Victoria wind and solar farms warned of curtailment

Print Friendly, PDF & Email
Ararat wind farm

Ararat wind farm: facing curtailment?

Owners and developers of existing and planned wind and solar projects in north-west Victoria have been warned that their facilities run a high risk of “curtailment” – a deliberate cut back of their generation capacity – because of the weak grid in the area.

The warning, given by the Australian Energy Market Operator, specially mentions three new wind projects and five new solar projects that are due to come online within the next year.

AEMO also warns of a likely further decrease in “marginal loss factors” – the key calculation that acts as a multiplier of the estimate output delivered by power generators – and advises that some projects may need to consider battery storage to avoid output being spilled.

The warnings about curtailment and reduced MLFs will need to be considered carefully by wind and solar project developers and owners, because of their potential to have an impact on project economics, and on the ability to source finance.

Marginal loss factors have already become an issue in north Queensland and western NSW, and to a lesser extent in Victoria, as the shape of the grid changes, and more renewables are built in certain areas, and major demand errors also shift.

Victoria is emerging as one of the next major crunch-points because of the strength of the state’s renewable energy target – 40 per cent by 2025 – and the relative weakness of the existing grid, particularly in the western part of the state.

The problem was identified in a major report that we wrote about last year – Victoria’s big renewable energy plans face major network hurdle – when AEMO warned that curtailment rates of up to 50 per cent could be experienced if the network was not upgraded.

AEMO grid constraintsIt pointed, in particular, to the 220kV line that links Ballarat, Horsham, Red Cliffs (a key connection spot between Victoria and NSW), Kerang and Bendigo, and this is again the focus of its latest update.

It notes that between Ballarat and Horsham there are already two substantial wind farms, Ararat and Waubra – and these will soon be joined by the 80MW Crowlands, the 204MW Bulgana and the 226MW Murray Warra wind projects – that may see their generation capacity limited.

“With these five wind farms connected, the (Ballarat to Horsham) line will, at times, be loaded beyond its thermal capacity,” AEMO says. “Associated network constraints may result in limitations on the generating capacity of these wind farms.”

AEMO says there has been significant interest in further wind and solar projects and also makes a similar warning to proposed solar farms in the area – the 110MW Wemen, the 100MW Bannerton, the 90MW Karadoc, and the 81MW Yatpool solar projects.

All of these are due to be connected in the middle of this year, and with new projects also being commissioned on the other side of the border in NSW, which influences flows coming south via Red Cliffs, then issues may arise.

“If further development proceeds as suggested, this will add to the thermal constraints in this area,” AWMO warns.

“The connection of new generation in (north-west Victoria) has contributed to Marginal Loss Factors (MLF) in the area decreasing, with some generators experiencing a decline in over 3 per cent from 2017/18 to 2018/19.

“It is likely that with the addition of more generation in the area loss factors will continue to fall.”

This is particularly important for project developers. Many will have – or should have – taken reductions in MLFs into account when putting together the business models for their projects.

The MLF is a calculation used to estimate how much a plant’s output actually reaches a destination. An MLF of 0.9, for instance, suggests losses of 10 per cent, so a wind, or solar plant will be credited for just 90MWh out of every 100MWh registered at the meter at the plant.

The MLFs also apply to fossil fuel generators, but the worst affected – peaking plants in regional areas – don’t care so much because they only switch on when prices are very high, so are guaranteed of big profits regardless of a reduced MLF.

AEMO says the problems will be exacerbated as the number of connections increase along these lines, and suggests that it will apply to smaller installations too.

“Where considered reasonably necessary for adequate system operation and maintenance of power system security, generators below 30MW will also be required to be included in central dispatch,” it says.

“The relatively low system strength in North-West Victoria may result in additional constraints and works may be required to ensure the stable operation of additional generation. These works may include installation of synchronous condensers or other plant to raise fault levels.”

It says grid upgrades are an obvious solution, but it is not yet clear that there is an economic case to do so, nor that anything could be delivered within seven years – when the VRET is due to be met.

So it suggests that some wind and solar farms consider battery storage, locating the project elsewhere, or even paying for its own infrastructure upgrade.

“There are a number of things that could assist in alleviating the network congestion, including:

“If your project is in its early stages, if possible consider alternative network locations, or investigating potential energy storage solutions to store excess electricity generated by your wind or solar farm.”

Ironically, some wind and solar projects are already doing just that. Bulgana and developer Neoen is proposing a 20MW/34MWh battery storage project to help deliver 100 per cent renewable energy to Nectar Farms’ new vegetable greenhouse as part of a $550 million project.

Ganawarra is also building a 25MW/50MWh Tesla battery to accompany its newly completed 50MW solar farm, while a battery storage facility is also being proposed for Ballarat, using Fluence technology. It will be contracted to EnergyAustralia.

These, however, will likely only address curtailment issues, and not necessarily marginal loss factors.

AEMO is also warning of imposing pre-contingent constraints that will apply to generators (or groups of generators) to limit the total lost generation for a single contingency to approximately 600MW.

The implication there is that it will take action if too much capacity is loaded on to one line, given the potential for a network fault to take out significant capacity, and the need to find an alternative source.

Such contingencies already exist for large fossil fuel plant – such as if a 500MW coal unit trips suddenly, as they are wont to do, particularly in summer. But this may be the first time a similar approach is taken for groups of wind and solar farms.

All of a sudden, the developers of wind and solar farms have a lot more to think about than the cost of generation and the striking of power purchase agreements, or the state of the wholesale market.


  • Greg Hudson

    It sounds like the 22kV line isn’t gold plated, but is plated with fools gold instead !

  • Tom

    Dunno about the choice of headline pic. I doubt Ararat wants to be represented by an old turbine with a pitch leak.

    • Hettie

      Some cows grazing on a green hill.

      • Tom

        One blade is stained brown with grease.

        • Alastair Leith

          thought you meant blade of grass for a minute!

      • Cheetaroo

        Milk cows. In Ararat? We only do beefers.

        • Hettie

          Don’t blame me. I am not the one who took the pic.

  • Malcolm M

    Why are so many developers focusing on Victoria’s north-west, feeding into a single circuit 220 kV line with a rating of 359 MW, rather than developing projects in south-west Victoria that would feed into the dual-circuit 500 kV line with a combined rating of nearly 6,000 MW?

    • Jonathan Prendergast


      • And wind.

        • Vox

          And excess population density, with many smaller landholdings and a lot of lifestyle plots.

    • Ken Dyer

      Victoria has always been coal centric as far as electricity is concerned. This means South East out of the Latrobe Valley.
      North West Victoria has always suffered. Internet access and mobile phone coverage up that way has also always been lousy.
      I remember talking to a local out Stawell way a few years ago. Being a city slicker, I asked her about internet access on her property. She said she would gladly get it when she eventually got electricity.
      Curtailment is a good sign that the decentralisation of the network with renewable energy is working. I will bet that the gold plated network out of the Latrobe Valley is becoming seriously underutilised.

    • Jon

      You have to also ask how these issues got past investor due diligence?

      • Alexander Hromas

        They most likely picked it up. The regulator was up till recently ignoring line losses as these generators were helping to stabilize the grid and so they got a 1.10 energy rating i.e. the energy that they generated was worth 10% more than its actual total sum. The figures may not mean that much anyway as these companies try to secure long term forward contracts on a large portion of their energy while leaving some over for short term speculative sales when prices are high. A weakness of our privatized system is that the transmission system owners see little value in upgrading a particular asset and if say their coal fired plant is in direct competition with a renewable on their network they have an excellent disinsentive

      • Vox

        The developers also take a view on these factors, and the banks are usually quite sanguine to ensure they do not lose out.

        Both curtailment and dropping MLF factors are usually included as model assumptions, and so the projects that go away (with finance), did so on that basis.

        That said, there are developers that look at the equations with a bit more sophistication than others, and that will make a large difference.

  • Peter F

    This is all good, it focuses development where the total cost of energy supply is the least. Eastern Victoria and rooftop solar will be advantaged and the grid will become more resilient as a result. Therefore we will ultimately have a cleaner, lower cost grid. If the northwest is so superior then it will pay developers to provide storage or grid infrastrurture

    • Tony Wilson

      Not so. Rooftop solar is horrendously expensive compared to utility scale generation. The only reason it is popular is because of our weird market structure such that it competes against retail power (~$350 MWh on average) instead of wholesale power (around $85 MWh). Further, eastern Victoria has relatively poor solar sites (less sunshine per year) and may have fewer good wind sites also (though probably still more than enough).
      The sensible thing to do is build modern infrastructure in the north and west. With good planning, this can be multi-purpose: not only to provide cheap power to the population centres in the south, but also to provide better linkages with NSW and South Australia. The better the grid, the further you spread the diversity of supply, and the more able renewable are to provide consistent power.

      • Peter F

        In Victoria the Transmission, distribution and retail costs account for about 70%+ of the cost of power most of these are avoided with rooftop solar.
        It is true that the solar resource is weaker in Eastern Victoria as is wind, but because the wind will generally be out of phase with wind farms in the north and West and transmission expenditure will be almost nil the reduction in transmission and backup costs will offset the lower productivity of the sites

        • itdoesntaddup

          The correlation of wind across Victoria is fairly high, with most reduction of correlation caused by delay of a few hours as a weather system moves through. There is no anti-correlation. The problem is real, and can’t be waved away.

          • Vox

            I think you’ll find there are very different weather systems affecting the coastal wind farms vs the mountaintop wind farms vcs the plains wind farms. This does make for a somewhat uncorrelated set of generation in most occasions.

          • itdoesntaddup

            I’ve actually looked at the statistics from BOM/AEMO in their renewables study (hourly wind traces by area), and done the correlations myself. Have you?

          • Vox

            Not myself, but I’ve seen analysis for that purpose using AEMO data and on-site wind data.

            All I can say is that there are regions that are surprisingly uncorrelated with very interesting wind patterns you don’t see on other wind farms.

          • Peter F

            I understand that anti-correlation does not exist but a delay of a few hours between McArthur and say Rosedale means that is a few hours less storage. A new windfarm at Rosedale with a MLF of 1 and a capacity factor of 40% is a better use of resources than yet another solar farm near near Mildura that has a capacity factor of 28% and an MLF of 0.85

      • David Osmond

        I disagree that rooftop solar is expensive. A 5kW system on your roof for under $5,000 after STC subsidies works out as about $8,000 pre-subsidy. That delivers solar with a LCOE of about 9c/kWh assuming 5% discount rate, a new inverter after 12 years and 25 year life. 9c/kWh is only a fraction higher than utility PV. And if it is mostly consumed on-site, then it helps reduce transmission losses. Rooftop PV is awesome.

  • Hettie

    The more people who get riled by stuff like this, the higher the likelihood of a change of federal government.

  • RobertO

    Hi All there is an AEMO ISP idea
    Kerang – Darlington Point path
    The proposed Kerang – Darlington Point path creates a new corridor for high power transfers between Victoria and
    New South Wales. This option would be most beneficial if large amounts of variable generation connect in Western
    Victoria. It can be combined with proposed upgrades from South Australia to New South Wales or Victoria (see
    4.3.4). This option includes the following work:
     Install a Sydenham – Bendigo – Kerang – Darlington Point – Wagga double circuit 500 kV line.
     Install a Wagga – Bannaby – Lower Tumut – Wagga single circuit 500 kV loop.
     Construct 500 kV substations at Bendigo, Kerang, Darlington Point, Wagga, and Lower Tumut (or expand existing
    substations to accommodate 500 kV plant).
     Install four 500/220 kV transformers – two at Bendigo and two at Kerang.
     Install six 500/330 kV transformers – two at Darlington Point, two at Wagga, and two at Lower Tumut.
     Install a phase shifting transformer on the Bannaby – Sydney West 330 kV line.
     Install an additional 330/220 kV 240 MVA transformer at Dederang.
     Install additional reactive plant.
    This option would increase the Victoria to New South Wales transfer capability by 2,100 MW towards New South
    Wales and 1,700 MW towards Victoria. The approximate cost of this option is $2,700 million.

    This is the full build idea but there should be short term build parts which could help move the new supply into NSW as we build stronger connections south to Melbourne.
    Other parts would also need to be started if this is to proceed. I suspect the timeframe is a little long.

    • BushAxe

      It’s more than an idea, one can read between the lines and see what the industry is pushing -new large capacity interconnectors between the eastern states and SA.

    • Peter F

      $2.7b would buy a lot of solar thermal or other storage. We can build all the renewables we need within 180 km of the population centres, like France and Spain are doing with hundreds of small wind and solar farms, where 90% of the energy is distributed at 66 kV or less and 70% of it is used within 150 km of the generator

      • RobertO

        Hi Peter F So a smaller 220 kV link from Kerang to Deniliquin will not help (and can it be built as a 500 kV system but used as 220 kV, for forward planning to link into Darlington Point 500 kV system).

        I agree with the idea of local supply and local use, but you also need to look at interconnects as part of the equation and how many do they have in Europe (and what Europe doing with new interconnects 45 billion euros is what their debating). In the AEMO report for Vic grid there was also some small $ spends that would improve this area.

        There are lots of little things that we as a country need to do much better than we currently do. Making Batteries in Australia in better than importing all our batteries we will need.

        • Peter F

          It is not an apples for apples comparison but the interconnectors between the NEM states are already far stronger than most European interconnects. The European standard is at least 10% interconnect capacity.
          I do think that upgrading Kerang- Deniliquin and NSW Qld a bit would help diversify the grid but it is more for reliability in case of transmission faults rather than moving much power around. Germany at the centre of the European Grid can only export about 15 GW of just under 90 GW peak production Victoria has export capacity of 2.3 GW vs peak production of about 10 GW. NSW can export about 3 GW vs 15 GW peak so we are about right.
          Of course there are many things we can do and there is not one silver bullet, making batteries could be good, converting all hot water services to grid controlled electric or direct solar, space heating to heat pumps are also very effective energy conservation and GHG measures which save users money and improve utilisation of grids and existing generation capacity

  • Has the fear of ‘gold-plating’ gone too far? Western & Northern Vic are where the lion’s share of the state’s renewable resources are and it’s not as if the transmission deficit in the area is anything new.

    220kv, 500kv and new 275/330kv interconnectors with SA, NSW or both states are all outlined from page 22:—Project-Specification-Consultation-Report_FINAL.pdf

    From the La Trobe Valley, there are 4 x 500kv circuits to South Morang, then 3x circuits to Calder Park and then 2x circuits from there to Portland. The AEMO planning doc linked above has a possible 500kv link from Ballarat direct to Calder Park as an option and interestingly enough if they’re worried about land clearing for the easement, the Western Freeway between Melton and Ballarat is relatively direct, wide and ripe to take on that role.

    If Vic’s going to go beyond the 40% RE target eventually (why wouldn’t they?), Western & Northern Vic is where most of that new generation is going to be located.

    • Alastair Leith

      Wasn’t the gold plating largely on the distribution networks rather than transmission?

      • Hettie

        What about all the gas peakers that get used about 6 hours a year? Aren’t they a big part of the gold plating?

        Anyway, surely the sensible response to having more power in a location than can be transmitted, would be to attract energy intensive industry to that area. You beaut EVs for the industry staff with free at work charging…..
        The clever can surely think of ways to use on site that power which can’t be shipped out.

  • mick

    audrey audrey audrey what did you tell them that for let them build then force upgrades on the government

    • Jon

      She has more integrity than that

      • mick

        mores the pity

  • George Darroch

    Why are we not building more poles and wires in this area?

    I know there is an obsession with “gold plating” but this isn’t gold-plating. This is basic improvements to a system that needs more capacity.

  • Jon

    There shouldn’t be too much sympathy for the developers / investors in projects in this area. The issues in this part of the network have been known for a long time. The efficient delivery of new generation to the market should take into account network constraints and MLF risk. It encourages the full utilisation of existing infrastructure before we build new.

    • Vox

      But there is an equation that AEMO needs to make when assessing future expansions: Where is the power delivered going to be cheapest?

      The the age of the Fossil Fuels, it was by coal mines, hence the large lines going to Hazelwood and Yalourn. Now it is in the western plains, where large wind farms and solar farms can be built on farming land, where hobbie farmers don’t complain of soured views and destroyed lifestyles.

      So if AEMO wants to deliver cheaper power, it will need to provide capacity for the places where the power can be generated the cheapest.

      • Jon

        AEMO doesn’t provide any capacity, just advice as to what the loss factors are. Investment in the past was made by governments using public funds, now it is the responsibility of private investors.

        • Vox

          I would suggest you read up on the RIT-T process undergoing currently.

  • Tom

    I could foresee this situation ages ago, and this is why I’ve been interested in 1414 Degrees, and why I’ll be participating in their IPO.

    It is clear that in many places it will cost more – possibly several times as much to upgrade the transmission infrastructure than to build the generation infrastructure. We’ve got this argument in Tassie with the proposed second Basslink, which is never going to happen for this exact reason.

    The problem with variable energy is that it might only be utilising the transmission infrastructure at 25-35% of its total energy carrying capacity: 100% of its power transmission capacity when it’s sunny & windy, but 0% when it’s dark and calm.

    That’s where on-site energy storage comes in. With thermal storage & regeneration (whether by steam or by Stirling engine as 1414 Degrees is using) it might cost 1MWh to store and regenerate 1MWh (50% efficiency), but new generation capacity is so cheap compared to upgrading transmission lines that it may well be worth it.

    Sure, batteries are 90%+ efficient, but how will this translate to LCOS?

    I expect that batteries will never have a role in grid-scale storage (unless there is a major breakthrough in chemistry, such as the Aluminium-Graphene-Air batteries which showed so much promise but haven’t yet delivered). They will have a big role in domestic storage, local network backup storage, and grid stability, but not powering the grid for an overcast winter’s week. Thermal + regeneration will almost certainly have a very big role, unless there is a widespread pumped-hydro boom.

    • RobertO

      Hi Tom You have hit the nail on the head with “That’s where on-site energy storage comes in!”
      You’re sitting on the largest battery system in Australia. Add to the system the wind (factor in higher than most other places). Add security to the system (and I want to add community benefit to King Island) I believe it will go ahead and it will not be a MW system but a GW system possibly in the 800 kV HVDC cable. Given that Hydro Tas and ARENA are investigating this I suspect that the profit factor will drive this (and efficiency is not a concern given the security it brings).

    • itdoesntaddup

      You may well be right about alternatives to PHES (the first problem being you have to get the energy to and from the storage site). Nevertheless, this is going to add significant cost, even if it is cheaper than building more long haul transmission capacity. It’s the problem of the resource being a long way from the demand centres, which was not really the case in the days when the demand centres grew because there was a local resource along with other locational advantages: ports for shipping, coastal milder climate, more water, etc.

    • TheTransition

      Very interesting. I’ve long had the suspicion that the transmission costs of large scale variable RE were way under-estimated. It makes Nuclear look more affordable even though the capital costs are much higher per $ of peak capacity. I suspect pumped storage near existing transmission infrastructure will win in Australia though. There are good arguments for a 2 GW facility 34 km from the old Hazelwood station near Mount Baw Baw.

      • Tom

        I’ve looked at that site. That’s exactly why the whole “Tasmania: Battery of the Nation” concept is never going to work.

        You’ve got that site in Vic using Thompson Reservoir as the lower dam and the Mt Baw Baw plateau as the upper dam which compared to potential Tasmanian sites would be as cheap to construct (still very expensive, but cheaper than Snowy 2.0). It would store more energy depending on how much land you’re willing to flood. And it’s very close to gigawatts of transmission capacity compared with building billions of dollars worth of new Basslinks.

        PS – I’m not sure how this gets linked to nuclear? Nuclear is never going to be anywhere near affordable, and it’s even more uselessly “baseload” than coal (it’s output remains constant while demand varies).

        • TheTransition

          Re Nuclear. You don’t need as much RE, interconnects or pumped storage for the same capacity. Fuel and running costs are very small (similar to RE running costs) very long lifetime (60-80 years). Anyway, the discussion about Nuclear is still at least 10 years premature in Australia. We’re not even close to 50% RE for electricity. Let alone replacing all the other Fossil Fuel burning we do to make our economy go. It’s after that I think all the costs you talk about will become more obvious to everyone.

          • Hettie

            Please wake up. It’s not the fuel costs make nuclear peak stupid.
            It’s the time and cost to build, the overheads like insurance against accident, the costs to decommission. All the reasons that nuclear plants around the world are going bust. All those costs have to be met from the selling price of the product, and power from renewables is now around one third of nunclear.
            Besides which, we don’t have ten years to wait.
            Look around you.
            Look at the failure of food crops in Africa. The results of droughts followed by floods in Australia, the increasingly severe and frequent extreme weather events.
            Yes, I know that the climate scientists say that it’s not possible to say that any one specific tornado or bush fire as caused by climate change, but they also say that the pattern certainly is caused by AGW.
            Work done last year to ascertain which of the many CC models is closest to accurate, by running them all backwards, feeding in the current stats, to see which produced an answer closest to the starting point, showed that the worst case model came closest.
            This is a climate emergency.
            Renewables and battery are by far the fastest generators to build. The switch over to EVs over the next 10 to 12 years will about double demand for electricity worldwide. The time nuclar takes from concept to commissioning is so long that by the time it is half built the LCOE will be so much higher than the general market that the build would be abandoned if indeed any financiers were dumb enough to fund one.
            Forget nuclear.

          • Tom

            @Hettie – absolutely.

            Nuclear in a 3-word slogan: Hinkley PPA price.

            Ok, PPA isn’t really a word, but 90 pounds (approx $150 AUD) per MWh is pretty hard for even rusted-on nuclear proponents to justify.

          • Peter F

            Worse than that Tom it is indexed for inflation from 2012 it will be well over A$200 by the time it opens in 2025,,,26…27… and keep going up

        • Ren Stimpy

          Tasmania doesn’t need any pumped hydro to be the “Battery of the Nation”, it just needs a lot more wind power. Every MWh of wind power Tassie uses allows their dams to stay stocked rather than deplete that MWh from the dams, and also to be restocked faster when it rains. Wind power is cheaper than hydro, and the wind in Tassie is excellent. The more wind they put into the mix down there the more secure their hydro power will be, and the more available for exports to the mainland their hydro power will be. They have a fantastic export opportunity here I would have thought an opportunistic Liberal state government would be all over it.

          • Tom

            @Ren Stimpy – I 100% concur.

            The only problem (well, not the only problem) with the Liberal state government is that they are so obsessed with privatisation of everything that they can’t possibly envisage the Tasmanian Government’s state owned enterprises building the wind power ourselves.

            Some would call them corrupt, I wouldn’t quite use that word, but everyone knows that they are up for a sweetheart deal if they can meet the right people and rub the right backs. So you’ve got a few “mates” of the Liberal party (rich farmers, the Chinese Communist Party, an American multinational) working hard to get their wind projects built on the best possible terms (in other words, the worst possible terms for the Tasmanian taxpayer).

            We’re talking high PPA prices, long contracts, free supporting infrastructure, and locking out competition in return for donations, dinners, positive PR, and cushy jobs for members once they leave politics.

            PS – I would also add single axis tracking PV into Tassie’s mix. It would “smooth” the RE generation curve, especially over summer months. The eastern half of our north coast has a good solar resource – better than Melbourne, and almost as good as Adelaide and Sydney.

          • Ren Stimpy

            Hmmm, I see.

          • itdoesntaddup

            Why waste money on building more wind when you can import surplus from the mainland via Basslink?

          • Peter F

            Because Basslink is not that reliable and if they had had 200 MW of wind which they are now building they would never have run out of power in the first place

          • Ren Stimpy

            Because when the carbon price was in place Tassie made a lot of money exporting hydro power to the mainland, albeit at an unsustainable rate which depleted their dams. With a lot of wind farms they could do that again but in a sustainable manner which keeps their dams well stocked. Keep in mind that power prices on the mainland have risen significantly since the carbon price was removed.

          • itdoesntaddup

            Keep in mind that wind generation has been growing rapidly in VIC and SA, so there is really no need for TAS to duplicate that, especially given the high correlation of wind output with VIC.

          • Ren Stimpy

            Tasmania doesn’t receive any of that windpower from VIC or SA, so yes there really is a need for TAS to duplicate that – high correlation of wind output or not. Every MWh of wind power that Tasmania uses allows a MWh of TAS hydro to remain stocked for other purposes, such as TAS firming and exports to the mainland when the wind is low.

            The carbon price years demonstrated that TAS can supply their own power with hydro AND export large amounts of hydro power to the mainland, but that depleted their dams a lot faster than normal during those two years, and there was a drought. With a lot of wind power TAS could do the aforementioned but in a sustainable way, ie. without depleting their dams at an unsustainable rate.

  • Alastair Leith

    Giles isn’t this the part of Victoria the Energy & Environment Minister, Lily D’Ambrosio was saying could get all this new generation without building more transmission? Hinting at storage or some ingenious constrained access that didn’t actually hurt their bottom line.

  • iain

    Can anyone provide a link to the AEMO report. RE should really provide sources for this type of article.

  • Rick

    Curtailment or system constrained output is the inevitable result of run-whenever-you-like generation increasing market share. Once the connected intermittent generating capacity exceeds the minimum demand it is inevitable that they will become system constrained.

    South Australia provides a good example. AEMO are forecasting the minimum demand in SA to be ZERO by 2024 due to the high take up of rooftop solar. That means that the grid scale wind generators in SA will rely completely on the interstate interconnections to have load. If other states have the same level of rooftop take-up as SA, all the grid scale wind and solar will have zero demand.

    To meet the current NEM demand with 100% wind, the capacity factor would fall to 7%. To get 100% from solar the capacity factor falls to 4%. That is for the minimum cost option based on present price of battery storage, price of wind generators and price of solar generators considering the variation in wind and solar insolation across mainland Australia.

    • Peter F

      But that is a straw man argument, no-one ever envisages a completely wind or solar grid. We do have quite large hydro backup (25% of current peak demand) so the system will never be more than 50% solar and 40% wind.
      If our hydro systems were optimised for peak power without new dams we can upgrade them to 10-11 GW without pumped hydro so we can get to at least 15 GW of wind and probably 20 GW of solar before system wide curtailment becomes and issue.
      Then there is the option of controlled demand for hot water, pool pumps, clothes washing and drying, water distribution, cool stores, space heating and cooling and even modulation of loads at metal processing facilities
      As electrification of transport increases smart charging of vehicles will be an extremely versatile tool for managing demand further reducing demand for dedicated storage.
      In summary if the system is planned correctly we can get to 60-70% renewables before there is major requirement for storage. That does not mean that in some areas it will be cheaper to install some storage rather than expand the grid, sometimes at the customer end and sometimes at the generator end, but that is the case even now as distributors in Queensland and WA are finding

      • Rick

        Wind generators are already system constrained in South Australia at 39% market share and they have access to 600MW of load in Victoria.

        The fundamentals are that solar has no benefit of scale and wind has small benefit of scale. Grid scale generators are severely hobbled by the cost of power transmission being 110% higher than the cost of generation. So anyone who can afford the capital will have their own rooftop generation. That means those without their own generation will be paying a higher unit cost as the system costs continue to pile up with addition of more and higher capacity transmission lines to remote wind and solar generators.

        High capacity transmission lines over long distances are expensive. All the additional cost has to be paid for by the remaining power consumers.

        You will find it educational to read through the paper on this link and absorb the information:
        It covers the emerging situation in Germany. It is based on actual system demand and wind generation at short time intervals.

        Without this level of modelling detail your hydro power capacity could be constrained by its energy capacity. Also, including Tassie hydro in the capacity would be a brave move given the history of BassLink. I guess it could be replaced but that is just more cost for the dwindling number of consumers to cover.

        When you look at this modelling detail you see why the Finkel report is so naive, relying on the Jacob’s Group fairy tale that the solar insolation in Australia gives 100% for 6 hours every day in June and wind generation can be based on the average capacity factor because it does not vary much!

        If you are operating you own wind or solar system off grid then you would have a better understanding of what is required in terms of capacity to meet a particular demand. Whatever you could do off-grid, providing you have suitable roof space on mainland Australia, will be lower cost than the grid can achieve using wind and solar generation.

        • Peter F

          As I said a fully renewable system will probably not have more than 40% wind without storage. Viola SA is at that point. SA is a special case in two ways. a) It has extremely variable demand as you point out and b) it has no hydro.
          Curtailment at the moment is not usually caused by excess supply rather by the mistrust of wind and solar to provide grid stability, therefore AEMO specifies that at least 3 gas turbines and sometimes 4 must be running. As grid stability can be provided by synthetic inertia, pumped hydro, batteries, flywheels or synchronous condensors, it is likely that the requirement for gas generation will gradually be relaxed and curtailment reduced.
          My modelling works on the basis that with a sufficiently diverse portfolio of low wind turbines minimum wind will be about 10% CF.
          The German situation is quite different to ours,
          1.Their peak demand is on cold still winter nights, ours is on hot windy summer days,
          2. Their current hydro capacity is 7-8% of peak demand, ours is 25% and could be increased to 30% by energy efficiency on one hand and re-configuring existing hydro for more power on the other.
          3. Their current wind capacity factor is 22% our newer wind farms are over 45%.
          4. They have a phenomenon called Zauberflote where they can get two weeks of dull windless winter days corresponding with peak system demand. We do not
          5. The average wind speed and solar insolation are both higher in Australia to the extent that a new wind turbine in Australia will generate 12-20 GWh per year. The average in Germany is 3.6. Similarly a new solar panel in Australia in winter will generate about 8 times what a 2010 vintage panel will generate in Bavaria in winter.
          6. The NEM is spread over 1,200,000 square km with quite different wind regimes almost all their wind is concentrated within the northern 150,000 sq km of the country. The probability of extended low wind periods here is much smaller

          • Alastair Leith

            “4. They have a phenomenon called Zauberflote where they can get two weeks of dull windless winter days corresponding with”

            On the SWIS in WA we do get winter wind droughts (of up to two weeks) with poor isolation. But not peak demand maximum.

          • Peter F

            Poor insolation in WA is nowhere near as bad as poor insolation in Germany. In a good week (week 21) in Germany last year solar produced 16% of power. In a bad week (Week 2) 1%

          • Alastair Leith

            I expect overcast/rainy weather with no wind is poor in both. I’ll lookup the data when I get a chance.

          • itdoesntaddup

            The real question is how much is the supply gap when the solar output is low. It may not be peak demand, but it is still demand that has to be satisfied.

          • Alastair Leith

            Hence grid modelling investigations. The field of solutions are infinite, but given a set of cost and other constraints there becomes a much smaller set of solutions. We then test various prices and mixes to evaluate sensitivities to cost for the various components of any given solution. It’s not much different to the linear programming we all learned in yr 8 maths class just more inputs and value ranges rather than exact knowns.

          • Peter F

            Apart from more variables it is not always (in fact never) linear but your point is valid.

          • Alastair Leith

            @disqus_lUQ6zHXr2B:disqus “not much different” … not counting complexity 😉

          • Alastair Leith

            And yes I could let you what that gap is on the SWIS in a typical year. But I’d have to look it up in the output data. We have some example graphs on in the Powerpoint file under executive modelling tab showing summer PV surplus (even with reasonable stroage) and winter wind droughts with PHES storage running out in a couple of days. So it’s either gas, biogas for ~10-15% of annual demand just in those winter weeks, or an overbuild of wind, PV, CST with TS, PHES and lots of surplus generation that could potentially have industrial uses, but maybe not given the intermittency/variability of surplus generation. Power2gas may come to the rescue I suspect. Price of local hydrogen production from electrolysis is constantly falling like PV ATM and it hasn’t really got a toe hold yet.

          • Peter F

            Every grid as overbuild. The old NEM had 53 GW of capacity plus thousands of standby/backup generators for average demand of 24GW and peak demand of 35GW. Around the world this is typical. The good thing now is that standby capacity is much cheaper per MW than coal plants. There seem to be quite a number of sites for pumped hydro for less than $1.20 per W. about 1/3rd the cost of a coal plant

            Seasonal storage is yet to be resolved but some storage, some energy efficiency and perhaps 20% gas for 2-6 weeks a year is not the worst outcome. Once the grid is 80-90% decarbonised we probably get far more bang for our buck improving other sectors such as space heating, transport and agriculture. Some recent numbers I have seen show that restoring the mallee across southern Australia would sequester 400 m tonnes of carbon a year. 2/3rds of Australia’s total emissions. It may even reverse the southern rain defecit

          • Alastair Leith

            Agree with most of that.

            I think that while the last 10% to get to 100% RE is seen as very problematic in the models, several factors will converge at that time in the most advanced grids to get that last 10% without too much extra capital outlay. Those factors include the incentive to keep deploying PV and storage behind the meter as grid prices increase (or even remain stable which they aren’t forecast to do for the next 5-10 years) and PV and batteries continue to slide down their respective learnings curves. This means overbuild will happen one house, one commercial premises, one industrial roof at a time, and thousands a week. Add in P2P energy trading and commercial building occupants with no access to PV/rooftop will be putting PV on industrial roofs and importing it.
            As more RE is deployed, more surplus gets spilled, and more wind farms will start adding battery storage. At some point that’s only economic to a point, not enough to cover a week of winter wind drought. So surplus is still increasing once the battery market is saturated. Power2Gas now becomes very viable, and it’s learning curve apparently has been as steep as PV in the last few years for on-site H2 manufacturing from grid electricity. This offers the potential for storage as H2 or a derivative fuel in vast volumes, that could cover a week of poor wind and solar generation, especially if it commanded a price premium in a market only it or fossil fuels could serve the peak demand of.

            Mallee gums also make great on farm cash drops for biofuel, while providing on-farm services like reducing dry-land salinity, reducing winter water logging and habitat corridors. I’m surprised to hear they can effect local rainfall. The Margaret River area is said to be significantly drier and warm, than 50 years ago in large part due to the massive deforestation of that area, especially the wheat belt side of it.

          • Alastair Leith

            Especially agriculture. 54% of national GHG emissions when the accounting is done right and a (more appropriate for the SLCPs in that industry) 20 year time-frame is used. See

          • Alastair Leith

            Germany also has loads of politics around N/S transmission lines over their beautiful Alps. Might never get another line though, even the timetable to discuss is on the decade scale I’m told by a German fellow who used to be at AEMO.

          • Peter F

            That is true but indirect routes are opening up. Like everywhere else, transmission planning is being up-ended and it appears that distributed resources and energy efficiency are deferring the need for the large infrastructure that was planned so better solutions can be thought through

          • Rick

            That gets to the basis of this post. It does not make sense to enhance transmission to enable remote wind generators better access the grid. System constrained output is the natural result of increasing installed capacity. Transmission costs to households is already more than the cost of generation so adding more costs just encourages more households to take the economic option.

            Anyone thinking it through, as I have done for my own on-grid and off-grid power supplies, realises that grid connected intermittents cannot compete with rooftop power on mainland Australia. Tassie is different because it can use wind and solar to conserve perched water.

          • Rick

            And you think Australia is different!

          • Rick

            South Australia has an effective battery of 600MW/UnlimitedMWh with its link to Victoria. Average price for household power in SA is 60cents/kWh, already higher than a rooftop solar/battery system. Without that link SA wind would be constrained much more frequently and prices would be even higher.

            Yesterday the Australian wind fleet was below 10% output from 9am to 9pm:
            This is actual data not hopium.

            With reference to your points.
            1. Agreed.
            2. It is not only the power available from hydro but also the energy. At present hydro supplies just 7.1% of Australia’s electric energy. The only way to verify the value of hydro capacity to run-whenever-you-like generation is to do an analysis of proposed system based on actual recorded data as Sinn has done in his paper for Germany.
            3. Please provide the annual data at billing intervals on the wind generators achieving 45% CF. It is not the average power output that counts but the variability as Sinn demonstrates.
            4. Australia also has long periods of low sun and windless days. It is not difficult to find long periods of low output:
            5. There is an unproven hypothesis that geographic spread across Australia will even out supply. The wind generation data indicates it is a myth. A single high pressure system can sit across the whole east coast. Also there is an incredibly high cost in transmitting large amount of energy long distance.

          • Peter F

            I think the argument is about semantics.
            1. Yesterday when wind in the NEM was below 10% for most of the day, it included two wind farms north of Sydney and when I checked they were running at over 30%, so when Saphire, Kennedy etc come on line CF will rise. Only about 30% of the active generators were low wind when Bulganna, Silverton, Coopers Gap Point Gelibrand etc come on line CF will rise further, even in the southern regions, so 10% minimum CF across the NEM is easily achievable.
            2. I agree hydro is energy constrained but the more wind and solar is on the grid the more hydro can be confined to low wind/solar times and therefore the more it can be relied upon for peak power.
            3. Australia does have long periods of low wind and solar output, but average demand on the NEM in Autumn is around 19 GW so if 35 GW of wind and 35 GW of solar and 8 GW of hydro operate at 20%, 10% and 70% for a week respectively that is still about 16 GW average. Biomass, waste to energy etc can easily provide another 0.5 GW. If in such a week gas has to run at an average 15-25% capacity that is not exactly the end of the world.
            As for annual data that is difficult to find but for quarterly periods look at
            4. Again semantics if you look at Anero there is plenty of time when individual wind farms are not operating at all but across a state it is never zero but often around 5% or less, across the NEM rarely below 10% so geographic smoothing is real. It is not perfect no-one claimed it was.
            6. At current capital costs a new coal plant has about $40-55/MWh in capital costs and $15-20 in O&M before variable operating costs so operating below capacity costs $65/MWh and yet col plants in Qld and NSW mostly operate below 65% capacity factor. i.e. they have 35% curtailment. This is accepted, why is it such a bad thing for wind and solar.
            Operating a wind or solar farm below capacity also costs about $55-70/MWh but they cost less to build. It is just a fact of life very few industries can sell all they can produce 24/7

          • Rick

            1, I am pointing to actual recorded data. The average capacity factor is meaningless you need to look at the variation over time as I have done.

            2. The more wind and solar that is on the grid the more often it is system constrained. That means lower capacity factor because it is now system constrained rather than weather constrained. Again you need to consider the situation for each billing period over at least three years to understand the interplay between intermittent generation and hydro available energy. Tim Flannery pointed out some years ago that Australia can no longer rely on rain, making hydro an unreliable source. It is just silly to think that the unconstrained capacity factor can be relied on as the basis for determining output from an intermittent generator. Adelaide’s rooftop solar is already being constrained by over voltage conditions, thereby incentivising battery installations.

            3. Looking at averages is a waste of time. The power system needs to be in balance for supply and demand every second of every day. Industry does not run on averages. It requires power to be available when called on.

            4. I do not consider less than 10% “real” smoothing from geographic spread. And that figure is speculative without the billing interval data to back it up.

            6. Wind and solar on the grid are progressively destroying base load. The consequence is the economic destruction of base load generation being replaced with high cost, fast response gas. So it is inevitable that the capacity factor for coal will fall. On the demand side, the higher wholesale price has already forced closure of heavy industry regarded as base load.

            Proper system modelling at billing interval detail will show that by the time mainland Australia reaches 40% market share of grid scale intermittents it will be cheaper for anyone with a roof to make their own electricity. That means there will be those who have lower cost power and those left on the grid paying ever higher unit price as the market diminishes.

          • RobertO

            Hi Rick You say Hydro is unable to rely on rainfal however we need more PHES and PHES uses about 5% of the water used by coal, and we only have about 22,000 that may be worth considering. They do not have to all be 250 MW units, we could use say 1000 units at 15 MW and different sizes of storage (Its a mathmatical equation of what we want built over the next 5 -10 years (or what ever time frame Australians want or need). It’s our FED Gov that should be leading Australia not standing in the road (or “Failure to plan is planning to fail”).

          • Rick

            You are right that it is a mathematical equation and I see many silly calculations based on averages rather than the true intermittency of wind and solar.

            No matter how it is calculated, the fundamentals in Australia means grid scale wind and solar cannot compete with rooftop. Any pumped hydro just adds more cost to the grid. There is no way to spend more money on grid storage and transmission and make power cheaper. South Australia had to fund all its latest grid enhancements from general revenue rather than hitting the diminishing number of consumers with higher prices. They are the leading State with disconnections:

          • Your second paragraph is a real doozy – six phrases, six misconceptions. grid scale wind and solar don’t need to compete with rooftop, different market. you can spend money on storage and make it cheaper because it introduces more competition, and may deflect network upgrades. south australia did not fund all its latest grid enhancements from general revenue, and the tesla investment has already probably paid for itself. that link does not say the state is leading with disconnections.

          • RobertO

            Hi Giles Thank you! Hopefully you can also answer this one. There are two SA to Vic interconnects, Murry Link at 220 MW (and the question is can it flow both ways) Heywood Interconnect is now 650 MW (is it still restricted to 600 MW)

          • Rick

            Grid scale wind and solar are already competing with rooftop and rooftop has priority scheduling because the only way it can be constrained is by system at voltage limit. The linked chart is from 2017 AEMO report on South Australia:
            It forecasts minimum demand to reach zero by 2024. That means all the grid scale run-whenever-you-like generators will not be able to export power at those times and the constrained period will increase in duration as rooftop increases penetration. The whole of Australia will be in the same boat as rising grid prices encourage accelerating rooftop penetration.

            Investment for grid scale intermittents is already drying up because the LGC price will tank as the RET is exceeded and no prospect of a higher target to tap into subsidies. No wind generator in Australia can pay for financing if there is no subsidy.

            There is no point in making grid upgrades to remote run-whenever-you-like generators because they will be uneconomic as rooftop and carpark solar, near loads, takes over. They can just cover O&M costs without subsidies but no new ones. Once the banks realise that their Capacity Factors can be severely reduced by system constraints the financing will dry up.

            This link gives details on Weatherill’s $550M State debt funded “energy plan”:
            The Tesla/Samsung battery was paid for under the same plan:
            Who knows what it cost and how long it will take to recover that cost, if ever. Given the massive wholesale price swings in SA there is some prospect of it recovering its cost on price arbitrage.

            I stand corrected on the the issue of which State is leading disconnections as I equated those on hardship programs to disconnections. The following quote is taken from the referenced link.
            “South Australia had the largest proportion of electricity customers in debt, with five out of every 100 residential customers holding an average debt level of $1596.

            South Australia had the highest number of electricity users on hardship programs”

            It reinforces my point that there are those who own a roof and those left to pay ever increasing price for grid power – OR NOT when they can no longer afford to pay. That goes for heavy industry as well. They were enticed to invest in Australia last century with the prospect of buying base load power at marginal cost. That was a fraction of the current wholesale price.

          • itdoesntaddup

            I see you are relying on Blakers’ hand waving assumptions. Perhaps you should look at the ROAM study:


            It finds a total of 516GWh of which nearly half is in Tasmania, with the individual sites being identified for further perusal. In fact, it
            turns out that a good chunk of these are already likely to be too
            expensive – see Figure 8.3, which shows a cumulative capacity against estimated cost diagram for those where the capacity is reasonably certain. If we set a cost criterion of 200 AUD/kWh of storage capacity then the maximum drops to little over 200GWh.

        • RobertO

          Hi Rick, What happens if they devalue the grid to keep people connected? What happens if the Fed Gov decides to support industries that use lots of electricity in areas that have issues with curtailment. Our Fed Gov has it head buried in the sand and they can not see past the coal dust and polution of FF in Australia. Even fuel stardards for light vechicles is a carbon tax in Australia (let alone our fuel supplies at record lows).

          • Rick

            All the remaining Aluminium smelters are already on government life support of some sort to keep them operating. A number have already closed down. It does not make economic sense to keep energy intensive businesses in Australia with high power prices.

          • Peter F

            That is true at current power costs. If we close them down, base load on the NEM drops by 3 GW and annual demand by 25 TWh That immediately increases the renewable share by about 3%.

            Alternatively Aluminium refining in WA makes perfect sense if they can match worlds best solar and wind prices at less than US 2 c/kWh close to the bauxite resource and reducing freight costs by a factor of three

          • itdoesntaddup

            Bulk freight costs for alumina by sea are trivial in the cost of smelting aluminium – just a few dollars a tonne in an aluminium price of a couple of thousand. That is why you find smelters where the power is cheap – e.g. geothermal power in Iceland (usually supplied with alumina from Brazil), rather than near bauxite mines.

          • Peter F

            True but even Iceland cannot match 2c/kWh

          • itdoesntaddup

            Not true: Rio Tinto were paying about 15$/MWh in Iceland. I’ve also seen lower than $20/MWh coal fired power costs for aluminium elsewhere. And your 2c/kWh remains as yet imaginary.

          • Peter F

            Ok maybe I exaggerated a bit re Iceland but you can’t refine all the Aluminium in the world in Iceland. $20/MWh for coal works on old coal plants but not on a new plant with current world coal prices. Many of these low prices were subsidised by other users so that the coal plants could avoid ramping down at night. Portland is a classic example, as a case of combined subsidies for propping up the minimum load on coal plants and “regional development”. This plant has been a drain on Victoria almost every year since it was proposed

            but the 2c is real

          • itdoesntaddup

            2 ¢ is only real when you have plant in production that hasn’t gone bankrupt. We do not have that yet anywhere, and certainly not in WA (which was your original claim).

            I think you forgot about China, the world’s largest aluminium producer? Aluminium is clearly not your speciality. Sub $20/MWh smelters can also be found in the Middle East, based on gas fired power.

          • Peter F

            Again $20/MWh is subsidised. In the middle east the best you can expect from a CC gas plant is around 52% efficiency. At world parity price of US$3/GJ, fuel cost alone is US$21 per MWh

          • itdoesntaddup

            Gas in the Middle East does not have a parity value with your “world price”. It is a stranded supply, that previously was flared for zero value.

          • Hettie

            But $20/MWh IS 2c per kWh.
            $20.00 is 2,000 cents.
            1 megawatt hour is 1,000 kWh.
            2 c per kWh. QED.

          • itdoesntaddup

            Quite right, Hettie.

          • RobertO

            Hi Rick, Like i said, our Fed Gov has their heads buried in the sand. They say stupid things like “Lets spend $500 million on the Great barrier Reef to make sure it’s corals can survive (by breeding heat tolarent corals) but $0.00 on the causes of heat. Let support Farmers to enable then to survive the heat stress at $100 million, but $0.00 on what causes the heat, all cause there is no such thing as “Climate Change” or “Climate Change caused by Mankind!”
            Their best claim is “We export jobs for Australians?” “We ship jobs overseas for our mates in big business?”

            Where is their plan on how to keep jobs in Australia?

            Surely somebody in WA could put together a plan to build a wind farm and a bit of solar (Pilbera ring a bell? Nah far too big and job numbers)
            Maybe somebody in NSW could plan a new Wind Farm in Northern NSW (or maybe near Broken Hill or somewhere in between ) and a bit of Solar. I am very sure there room to move on this idea. If the Fed Gov was serious about jobs then RE has lots to offer.

          • Hettie

            Lots of wind and solar going up in Northern NSW. Around Glen Innes, Inversely district. Huge turbine blades going up the NE Hway. Sapphire wind farm.

          • itdoesntaddup

            Please calculate exactly how much temperatures would be lowered in the sea at the GBR if humanity was extinguished from Australia altogether.

          • Hettie

            You want to know? You work it out.
            Then when you get it wrong you’ll have only yourself to blame.

          • itdoesntaddup

            I’ll tell you Hettie – a very close approximation to at least 5 decimal places is zero.

          • Hettie

            There are no prizes for being a smart arse, IDAU, and although you sometimes talk sense, you are just a smart arse far too often.

          • RobertO

            Hi itdoesntaddup, Why straw man argument. Australia should do no thing about Climate Change because we do not contribute to Climate Change in any way shape or form.
            I run a 20 liter hot water ern so that the staff can make coffee when they arrive at work to have breakfast. After every third person removes a cup of hot water if I am on duty I put a litre of cold water in the ern to fill up the ern. It takes about 2 and 40 seconds to return the ern to boil yet one of the main complaints when I am not there is “we ran out of hot water”. There are over 100 staff that drift in for breakfast. I keep telling people to put small amounts in the ern, 1 litre at a time so as not to hold up people.

            It up to our Fed Gov to lead Australia

          • itdoesntaddup

            Strawman comment. The discussion is about the effect on the temperatures at the GBR. Now, you may not know that the oceans have about 1000 times the heat capacity of the atmosphere, which means that the same amount of heat that raises the atmosphere by 1 degree, raises the ocean by 0.001 degrees. Since Australia accounts for just over 1% of global CO2 emissions, if it were to disappear altogether emissions would fall by 1%. 1% of 0.001 is 0.00001.

          • Hettie

            Closer to 2% domestically, and vastly more than that from exported coal.

          • Hettie

            That’s why Gupta is fuelling Whyalla etc with renewables.
            It’s cheaper that way. So the aluminum smelters should be rushing to solar and wind to stay viable.

          • Rick

            Whyalla is another smelter on government life support.

        • Peter F

          Rick further to my comments below I skimmed that German paper it is a fantastic piece of work however it does ignore technology change and existing hydro. He also left France out of his regional grid, not only does France have nuclear but it also has between two and three times the solar, wind and hydro potential of Germany. This does not mean he is completely wrong but even in three years the divergence from his predictions is obvious. For example Germany has installed an entirely trivial amount of new storage and yet wind and solar are already at 28% of generation and will reach 30-32% by 2020
          Some examples of omissions
          1.already old windfarms are being re-powered with turbines which have twice the capacity factor of earlier ones. In effect they generate for at least double the hours, That directly impacts the storage requirements
          2. I would assume that in Germany they will keep much of their gas power infrastructure but powered by an increasing share of biogas and hydrogen/ammonia which can easily be stored for months. They will still use natural gas for a long time
          3. Much more of German wind power will come from the North Sea and later the Baltic which have higher capacity factors and again reduce the need for seasonal storage
          4. While ground mount solar is notionally illegal. It is not illegal to re-purpose waste land such as landfill sites, old lignite mines, railway rights of way etc. That means there is in excess of 2,000 sq km which can be used for solar, that can generate about 100-200 TWh per year or 30% of Germany’s electrical demand.
          5. More storage means that the remaining gas plants are more likely to run in combined cycle mode, reducing gas as a share of primary energy.
          6. The regional concentration of wind in the north and solar in the south is breaking down. Tall low wind turbines are becoming economical in the south and solar is becoming economical as far north as the southern regions of Norway. The increasing diversity of supply will also reduce storage requirements
          7. While batteries will probably remain a small part of the total system and initially only cover day to day smoothing, it is likely that in 10-15 years time Germany will have 300-400 GWh of battery storage

          • Rick

            Wind and solar reached 22% in Germany in 2017 not 28%.
            The investment in run-whenever-you-like generation is in rapid decline as the subsidies dry up:

            Without massive storage, 30% market share is the end game in Germany. Storage comes at a high price so Germany will not get beyond 30%.

            With subsidies ending there is no incentive to repower existing sites:

          • Peter F

            It is 28% YTD and to the end of April it installed 1.2 GW of wind and 0.6 GW of solar following on from 6.6 GW of wind and 2.3 GW of solar for 2017.
            Some new windfarms have been contracted at zero subsidy.

            Again the environment here is much better for wind and solar than Germany because of higher insolation and wind speeds, lower land costs, higher base prices for power and much more hydro backup 25% of peak load vs 6%

          • Rick

            It is exactly the environment here in mainland Australia that means households are installing their own solar at accelerating pace. Even in June there is enough sun to make solar/battery economically viable against a grid price of 60c/kWh.

            The grid scale wind and solar cannot compete with rooftop as it is severely hobbled by the transmission cost. It makes no sense to have grid scale intermittent generation as it is accelerating the demise of the grid. It is accelerating the demise of heavy industry and making power more expensive for those not in a position to own a roof.

            The down trend in consumption in Germany indicates the demise of its heavy industry. It is increasing imports of energy intensive goods like steel.

          • Peter F

            Rick I agree rooftop solar has many advantages, but rooftop solar + local storage is still more expensive than grid supplied wind/hydro.
            The only heavy industry that has closed because of high power prices was Point Henry, even at realistic brown coal prices it was old and uncompetitive anyway just like most of the refineries in the US and France which relied on cheap nuclear+hydro have closed.
            Germany’s industrial sector is as big as it ever was, a a far larger share of the economy than the US for example. Even though the steel intensity of Western economies is falling. German steel production has basically stayed the same since 1978. This is a far better performance than almost any other western country.
            I agree for low areal intensity users like private homes rooftop solar is a good solution but you are just throwing all this other stuff into your argument about heavy industry that is just not true.
            Even for private houses it is cheaper for almost everyone to be connected to the grid. The total value of the transmission and distribution grid is about $90 b which works out at about $5-7,000 per household. If we added 70GW of wind and solar that takes the total investment to $16-18,000 per household. Most households can be net zero with solar for about $6,000. so we need a total of $24,000 per household to have most houses and small businesses net zero.
            To be off grid it costs $40,000-70,000 per small customer and most of the summer power is wasted. Do you want every household to spend $50,000 or $25,000
            A modern grid designed today would be a mesh network not a tree like we have now so it could be a lot cheaper, wind and solar farms would be smaller and close to demand but there would still be a lot of transmission to CBD’s and industrial facilities.hospitals etc

          • Rick

            In addition to Point Henry; Kurri Kurri closed in 2014, Bell Bay has reduced throughput in the last two years due to power availability (likely to be same in 2018 with BassLink out), BSL has reduced throughput due to high power prices and Portland is on Government support. Portland would have shut down after the 2016 power loss without government bailout. Tomago and AGL are planning a dedicated gas turbine for the smelter that will operate outside the NEM. It will obviously have scheduling priority over wind and solar so it will not be system constrained – just run at steady, efficient output without the ups and down imposed by the usual priority scheduling of run-whenever-you-like generators. This is another way that heavy industry will insulate itself from the high cost of intermittency in the NEM.

            With regard to distribution you are now supporting my point that we do not want to be building expensive interconnections to remote solar and wind generators. We already pay too much for transmission and distribution assets, including metering.

            Right now just solar is the best return for households but that is unsustainable. As soon as the rooftops are being regularly voltage constrained households will be looking at batteries to get more value from their solar investment. The solar is already sunk cost so no longer in the equation when evaluating battery options.

            Your terminology with regard to “summer power is wasted” is not accurate. To be accurate, the system requires overbuild in generation because the storage capacity is not infinite. You need to ask yourself – Why is this different to a wind and solar supplied grid? The grid also requires MASSIVE overbuild in generation because the storage is limited. That was the point of the Sinn paper. If storage was not limited then we would just shift the whole system to hydro.

            The most economic household system would have a small generator, probably gas fuelled as many homes have gas for heating. That can dramatically reduce the size and cost of storage. But as we are being constantly advised the battery costs are in free-fall!

            You may find this an interesting read:
            It demonstrates in a step-wise analysis why you cannot rely on averages when dealing with intermittent generation supplying an on-demand system.

          • Peter F

            Re the Victorian paper and working backwards up your post. No-one is proposing a 100% solar powered grid. All this simplified one or two factor analysis is just simply misleading. I don’t know why people waste their time on it.
            As Alastair says it is a 5+ dimensional optimisation problem. Cost is a dominant variable. Rooftop solar for half your power is cheap. 99.99% reliable solar power with storage is not.
            Brown coal power for 50-60% of demand is cheap. A 99.99% brown coal powered system would be very expensive.
            In the Australian context wind is cheaper than hydro, either per MW or per MWh so a wind/hydro system is cheaper than a hydro only system
            Thus at the current state of the art, the optimum target for Victoria in about 10 years time is probably- Energy efficiency -15-25% less energy use offset by electrification of space heating and early phases of transport electrification. Then generation would be 30-40% wind 10-20% utility solar 20-40% rooftop solar 7-10% hydro and 5-15% gas and residual coal. There might be 1-2 GW of pumped hydro and 2-3 GW of batteries however demand response will also contribute, solar thermal, wave energy, power to heat, hydrogen fuel cells or hydrogen gas turbine CHP plants will all come into the mix, nobody knows but at the moment you can let the market rip. There will be rooftop solar, utility solar and wind. There probably won’t be any market motivated significant transmission upgrades. Every year people will look forward and find that the market balance is changing and alter their investment priorities.
            Kurri Kurri closed before energy prices took off as did Point Henry. There is a worldwide oversupply of Aluminium driven by Russia and China and these plants were relatively older and inefficient. They closed, that what happens

          • Rick

            The market is NOT deciding. It is highly distorted by government regulation ensuring massive transfer payments to grid scale wind and solar. The poor people who do not own a roof are paying dearly for the income to the owners and financiers of the grid scale intermittent generators.

            Intermittency is a huge cost to all the coal generators and electricity consumers.

            Without prospect of increased RET, funding for grid scale intermittents will dry up.

          • Hettie

            Are you quite sure that your information is up to date?

          • Oh, do explain the “massive transfer payments” for new wind farms such as Stockyard Hill, Cooper;s Gap, Silverton, Hornsdale, Sapphire wind farms. There’s about 2GW of wind capacity. Tell me about the subsidies.

          • I’ll answer for you. There are none. In the case of Hornsdale, Sapphire, and Crookwell and Ararat and Coonooer Bridge, there are none, because the LGCs are cancelled. In all the others, the LGCs have no value because the combined price of wholesale electricity and LGCs is vastly below the current market price of electricity.

          • Rick

            The LGCs are not cancelled they will remain on the register until surrendered – some may never be cancelled if there is an excess. It is a way to manage price as the RET is reached or exceeded. Without an increase in RET the LGC market would be flooded and they would be worthless anyhow.

            The wholesale price of electricity has risen threefold due to growth in intermittency to the point where the unconstrained LCOE from intermittents is now lower than the wholesale price:

            This article is about “curtailment” or system constrained output. It is inevitable that wind generators will be increasingly subject to system constraints. The LCOEs are being worked out on unconstrained capacity factors and financing based on projected earnings with no system constraint. Wind generator proponents are surprised when their output becomes system constrained. Wind is only supplying 6.6% of Australias energy and system constraints are already impacting on output. Once proponents and financiers realise their generation will be regularly system constrained the projects will dry up.

            Already the high price of grid electricity is encouraging rapid take-up of rooftop solar further increasing intermittency. The sad fact is that the poor, without ownership of a roof, are further disadvantaged as they have to bear the cost.

          • Rick

            Current price of LGCs is $82/MWh. That is the transfer payment from every power consumer for all energy produced by a wind generator.

            That means those who do not own a roof for their own generation are obligated to make a payment to the entity holding the rights to the LGCs via their electricity retailer.

          • That is completely and utterly false. $82/MWh is the spot price, which represents a tiny fraction of the market, and certainly does not represent the transfer price for “all energy produced by a wind generator.” It is paid by those retailers who didn’t bother to invest in renewables. Their customers should ask them why.

          • Rick

            The linked article gives some detail on the massive transfer payments (currently $2.5bn annually and rising) from poor consumers to the proponents of run-whenever-you-like generators.

            And this direct payment is a fraction of the cost of intermittency.

          • That link doesn’t work, and The Australian is full of rubbish about the RET, and demonstrably so. They simply do what you do, multiple a market price by all the generation, which as I have explained bares no relation to the cost.
            Please do explain the cost of intermittency. you say $2.5 billion a year is “a fraction of the cost”. So let’s say it is $10 billion. explain where that is coming from, and how it differs from the existing back-up installed to cater for coal and gas outages and trips and variations in demand.

          • Rick

            Sorry for the paywall – This is a link to the original paper:

            Intermittency destroys base load. The best example is in SA as it has the highest proportion of intermittent generation. There is 1600MW of wind generating capacity in SA to meet a typical daytime demand of 1000MW. With wind swinging from zero to potentially 1600MW in hours, the state can go from a load of 1000MW to exporting 600MW (prior to 1200+MW stability constraint). Low cost coal plants cannot follow load swings at that rate so they have lost base load equivalent to the swing. The 600MW link to Victoria made Hazelwood uneconomic.

            With 4000MW of wind in the NEM the base load has been reduced by a value approaching 4000MW. That means that portion of the load is now being supplied from high cost, fast response gas rather than low cost coal.

            The consequence is skyrocketing wholesale price:

            Then there are all the additional costs to achieve system stability like demanding SA gas plants stay connected. Adding further demand on the gas supply. Buying diesel generators to cover for the loss of low cost base load coal generators because they lost their steady load and became uneconomic. Buying a big battery to provide fast response to help maintain system stability because the wind can change quickly.

            The next stage in this unfolding disaster will be the realisation that project financing based on LCOE of wind generators and average wholesale price is flawed as the system increasingly constrains output and when wind output is high the wholesale price will be at its lowest and often negative. The bleating about “curtailment” will be incessant.

            Only 6.6% of Australia’s electric energy is from wind. And it has already contributed mightily to a threefold increase in wholesale price as well as other costs sourced through retail transfer payments, added cost of poles and wires and direct government funds.

          • As I suspected. That BAEconomics report is garbage, and the main reason such drivel is repeated in The Australian. As is most of the rest of our post. To suggest 4,000MW of coal has been replaced by gas is absurd. The Tesla battery has already paid for itself by smashing the gas cartel on FCAS. Agree diesel gen-sets were not needed, and they haven’t even been switched on, and probably won’t. Haven’t got time right now for the rest.

          • Peter F

            Ok, within the constraints of current regulations the market is deciding between technologies, existing coal, gas and hydro or which new types of generation and storage to add.

            Average demand on the NEM is 21.7 GW. Annual hydro supply is 13-15 TWh an average of 1.5 GW leaving 20.1 from coal and gas. There is 23 GW of steam plants and 3 GW of CC gas plants i.e. if the steam plants average 85% CF and CC Gas 70% they can generate 21.5 GW so there is no annual capacity constraint.
            Considering peak capacity there is 23 GW of steam, 8 GW of hydro and 6 GW of OC gas/diesel 37 GW capacity against peak demand of 31.6 GW.
            The reason new generation gets a guernsey is because it is cheaper and cleaner.
            As for huge costs. Coal power production in the NEM has declined from about 170 TWh in 2010 to about 145 TWh last year or about 15%. Coal capacity has declined by 5,000 MW or about 20% so coal plant capacity factors have actually risen. Similarly gas capacity factor has also risen so how can you say that renewables have imposed huge costs when the plants are actually getting higher returns on their fixed assets.

            If the poor people who don’t have solar are being disadvantaged by subsidies to large scale renewables how does this happen. If all large scale wind and solar which generate less than 7% of total power earned subsidies equal to double average cost of coal and gas that would add up to a total price increase of 14%.
            However they aren’t subsidised by anywhere near that much, in fact when winds are strong power prices fall and the REC market is a balaning market so a large fraction of large scale renewables don’t get any subsidies so as the ACCC has said all green schemes add about 6% to the bill. Of that about half is attached to rooftop solar, so subsidies to large renewables are 3-4% of the bill at most.

            Most of the plants being built now will not receive significant revenue from RECs because
            a) the scheme ends in 2030 after 1/3rd of their life
            b) even a year ago the futures price for RECs from 2020 was $50, now about $20 so the NPV of REcs over the life of the project was $9-11
            c) There is more than enough capacity being built so the last people in the queue will get virtually nothing anyway

            In my opinion new gas plants will add short term security but will force out older less efficient plants will close because utilisation is so low and thus new gas and coal plants will actually provide very little extra security

          • Hettie

            Oh Gawd. Not you two again!
            Why don’t you just exchange email addresses and bicker in private?

          • Peter F

            Sorry Hettie. You are right -:. Enjoy the rest of your day

          • Hettie

            LOL No hard feelings.

          • Rick

            Intermittents have priority access to the market. There are destroying base load and making low cost, slow response coal uneconomic. They have caused a huge increase in wholesale price through fast response, high cost gas being the only suitable reliable generation.

            All other generators are required to submit capacity for scheduling. The so-called semi-scheduled generation is a nonsense term. IMore appropriate terms are “random generation” or run-whenever-you-like generation.

            The wholesale price rises are primarily due to gas being substituted for coal, which is a consequence of intermittency, as well as having to maintain this plant on standby at the vagaries of the wind.

            This article is about “curtailment”. So far intermittents have not achieved the level of market penetration that is affecting the economics through being regularly system constrained (curtailment). The 1200+MW constraint in SA is occasionally reached.

            Whenever so-called curtailment occurs we get bleating from wind generators that they are being treated unfairly. It is NOT UNFAIR but rather the inevitable consequence of the installed capacity, market demand, random generation and negligible storage. It will be a constant feature as market share of intermittents increases. They need to understand it and get used to it.

            Once proponents realise that unfetted capacity factors are not a valid method of assessing economics of intermittents we will see the investment dry up.

            All unsurrendered RECs are held on register by their respective owners and I expect there is some hope of them being of value in the future. There is considerable political push to increase the RET to keep the value in this massive transfer payment from the poor. The investment in wind died when Abbott threatened the RET and we see investment in Germany falling as their subsidy is winding down.

            Look at how ERM used the REC system to maximise benefit. From their 2017 annual report:
            “For the 2016 compliance year, the Company chose to achieve compliance with the Renewable Energy Target scheme by paying the Clean Energy Regulator $123m, in lieu of surrendering 1.9 million Large-scale Generation Certificates (LGCs). This enabled the sale of existing LGC inventory into the market while prices were high, with the Company reaching agreements to procure lower cost LGCs in the future.”
            And the poor people pay!

            This sort of financial fiddle leads me to conclude that LGC holders place value in whatever is held on the REC register.

          • Hettie

            Rick, to me this story of yours appears to be full of holes, but I will leave it to the better informed to point out what those holes are, except to say that, as I understand it, the gas peakers are often called on when a coaler falls over, as they do increasingly often.
            They fail without warning, dropping hundreds of Megawatts from the grid. Unlike renewables, whose intermittencies are highly predictable, and in each case the loss of power is mostly far less than occurs with coalers.
            (I don’t like that sentence, but can’t seem to make it clearer.)

          • RobertO

            Hi Hettie, another point to note is that ERM had decided not to acquire LGC as all retailers were required to do. They sold their product Electricity without looking at how they could get LGC by investing in RE. As with most companies they will always put the best spin on any report or publication and Rick is a cherry picking to implying that ERM had 1.9 million LGC and that they later on sold. Part of this mis-information is that there were Tax reasons to pay rather than buy LGC’s. In simple terms ERM made the decision that was in their best financial interests, and then cover their decision they applied spin (it’s true but not 100% true as Rick contends). Yes they had some LGC but they got more value from them by paying the fine and later acquiring more LGC to surender later.
            LGC also finish in June 30, 2020. After that date retailers are no longer required to buy or surender LGC’s. RE generators can still make LGC’s after that date but whom will pay them for their work.

          • RobertO

            Hi Hettie, Another point about ERM is that the main owner is a very big supporter of RE, see this Along comes COALition and babbott and let get rid of RE (nearly worked), then as a reward for this NSW COALition devalues Vales Point Coal Power Station (VP) from $770 million to $1 million. (I cannot prove it was a reward) but same owner got 14 mates together to buy VP and last year when the financials were released for VP he had bought the 14 Mates out at $2.85 million each share. VP has now been revalued at $730 million.

  • RobertO

    Hi All, So where does the Hydrogen cycle go? Do we need to change our ideas of how we pay these intermittent generators (RE) as David Leitch has suggested. Do we take curtailment as a means to suggest the H2 cycle somehow (or what does the future hold). I still think that the prize in not cheap electricity, but the prize is survival of the human race on this planet and the by-product is cheap electricity. Lots of arguments below are over the size of the generator fleet in Ausrtralia and how and what we need (and lots forget that we need to include Transport as well). One good point is that there is no right answer to all, but there is one wrong answer for all, and that is we do not need any plan on where, when, why and how for all the things that may be needed for a RE based grid, transport solution and our agriculture changes needed

    • Hettie

      Nailed it, Robert.
      If wind and solar energy would otherwise be wasted, and the terrain is not suitable for pumped hydro but there is a water source handy, H2 production seems like a plan.
      The gas is transportable as ammonia, or a plant to make fuel cells for cars might make sense, probably lots of other possibilities, though I do like the idea of aluminium or steel smelting.
      If all FF subsidies are withdrawn, there’s several billion $ for venture capItal to establish new industries.
      It’s good to dream.

  • John Mc Clelland

    Why is there talk of curtailment,surely before every Tom Dick and Harry decides to put up some of power producing plant they have to get permission, there must be a government dept that is in control of this situation, why has all this not been worked out before hand, they must have known what the power grid in these areas could not handle, yet they let all these power companies just go ahead and produce electricity that might not be used, this is typical of a government run dept and the attitude of she’ll be right mate we’ll cross that bridge when we come to it, and to top it of these government yobo’s are getting paid mega bucks to not do there job.