The South Australia government has announced that its 100MW battery storage tender – which it says is the world’s largest – has been won by Tesla and French renewable energy developer Neoen.
World’s largest lithium ion battery will be installed in #SouthAustralia under a historic agreement between #Neoen #Tesla & SA Gov! pic.twitter.com/GcfrwOzD9g
— Jay Weatherill (@JayWeatherill) July 7, 2017
The 100MW/129MWh battery bank will be built at Neoen’s huge Hornsdale wind complex near Jamestown, where the last stage of a 309MW project is currently being completed.
Premier Jay Weatherill said the Hornsdale Power Reserve will become the state’s largest renewable generator, and while the lithium battery would be the biggest in the world.
“South Australian customers will be the first to benefit from this technology which will demonstrate that large-scale battery storage is both possible and now, commercially viable.”
The announcement was made jointly with Tesla founder and CEO Elon Musk, who flew into Adelaide for the announcement. Musk said the installation would be three times bigger than the next installation.
“This is a chance to show you can really do a heavy duty, large-scale utility battery and battery system, and South Australia was up to the challenge. If South Australia is willing to take a big risk then so are we,” he said.
Financial details of the bid were not revealed, but Neoen deputy CEO Romain Desrousseaux said the company had been “aggressive and very positive” in its bid with Tesla. He told RenewEconomy Neoen would provide all the equity for the project, and then seek bank finance once it was up and running.
“There was no time to talk to the banks,” he said. “What we are showing today is that this is only the beginning for more renewables. Everyone wil be looking at South Australia, hoping to replicate this at other projects. We hope to replicate it at many other projects.”
The addition of the storage facility at the biggest wind farm in the state will add to the state’s energy security – providing quick reactive power in the events of faults, and smoothing the output of wind – it will also add much needed competition into the state’s energy market.
The arrival of a “dispatchable” renewable energy generator means that prices will no longer be controlled only by a handful of gas-fired generators, who have been largely responsible for the state’s soaring gas prices. It will also likely mean new rules that threaten to restrict wind output will also be dropped.
The tender was flagged earlier this year after the series of blackouts and load-shedding that afflicted the state following a series of extreme weather events and equipment failures, and after a billionaire tweet exchanged prompted by Tesla’s claim to be able to fix the problem in 100 days.
It is the largest battery storage tender in the country, although Victoria is also seeking proposals for two 20MW battery storage facilities with up to 100MWh of storage, and a 20MW/34MWh storage facility will be built alongside a 196MW wind farm powering the Nectar Farms glasshouse facility near Stawell.
Other governments are also looking at battery storage, including Queensland which has flagged it as part of a 400MW tender for new renewable energy projects, and the Northern Territory last month allocated a tender for a 5MW battery storage facility to Vector Energy, using LG Chem.
During the billionaire tweets between Musk and Australia’s Mike Cannon-Brookes, Musk promised that if the facility was not built in 100 days, it would be done for free. That offer still stands.
“It has been agreed between Tesla and the South Australian Government that the starting date for the 100 days will be once the grid interconnection agreement has been signed,” Weatherill’s statement said.
“After leading the nation in renewable energy, the 100MW / 129MWh battery places South Australia at the forefront of global energy storage technology.”
The battery will operate at all times providing stability services for renewable energy, and will be available to provide emergency back-up power if a shortfall in energy is predicted.
Weatherill said Neoen was selected on a merit basis after a multi-stage procurement process attracted around 90 responses to the Expression of Interest, with 14 proponents invited to supply, and 5 shortlisted for detailed assessment.
Those that expressed interest are believed to have included all the leading battery storage manufacturers, including LG Chem, AES, Kokam, and others, and developers such as Zen Energy, Carnegie Clean Energy and AGL Energy.
Weatherill said the consortium “provided a highly competitive commercial offer with the best value for money” although details were not immediately available.
“Neoen and Tesla have a track record in comparable scale projects, and are committed to deliver on time at the lowest cost with a suite of value-adding initiatives,” the statement said.
“I’m thrilled with the selection of Neoen and Tesla, whose experience and world leadership in energy security and renewables will help South Australia take charge of its energy future,” Weatherill said.
Neon’s Desrousseaux said the Hornsdale Power Reserve will become home to the largest lithium ion battery in the world, and would take the company’s long-term, direct investment in South Australia to almost $1 billion since 2013.
The company also recently announced that 196MW wind project and 20MW/34MWH battery storage installation in Stawell in western Victoria to power the country’s biggest glass-house for vegetable growing and supply the grid. (That $563 million project has been ignored by most mainstream media).
The company has also been an aggressive bidder in Australia, winning contracts for all three stages of its Hornsdale wind farm from the ACT government, under its 100 per cent renewable energy target, and three solar farms in NSW in the ARENA large scale solar tender.
Desrousseaux said the battery storage technology will demonstrate that large-scale battery storage is both possible and now, commercially viable.
“Together, the South Australian Government, Neoen and Tesla will demonstrate that renewables can provide dependable, distributable power that will turn a new page in Australia’s energy future.”
Giles Parkinson is founder and editor of Renew Economy, and is also the founder of One Step Off The Grid and founder/editor of The Driven. Giles has been a journalist for 35 years and is a former business and deputy editor of the Australian Financial Review.
Really curious of the details and cost……
I doubt you are Robinson Crusoe in that.
Powerpack 2 pricing has been “said” to be about US$400/kwh. At that rate, would be around A$70M for the 129MWh battery, plus connection and other infrastructure. Plus postage (express).
Don’t forget the GST!
Get it back at the end of the quarter.
It would be U$250/kWh for the Powerpack.
Which is in the ball-park of the minimum cost base for keeping additional gas generators running to cover a couple periods on some days.
and further under the NEM rules allowing $14,000MWh for supply during crunch periods it doesn’t take too many hours averaged over 365 days to take the matters from sensible 99.9% of the time to insane as demonstrated last summer…
In any case now that South Australia is maxing out on wind, having the battery provides a place to stash the excess power and then provide ancillary and instant response services to the NEM.
That’s per kWh. Powerwall 2 has max continuous power rating of 5kw. Based on A$8000 that’s say A$1600 Kw. My guess is A$80-a$100 m
Can’t correlate the price between the the Powerwall and the PowerPack, different economies of scale.
Some time ago it was said that the cells in Tesla batteries cost under US$190/kWh, so US$250/kWh for the whole unit would be about right. Plus shipping, plus taxes, plus stupid high mark up by any Aussie company that deals with it just because it’s government funds…..
Hmm I was one of the ones talking about the cell costs,which closely match numbers projected by the excellent Argonne labs Batpac C.
However I’ve done some research on costs, at least late last year, for utility scale lithium packs. Still think my number is a fair guess. I expect we will find out in time.
Figures quoted on Radio National this evening were $50 million for the 100 mW and 129 mWh output.
That would make it about A$250/kWh then. Certainly getting better.
Plus all the sophisticated electronics in the inverters that handle not only the conversions between DC and AC but also adjust the frequency and phase of the supply to help with grid stability.
Most of the commercial inverters from the big companies like ABB, SMA etc can do all this. That’s par for the course in this field. It’s just that the national networks seem to have no clue of the benefits of this technology. 20th century thinking will fail when 21st century solutions are required.
The point is the electronics are not free – they’re quite expensive, and will add significantly to the installed cost.
In almost all cases, unless specified otherwise, a grid connected energy storage system includes the electronics (inverters) to convert AC to DC and DC to AC. Certainly the case with Tesla. So when Elon Musk quotes a price for a given size system it will include the electronics to make it work.
For any given project, there is what is known as balance of system. Meaning all the other components needed to be operational. In this case there will be the civil works in preparing the site, concrete slabs to mount everything on, cables and conduits to connect everything together, and perhaps the most expensive part, the transformers to adjust the AC voltage from the LV (415Vac) of the ESS to the MV or even HV available at the site. Then there’s security, monitoring/control systems, and metering.
Just wrong. Tesla prices quoted for their powerwalls exclude the inverters and installation. If Musk says batteries at $250/kWh (as he did) that is what he meant.
Mate, how about you interpret both those articles correctly. Neither contradict anything I’ve written.
The Reuters article said Musk quoted $250/kWh for systems over 100MWh in size. That is US dollars.
The Forbes article clarified exactly what I said about balance of system components.
Elon Musk and Tesla stated prices are for the components that they supply – the batteries and the inverters. It is always like that.
It’s about US$400/kwh fully installed based on Musk’s “we’d lose about US$50 million if we missed the deadline” statement. (With one significant digit.)
I think $250/kwh is the price for the parts, delivered on pallets, buyer pays shipping. The installation labor and transport makes up the difference.
Musk has tweeted that he would be $50 million out of pocket if he doesn’t install the batteries within 100 days. So the total cost of the project must be more than that.
The price that Musk quoted sounds like the production cost. From what he has said, he is offering the batteries at a ‘close to cost’ price because this project has drawn such huge publicity he sees at as a good promotional opportunity for Tesla. If true, that will be good value for SA.
I’m really interested in the difference between the cost of this project and the cost of the next large scale battery project. Hopefully more of these projects will follow quickly after the hype around this one – and the costs will be headed in the right direction i.e. downwards.
Likewise. This is the sort of project that will demonstrate the usefulness of having storage on the network, increasing the likelyhood of more following. And getting a decent volume on the market will all help bring the price down.
All the losing tenderers now know they were too expensive and will have to drop prices for any future tenders. Bodes well for Victoria.
For context, in SA demand on typical days is in the range of 1-2 GW, while peak demand during summer heatwaves can go up to 2.5-3 GW.
@disqus_O5xhE4v303:disqus
Your post is what they call “disingenuous” insofar as the blackout last September was due to high voltage transmission towers being blown over. Any power system, including that in NSW and Vic and everywhere else on the planet would experience blackouts under similar circumstances.
Also “in context” the demand in SA and elsewhere is largely met by existing supply sources (gas, solar, wind, and coal-based power via the interconnect). However, the gas peaking plants are switched on to meet peak demand (that’s why they’re called “peaking” plants.). Tesla’s battery bank will be like a gas peaking plant, used on those infrequent occasions when required.
The blackout was due to the response of the generators along the transmission lines to the voltage changes. The point of the inquiry was that the wind generators probably would not have had to shutdown if their settings were different.
Now , assuming that the 100 MW battery had been in place at the time and that its response settings were different to the wind generators, then it might have been able to supply 100 MW of power when the wind generators went down. Would this have made enough difference to keep the whole system from going down?
Note that the battery is going to be at Hornsdale, so it is on the same transmission lines of most of the wind generators. So, maybe, it would also have been out of action. You would have to look at what transmission lines were still available.
Did they rebuild the transmission towers with the same wind loading specifications? I would hope their wind specifications were somewhat increased?
Survey says nope.
@WR,
You mean, like, “let’s see, worst case scenario”, if North Korea lobs a nuke on SA, and blow out all their wind farms, will batteries be enough? Clearly not. Okay, let’s not install the batteries.
Lol. You are over reacting to my posts.
I’m all in favour of adding storage. For a 100% renewable supply in the state, based on current demand the state will need storage that can supply about 2.5 GW of power and about 15-25 GWh of energy, depending on the mix of small-scale and utility-scale storage.
I was not making a judgement in posts, just providing context.
Suggest you read CSIRO report, might help you get a grip on storage needs.
The CSIRO model is fairly similar to what I have suggested. For a 100% renewable supply you would expect to have storage capacity of about 50% of daily energy use if the system was designed to work as efficiently as possible. For SA, that would be about 15 GWh of storage based on current demand.
However, I’m anticipating that the current demand profile will change with better electrical energy efficiency for current uses but an increased use for charging electrical vehicles. Of course, the EVs would have a large amount of storage.
That is why I’m anticipating that total storage will end up being somewhere between 50%-80% of average daily demand.
WR your 50% figure is probably true if you wanted SA to operate without the Heywood link. As long as it is available, SA still has room for more solar and wind – more will be built as long as the revenue from certificates is higher than the cost of construction, as long as it can be reliably dispatched. Already we are seeing curtailment of wind, which will start to hurt and soon limit finance for additional capacity. Storage enables not only dispatch to take advantage of arbitrage in the evenings, but creation of certificates. I can’t see that being done by batteries except at the fringe, altho’ these guys have first mover advantage: they are effectively buying at $0 and selling against OCGT at $150-200 – even $400/kWh works at those prices! SA will need to get on with 500-1,000MW of pumped hydro on the numerous sites available.
Storage would need to be much higher than the figure I gave if there were no interstate inteconnectors.
The amount of storage will be determined by the relative costs of adding new supply that is needed to meet demand, whether that be wind, solar (PV and other), biomass, hydro, etc., compared to the cost of new storage. If adding more storage is cheaper than adding more supply from other sources, you will get more storage.
Read some of the models by the CSIRO, Blakers, Mark Diesendorf, and the AEMO and you will see that the figures I have given are pretty realistic.
With regard to cost of energy purchase for the battery, the price would only be zero if the output was going to be curtailed. Otherwise, it is the opportunity cost of the current market value.
It’ll be done by batteries. I predicted the financial dynamics a year or two ago. Buy at $0, sell at $150…. Batteries are a highly-profitable, quick-return industry until there are so many that there’s no curtailment of wind (they get quick payback).
And once there’s no curtailment, at that point more wind and solar start gettting built. Rinse and repeat.
In the time it takes to build one pumped hydro station, we’ll have dozens, nay hundreds, of battery installs
How do you handle seasonal demand variations? How much power has to be curtailed, and how much storage do you need?
See the model that I linked above.
The model is fairly simple but it’s probably quite accurate because the current distributions of PV and wind in the region that I used for my data sources is probably quite indicative of the future distributions of these sources for the region. The main uncertainties in the model would be the future distribution and capacity factors of solar in the region. The daily demand profile would probably also change with much higher daytime demand to take advantage of any significant amount of solar generation.
The cost formula used to optimise the model inputs are based around some hypothetical average relative costs of the different technologies over the next couple decades. The formula assumes that wind and solar will have approximately the same LCOE based on their relative capacity factors for the region. The relative cost of storage assumes that 2.5 kWh of storage has the same cost as each kW of installed solar. ie. If solar costs $1000/kWh than storage would have an average cost of $400/kWh. These are installed costs. The cast assumptions can be easily changed to model different scenarios.
Hi Giles, please link.
https://www.csiro.au/en/Do-business/Futures/Reports/Low-Emissions-Technology-Roadmap
@disqus_O5xhE4v303:disqus
100% renewables?
I wasn’t aware SA were scheduling 100% any time soon.
In any case, within the NEM (SA, Vic, NSW, Qld) with sufficient scattered wind farms (even before we consider hydro and solar), the grid supply gets to look like “base load” 24/7 (with flat supply) as was found by Stanford University a few years back in regards to the USA network.
Arguably with sufficient wind farms across Qld, down the East coast, and SA, batteries would be unnecessary.
I’m an optimist but not that optimistic!
Hydro will have to be used as back up with or without pumping.
That’s incredibly optimistic. I’ve never seen a model for Australia that shows anything like that.
You can check my model for SA-Vic if you want some context.
We’ll have to have 100% renewables at some stage or otherwise we will eventually run out of non-renewables. That’s why they’re called “non-renewable”.
Might as well start now.
words of wisdom Tom, actually, knowing their definition, but some need to be told a few times for it to sink in.
In an 100% Renewable grid some form of storage will always be needed and even if 50% RE storage makes things more reliant.
I think you might want to check up on that with some real data. You might also want to consider how much costly long distance grid connection might be required even to begin to make it feasible to get some benefit in the cases where there is potentially some surplus in one area that could go towards a deficiency in another.
Stop wasting everyones time with your off-track and negative woffle…
South Australia has a peaking capacity problem and the battery proposal will likely cover that for next Summer…
Sorry Steve150. You need to read the AEMO final report on the Black System Event. The system had already gone black when the towers were blown over.
@kevandaly:disqus
Nope, don’t need to read the AEMO, other than to recognise they’re way behind the Europeans who have oodles more wind (percentage wise) and they don’t have the problems we saw in SA.
System settings, being inadequate, is no reason to demonise wind, or renewables.
We should just transport all of AEMO personnel off to Europe with the caveat “you can’t come back till you get yourselves up to speed”. You know, for them to earn the dollars we’re paying them.
Where do you get that myth from? Most of Europe operates as one synchronous grid, in which wind is a rather small element. It benefits from all the grid stability and backup provided by fossil fuels, hydro and nuclear generation.
EU wide electricity generation was 3,247TWh last year, of which 300TWh was wind and 112TWh was solar.
SA is right in the forefront of the experiment with high penetration of renewables, especially every time its interconnects with Victoria trip out.
care to point where in the final report it mentions that the system was black before the the towers blew over?
So it will be able to absorb 10% of wind generation in SA for use when the wind doesnt blow or a gas plant fails to start or just fails.
Nice, pretty good for the first of many infront and behind the meter.
Nope, Tesla systems produce 1/4 of their total storage for 4 hours.
i.e. a 100MWh system can produce a peak amount of 25MW for 4 hours. They are based on the Tesla utility grade PowerPack system.
So 2.5%, not 10%.
Tesla have specified the system as 129MWH of storage, with a peak output capacity of 100MW. Most of its use will actually be to supply or absorb power for a few seconds at a time to help stabilise the grid. Occasionally, that might extend to 10-15 minutes while a gas generator gets up to speed.
OK, still trying to figure out this mismatch.
I think Tesla can up the peak power simply by changing the way the packs are wired together. 50MW for 2 hours should be as easy as 25 MW for 4 hours. 100MW for 1 hour should be viable with a 100MWh system.
In respect to the big storm event, nothing economically practical would have likely covered the initial brown-out from major/multiple network failures.
However, when during LOR1/2 events as my previous comment highlights a 100MW of instanteously available capacity buys time to get other capacity on-line and/or run the network in a more efficient manner overall.
Well, if you exclude using fossil fuels and hydro instead of solar and wind renewables. A zero renewables grid would have had no trouble at all. It would probably have coped with the loss of the blown over transmission lines as well.
Bullshit. An all-fossil centralized-power-plant grid would have crashed completely, immediately, and the transmission line losses would have kept it down for WEEKS.
You forget that some of us have actually dealt with such things in other countries.
Distributed power is more robust than centralized power.
A sensible grid for SA would have something like 3.5-4GW of capacity (assuming no power imports), split across at least 10 gensets and multiple locations. There is no reason for any of them to have crashed in 80kph winds of your little “big storm”. Even if one did, the others would have easily taken up the slack. The grid would be decentralised sufficiently to provide a robust supply, with alternative routing available between generators and demand for much of it.
As it was, it was the thermal plants that were last to trip out in the big blackout: they proved their reliability while windfarms were shutting left, right and centre. The thermal stations only tripped when the load was entirely thrust on their inadequate capacity.
So in short your idea that wind is more robust is fair bunkum.
I think you need to read the AEMO report and get some actual facts on the events leading in the september system black, several points you have stated are incorrect. Then go and read the report on the march substation fire at TIPS where 500MW of gas generation tripped out almost blacking out the state again.
I have read the report. You’ll find it here:
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Market_Notices_and_Events/Power_System_Incident_Reports/2017/Integrated-Final-Report-SA-Black-System-28-September-2016.pdf
Perhaps you should give it a go yourself?
You make the point by referring to the TIPS fire. Because there was no alternative spinning reserve, the blackout risk was very real. Wind can’t provide spinning reserve. Musk’s battery can only offer a limited alternative. The SA system should be robust to the loss of its largest power sources – the Heywood Connector – but it is a long way from that.
I actually lived through the Great Northeast US blackout, where all the fossil fuel plants tripped and made a mess. Hydro, wind, and solar came back early (at the time there was very little wind and solar).
I know what I’m talking about and you’re full of shit.
As BushAxe points out, you’re also full of shit on the South Australia case, where the fossil fuel plants tripped and shut down quickly. And one never started at all.
I don’t talk to people who are full of shit. So after pointing out your shittiness to the general public, I am reporting and blocking you.
Chill, he’s a FF troll.
Temper, temper! I suggest you read tha AEMO final report which states quite clearly that the conventional generators stayed up after the wind shut down. It’s here:
https://www.aemo.com.au/-/media/Files/Electricity/NEM/Market_Notices_and_Events/Power_System_Incident_Reports/2017/Integrated-Final-Report-SA-Black-System-28-September-2016.pdf
Double what was installed in australia last year @52Mwh.
Thank you Giles, best news I’ve heard all year.
IMO 100MW is a drop in the power ocean and if that 129MWH storage puts out its max power it lasts…at best….about 80 minutes. We use GWs and GWHs of electricity. So basically it makes no measurable difference in either power output or duration in the case of a power outage. It’s a step in the right direction, but VERY small.
I think your numbers are out….on the generous side. No way 80 minutes of SA demand.
Given that the unit cannot provide anywhere near the SA demand anyway, I think….no I said…that it would be 80 min at 100MW.
You’re right – it is a drop in the energy demand ocean, but it’s the first of many drops. In a place like SA where mountains and rainfall don’t allow for much in the way of pumped Hydro, batteries are a very good option, even if they’re expensive. If Snowy 2.0 gets a wriggle on, we’re well on our way to a renewable energy future.
This will open up a pretty good supply chain, and spur the competition on to better offerings in future i.e. grid battery installs here of similar magnitude will quickly get less expensive from here forwards.
I live in WA so yeah I know about non0-hydro capabilities :). I think my concern is mainly that people will not realise how small this thing is and expect it to answer everything.
Sandgroper here too. And yes it will want to get cheaper for the next one too; this is a billion dollar project!
Where did you get the billion dollars from?
It only costs 250 million to put that much battery storage on houses.
Sorry, I was referring to the total SA energy spend, including the extra generation and infrastructure being added to the system.
Sorry, butr another comment. I realised why I am so concerned about all the hype. The power wall has triggered a lot of stuff about going off grid and for the average household that is just not on at 10KWH,…….think 4 days of heavy cloud and rain etc.
I am pretty sure I can go off grid in a couple of years and I have a high-usage household with a small roof.
On some days I’ll have to turn off most of my discretionary electricity-guzzling equipment. But I think I can cover heat and lights, which is all that really matters…. around here we get grid outages sufficiently often I’m used to dealing with having the rest of the stuff out, you see.
I am also pretty sure it’s not going to be cost-effective to go off grid. But it seems technically feasible.
You are pretty sure. Show figures.
I am under no obligation to provide my personal analytical spreadsheets to you. They add up.
Buying a Powerpack for personal use is kind of expensive, though. Definitely in “hobby for the rich” price levels right now.
Well I guess we are under no obligation to take any notice of statements made without any visible foundation……
Well, you’re under no obligation to, if you’re a complete idiot.
I mean, it’s easy enough for you to do the math, but I don’t feel like disclosing the specs of my house, which is a private matter.
I should note that my house is superinsulated so the heating and cooling costs are rock-bottom cheap to start with. That’s important.
There’s an alternative to installing more batteries – install a small generator. Have this going for a couple of days twice a year during those said periods. It might cost a few thousand to buy and hook up, and it might be a pain having Jerry cans around the place, but it would be much cheaper than expanding your 10kWh battery to 40 or 60 kWh.
PS – a personal PowerPack would be awesome though. No more problems charging your car.
My figures show it’s cheaper with a 5kW solar system and a powerwall than what we pay for power, let alone the $1/day to be connected. Your mileage may vary, but as soon as they release the off grid powerwall, I’m off.
https://uploads.disquscdn.com/images/b75b62e2c589e68a6c3b348c32ef84b446954782c56577696a602846954e7dbc.png
Plenty of water in the sea though Chris. Are you aware of Energy Australia’s plan for sea water PHES?
Yes, and it certainly looks very feasible, however there are some very high conservation value areas which would be negatively affected by seawater storage if the reservoir were to leak. One way around that may be to desalinate the water going into the upper reservoir and have a suitably sized lower reservoir to contain the outflow. Adds to the already substantial costs though…
Highly unlikely it would leak. When was the last time you heard of a dam failing? Depending on the height above sea level this dam will be built, it could be dug out of the ground or partly so. The most pressure is at the bottom of any dam, so if all or most is under ground level and engineered correctly, I doubt there will be problems.
Of course it’s unlikely, but the result if it did would be catastrophic. The dams would have to be lined completely as the salt would leach its way into the surrounding landscape. The dam may not be breached, but the salinity will kill all plant life around it. Moreover, the land around Commissariat Point is quite stunning, and I doubt people will take kindly to a bunch of concrete reservoirs on the hills. Wind turbines can be enough trouble as it is!
I’m sure the geo techs and engineers will make sure it was lined.
The battery is meant to buy enough time to get a gas unit up to speed and improve overall system efficiency by filling in the capacity increments on large gas units and momentary wobbles from wind and solar.
Makes perfect cents..
True from a MWh point of view, but its FCAS potential is huge. The rule that 4 gas generators need to be generating if wind is generating over 1200MW must surely be abandoned once this battery is up and running.
This year so far there has been an almost continuously high wholesale energy price in all states, but soon in SA there will be times when the price of energy goes very low as more wind generation is being produced than demand for it. These ultra-low-price periods will become longer, more frequent, and more widespread as time goes on.
Arbitrage (buying cheap energy and re-selling it when expensive) has been a nightmare so far this year – nobody knows what “cheap” is as many days it is over $80/MWh for days on end. However, once these ultra-low-price periods start appearing, arbitrage will become easy, and very valuable. Grid-scale batteries will absolutely take off once this happens – no subsidies required.
The “4 gas generators” protocol is not about FCAS but fault current (system strength) Tom. I doubt that the battery installation by itself will change things on that front. But it will presumably be set up to provide FCAS regulation and perhaps even Fast Frequency Response so may make a difference to the price of FCAS and some impact on inertia requirements.
I think it will as normally they will run a few TIPS units at 20% during high wind generation, so the battery will easily provide the same level of fault response.
Blind Freddy
Working as a grid controller for 30 plus years I say the following. This system, if available in my experiance would have averted almost all of the critical system disturbances I had to deal with on shift.
It would not have solved the problems we had after the Newcastle earthquake in 1989. It would have prevented a state separation I experianced in 2001.
Out of about 100 serious frequency disturbances I experianced from 1982 to 2015, almost all were caused by failure of coal base load generators. I believe that a system as described in this article, if available in each state, would have averted all these situations at a fraction of the cost of the methods we did use.
Apart from this, the system as described, will sit in the market in an optimum way and scrape in top value from the wholsale market.
Could you elucidate a little more on this? I’d be interested to hear some numbers. I’m no expert, but it seems like having 100 MW on tap for about an hour wouldn’t have much impact.
It’s 30MW, the limit on Tesla batteries is 1/4 of the total capacity delivered across 4 hours. You can use lower peak capacity and extend the lifetime but you cannot extract it any faster than over 4 hours.
No, the maximum output is 100MW as specified by Tesla: that’s their design for this installation.
They also said they are using their Powerpack2 systems, and the datasheets for those systems say they store 210kWh per unit and have a peak output of 50kW. I doubt they plan to redesign these systems in the 100 days they have to produce a solution. Using those systems a 100MW system would store 420MWhr of energy. So somethings amiss in the info we have on the system.
Yeah, it looks like we don’t have proper specs available.
You know what? I suspect it does have 420 MWh of storage capacity. The reported 129 MWh is probably the spec requirement. I betcha it’s actually overbuilt, and that Neoen knows this.
If the cycling is kept shallow the batteries will last much longer. So instead of a 10-year warranty they may be running with a 25 or 30 year lifetime by shallow cycling…
Tesla have designed and specified the system. Why do you doubt their information? Remember, they have been working on this for months already. When did Musk first make his offer? Back in March. Do you think he did it without any prior study? How long ago now is the big storm blackout? Last September.
The power output depends on the inverter, not the battery. You can think of the specs as consisting of a 129 MWh battery with a 100 MW inverter. Undoubtedly they can modify their inverters to provide those stated specs.
A full spec has not been given. Here is a description of Tesla’s installation in San Diego which opened in February:
https://arstechnica.com/science/2017/02/as-ca-bill-aims-for-100-renewable-by-2050-utility-starts-30mw-battery-system/
Largest grid-tied lithium ion battery system deployed today in San Diego
Megan Geuss, 25 Feb 2017
“On Friday, Southern California utility San Diego Gas & Electric (SDG&E) held a small press conference in Escondido to show off its brand new energy storage facility, a 30MW battery system capable of storing 120MWh of energy, which can serve 20,000 customers for four hours. ”
So clearly the “30MW” is not nameplate capacity, but what is delivered for 4 hours, while the SA system is 100MW for 80 mins.(?) Like harrysnape suggests, the battery installation must have a considerably higher nameplate capacity to deliver that. But it is what is delivered that the customer pays for and which contracts specify.
The San Diego system began construction in December and opened 25 Feb 2017, so this is why Musk is confident he can deliver. Presumably the SA system is more or less three times the SDG&E system?
But note that this “world’s greatest” only applies to Lithium-ion as there are bigger installations: Kyushu Electric Power Co., installed a 50 MW, 300 MWh sodium-sulfur battery facility that went into service in March 2016.
As a matter of interest.
In the coming sumer we have peaks of about 2800 MWs on the SA grid.
If the largest generation site (Pelican Point) tripped , ie. 487 mws, the battery could instanly respond. The frequency would otherwise fall to 49.7 hz but the battery response would maintain the frequency above 49.85 hz. Doesnt sound like much but it is a significant difference and
changes the nature of the response. Minor alterations in the VIC interconection flow would quickly restore to 50 hz.
This would then require market response to prepare for the next worst case possibility. The battery would not need to provide more than 5 or 10 minutes of 100mw and could be placed back to zero output for further problems that may happen.
SA is also leading the way now in load side response. The contribution will be very significant as it will be targetted effectively for security across the whole day and not like the old hot water functions, same time same place every day achieving nothing for security.
I appreciate the detail, Cooma Doug.
Me too. Thanks.
By making 100MW available at milliseconds notice should reduce the need for the current high gas usage with renewables curtailed?
Giles it might be worth an article or two on the comparisons of the different technologies, i.e. wind, hydro, gas for the grid support services they all provide relative to their generating capacity?
Yep, good idea. I think it generally agreed that 100MW of battery storage would have arrested those voltage swings experienced before the system black, and given AEMO time to Think of Something. The gas generators were so slow the horse had bolted before they could react. after 6 seconds they were still going in the opposite direction to where needed. their blushes were saved by settings on wind farms.
do we know on what basis revenue is derived by the batteries? is it simply inside the gate of the wind farm and their sales of power to the grid, or is it paid to provide services other than straight MWh to the grid and if so, what are these items/rate?
no we don’t. there is some support from s.a. government and some nature of a contract, but the details have not yet been released
hopefully they release details would be interesting to see; far easier to value a battery behind the meter for a user, than to quantify and determine a remuneration scheme for the value it provides to a wider network, some of which would be subjective as to which element of the network was providing said function until an emergency situation when it could be determined
Pricing for the services it is likely to offer is simply undeveloped on any detailed metering basis. Grid batteries installed elsewhere mainly have an agreed annual revenue, and an agreed level of service in terms of speed, capacity provided and duration of response (per incident, and obviously some limits on the total call on the batteries in a series of closely spaced incidents). There may be a penalty for failure to meet agreed response.
464.4 GJ battery!
So what will happen to that plan to invest in more gas generation?
Nothing, I would assume.
The battery isn’t a generator so the two fulfill differing roles.
Ah but the battery is a generator. It is collecting energy that would otherwise go to waste due to low demand or non-“firmed” generation, for later/batched supply. It’s output is an addition to the current generation that makes it to the grid.
See this is the problem here. Currently there is no category for storage in the network.
The NEM really needs a new category – “Network Storage Service Provider”, for energy storage. Then it doesn’t have to pretend to be something it is not to be able to participate in the NEM. And when storage is correctly identified and treat as such, it will be able to fully realise all the benefits it can provide.
Yes “Service Provider” aka supplier. Without getting too caught up in semantics, isn’t ‘supplying’ the only type of interaction this thing will have with the NEM? Adding energy that would otherwise have been wasted(unavailable) to the available pool of supply?
For all intents and purposes it seems just like a generator.
No. Yes it can “supply” power. But it can “absorb” power at the same rate too. And this is important because not only can it absorb and store excess energy from the adjoining windfarm when it might otherwise have to be curtailed like we saw a couple of weeks ago. But it can also absorb some from a gas plant when it could be the difference between having to shut down for a few hours or not. Generally they would rather keep going at low prices than shut down and re-start, loss in efficiency doing that. And the other thing is they could absorb energy from the inter-connector from Victoria and put it back in when a gas plant might otherwise have to start. It is entirely possible that it can prevent some gas from having to run sometimes. That can be a big saving.
So being able to absorb as well as supply are just as equally beneficial traits.
You’d think that AEMO would already have something in place that would cover the function of Tumut 3 and Wivenhoe. I wouldn’t be surprised if they don’t though.
Good point. At that is exactly why these plants aren’t utilised anywhere near as well as they should. Because they are only treated as a generator.
I’m glad you brought that up because it highlights my point nicely. Thank you.
It’s not a generator but it is supply … and demand, as well, but not at the same time.
good question. it was always going to be the case that 100MW of battery storage was going to be needed and the gas plant, but i wouldn’t be surprised if they cancel the gas plant. they already reviewing the energy security target. time will tell.
Although a 100MW battery is a good start, it won’t be enough for a prolonged high demand scenario. More storage will be needed and I’m sure this will occur in time.
Exactly and that is why they are going to build pumped hydro using sea water.
More storage or more capacity in form of a mix of renewables. There’s certainly room in SA for more energy from RE and capacity, until the Heywood is permanently shipping in the export direction for SA they have room for more.
Unless they are looking at those GT/battery hybrid units you wrote about earlier this year, which would tie in with the concept of improving grid stability by having them located at strategic points across the state.
In the short term (this summer) pelican point will have gas to run at full capacity. There is also more wind (horncastle). In a year or two agl’s reciprocating engines will add additional felexible on demand supply plus there is more wind and solar. I expect there is additional behind the meter storage.
What are the chances they’ll cancel the gas generation and build another $360M in battery storage, that would be the ultimate outcome wouldn’t it?
Congrats to Jay Elon and Neoen. Kind of leaves Turnbull and Joshy left like two shags on a rock.
It leaves the idea of conservatism out to dry. It is conservatives and followers of hard right ideology that oppose renewable energy. They have held back renewable power generation in Australia. Soon we will have same sex marriage in Australia and that will be another major blow to the movement. This news is emblematic of the failure of conservatism.
Renewable energy debate
“Public policies can be developed and implemented that put our international commitments on climate change first, climate science first, and public health first, by moving to energy efficiency and renewable energy sources.” – David Suzuki
“What we truly need is to start spending much greater sums of public
money, that is, energy allocated for public purposes, to jump start the
creation of a sustainable, renewable energy society. We don’t have much
time to prepare.” – Kurt Cobb
“A sustainable energy policy must be based upon a dramatic reduction in the use of fossil fuels and a search for alternative, renewable energy sources such as solar energy, wind and wave power. These are by their very nature
sustainable and can be treated as ‘income’ rather than ‘natural capital’.” – Andrew Heywood
“We need bottom up democracy. We need small-scale economies, and small-scale technologies powered by renewable energy. We need smaller communities, structured to be self-sufficient, all tied together by high speed monorails. We need gardens and parks in our cities instead of cars. We need social halls, not shopping malls. And we have enough energy remaining to do
this, if we act now.” – Dale Allen Pfeiffer
“True individual and family security will come only with community
solidarity and interdependence. Living in a community that is weathering
the downslope well will enhance personal chances of surviving and
prospering far more than will individual efforts at stockpiling tools or
growing food. Meanwhile, nations must adopt radical energy
conservation measures, invest in renewable energy research, support
sustainable local food systems instead of giant biotech agribusiness,
adopt no-growth economic and population policies, and strive for
international resource cooperation agreements.” – Richard Heinberg
https://uploads.disquscdn.com/images/0908ddf2bf8efd7ea54ab5070f374e81b477b2bf73456d1f27efd9d8f48356e0.png
“I found a moonrock in my nose.” – Ralph Wiggum
as much as renewables have a role to play, i wouldn’t go to David Suzuki to get power network policy, that’s not his area of expertise is it? and Pfieffer is a geologist, again not a great reference point for how to supply power to a society
And equating renewables as a hard connection to left/right idealogy whilst simple/easy is not always that straight forward, as attractive as that attack point is to some on the left
The best description of right-wing ideology is that they want a return to aristocracy: those who are currently in power should remain in power forever. They think that the current incumbents who are rich and powerful were put there by God and should be kept there forever regardless of any facts. This has been the meaning of “right wing” since the term “right-wing” was coined during the French Revolution, and it hasn’t changed one bit.
As part of this, right-wing ideology is fundamentally opposed to science, opposed to reason, opposed to rational thought, opposed to evidence, and opposed to reality, because all these things might challenge the dogma that those who are in power now should be in power forever.
There is a hard connection between right-wing ideology and opposition to renewables. And there will be until the fossil fuel incumbents are all dead or powerless, at which point suddenly the right-wingers will back the *new* powerful/rich people.
(Left-wing ideology, on the other hand, can either support or oppose renewables or be indifferent. Same with “centrist” ideology.)
I should note that I do not believe that right-wingers deserve to be called “conservative” — they are not conserving anything. They are radical extremist reactionaries trying to disrupt things. I guess when aristocracy was the status quo, they were conservatives, but democracy has been the status quo for 100 years, and they haven’t been conservative for 100 years.
All true conservatives are now found on in “left wing” politics.
Seems to me that it has been the green ideology that has steadfastly refused to recognise that a grid that relies heavily on wind and solar is subject to high levels of blackout risk. Or maybe that’s the idea?
Even hydro power requires careful harvesting of the resource, so as not to be caught out by several years of relative drought.
I think you’re missing the point on renewables. The energy market operator thinks that we can have up to 40% renewables in our existing grid without thinking about complementary technologies. This article is about the recent installation and contracting of complementary technology which will increase the percentage significantly. I expect that we’ll see a lot more storage of various kinds coming online fairly quickly. As far as hydro and pumped hydro is concerned, it’s the perfect complementary technology for renewables, where you can actually build up capacity when renewables are generating excess power.
wow, if you say so… your bias shines through strongly. Left can do no wrong and can have differing thoughts on matters, right is a singularity of thought. I wonder if you realize that what came before renewables for power generation was based on scientific principles, maybe some rational exploration basis and some sound rational business building principles…. hmmm back to the drawing board
You should keep your far left views out of conversations about real life things like power supply
You just showed your idiotic, tribal bias. Did I even say anything good about “left wing” politics? I didn’t.
it was a fully balanced comment with no hue cast on either side of the political spectrum now that i re-read it… oh no it wasn’t, it listed a whole bunch of negative stereotypes against the ‘right’ of the spectrum and a few positives about the ‘left’ and also mentioned very distinct political themes like the aristocracy and the french revolution
seeing as you know the ENTIRE history of the energy business, i’ll leave you to your post doctorate research. Hint, probably don’t propose generalisations such as right wing people are anti-science or anti-logic, it’s simply not true… that is what shows your tribal bias
Now it is time for the regulator to price the short term supply response of batteries. So much faster than coal fired power.
Correct
Depends what you mean by supply response. If coal fired generation is already spinning, and operating at say 500MW out of 600MW capacity, the extra 100MW can be there pretty fast. If you want to start the boilers from cold, that is a very different matter.
Would this be enough power to restart the network if we had the whole thing fall over again ? Assuming the infrastructure connecting it to the grid was up of course.
Powering 50000 to 60000 homes for an hour or so?
No I mean would it be enough power to get the other gas generators online. Black starts are quite tricky apparently. You need power to get the generators started. A bit like how my gas stove, heater and water heater need electricity to run. According to AEMO we had systems in place called System Restart Ancillary Services that are supposed to do this job but fell over when called upon last year.
I would have thought it will be an excellent restart source with it’s ability to inject a large amount of power into the transmission network very quickly as well as providing FCAS and then absorb/supply power as you bring generation/load online.
Thanks
What? They can’t just use an idustralised sized match?
Ah I can just see 100 guys with something like a giant battering ram lit at one end marching towards the generator.
Yes, this is plenty of power to “black start” the gas power plants (assuming the grid connection is up to the task). Batteries are ideal for black starts, actually.
At last some sensibility – Well DONE S.A…. go and tell Turnbull to get back here and fix his mess.
What a great weekend we have now. Up yours, our right wing climate denying leaders.
Finally, a sensible move. Now we need to see the end of the gas nonsense.
Either pumped hydro or solar thermal can take up that role.
Natural gas is just another highly polluting fossil fuel which like coal and oil needs to be kept in the ground.
When we have more renewable generation than can be used by the grid or
storage then time to crack water into hydrogen for use where it makes
sense.
lng is living its golden years what r u talki’ bout bud?
I’ve been reading some of the news about this on the ABC website.
It looks like Tesla is supplying and installing most of the hardware and that Neoen will be running the system and trying to make money from the frequency control and ancillary market, and also from arbitrage.
In addition, the SA government appears to have bought rights to using at least 70% of capacity (70MW) to help meet peak demand during exceptional events such as summer heat waves.
I guess we’ll find out more next week.
Interesting times… but how long will 100MW last if the wind doesn’t blow and the sun doesn’t shine?
look buddy, u can not feed all country with li-ions, gotta? they die….. and when temperatures goes up, they die much more quickly.
on the other hand u have nuc. , lng, and coal. u can use all of that for 50 fucking years, 50 fucking YE-ARS!!!!! yo know wha i mean?
have a nice day buddy
Long enough to get the gas generators up and running. This is only a first step, there will be more batteries installed in the future.
If the sun never shines, you have a bigger problem. (The death of all life on earth, etc.) Don’t worry about it.
Yeah my thoughts too lol!
Hi Giles and fellow posters. There’s some discussion online with the naysayers saying this system can’t deliver 100MW from only 129MWh storage as Tesla’s Powerpack is designed to deliver only a quarter of its storage capacity for around 4 hours, which would amount to just over 30MW max. output in this instance. This despite all the publicity around this announcement, including Tesla’s own press release, saying this system will be able to deliver 100MW for around an hour. Is there anyone out there can help with the technical aspects? Can the existing Powerpacks be tweaked so’s they can deliver a higher output for a shorter time? Or are they not the same batteries as used in other Tesla products? I’ve searched and searched online but haven’t yet been able to find any technical details on the project. Cheers.
100MW is what the battery will deliver max, so if discharged at this rate for 1 hour, it will produce 100MWh and will be exhausted. However, if discharged at say a rate of 50MW, it will last 2hrs or more as the rate of discharge is less and so able to give up more energy.
So if Tesla’s 100MW battery is discharged at a slightly lower rate it can produce more energy.
Cheers, solarguy. I understand that’s what is being claimed for the project, but there are several people objecting to those numbers online, saying Tesla Powerpacks can only deliver a maximum of one quarter of their capacity for up to four hours, that you can reduce the load and extend the battery life, but can’t increase the load beyond that one quarter capacity. That was also my understanding of how Tesla Powerpacks function prior to this announcement. That’s why I’m wondering whether the S.A. project is using a newly developed battery, or perhaps different software or anciliary equipment, to be able to extract roughly 75% of it’s storage capacity at a given moment, as distinct from the usual 25%.
So Tesla’s Powerpacks have cells, assembled into modules, and the modules are assembled into Powerpacks, and then the Powerpacks are connected to an inverter.
The limitation on peak power rate appears to be at the high levels: most likely the inverter and the wiring to the inverter, maybe the Powerpack level, or *possibly* the module level.
The cells are definitely good for this power rate. I am guessing based on the South Australia announcement that the modules are also good for this power rate. They may have beefed up the wiring inside the Packs and from the pack to the inverter, and they have definitely beefed up the inverters.
Thanks so much for that, neroden. Exactly what I was after. Cheers!
100MW is peak output capacity. Most of the time, the battery will only supply power to the grid for a few seconds at a time to cope with sudden variations in wind output or in demand. Sometimes it will be used for a period of minutes while other generators are brought on line. It isn’t really designed to be run right down – full cycling of the capacity would lead to deterioration of the batteries.
After investigating, the Powerpack battery cells technically capable of pushing out power at the rate stated. The standard commercial model is probably software-limited to a lower rate of output. I’m pretty sure they can deliver a higher output for a shorter time just by tweaking the wiring and the inverters.
The systems are customised to suit the application, Tesla is a bit vague on details but if you look at the other big manufacturers like LG, Samsung, etc they will build the system for what is required ie short term grid stabilisation or longer term load shifting. This is why we will have to look at specs for each system when comparing costs as systems that have larger inverter capacity will obviously cost more for the same MWh capacity.