermined by excessive wind generation and which would still be running if AEMO hadn’t been asleep at the wheel and had the currently implemented rules to maintain stability of the SA grid in place.

          • Ren Stimpy 3 years ago

            The 1980s was also the decade that brought us MacGyver and the mullet. Just saying.


          • BushAxe 3 years ago

            Northern would have closed at some point all it did was postpone TIPS A from being mothballed. The increased capacity of the Heywood interconnector killed off Northern’s ability to compete for industrial contracts. Alinta also had the choice to build a replacement gas plant but chose not to (although it’s now looking at building one at Templers).

        • Rod 3 years ago

          $700 million you mean? Or are you conveniently forgetting the remediation costs.
          $700 million for at best 3 years extension doesn’t sound very smart to me. Public policy disaster only for those idiotic enough to believe the crap printed in a Mudrake newsrag.

          • Electric Boogaloo 3 years ago

            Where are you pulling $700 million from?

            Alinta asked for $25 million over three years, primarily for maintenance work on the Leigh Creek rail line.

          • Rod 3 years ago

            Find the document. Not the redacted version the advertiser printed,
            Hazelwood’s remediation costs are now in excess of 700 million.
            Northern would be around 500 million and they are yet to find a solution for the fly ash.
            Accepting the “offer” would have meant SA taxpayers being saddled with the clean up.

          • Electric Boogaloo 3 years ago

            You’re working from the assumption that a site cleanup is even necessary. The fly ash issue was dealt with by capping the dry pondage with soil.

          • Rod 3 years ago

            I’m sure the residents of Pt Augusta will be pleased the fly ash issue is sorted. Oh and look even a mention of the Pt Augusta clean up costs. Add in Leigh creek costs and…………..


          • Electric Boogaloo 3 years ago

            Didn’t you read the article you linked? It clearly states that the dry pondage has been capped with soil like I said.

            Interesting how you haven’t replied to my comment pointing you at the specific table in the document from which I sourced my statement about there being a decade worth of coal reserves at Leigh Creek.

            Can you point to anything that backs your claim about there only being a few years worth of reserves?

          • Rod 3 years ago

            Didn’t YOU read the article? The vegetation planted is too sparse and the lack of rain means the dust is still blowing. This issue is not going away.
            Re the coal: I can’t find any proof, but from discussions with someone elsewhere, the easy to reach, higher quality coal had about 3 years worth left. Moot point anyway, the coal extraction and carriage costs meant they could not compete with Vic imports.

        • MikeH 3 years ago

          Dream on. Your comments are a coal booster’s fantasy.

          Alinta closed Northern because in the words of its CEO

          >”The reality is, the technology we are using here is old, the cost structures are high and there’s no longer a place for us in the market,” Mr Dimery said.

          They offered the plant to the SA government but as a ruse to escape clean up costs.

          >SA Energy Minister Tom Koutsantonis said despite the apparent free deal, the Government would have taken on huge costs.

          >”Alinta would have walked away without having to pay any of the money for the clean-up of the mess that they had incurred and legacy liability they had taken on which is worth hundreds of millions of dollars,” he said.

    • MacNordic 3 years ago

      They are not:
      Quote from the article above: “…In that instance, AEMO estimates that the big battery reduced the cost of regulation services by about $3.5 million during the five-hour period in which the constraint bound….”

      Add the additional events and the arbitrage in& you have far lower prices for all.

      If you have a cat in the house, it is rather unlikely the mice will keep dancing on the table for long… Unless you have a Garfield, that is. Wouldn´t call HPR a Garfield, though. Bit too nimble…

      • Electric Boogaloo 3 years ago

        According to the graph of FCAS prices in the article, they are.

        The evidence is clear: the cost of SA retaining the Pt Augusta power station was significantly less than the costs incurred by SA since in both elevated electricity prices and government spending to fix the problem it created. To state otherwise is absurd and is indicative of innumeracy.

        • Rod 3 years ago

          Elevated prices. Now what would the cause be? Privatisation apparently cost SA 2 Billion.
          To state otherwise is absurd and is indicative of innumeracy and idiocy.

        • DJR96 3 years ago

          Mate, no point lamenting the past. Look forward.

          Do you really think power prices would have been any cheaper in SA if Northern had somehow continued operation.

          No point throwing money at a short term fix on something that only had a matter of years left in it anyway. Far better to invest in longer term solutions that will better suit the system into the future.

          • Electric Boogaloo 3 years ago

            The problem with looking forward without “lamenting the past” is that if we don’t learn from history then we are doomed to repeat it.

            Yes, power prices would have been cheaper. It is indisputable. Only someone who refuses to look at the past could say otherwise.

        • MacNordic 3 years ago

          Whoa, mate.
          May I request you to stay courteous, please.

          Both charts 16 and 20 show a significant lower price post- HPR power up than the year before. YoY comparison shows that rather well…
          If you would like to compare to 2016, you will notice the upward trend in the run- up to the closure of Northern in May, with the FCAS spending in Q2 16 being on par with Q1 18.
          As these are still early days, one would have to wait for the first year of HPR operation to be complete for any robust evaluation.

          As lack of competition (nearly) always drives up prices, the introduction of more providers will likely do wonders for the price level – well into the future, not just for the near term, regardless if that may mean two or ten years (the varying statements of remaining coal reserves for Leigh Creek).

          Taking old power stations or industrial sites onto the books is nearly never a good idea- unless you can securely shift the cleanup- cost to somebody else. The real end cost, that is.
          Just ask around in the UK or Germany. Or in the US with their Superfund sites…

          • Electric Boogaloo 3 years ago

            Pointing out reality is discourteous?

            A whole lot of waffle, but nothing to rebut my claim…

          • neroden 2 years ago

            The US has laws obligating *every previous owner* to pay for Superfund sites. The only escape is bankruptcy. Unfortunately, a lot of the sites were owned by bankrupt companies. Or by the US government directly, in the case of the nuclear sites which are some of the most horrible and expensive (Hanford, Savannah River, Idaho, etc.)

          • MacNordic 2 years ago

            Thanks for the info – didn´t know that tidbit!

  2. Ren Stimpy 3 years ago

    It looks like if the $/MWh price feedback to the battery’s operator was faster, the arbitrage value would be better utilised, as the battery’s generator mode could be better matched to the price line (red line in Fig. 18). The move to a 5 minute settlement period can’t come soon enough.

    • John Smith 3 years ago

      Fig. 18 is showing the aggregated behaviors over the whole quarter. So, while it *looks* like the battery may be missing out on fully capitalizing on the highest price peaks, that’s likely just an artifact of the fact that the price peaks may not exactly correspond to the bulk of average daily experience. Super peaks in price will distort the “average price” but the battery can only ever discharge at its maximum power. So, multiple days of 40MWh at the normally higher afternoon price of, say $200/MWh swamps one day of the same 40MWh at $14,000/MWh. But the average price graph will show a big spike.

      The point about responsiveness and settlement is still true, it’s just that this data/graph isn’t the best representation for showing that. 2021 is too far away. 🙁

  3. rob 3 years ago

    is ren stimpy back?

    • Ren Stimpy 3 years ago

      Always been. Is rob back?

  4. ashentegra 3 years ago

    The clear advantages of wind + battery make building anything else – including solar – an exercise in foolishness at this time. The economics are crystal clear.

    Onshore wind is the cheapest electricity generator. It’s progress in improvement in production efficiency continues to match solar generation and maintain it’s cost advantage (See: Wright’s Law). GE’s latest turbine generates 4.8 megawatts and has a 158-meter rotor on a tower 240m tall. It is enormous, designed to catch the stronger, more consistent wind higher in the air, making it so much more productive and efficient than smaller designs. One would be crazy to settle for less.

    Wind + Battery can time-shift midnight to dawn generation that is currently virtually given away and sell at peak late-afternoon consumption and prices. Solar + Battery can also time-shift from, say, peak-sun 10am to 2pm surplus production, to late afternoon/evening peak consumption as well. But S+B arbitrage benefits are minimal compared to W+B. Peak-sun is a time with reasonable prices – adding a battery to solar to pursue peak prices lacks the very major uplift in prices W+B enjoys. It would not be hard for W+B to erase even the arbitrage from peak-sun to peak consumption and eat S+B’s lunch. See fig 18 above.

    The war on renewable energy has been won. Both solar and wind (+ battery) have the capacity to prematurely close all Australia’s fossil fuel generating kit, simply on price.

    We let a thousand flowers grow in seeking a solution to climate change and air pollution. Now it is known: W+B trumps S+B.


    Disclosure: IFN.ASX

Comments are closed.

Get up to 3 quotes from pre-vetted solar (and battery) installers.