Battery storage booming, but even Tesla struggling to cash in

Battery storage booming, but even Tesla struggling to cash in

Print Friendly, PDF & Email

Grid-scale battery storage is taking off in Australia but it is still a struggle for investors to get value from their services. The contracts for the new Victoria big batteries explain why.

Print Friendly, PDF & Email

Grid-scale battery storage is taking off in Australia, but it is still a struggle for investors to get value from their services. The contracts for the new Victoria big batteries explain why.

Australia is at the start of a big boom in battery storage. The Tesla big battery is up and running, and by the end of 2019 will likely be joined by another dozen or so installations of varying sizes. Some bigger and some smaller.

The big test for investors and developers for battery storage is how to get their money back. Despite the speed and flexibility and versatility of the technology, there is actually no market for many of its service.

And that explains, why the two batteries to be built in Victoria this year – the 25MW/50MWh Tesla battery to be co-located near the 60MW Gannawarra solar farm near Kerang, and the 30MW/30 MWh Fluence battery at the Ballarat terminal station – relied so heavily on state and federal subsidies.

As we wrote last week, Tesla highlighted the problem in its recent submission to the National Energy Guarantee, noting that 40 per cent of the services it has delivered since it started operation in December have been for free.

That’s because there is no market for the many of the services it provides, particularly the stunning speed of its response. Even the institutions don’t seem to recognise what the battery storage has shown it can do.

Tesla has already shown that its presence has delivered significant value to consumers and the grid, by smashing the cartel on FCAS (grid services), keeping prices down in wholesale spikes, and adding to grid security.

There are estimates that the Tesla big battery has delivered value worth tens of millions of dollars in just its first three months of operation, but it has only been able to access a portion of these benefits.

The energy market rules, Tesla, the Australian Renewable Energy Agency, and many others note, are designed to suit Australia’s ageing fossil fuel fleet. The rules simply haven’t caught up with technologies like battery storage.

That perhaps explains why the owners of the new Tesla big battery to be built at the Gannawarra solar farm and the Fluence battery to be built at Ballarat have signed away market revenues for the next 12 to 15 years.

Under a complex financing arrangement, EnergyAustralia will take the market risk, and will operate the entire charging and dispatch from the two batteries in exchange for a fixed $50 million fee.

That $50 million fee will be shared between the owners of the Tesla battery (Edify and Wirsol) and the Fluence battery (Ausnet), and that will be their lot until the contracts expire in 2030 (Tesla) and 2033 (Fluence).

Ross Edwards, the head of trading at EnergyAustralia, says the fixed contract – along with the hefty up-front support totalling $50 million from the Victoria government and ARENA – was necessary to get the batteries built.

“It’s still early days for battery storage in this market …. And this will bring forward an investment that would not happen otherwise,” Edwards told RenewEconomy.

(It should be noted here that the unrewarded value of the two batteries – as noted by both the Victoria and federal governments – would be reduced costs for consumers, greater energy security, and higher penetration of renewables.)

Edwards says EnergyAustralia will explore a range of value streams in the market: capacity (selling at times of peak demand), arbitrage (buying low and selling high) and FCAS (grid services).

“Part of our approach is to understand how these (revenue streams) land in the future,” Edwards tells RenewEconomy.

“We need to understand these things better … how they will overlap, or if they will be totally exclusive …. and we need to understand the risks.”

“There is only so much modeling you can do to understand what these things will deliver. Most reliable modelling will look backwards.”

John Cole, the head of Edify Energy, which with Wirsol Energy will jointly own the battery, alongside the nearly complete 50MW Gannawarra solar farm (pictured above), making it one the biggest such installations in the world.

He says his company will likely be “under water” on the transaction, even with the big grants from the Victoria government and ARENA.

“We are not doing it for the money, we’ve done it for the learning,” Cole tells RenewEconomy.

“From a strategic perspective, and for understanding how the batteries will work, we will learn a lot. They – the batteries – will be a fundamental part for the future.” Indeed, he  says they will be a category killer for the energy industry.

“I’m a big believer in lithium-ion, I think it will follow the same cost curve as solar, and that has fallen 80 per cent since I have been involved in the industry.”

Ivor Frischknecht, the head of ARENA which has agreed to match Victoria state funding and grant $25 million to the two projects, says the issue is one of costs, which are coming down, and markets, which are yet to develop.

“The market is still fairly immature,” Frischknecht tells RenewEconomy.

Costs are coming down, but this will largely be driven by technology and manufacturing developments overseas, which will in turn be delivered by scale of production, as happened in the solar industry.

The other side is revenue, something that is within Australia’s capacity to influence.

Frischknecht points to potential markets such as fast frequency response, fault current and black start capability – which batteries can deliver, but have no access to under current rules.

“The revenue side is still early in terms of development, and that in addition to the cost is the reason that (these developments) need substantial subsidy.

“That’s why we want to be involved. What we want to do with these projects is to put the collective learnings to the various regulatory bodies and regulatory initiatives.”

As for costs, Frischknecht says around 30-40 per cent comes in the “non battery-pack costs,” such as cost of installation, software and engineering.

“That has a huge ability to come down … we saw that come down by a factor of five in large-scale solar … and they will come down with storage with volume.”

Other issues involve regulatory issues such as their status as a standalone facility, or as part of a solar or wind farm. Tesla has made a similar point. The differences in fees and network costs is potentially significant.

“There are some big gnarly issues that have to be sorted out,” Frischknecht says. “We don’t necessarily have the answers yet. Once we do, that will increase the value of battery storage.”

Print Friendly, PDF & Email

  1. howardpatr 3 years ago

    Don’t let the AEMC go near this issue.

  2. Andy Saunders 3 years ago

    Bad that EnergyAustralia controls it – means the oligopoly is extended/enhanced.

    • BushAxe 3 years ago

      Disagree, they’ve put a significant chunk of the funding up which shows they’re willing to take a gamble on new technology.

      • Andy Saunders 3 years ago

        But they will add the capacity to their existing plant and bid in their overall best interest (which includes significant gas plant in SA) rather than in the best interest of either the battery itself, or the NEM.

        • Alex Hromas 3 years ago

          Running the gas plant is more expensive than the battery especially over short intervals

          • Andy Saunders 3 years ago

            Of course, but let me give you an example. Let’s say the battery got charged at say 20 dollars per mwh average, and that as an independent plant it might bid into the energy market at say 200 dollars.

            If controlled by an entity with a much bigger gas plant, they may well be inclined to bid at say 1000 dollars for both the battery and gas plant. They don’t want to undercut the price…

            Cost to generate has little to do with it.

  3. JohnRD 3 years ago

    What we need is contracts to supply capacity with relatively minor payments per kWh to reflect operating costs. In some cases the capacity charge may vary depending on degree of readiness. What is used when would be up to the power system operators
    Such an approach will give investors the confidence required to encourage very competitive bids that don’t have to take account of uncertainity.

    • neroden 3 years ago

      No, that would be silly. Because capacity is worthless.

      What we need is payment for kWh when it’s needed — on a second-by-second basis. Different price each second, or even millisecond. This may not be viable but it can be approximated. You need a decent bidding system rather than the rigged “rebidding” system currently used.

      • JohnRD 3 years ago

        Neroden. The actual cost of producing power is dominated by fixed costs not variable costs. Paying per kWh creates investor uncertainty unless the system guarantees to pay for a given amount of power per yr whether the system uses that power or not. Investor certainty helps bring the price of power down.
        One of the SA blackouts last yr occurred because a gas generation company didn’t think it was worthwhile starting up for what looked like a short peak. Capacity contracts avoid this problem.

      • john 3 years ago

        Very true i have a gen set or i will switch it off or i can now bit at a higher price in any other area of business this would be looked at as rigging the market.

  4. john 3 years ago

    As per usual the situation is the people in charge are not up to speed especially the speed of a battery which can switch on in thousands of a second instead of minutes with the old system in place.
    Hint bring the system of payment into the 21 century get rid of the 18th century knowledge.

    • itdoesntaddup 3 years ago

      You mean change the laws of physics?

      • john 3 years ago

        No the present rules as per under.
        There are estimates that the Tesla big battery has delivered value worth tens of millions of dollars in just its first three months of operation, but it has only been able to access a portion of these benefits.
        What has to change is the rules that were written as in the 30 minute rule, we are going to 5 minutes in 2 years. Why?
        The speed at which batteries can operate is not recognized once we have 5 batteries all of 100 MW size then they will stabilize the grid in less that a 20 thousand seconds.
        In fact I expect we will see many more than just 5 perhaps 20 or so then frequency control will be a forgotten item.
        Just remember when the first solar inverters were put in place they could not work because the voltage and frequency was all over the place.
        Nothing wrong with the inverters it was the pathetic old style system in place.
        As a recent article pointed out the NEM has a last place position against international grids on frequency.
        Link to frequency problem with the East Coast Grid.

        • itdoesntaddup 3 years ago

          5 minute vs 30 minute settlement has nothing to do with FCAS 6 second and 60 second markets, or even the 5 minute raise one (since for FCAS the 5 minutes relates to the period following a disturbance, not a settlement period). Equally, there is no point in batteries having sub second response times any more than there is in having a car that does 250km/h and no race track. In fact, switching full power in a single step is destabilising for the grid, introducing all manner of harmonics (as you would realise if you have ever looked at Fourier analysis). AC circuits operate with overlaid exponential and sinusoidal behaviours, and stabilisation needs to reflect those realities. The 6 second market providers are actually providing power that ramps up to full capacity within 6 seconds – not a 6 second delay before anything happens. If you actually look at the chart Tesla provided in their submission about the NEG, you will see that it shows a smooth envelope of switching over several seconds – much as would be delivered by a more conventional FCAS provider. There is nothing particularly special about it, other than the special pleading for more money.

        • itdoesntaddup 3 years ago

          Incidentally, your article link shows the “best” performing grids are Russia and China – neither of which has significant renewables production and both of which can call on large volumes of hydro for stabilisation. The chart is actually just a sample taken during January, which is therefore not guaranteed to be representative: why the Malay/Singapore system should have been running apparently over frequency is not addressed. In the EU, they have had a problem with the grid now running consistently below frequency (cumulatively, over six minutes since mid January) because Kosovo and Serbia refuse to provide (or purchase) between them the power they have been consuming. What the article fails to explain is why payments rose to $200m given the earlier experience:

          Table 1 FCAS Payments from 2010 – 2014
          Year Payments
          2010 $30mn
          2011 $38mn
          2012 $25mn
          2013 $23mn
          2014 $30mn

  5. Askgerbil Now 3 years ago

    It seems there are a number of component makers and no one assembling a product that is actually useful. Batteries … as useful as bicycle handlebars by themselves. All that is missing is a couple of bicycle wheels, a bike frame, pedals and a bike chain.

  6. Jon 3 years ago

    In “the rule makers” defence it’s just as bad (probably worse) to rush in and make a new rule based on a very short observation time of a big example of a new technology in a rapidly changing market.
    Within 12 months there will be multiple batteries interacting into the grid to get a better insight how they will react in concert.
    The rule makers also need to consider how the 1,000s of domestic batteries are going to interact and be compensated for that interaction if they are made to work in concert or singularly.
    Pumped Hydro can deliver these services as well and wind farm trials in Tassie are showing promise.
    If in 2 years we don’t have a rule change it’ll be pretty poor but not having one less than 4 months after the biggest battery in the world was commissioned in a relatively small grid is not surprising.
    It’s better to be a little slow and get it right than screw it up quickly and pay for it for a long ng time, we’ll be the ones paying for it.
    It’d be interesting to see how batteries are being compensated for this in other networks.

    • Mike Westerman 3 years ago

      Jon – I think the frustration with the rule makers is that the forces of change have been recognised as coming for some time now, but the regulator has listened too much to the politics of energy rather than the science. What is indisputable is that we have gone from:
      – few sources of energy to literally millions
      – highly monitored and remotely controllable sources to largely unregulated
      – largely continuous sources to largely intermittent
      – largely “electron-mechanical” sources to largely “electronic” in terms of synchronism
      – relatively inflexible, reserve driven operations to highly flexible, instantaneous operations
      – high carbon emissions intensity to very low intensity

      Clearly the current energy only with limited auxiliary services and separate “bonuses” for emissions reductions is simply inadequate, but this should have been acted on some time ago. A well designed market ought to have enough light handed “incentives” for generally good behaviour but strong rewards and penalties for aberrant behaviour. And it places these at the feet of the participant who can most efficiently contribute or obviate them but with the least incentive or scope for gaming. On all these counts the NEM is overwhelmingly defective and all the NEG does is compound this fact.

      • itdoesntaddup 3 years ago

        Perhaps SA should operate as an island grid, with their choice as to what to do when the wind isn’t blowing. The reality is that most of the NEM remains coal dominated, with grid inertia time (measured as stored energy divided by power output) comfortably above the 6 second fastest FCAS response market. Faster response is only needed as and when grid inertia drops with rising renewables penetration. Until then, there is no market for it.

        • Mike Dill 3 years ago

          IF we can put a AC-DC-AC converter on the NSW interconnect (to remove the frequency issues with the NSW coal plants) and run the interconnect at the thermal capacity (which we do not do now due to balancing issues with coal plants), we should be able to isolate the SA grid and keep it better regulated and more reliable.
          There should be no reason why SA should continue to be abused by the national energy market.

        • Mike Westerman 3 years ago

          Why should any region ignore the economic benefits of interconnection? Vic benefits from Tas and SA, while both those states benefit from Vic.

          The reality is change is happening quite quickly – sure, we can follow your implication of doing nothing till “the market” dictates by which time we will be in a worse mess than we are at present. Markets in fact have limited bandwidth – they clear imbalances over short time frames and diversity ranges, providing the need for more deliberative processes.

          • itdoesntaddup 3 years ago

            The Tesla attitude seems to be much like the hammer that regards everything as a nail. To allow one voice to dominate this conversation about how to provide for much higher renewables penetration is I suggest a folly – and an expensive, highly subsidised one at that. I was trying to work out how much skin in the game is left when you have the State of Victoria and ARENA providing A$50m towards 80MWh of batteries, and Energy Australia another A$50m in revenue guarantees – and still the projects are under water.

          • Mike Westerman 3 years ago

            Quite possibly – despite the hype, STATCOM with ES has a specific role but are expensive if used for the type of storage needed in the NEM, and the control systems are not there for sidespread use ie the grid will continue to rely on inertia to slow RoCoF for stable recovery rather than active control. We can expect more active control of asynchronous generators altho’ most wind generators just want to run at 100% to earn income from energy sales!

            What’s more, when battery manufacturers can make much bigger margins, other than these few subsidised projects, in the much bigger markets of EV and behind the meter storage, it will provide an obvious price support so that I doubt we will see the sort of price drops evident in solar – nothing to drive it for quite some time yet.

            But all this doesn’t get away from the need for AEMO to be allowed to plan the future of the grid without all the political ignorance on show trying to stall the inevitable changes coming.

        • john 3 years ago

          They do not have enough distributed RE to do that.

      • Jon 3 years ago

        I’m not disagreeing that some rule changes should have been implemented by now, e.g. the 5 minute rule.

        I’m saying that what a large amount of battery/inverter technology can do in the grid is still very much being learnt.
        I can actually see the model of reserved capacity being paid for that then does the very rapid response work as an unpaid service being a good model and the market disappearing.

  7. Mike A 3 years ago

    This is a great articles explains something important. Pity ministers don’t know this.

  8. MarkH 3 years ago

    Article heading contradicts previous articles saying how much money the new Tesla battery has made and furthermore how much it has saved consumers in capping prices.
    So which is it ???? I understand the point is that some of the capabilities, especially speed, are maybe under valued by the stupidity of the current regulations, but if going by you’re own sites previous articles there is still plenty of money to be made within the current framework, particularly within the environment of price gauging that the existing oligopoly has been allowed to get away with. But ultimately the fast acting market is of limited size and to expand energy storage it will need to compete with cost of supplying peak electricity, in order to continue to make room for more renewables. The time is now, we should encourage peak shaving storage technology, not excuse it and claim it should be justified by something else as well. The cost for this service alone is justified from what I’ve seen vs new build ff peaker plants. Bring it on.

Comments are closed.

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