This piece is clearly written by someone, who as no glue about the grid. (or wants to make a buzz out of nothing special)
Battery power is always faster then generator power. That is the reason why every data center has battery backup for immediate reaction and generator power for the longtime backup. Normally, the battery size is designed that the minimum time is at least twice as long as to start the generator.
Exactly that has happened with the grid. It is also not related to some contracting. Because the whole grid is connected. So the battery will always kick in, it does not matter where it is located.
Actually synchronizing the grid is tough task. A friend and colleague from my University started in this business after he made his Master Thesis tens of years ago.
By monitoring the frequency of the grid you know if there is too much energy produced (too less load, frequency to high) or vice versa. Everybody basically can do this from home. There are some maker projects and personal projects who do that.
> (or wants to make a buzz out of nothing special)
It's important to note it is something special. The battery works, and it works better than the existing solutions.
This is very important and big news in Australia, because the coal lobby is so strong and powerful they have been telling the public for a year now the Tesla battery will not work and is a waste of money. Obviously some people such as yourself understand how the thing works, though I would bet a massive percentage of "people on the street" have no idea, and only understand what the coal lobby is telling them.
It's important to take every opportunity to educate the public about the real facts, and this event is a fact. The sooner the public learn the truth, the better.
The fact that Australia has the largest functioning lithium-ion battery in the world, and yet the Federal Government of Australia has done nothing but ridicule and belittle its usefulness should give anyone a clue as to the power of the coal lobby in Australia.
Any news of the flexible uses of this battery should be published far and wide just to blow away a lot of the misconceptions about 'anything other than coal and gas' that are rampant throughout much of Australia's mainstream media.
On a broader note, it is important that in the age of (mis)information, that genuine advancements/improvements on tech are advertised properly.
This is difficult because taxpayers often take a dim view on their money being spent on marketing government programs. Unfortunately with large companies advertising directly against programs that often run contrary to what is in the public interest, it is arguably in the public interest to market public sector programs.
This issue being what is the difference between this and tax payer funded propaganda...? It's a tough area of discussion by any stretch, but its become clear that advertising/informing the public is kinda needed to offset the campaigning against public programs from large industry.
“Unfortunately with large companies advertising directly against programs that often run contrary to what is in the public interest“ - uh? I think you meant the opposite of what you wrote. ;)
A few seconds faster, but with less than 2% of the required power.
What are the claims from the other side about why "the Tesla battery will not work"? If the claims are "it won't provide enough power to matter", this data seems to support those claims, unless I misunderstand how the 9MW provided by the Tesla battery (the HPR) resolved the sudden loss of 560MW.
As I understand it, the article is about the frequency stabilisation market. The existing reserves are able to pick up the loss of energy eventually, but reliability of the grid depends on more than just total reserve power. That is why the frequency stabilisation market exists. As the article points out, no one was expected to lose power here.
Power systems are highly nonlinear. At any given time there are power generators spun up and able to deliver more power. In the case of hydro, for example, vanes can control the flow of water. More water at one generator means everyone's turbines can spin slightly faster. Throughout the day, many such adjustments are made to match the load with the supply, e.g. with predictive adjustments every 15 minutes, say. But the systems can only respond so fast and each station can only pick up the frequency so fast. When a large enough generator drops out, or load suddenly increases (lots of people turning their kettles on during an ad break on telly, for example), numerous things can happen, including increasing supply, reducing loads and tripping breakers. In the worst case, some consumers lose supply. The advantage of electric batteries, being pointed out here, is that they can react in milliseconds, helping to stabilise the supply. This can prevent those corner cases where dropping consumers from the grid, due to cascading failures (highly nonlinear events), is the alternative.
The issue is that it helped stabilise the frequency which prevents circuit breakers kicking in and possibly taking South Australia off the air like it did previously. In other words it did exactly what it was bought to do.
I don't think it's making "a buzz out of nothing special"... you compare it to every data centre with battery backup which is a standard thing, but scaling that to apply the same principle at a full grid scale rather than a single data centre is not nothing special.
Something that was interesting and buried down in the comments is that they're supposedly contracted for 35MW of fast-response power to maintain grid stability based on engineering of current generators/etc.
If that's true then Tesla is already providing 1/5th of that capacity today.
I think the reason that the battery put out the amount it did was that that was size of the local shortfall, 1,000 miles away from the problem. You can see that at other times, the battery has put out its full 100MW rating.
Tesla's battery is better at doing what it was designed for, compared to coal plant(s) that are better at doing what they're designed for. In conjunction with one another, both technologies work incredibly well.
In addendum to this, I'd like to mention that things that are designed for a specific purpose are good for that specific purpose.
And in general, things that are made to work well together work better than things not made to work well together. This also applies to things that happen to work well together, despite not initially being designed to.
This response by the battery is really interesting. And it is certainly helping the Australian grid.
However, it is worth noting that the Australian grid, South Australia's in particular, is so woefully mismanaged that almost anything goes. South Australia is in the running for the most expensive electricity in the world [1].
As far as I've been able to tell, we aren't building coal or nuclear because there are outrageous political risks (opposition party & green lobby policy will probably make them uneconomic in the next decade); we aren't building anything else because because the economic option is coal. I think there is some sort of fracking ban that seems to have caused a gas shortage - I don't really understand that.
Long story short; Australia has looming electricity generation problems due to government policy and this battery is helping, but it is not fixing them.
> we aren't building coal or nuclear because there are outrageous political risks
This is a one sided way of putting it. The people who build coal plants in Australia say that they've stopped doing it because of climate change. The plants need to run for decades to pay off the cost of construction. The business people expect that, whoever is in charge of Australia, power plants will have to stop burning coal before then.
The people who build nuclear plants say that there aren't enough Australians to buy their electricity, so anyone who tried to sell nuclear power in Australia would lose money. Again, regardless of the politics.
I guess in another market somebody would bet on the nuclear and hope to draw people with lower prices. Lets see how deep the confections are.
But since build those monsters is such a long process and the approval alone is so difficult already. If you want to develop something new, you can add a couple years to the project.
The hope is China, they are selling mid-sized pebble generators and they should start producing them pretty soon for delivery in a couple of years.
> we aren't building coal or nuclear because there are outrageous political risks
Wrong. We aren't building coal or nuclear for reasons of economics.
The high electricity prices are partially due to gold-plating the poles and wires due to very odd 'oversight' agreements in which any investment in poles and wires infrastructure had a guaranteed return on investment - it allowed the infrastructure owners to increase their charges to cover the poles and wires investment, almost no matter how much they invested, so they went LARGE.
High power prices are also partially due to the generators gaming the market in times of high power usage and low wind / solar generation. Making use of loopholes in the 20-year old rules. Loopholes that didn't exist when everything was coal, but caused by the fact that the rules haven't been updated to reflect the inclusion of wind and solar into the generation mix. The "Market Operator" had been asleep at the wheel for years.
Politics is getting in the way of building Wind and Solar , not the other way round.
If you're including regulatory oversight (or lack of) under the 'management' heading, I'd definitely agree.
"Since 2009, the electricity networks that own and manage our “poles and wires” have quietly spent $45 billion on the most expensive project this country has ever seen. Allowed to run virtually unchecked, they’ve spent vast sums on infrastructure we don’t need, and have charged it all to us, with an additional fee attached. The spending was approved by a federal regulator, and yet the federal government didn’t even note it until it was well underway." [1]
> As far as I've been able to tell, we aren't building coal or nuclear because there are outrageous political risks ...
That adjective 'political' is not needed there, and serves to distract from the bigger, actual risks attached to those obsolete technologies. (With my usual caveat that I'm assuming people using the word nuclear mean only fission.)
> .. we aren't building anything else because because the economic option is coal.
Is it technically still an 'option' if you think there's only one?
Anyway, we (Australia(ns)) are not special.
Coal is not a sensible choice -- economic or otherwise -- anywhere on the planet.
Unfortunately many places, including here, haven't comprehended that yet. Or rather, the people that run the country -- both within and outside of the government -- prefer the extant power generation and delivery infrastructure for a variety of fairly obvious reasons.
Nuclear is obsolete to what? As far as I understand, nuclear is the replacement for coal, nothing else comes close to being able to provide a stable constant supply in large quantities.
What else can take the role of supplying the bulk of electricity to grids that doesn't also have it's own environmental concerns?
What else can take the role of supplying the bulk of electricity to grids that doesn't also have it's own environmental concerns?
Gas fire plants
Hydro
Pump Solar
Liquid salt storage
Geothermal
Wind + Large Scale Solar + Storage
Tidal
Decentralized PV solar plus decentralized storage.
There are enormous numbers of combinations of solutions which don't involve coal or nuclear. Both coal and nuclear power are slow to respond to changes in demand (or supply problems), and both are very expensive centralized solutions.
Much more efficient solutions can be built using smaller numbers of smaller, faster, cheaper power stations (ie, gas fired) plus a diverse combination of other sources.
Gas: Shits out CO2 and others, short lifespan, order of minutes to respond (less if forewarned)
Hydro: Used in the UK to provide peak backup (however thats pumped storage not generation. given the droughts and irrigation usage in aus, not the best idea) requires time to warm up. Has large environmental impact unless using a natural dam (~60 seconds to respond)
Liquid salt storage: See above, less efficient, slow response time (unknown response time. Suspect in the order of tens of minutes, as heat has to be transfered into water for steam, better as base load)
Geothermal: needs the correct rock formation (again, unknown response time, better as base load)
Large scale solar + wind: Storage is the problem. The amount of lithium required to proper storage is utterly vast. (depends very much on the storage medium and controller, also the amount of wind solar in the preceding days)
Tidal: requires the right geology, can have massive environmental impact, power comes when it wants, but is predictable.(Has some ability to provide backup outside of high/low tide. )
Decentralised x: horrifically expensive, however the cost can be passed directly to consumers. Keeping it all in sync requires decent infrastructure. (without a new and secure mechanism unlikely to provide grid backup, but could be used to dynamically cut load. )
In conclusion nuclear provide a brilliant base load, which can be used for things like charging the various storage devices, or allowing for solar to not work at night. Its relatively cheap across an entire life cycle, and surprisingly dependable. In terms of environmental impact, significantly less that the constant low-level pollutants that are emmited by coal & gas, and the environmental changes caused by hydro. Also Lithium mining has a large environmental impact too, as does PV production.
nuclear... relatively cheap across an entire life cycle
> Between 2002 and 2008, for example, cost estimates for new nuclear plant construction rose from between $2 billion and $4 billion per unit to $9 billion per unit, according to a 2009 UCS report, while experience with new construction in Europe has seen costs continue to soar.
And if you read the studies they require short term peak fossil fuel standbys.
"The real challenge is to supply peaks in demand on calm winter evenings following overcast days. That’s when the peak-load power stations, that is, hydro and gas turbines, make vital contributions by filling gaps in wind and solar generation."
In the UK that translates to gas/diesel standbys which in the case of the latter is horrendously polluting
Now, if you look at all of those studies, they make the assumption that there there is gas/hydro to act as backup. that is a base load.
The NRDC link ~25-50% of the power comes from Gas, which puts out CO2. Then there is biomass, which is also environmentally problematic in its current form. (https://www.nrdc.org/sites/default/files/styles/full_content...) Granted that does not take into account grid scale efficient storage.
But there is a base load, you can see it in the graphs. They almost exclusively use gas to plug that gap. If you want to reduce carbon output, that has to go.
Now, if I was to be in charge of a country, my route to energy security with minimal environmental impact would be thus:
1) insulation at a massive discount/loan.
2) diverse renewables, but not at any cost, and not on farm land. (UK based, so tidal, solar, wind)
3) small scale nuclear. Molten salt or similar with a negative feedback loop
4) Domestic & industrial power storage/generation incentives. If you can adjust the demands at source it greatly reduces the highs and lows.
Why in the UK is something like nuclear needed? Because if we are to reduce carbon output, we need to heat our homes with electricity instead of gas, that triples the power demand at a stroke. insulation standards of 0.6U or less would reduce this significantly. However its still more power than can be feasibly harvested with rooftop solar/wind/tidal alone.
Ah, the idiotic libertarian response. Just the other day I got a small cheque because I'm an investor in a local solar scheme.
Renewables work, have very modest maintenance requirements, and the intermittency can be managed. It's ridiculous that Australia isn't getting at least 30% of its energy from solar.
"Got a check" isn't the same thing. Somebody subsidized that check I think. We also have cities subsidizing aluminum recycling, which pays nobody and has no ecological benefit.
I'm just practical about what theatre I endorse, and what I skip.
Of course, but the coal is on a favourable tax regime and the nuclear plants won't be built without subsidy either. Hinkley Point C is guaranteed a price of £92.50/MWh regardless of the market rate.
Aluminium recycling is actually has lots of ecological benefit. It uses about 95% less energy to recycle aluminium than it does to produce it. In terms of subsidies to recycling you've chosen the item that would always be recycled no matter what.
> Nuclear is obsolete to what? As far as I understand, nuclear is the replacement for coal, nothing else comes close to being able to provide a stable constant supply in large quantities.
The article pretty clearly demonstrates that grid-scale battery storage systems are able to provide very fast reaction times.
More battery storage systems, along with other forms of energy storage such as Solar-Thermal and pumped-hydro can take daytime generation and make it usable at night. They're proven workable grid-scale technologies.
Large widely distributed deployments of Wind generation can also provide a lot of capacity. For those who claim wind doesn't always blow - well, it's always blowing somewhere.
Having a lot more individually smaller generators plus energy storage will inherently permit a more stable grid - one smaller generator tripping offline doesn't cause a massive issue.
And a fracking ban didn't cause the gas shortage (whoever told you that has a pretty serious agenda!). Australian gas reserves are the second biggest in the world and relatively easy to access. The "problem" is/was that they are tied up in long term export agreements.
Yeah, the contracts are ridiculous. Santos et. al. are locked in to exporting too much of our production to countries like Japan (more than they even need anymore I beleive), there isn't enough of the gas produced being sold domestically so prices locally have jumped significantly, but at the same time the international price has dropped so I don't think the export is even breaking even, and I've read that at this point we could import the surplus back from Japan (that we sold them originally) for less than what we pay on the domestic market!
This is a huge problem because gas is required for a lot of our industry, but also because of where it sits in the energy pricing mix, the gas price almost entirely determines the price of electricity - https://www.macrobusiness.com.au/2017/07/truth-behind-electr...
The economists at MacroBusiness (I linked an article from their site above) believe a gas reservation policy (requiring the producers to sell a larger portion of their supply on the domestic market) could fix a lot of this.
> I've read that at this point we could import the surplus back from Japan (that we sold them originally) for less than what we pay on the domestic market!
What's sillier is that apparently we wouldn't actually import the gas. I've spoken to people associated with the gas industry who have told me that we would actually buy a future allotment of gas from Japan, which would be pretty much directly supplied from Australian providers. In this case the Japanese companies basically become re-sellers who can undercut the local cost of buying the same gas, without that gas ever actually leaving the country.
If this is true, it doesn't surprise me. Australia has a terrible track record for taking advantage of primary resources. We've done similar things with iron ore, oil, gold, diamonds, coal and uranium (and probably pretty much everything else we dig out of the ground).
My understanding is that no one wants to build new coal power plants because renewables are going to be cheaper regardless of any government subsidies.
This is not the first time I've heard Australians complaining about the green lobby.
From what I understand one of the major problems Australia is the sheer distance the electricity has to travel and the infrastructure required for such a small population.
Solar power seems such an obvious choice, yet many Australians seem to be against it.
In Australia, being green or anti-green is a big part of politics. The party currently nationally in power is anti-green.
Interestingly, because of the combination of great solar conditions and long distances, if you want to build a house in the middle of nowhere, it's cheaper to buy solar and a battery than hook into the grid.
I think many Australians would appreciate that we are in a good position to exploit solar potential, but pitched as part of the green/renewable solution, it's quickly tied into tribal politics and people get a bit silly.
I don't think anyone (except the coal lobby) is against solar power.
It does have a problem that we do have nights here in Australia, and it is only recently that batteries have got close to the break-even point in terms of cost.
We also have large amounts of wind power, but again it comes at the cost of not being on 100% of the time. Again, storage can solve that as it continues to drop in price.
The takeaway (if I'm reading the graphs correctly) is that the Gladstone coal unit took over 3 minutes to respond to the Loy Yang failure. It was contracted to respond within 6 seconds, but couldn't. Fortunately the Tesla battery was able to respond within milliseconds and cover the gap. It's a good bit of PR for Tesla, and could have some negative implications for coal or gas fired backup units.
Where did you get the three minute response time number for Gladstone? It is not backed up by the article, it only mentions that it should respond within six seconds, but that "Data from AEMO (and gathered above by Dylan McConnell from the Climate and Energy College) shows that the Tesla big battery responded four seconds ahead of the generator contracted at that time to provide FCAS (frequency control and ancillary services), the Gladstone coal generator in Queensland.". I would love to see more information on the actual impact that this had.
edit: Updated with the response time from the article
While we're extrapolating meaning, I noticed that the first graph shows the HPR kicked in a few seconds before the frequency stopped dropping. Does that mean that the HPR did little to nothing to arrest the fall in frequency, and that it was the Gladstone generator, which the article says activated four seconds after the HPR, which deserves the credit?
4 seconds faster (within milliseconds of the frequency drop event occurring) than the coal plant contracted to provide frequency response services.
"But in reality, the response from the Tesla big battery was even quicker than that – in milliseconds – but too fast for the AEMO data to record.
Importantly, by the time that the contracted Gladstone coal unit had gotten out of bed and put its socks on so it can inject more into the grid – it is paid to respond in six seconds – the fall in frequency had already been arrested and was being reversed."
TL;DR The reserve system did this to show its skills. It can respond faster than any existing base load generators, and can derive revenue not only from providing energy on demand, but also grid services (in this case, frequency response).
When a contingency happens (loss of equipment such as a generator), units respond pretty much immediately due to governor response in the unit itself (good ole' inertia). Most RTO/ISOs in the USA (I know AEMO is in Australia) have a very large amount of power already ready to go if a unit trips offline. Using batteries to help with this is being worked on. They already exist, but new market rules are being drafted in order to allow them to be taken advantage of and properly compensated. The term is stored energy resource. Google that term in conjunction with ISO-NE, PJM, MISO, SPP, NYISO, CAISO...etc.
The requirements were relaxed because the intention was to have an entirely market-driven mechanism for frequency control. The system reliability impacts were an unintended consequence.
It's worth pointing out that this battery only solved a second order problem: what to do when one thing unexpectedly breaks. The big technical problem in Australia is what to do when the weather forecast says that a whole lot of things will stop generating power at the same time, and it would take a very large battery to help with that.
An early test of the battery control system and it responded!
What the article does not say is that the lumbering coal fired units also responded within milliseconds due to the enormous inertia of their turbo-generators. This happens before their throttles open purely due to the physics of the situation.
>the Tesla big battery responded four seconds ahead of the generator contracted at that time to provide FCAS (frequency control and ancillary services), the Gladstone coal generator in Queensland.
I find it curious that a coal plant would be contracted with keeping the frequency stable when both hydro-storage and gas power plants provide better response times than coal. Is it normal in Australia that coal plants do grid stabilisation?
Gas is tied up largely in export contracts, and the whole thing about Australia being both somewhat flat and dry means hydro basically doesn't exist (it exists only in the literally 1 place on the mainland continent we have any "significant" snowfall).
The biggest surprise to me was that a plant in Gladstone (Queensland) was contracted to prop up a plant in Victoria. These two places are 1375 miles apart. Then again, I guess the linky bits between the wires must be pretty high-speed, so the speed-of-light propagation of the grid conditions wouldn't take too much extra time.
I'm not a grid-scale engineer. My guess is that these batteries work by pushing more electricity into the grid when voltage / frequency is low; and putting less electricity into the grid when voltage / frequency is high.
Out of the wall in the US we get 120v @ 60Hz (elsewhere is 220v to 240v @ 50Hz). If the load becomes too high, the frequency starts to drop.
Inductive (motors and such) loads do not fare well when the frequency drops which can cause serious damage. Capacitive or Resistive loads can handle a bit more fluxation, but the whole system is built around the idea that voltage and frequency are fixed. Fluxations can cause grid failures, brownouts, or damage to consumer electronics.
AFAIK, that is precisely what is happening here. The batteries are dumping energy into the grid to bring the frequency back up to where it should be. What's cool is that Tesla was not contracted to do, but did anyway. Good PR.
Maybe significantly cheaper though. Our generators often advantage of these events to gouge the grid by having 'unexpected maintenance' and restricting supply to force the market operator to buy their capacity at a much higher price. Having this battery act in a way that isn't profit motivated could actually change this market a lot.
What I gather from the (incomplete) data, indeed the backup plant would have fired up in about 6 seconds. Isn't it wasteful/inefficient/expensive though, to fire up a whole plant if just providing a few seconds of electricity from a battery is enough to bring a grid back to nominal frequency?
The reason the frequency dipped is because not enough generating power was online to supply the demand. In order to bring it back up to nominal frequency and keep it there, something had to keep supplying the extra power required for as long as it was being used.
Water is wet and using a battery to cover high speed changes in supply/demand works exactly the way you'd expect at the car, datacenter and grid scale.
But in an industry like power generation / "the electric grid," the obviousness of this might be outweighed by momentum (and money?) for the status quo. Especially in regulated services like these (roughly speaking, and depending on the implementation, the industry behaves like a monopoly with more/fewer government-imposed controls), the message and the politics are important to line up, or else a superior technology might only be adopted decades after it's actually viable.
So you might find it to be fluff, but it's good for people to read actual in-the-field success stories.
It's not obvious to Joe Voter who wonders why the South Australian government spent millions of dollars on this battery boondoggle by some American millionaire.
The more the message gets out that the battery is working, today, and is already covering for shortfalls / screwups by large scale (non-renewable) generators the better.
Battery power is always faster then generator power. That is the reason why every data center has battery backup for immediate reaction and generator power for the longtime backup. Normally, the battery size is designed that the minimum time is at least twice as long as to start the generator.
Exactly that has happened with the grid. It is also not related to some contracting. Because the whole grid is connected. So the battery will always kick in, it does not matter where it is located.
Actually synchronizing the grid is tough task. A friend and colleague from my University started in this business after he made his Master Thesis tens of years ago.
By monitoring the frequency of the grid you know if there is too much energy produced (too less load, frequency to high) or vice versa. Everybody basically can do this from home. There are some maker projects and personal projects who do that.