#1 cleantech news, reviews, & analysis site in the world. Subscribe today. The future is now.

Clean Power

Published on July 21st, 2014 | by John Farrell


Utilities Cry “Fowl” Over Duck Chart And Distributed Solar Power

July 21st, 2014 by  

placeholderThe rapid changes to the electricity system being wrought by distributed solar have utilities crying out, and they’ve poured much of their distributed solar angst into a chart being shared throughout the energy nerdocracy – the duck.

The Duck Chart, Showing Net Supply/Demand on the California Grid in 2012-13, Forecast through 2020

the infamous duck chart electricity demand with solar California

Until 2012, daily energy demand looked like a two-humped “camel,” with peaks mid-morning and early evening. Utility operated power plants supplied most of the needed energy. But the substitution of local solar power to meet local energy needs affects the demand for mid-day energy from the grid. The daily demand curve transforms, from a camel (orange line) to a (forecast) “duck” (bottom green line).

The duck is the perfect vehicle for utility complaints because it casts the growth of distributed solar as a major technical problem (an area where most policy makers defer to utilities) rather than an economic one, where utility complaints can be contrasted with their customer’s desires for more local control over their energy use and costs.

The utility companies crying “fowl” highlight a particular part of the duck chart: the dramatic ramp up in power generation on the light-green 2020 curve that happens in the late afternoon, as energy produced from solar wanes but energy demand rises. In the traditional grid operating model, accommodating this ramp-up in energy use requires a lot of standby power from expensive to operate, rapid-response power plants.

Evidence suggests utilities are crying “wolf,” with several experts poking large holes in the utility argument. The Clean Coalition and Regulatory Assistance Project have both offered numerous strategies utilities can use to “flatten the duck” or “teach it to fly:”

  • Target energy efficiency measures for the “ramp up” period
  • Orient solar panels to the west to catch more late evening sun
  • Substitute some solar thermal with storage for solar PV [I’d suggest adding storage to PV also works]
  • Allow the grid operator more demand management via electric water heating [already done extensively by rural cooperatives in Minnesota]
  • Require large new air conditioners to have two hours of thermal storage accessible to the utility
  • Retire inflexible generating plants (read: coal and nuclear) that need to run constantly in off-peak periods
  • Concentrate utility demand charges on the ramp up period.
  • Deploy electricity storage into targeted areas, including electric vehicle-to-grid
  • Implement aggressive demand response programs (subscribing more businesses and homes into programs to shed their energy demand at key periods)
  • Use inter-regional power transactions
  • Selectively curtail a small portion of solar power generation

In other words, the technical challenges of the duck are manageable, largely with existing technology.

The economic problems for utilities – stemming from an outdated business model – may not be so manageable.

More solar generation serving peak afternoon loads will out-bid competitors – utility-owned peaking gas power plants – because solar has zero fuel cost, so utilities will lose money even as customers save money. By 2020 in the California example, solar production at noon will also be sufficient to cut into power usually provided by so-called “baseload” power plants (e.g. coal) that are only economic when operated round-the-clock. The problem for utilities is that they’ve continued to pour money into a twentieth century grid system (inflexible, centralized power plants with long-distance transmission lines) even as the grid has been transformed by distributed, local renewable energy.

Utilities point to the duck chart as evidence that renewable energy development should be stopped on technical grounds. But it’s in the belly of the beast we find the real utility problem: energy demand displaced by solar represents lost market share for utilities as their customers switch to sunshine.

To paraphrase Mark Ferron, recently retired from the California Public Utilities Commission: it’s open season on the utility’s duck.

Image credit: Matt 

Tags: , , , , ,

About the Author

directs the Democratic Energy program at ILSR and he focuses on energy policy developments that best expand the benefits of local ownership and dispersed generation of renewable energy. His seminal paper, Democratizing the Electricity System, describes how to blast the roadblocks to distributed renewable energy generation, and how such small-scale renewable energy projects are the key to the biggest strides in renewable energy development.   Farrell also authored the landmark report Energy Self-Reliant States, which serves as the definitive energy atlas for the United States, detailing the state-by-state renewable electricity generation potential. Farrell regularly provides discussion and analysis of distributed renewable energy policy on his blog, Energy Self-Reliant States (energyselfreliantstates.org), and articles are regularly syndicated on Grist and Renewable Energy World.   John Farrell can also be found on Twitter @johnffarrell, or at jfarrell@ilsr.org.

  • Human 256

    This is not manageable. Some of your suggestions simply won’t work and others will only make a little difference on the margin. You need to realize that this is a large problem.

  • Human 256

    Orient solar panels to the west to catch more late evening sun

    That is done in California. However, think for minute. It will only make the curve steeper when the sun goes down.

    • Bob_Wallace

      No, it shifts the need for non-solar input later into the day.

  • Randall Coles

    Just look at OG&E “smart hours: program where rates spike between 2pm and 7pm where you are encouraged to cut your electric to near zero to save on money.

    The electric companies own program causes you to spike your load around 1pm to precool the home and then ultra low usage until 7pm and then over the next 2 hours your cooling and electric usage spikes then tapers off at night.

    This makes for the same duck curve they are using against solar.

    Its all about the money and always will be. Power plant can ramp up or down at a given notice so this argument is pointless and is all about the money.

  • The chart to the right reflects the average cost per kilowatt hour from 2001 to 2011. New Jersey residents are currently paying over 19 cents per KWH between delivery and supply charges. Sun Up Zero Down can actually lower your cost per kilowatt by 30-40%. Your new rate will be locked in so you will no longer subject to rate increases.

  • Steven F

    Everyone should keep in mind the graph is for March (winter). Few air conditioners are in use and winter winds are very intermittent. The demand increase at 6 PM is due to people driving home from work and turning on the home lights.

    In the summer the grid demand profile is very different. In the summer the inland rising hot air pulls in cool ocean air. So the wind starts up in late morning and gradually increase until sunset and gradually die down at night. The wind is higly predictable in the summer. Also the sun stays up longer so home lights come on later resulting in a reduced evening power demand surge.

    The graph is not cherry picked. It is a typical winter demand profile for California. The demand increase at 6 PM can be addressed by:
    1. increasing the use of LED lighting in homes. This is actually happening.
    2. convert street lights to LED lighting.
    3. increased use of battery storage at home , businesses ,and on the utility grid. The batteries would charge up when solar is at its strongest and then release that energy at night.
    4. Increased use of Geothermal and solar thermal with thermal energy storage. These power plants can operate 24 hours a day 7 days a week.

    With solar increasing rapidly in California now utilities will have to adjust. In California utilities own very few power plants. Utilities buy most of the power from independent power producers. They can start by increasing power purchases from geothermal and solar thermal (with storage) and by installing utility scale power storage systems.

    • eveee

      Yup. All the CA wind is pigeonholed in two tiny places where wind funnels through the passes. CA is not the best place for wind, but the summer diurnal matches the solar output so well. California winter wind is very low. Midwest is just the opposite, and it has much better regional distribution. I question how much storage is needed. There are far fewer peak hours and a lower peak.The effect of wind and solar in Germany reduced the market for nighttime to daytime pumped storage. Solar and wind result in a surplus of gas peakers. For now, its cheaper to operate them less than to build storage. Each place is different and cooperation among wider geographic areas is increasing.


    • eveee

      The true California renewable energy challenge is seasonal, not daily. WInter is not a good time for solar or wind where wind is currently located in California in the passes. In order to get a good mix, a wide geographical area and a better mix of uncorrelated sources, or more sources correlated with load are desirable. Some Stanford papers indicate a mix of offshore and wave energy would be good. Just don’t know the anti-correlation to onshore.

  • Brian Delight

    What about CSP plants with thermal storage? These are a great idea in sun rich regions, and will really help that evening peak, but I am not sure about the cost.

    There are a lot of good solutions but everyone of them is going to cost. Its not just the utilities that pay, its everyone, the poor, the business owner, the overall economy, the average American trying to make a success in this busy, busy world.

    • Bob_Wallace

      We don’t yet know what the cheapest storage solution will be. There are some promising technologies that are yet to be proven.

      Thermal solar storage is limited. All it can store is heat generated in that plant. Other storage technologies are more “generic”. Batteries, CAES, pump-up hydro, and other technologies can store electricity from any source.

      That means that those other technologies can cycle (charge/discharge) more frequently than thermal solar which is limited to one cycle per day (sunny days only). Generic storage can carry late night wind power into the morning peak and daytime solar into the evening peak, earning twice as much revenue.

      It’s a bit more complicated due to things like overall efficiency, cycle life and some other stuff. But, in general, more frequent cycling is going to be cheaper.

      • Brian Delight

        CSP with thermal storage would be ideal for the Duck Chart dilemma. Store energy during the dip and release it during the evening peak. It could be an ideal solution to this specific problem, even with one cycle per day… but still not sure what the cost is.

  • JamesWimberley

    Can somebody tell me what the allegedly incompressible 18 GW load at 3 am consists of? California does not SFIK have aluminium pot lines or electric-arc steel mills, the kind of electricity-intensive heavy industry that runs 24/7. Water, sewerage, hospitals, police stations, street lighting and other 24-hour services, plus pumped storage, don’t at first sight generate that level of demand. Night-time heating? Or is night-time power so cheap that nobody bothers?

    • TCFlood


      Just a few relevant numbers: CA population is over 38 million, pumped storage is over 3.4 GW installed, it takes 1.2 GW just to pump water in the aqueduct from the bay area to LA.

      In general the 3 a.m. “base” demand is more typically 20-25 GW depending on the season according to posted CAISO graphs.

      Also, CAISO, AFAIK, includes utility scale solar and not distributed. Does the “peaking duck” graph purport to include distributed?

    • Steven F

      Many people believe there are few factories in California. That is not true. There are many in California and many and they make many different things. also California is where the the semiconductor industry was born. Computer chip factories need a lot of electricity to run vacuum pumps, RF generators, robotic systems, and other electrically powered systems. there are also oil fields in California and electric pumps are used to pump the oil out of the ground.

  • John Howe

    Globally, humankind does not have a choice — we must make the transition to clean local energy, within a very few decades, as an environmental imperative to avert the worst consequences of climate change. Thus, the challenge to reframe this debate is of paramount importance. Instead of letting utilities offer excuses as to *why* this can’t be done, cleantech proponents must advance solid technical strategies showing *how* it can be done. Kudos to John Farrell, to RAP and the Clean Coalition for pointing the way. Now let’s go farther and think about all the ways we can apply new technologies and use the coming surpluses of
    clean energy, as costs and demand continue to fall, to actually reverse a
    century of climate damage from fossil CO2 emissions.

    • Mint

      The duck curve isn’t an argument showing more renewables are impossible. No matter what the demand curve, it can be supplied with today’s technology.

      It’s merely an argument saying solar’s economic value is limited to saving fuel in FF plants that we still need to keep available.

      • Bob_Wallace

        Well, gosh, isn’t that a terrible thing?

        Saving all that fuel. And slowing global warming?

        Gall-lee, Andy, what are they a-thinkin’?

        • Mint

          I explicitly said economic value.

          ‘Saving all that fuel’ works out to $0.03, plus 0.5kg CO2, per kWh of PV solar produced. I personally think humanity has better ways of spending $1 trillion of goodwill than averting 0.01 deg C of warming, so I’m not gonna add a ludicrous social cost to the CO2 saved.

          Anyway, I stated my point. After a few percent penetration, solar no longer reduces peak demand, and beyond that the duck curve shows that the total cost of electricity generation goes down by only 3c per kWh of PV generation.

          • eveee

            Nope. Looking at it wrong. The grid is not one source. Take a look at the chart you gave. Wind nicely times with solar. Its diurnal. You missed that. That’s why wind plus solar is still lowering the evening peak ten years later. No, the peak does not just stay where it is. And you are ignoring the combined effects of multiple sources. That paints a distorted view of the picture. FUD.

          • Mint

            Wind is not a reliable peak reduction. I’m gonna reply above where you make the same claim.

          • eveee

            You clearly are not a student of wind energy. Diurnal marine cycles are well understood in meteorological and wind energy circles. They time very well with sunrise and sunset. So does solar. Lets not get into the whole one day I saw thing. We are talking probabilities. Correlations. Amounts. Since the peak is shifted to later, and wind correlates with those evening later peaks… that is the crux of your whole fallacy. Starting with ignoring multiple renewable sources and extending to not understanding the correlations. You are grasping at every point to inject FUD. Start by reading about land sea breeze.

          • Mint

            Let’s not get into “one day”? Do you not know what the standard for reliability is for the electric grid? It is literally ONE DAY per decade.

            Mere “correlation” isn’t good enough unless the coefficient is 0.999+, and you couldn’t even cherry pick a source showing close to that.

            You keep showing your ignorance about the grid time and time again, and your own sources keep proving me right.

            NREL WWSIS shows the marginal savings of renewable energy is 3c/kWh, that for hundreds of hours per year system-wide wind output is very low, and that CCGT uses more fuel per kWh when run at lower capacity.
            Your yes2renewables link above shows that it only takes a few hours per month for peakers to rip you off enough to have a big impact on overall prices.

            Bob also knows that deep backup is needed to stick around, so why do you deny it? Do you know why he does? Because like me, he actually reads studies. And no, these studies are not FUD.

          • eveee

            You keep pretending that a comparison of renewable reliability means you must look at just one source. If your arguments were true, Germany would not be the most reliable source of electric power, far greater than reliability in the US. Just another canard.

          • Mint

            Where? Where did I look at “just one source”? All my examples use data for grid-wide totals of thousands of solar & wind sources.

            Germany is not shutting down FF generation capacity, due to its renewables, which you keep claiming is possible without proof.

            Speaking of Germany and combined output:
            “The maximal daily sum of Solar and Wind production was 0.58 TWh at 31.01.2013
            The minimal daily sum of Solar and Wind production was 0.022 TWh at 16.01.2013”

            Hourly analysis shows how little solar and wind helped during the seasonal peak demand in January (see page 156).

            Stop making strawman arguments. I never said renewable energy makes the grid unreliable. I said unreliable renewable output (i.e. rare dips) needs to be covered by FF. Even if it’s as rare as 1 in 100 days, it needs to be covered.

          • Bob_Wallace

            Nov 2013

            “At the beginning of the month, Germany’s Focus Magazine reported (in German ) that the German Network Agency plans to allow 12 conventional power plants to be shut down.

            … the new record-low wholesale prices
            spell doom for all conventional power plants, which are running at ever lower capacity factors. Utilities have responded by applying to
            have their plants shut down. They cannot do so on their own; first, Germany’s Network Agency has to give its approval based on an estimation of how critically a given power plant is needed to ensure supply quality.
            At present, some 28 power plants with a collective capacity of 7,000 megawatts – roughly equivalent to the capacity shutdown in Chancellor Merkel’s sudden nuclear phaseout in March 2011 – have been submitted for decommissioning.

            In some cases, the firms placing these submissions really just want money; if the Network Agency finds that the power plants are to crucial to be shut down (usually, it’s a matter of geography), the firms can demand special compensation to ensure profitability. Essentially, these are capacity payments ….”

            March 2014

            ” E.ON Friday said it plans to shut down one of its nuclear reactors seven months before its operating license expires because subsidized renewable energies and a tax on nuclear fuel have eroded the plant’s profitability.
            Germany’s largest utility by market value said it has requested approval from the country’s grid regulator and a power network operator for shutting down the 1.3-gigawatt reactor Grafenrheinfeld as early as May 2015.

            The rising amounts of power from wind farms and solar-power plants coming on the grid have resulted in a capacity glut that has eroded wholesale power prices and the profitability of large power stations .

            So far utilities have mothballed mostly gas-fired power plants. These power stations have been hardest hit because they are designed to meet peak daytime demand which is when wind- and solar-generated electricity is available.

            E. ON’s announcement to retire a nuclear reactor early, however, is the first time that a utility is shutting down what was widely seen as one of the industry’s cash cows.”
            http://online.wsj.com/news/articles/SB10001424052702304418404579467023883510280 .

            ” unreliable renewable output (i.e. rare dips) needs to be covered by FF”
            Periodic supply shortages need to be covered by something. Right now we’re using fossil fuels because we’ve got ’em.

            That does not mean that the only way we can deal with supply shortages is by burning fossil fuel. We could burn biofuel or use stored power.

          • eveee

            Thats what happened in Presque Isle. MISO is keeping an old coal plant open that the owners wanted to shut down. They claim they need it for reliability. They need to run some power lines out there from Duluth or something, Upper Peninsula is pretty remote.

          • Bob_Wallace

            Sorry about the links garbage.

            I’m saving this one (not cleaning up the links) so that I can show an example of the problem of replying from email to the Disqus folks.

          • eveee

            One source of energy, not citation. I hope that is clear. So look at the curve. Don’t argue with me. Argue with the curve. That curve you see has a really compressed scale. Take a look at caiso today or tomorrow and notice. The solar goes down at 6PM, the wind goes up. In the morning, the wind goes down, the solar goes up. Just look at it.

          • eveee

            You didn’t say renewable energy makes the grid unreliable? You said the correlation between wind and solar has to be .999. You said there has to be some extremely high level of reliability, like 1 hour in 10 years. Whatever. Now you say renewable output has to be covered even if its rare. But the whole point is that if it is rare, its less expensive to cover, because if its rare, less reserves are needed to cover it. See. Follow your own logic, man. And peak power is worth more. Solar follows that. That is why it is worth more than retail. It does not just displace fuel. Which is where you started. It displaces reserves and expensive gas peakers. That is why those are going out of business in Germany. You need to stop worrying about wining arguments and make sure your ideas correlate to what is actually going on. Like sure enough today, and every day since you made your first comment, wind dies down in the morning when solar comes up, solar dies down in the evening when wine comes up. And it does not have to have anywhere near the reliability numbers you claim in order to get value for capacity and peak power credit. It nicely reduces peak power in the summer, and that is what determines outlays for capital equipment to meet peak demand the next year. It lowers costs.

          • Mint

            .999 correlation is needed only IF you want to shut down major amounts of FF capacity.

            You are the only one making that claim. Bob knows we need “deep backup” even with storage. Budischak knows it. Germany knows it. Pro-wind experts on cleantechnica know it.

            But the whole point is that if it is rare, its less expensive to cover, because if its rare, less reserves are needed to cover it.

            On what basis do you make such a claim? Whether you need 20GW of reserves once a year or 20 times a year, construction and staffing costs remain the same. You only save fuel.

            Germany doesn’t have peak load in the summer. And California net load has been during hours of minimal PV for years now. Yes, I admit that PV initially reduced the net load peak by a bit, but it isn’t doing so any further now.

          • eveee

            FInd a quote where I said

            “.999 correlation is needed only IF you want to shut down major amounts of FF capacity.” You can’t. I never said it. So I guess I can’t answer what basis I made the claim on. Try to quote correctly, please.

            The next quote you did successfully. Now for reserves 101. Reserves are calculated on the basis of probabilities. In particular, wind is forecast and probabilities for reserves are a function of the accuracy of the forecast. Since you asked, I thought I might explain why the frequency (and probability) of events dictates reserves. But being a mathematician and all, you should already know that. 🙂 And no, I am not worried about the basis of making the claim, since I have read the peer reviewed papers and no I am not going to provide you with a reference for one because its time for you to do some homework. Stop arguing with a graph you contributed that contradicts you. Admit it. You screwed up. Care to post another caiso.com graph? Like todays? Maybe wait until spring or fall. But the summertime peak is what counts. Not trying to make debate points.

          • AldivosTarril

            > “Germany is not shutting down FF generation capacity, due to its renewables”

            German electricity supply 2007 – 2013: coal down 11 TWh; gas down 12 TWh; nukes down 44 TWh; renewables up 59 TWh. https://pbs.twimg.com/media/Bdydb4UCMAAiT1A.jpg:large

            So, you reveal deep ignorance on the basics of Germany’s Energiewende.

            > “renewable output (i.e. rare dips) needs to be covered by FF”

            Or by hydro, or biogas, or biomass, or HVDC connectors, or hydrogen produced by solar / wind, or geothermal, or wave, or tidal, or grid-scale storage, or demand-side control, etc. etc.

            Your beliefs seem to be built on ignorance and fossil-nuke propaganda.

          • Mint

            I said CAPACITY, not energy.


            Look at page 156. That week in Jan, Germany needed all of its conventional capacity and still had to import, because renewables had very little output.

          • AldivosTarril

            Of course you *want* to talk about “CAPACITY” and cherry pick one week in January because actual electricity production and the overall results don’t support your anti-clean energy ideology. Germany continues to prosper due to its investment in clean, sustainable energy.

            Correct. Demand is falling as the efficiency component of the Energiewende delivers outstanding results as expected.

            Coal use would have fallen much faster except for high export demand from neighboring countries. The fossil-nuke shills and ideologues try to paint this as the fault of renewables – which is a lie like most everything else they produce.

          • eveee

            Correlation isn’t good enough unless its 0.999? Oh. Sorry. I didn’t realize that. Thanks for the information. Please explain how Minnesota came up with a value of solar greater than the retail rate. How do utilities do that now? They have reserves. Everything requires reserves. If what you said were true, Denmark and Iowa, not to mention Germany would have the most unreliable electricity on the planet. Instead, Germany has the most reliable electricity. The grid managers have to guess when a large nuclear or coal plant is going to have an unplanned outage. Whats the correlation between that and the rest of the sources? How does that famous goal you stated get met? Think about what you said, with a little less emotion and a little more calm, clear thought after you calm down. You said peakers make a big impact on prices. So true. Solar reduces the amount of reserves and peakers used. Thats why its worth more than retail. Its displacing the most expensive electricity.

          • Mint

            Have you ever done any statistical math? Do you know what the binomial cumulative distribution function looks like?

            When nuclear, coal, natural gas etc have outages, it is UNCORRELATED. That is extremely important, because it means it is next to impossible for 68GW of capacity to be unable to output less than 1GW average over a day, as German solar/wind did on Jan 16, 2013.

            You wouldn’t need 67GW of reserves to backup 68GW from nuclear/coal/CCGT. If we had sixty-eight 1GW plants with 5% equivalent forced outage rate (and that’s poor by western standards), then 99.95% of the time they’ll be able to output at least 58GW, so only 10GW of reliable reserve would be needed.

            Do not equate reserves needed for intermittent and traditional sources. Germany has reliable power because it is NOT taking your advice of shutting down FF as renewables increase.

            (Before we get into another semantic battle, what is your definition of reserve? FF plants span the gamut of capacity factors from 1% to 80%+)

            You said peakers make a big impact on prices. So true. Solar reduces the amount of reserves and peakers used.

            Additional PV stopped lowering the net load peak in California a long time ago (proven by data). Neither solar nor wind showed up during some of Germany’s highest demand days (proven by data).

            The only thing I will agree on is that ever increasing solar will reduce the number of hours that peakers are needed. But that does not prevent those peakers from charging higher prices during the remaining hours (or equivalently, as Bob’s link shows, threatening to shut down if they don’t get higher capacity payments).

            This is how product pricing works in all competitive commodity markets. Equilibrium unit price = (fixed costs + volume * unit cost)/volume + small profit. If volume goes down for any company, long term price goes up (barring any efficiency improvements). I don’t know why you think electricity is a big exception.

            After a few percent penetration, solar displaces fuel only, and much of the expense you think is averted simply gets crammed into the remaining solar-less hours.

          • Bob_Wallace

            “Germany has reliable power because it is NOT taking your advice of shutting down FF as renewables increase.”

            Actually, Germany is both reducing their fossil fuel capacity and reducing the amount of fossil fuel they consume.

          • Bob_Wallace

            “After a few percent penetration, solar displaces fuel only”

            So friggin’ what? Eliminating fossil fuel use is what we’re after. Solar and wind, along with hydro, geothermal, biomass/gas, tidal and storage can do that.

            In the near term we will continue to use fossil fuels as there is no tolerance for cutting electricity use by 90%.

            As time goes on we will use less and less fossil fuels until we get to the point where we are using less than 5% of what we now use. We can live with that.

            And we may even find that we can develop cheaper ways to produce that last ~5% other than gas plants. Adding extra turbine or two to a hydro plant may be cheaper than gas plants. Or cheaper to add some more storage tanks to a flow battery facility.

            It feels to me that you’ve latched onto a tiny bit of relatively unimportant information and you’re riding that tiny pony into the ground.
            You may be right. We might have to keep a significant amount of gas capacity on standby to be used a few hours a year. But with the low cap cost of NG plants it’s affordable. Plus we likely already have all the NG capacity we need. And our current electricity prices include the cost of running gas plants, on average, less than 30% of the time.

            Continuing to use fossil fuels for over 10% of our electricity is not affordable.

            Replacing worn out coal and nuclear plants with new nuclear plants is not affordable.

          • Mint

            So friggin’ what? Eliminating fossil fuel use is what we’re after.

            Your position is very different from that of eveee. My debate with him is about cost in today’s market (i.e. without a price on carbon).

            Your premise is that we must reduce FF down to 10% of electricity generation, period. That leads to different conclusions and different lines of discussion.

          • eveee

            No. Solar does not stop getting peak demand when it is a few percent. Stop exaggerating and provide some math and links. You are making serial false claims. I caught you on the California graph that shows diurnal wind lowering the net peak at 7PM. Or just keep digging your hole….

          • eveee

            You sure about the uncorrelated nuclear outages? If you were, and you were a mathematician, would not say no correlation. You would use a number. I don’t want to get into pure math, but isn’t a situation with a discrete event system modeled on power outages without replacement? I mean you are not going to put a power plant back in the pool mathematically or otherwise.

            “If the sampling is carried out without replacement, the draws are not independent and so the resulting distribution is a hypergeometric distribution, not a binomial one. ”

            You seem awfully sure of yourself. So how to explain the domino effect of power outages that has happened on high demand days. Hmmm. Oh that. And how to explain the double outages at various times and places…. Hmmm Oh that. OK. Reserves are needed for … outages, yes. OK. Now what are peaking power plants needed for. Oh. They are used to meet the demand variation that nuclear cannot. So don’t get necessary peaking plants get mixed up with reserves. Both nuclear and renewables need reserves. But peaking is a different matter. Solar does that without trying. Baseload is base load. The most flexible nuclear in the world is not as flexible as peaking as plants. Like a broken record, you keep ignoring that wind lowers the new net peak at 7PM. Solar keeps lowering the daytime peak. (In California) How could it not? Your logic is .. faulty. Don’t go trying to weasel out of it by switching to Germany. You cited California, so California it is.


            In the Great Blackout of 2003, 22 NPP went offline. LOOP is no joke for NPP and restart was a long time.


          • eveee

            Wind in California lowers the evening peak. The ducks bill. Its diurnal because there is a land sea breeze. That meteorological weather phenomena is very reliable. Stop distorting my comments. `Read it. It says evening peak. Solar ends or falls off around 5 to 6PM in summer. Wind picks up at that time. They dovetail. Its been doing it for months. Most of the time daytime summer winds are low. Occasionally, the solar peak is accompanied by high winds. Then wind participates in the daytime peak, too. In California, the peak wind corresponds with both the seasonal peak demand, and the ducks bill created by solar. The result is a lower ducks beak. And the whole curve goes down as wind and solar are added.

          • Bob_Wallace

            I suppose it does not good to go through the argument with you again. You’ll simply ignore things you don’t like.

            But, heck, I’ll throw it out.

            If we use our NG plants less they’ll last longer and we will also save replacement costs.

            I’ve no comment on your apparent disregard for climate change.

          • Mint

            What you call an apparent disregard for climate change is due to me treating people in third world countries like humans. It’s ridiculous how little it costs to save lives and suffering there.

            Western society unfortunately has a very limited appetite for helping global causes, and regardless of your perception of how much damage GW is doing to poorer countries, putting the same dollar towards foreign aid helps them orders of magnitude more.

            I’m not a climate change denialist. That figure arises from IPCC warming projections. I’ll spell the calculation out for you if you ask.

            And as someone who doesn’t subscribe to right-wing supply-side nonsense in the modern demand-limited economy, I can’t buy the economic models used by Nordhaus to calculate the SCC.

          • Bob_Wallace

            Right now we need to cut our GHG emissions or we’re going to screw people in developing countries something fierce. They do not have the resources to adapt or to migrate.

          • Mint

            Right now developing countries need vaccines, water infrastructure, schools, infant nutrition, agricultural development, and a ton of other things. They ARE being screwed right now to the tune of millions of deaths per year in a way 0.1 degrees of AGW could never get close to approaching.

            Just try some math.

            They generated 29.7TWh of solar energy from 35.6 GW of capacity. Optimistically assuming minimal degradation over a 35 year avg lifetime, let’s make that 1000TWh. Let’s generously assume that it all displaces coal (1kgCO2/kWh), and we get 1 billion tons of CO2 averted. The IPCC tells us the world’s 35B tons of CO2 per year is causing ~0.2deg/decade warming. So what do we get?

            36GW of German solar prevents 0.0006 degrees of warming.

            I’m sure developing countries are jumping for joy that Germany spent tens of billions to do so, when instead that money could feed 100 million infants for a few years (that’s not a typo), saving them from permanent malnutrition disabilities or death.

          • Bob_Wallace

            Get back on topic.

      • eveee

        We need to be careful to talk about amounts. What was an all day peak has become a lower, shorter, evening peak. There is peak benefit for most of the day, a longer time than the evening peak. It is true that once there is a lot of solar, and with the daily peak greatly reduced, wholesale rates drop. That is proved in Germany. So whats the problem? Low rates? Too much generation? As a few have pointed out, places that have had summer air conditioning brownouts are starting to see that disappear due to solar. Solar is an ideal natural peaker. With wind it reduces the need for storage and/or reserves, not increases it. It is not limited to reducing fuel costs. We know when solar happens. Its very predictable. We are not at 20% renewables in every US state, or any large region in the world. That talk is just FUD. Fear, uncertainty, doubt.

        • Mint

          Germany has some of the highest electricity rates in the western world due to a very substantial FIT program. You can’t ignore the cost of FIT programs when looking at wholesale prices. Ontario is a less extreme version: 3c/kWh wholesale, but 9c/kWh total cost after paying for PPAs.

          I admit that getting a few percent energy from PV was beneficial in reducing the peak. But after that? Nope. The net peak remains in the evening, and I’ve shown you hard data.

          A ‘shorter peak’ only means higher prices for the remaining hours, as the plants generating during those hours run at lower CF. I don’t understand why you think there is some magic going on. If the only saving is fuel, then that’s all you’ll see in the prices. These aren’t iPhones with massive profit margins. Electricity is a commodity in a competitive market with low average profit, and if PV only saves fuel, then that’s all you’ll see reflected in generator prices.

          The retail utility, OTOH, is a regulated monopoly that passes grid and generation prices onto the consumer.

          Wind is another topic that I won’t derail this PV/duck-curve discussion for.

          • Bob_Wallace

            Germans pay more in non-utility taxes as part of their retail electricity cost than FiT caused taxes.

            Let’s look at how costs break out for retail customers in Germany…

            In 2013 the average household electricity rate is about 29 € cents / kWh according to the BDEW (Energy industry association).

            The composition:

            8.0 cent – Power Generation & Sales

            6.5 cent – Grid Service Surcharge

            5.3 cent – Renewable Energy Surcharge

            0.7 cent – Other Surcharges (CHP-Promotion, Offshore liability,…)

            In addition there are some taxes & fees that go straight into the governments budget:

            2.1 cent – EcoTax (federal government)

            1.8 cent – Concession fees (local governments)

            4.6 cent – Value added tax (19% on all of the above) – (federal, state & local governments)

            So 8 + 6.5 or 14.5 euro cents go to electricity purchase and delivery. It’s only 0.5 euro cents higher than the EU27 average.

            6 cents for investment in a clean grid. Temporary. And greatly lower electricity in their future as wind and solar farms pay off and run close to 1 cent per kWh.

            8.5 cents per kWh for other taxes.

          • eveee

            Nice job of slaying myths, Bob. That one is a medusa.

          • Bob_Wallace

            It has an ugly twin. The high retail price of retail electricity in Denmark.

            Taxes, also, too.

          • Mint

            LOL what myth? Bob just proved that they’re paying 5.3c on every kWh to pay for <20% electricity from solar & wind.

          • eveee

            And what if they are? That was their choice and they paid for solar when it was more expensive. Now its cheaper. In the US there is no PTC for wind. It isn’t stopping Warren Buffett from investing in it. He knows he can make good profit from wind without any subsidy. You failed a long time ago when you made the first false claim. You just keep doubling down on more false claims. Wind is under 5c/kwhr in the US without any subsidies. Cheaper than coal and nuclear. Yet coal and nuclear have subsidies. Nuclear has a PTC. Yet you don’t complain about subsidies for nuclear. Seems hypocritical.

          • eveee

            What? Start by looking up the merit order effect. You have no concept yet. It does pertain to the duck. And MOE is not magic, its math. Here is some instructive reading.


          • Mint

            So what? That does nothing to address how wholesale prices are misleading if you ignore PPAs.

            Nuclear has the same effect on wholesale prices due to its low marginal price, but I don’t see you extolling that virtue of nuclear and saying it lowers costs. It doesn’t, and neither does Germany’s renewables.

            Retail prices can rise even when wholesale prices drop. I don’t have German market details, but look at Ontario as an example:


            The “global adjustment” is there to pay for all the PPAs, including nuclear at 5.7c/kWh, wind at 11.5c/kWh, solar at 39.6c/kWh, etc. Wholesale prices mean very little to the consumer.

          • eveee

            How can nuclear lower wholesale prices when it comes lowest on the merit order? It does lower wholesale prices or even make them negative when there is little demand, yes. But it has little effect on peak prices and next to nothing on the amount of peakers needed. That is dictated by load variation. Nuclear has limited flexibility to follow loads and loses even more profitability if it tries to follow loads. I hate to mention it, but nuclear is off topic on a thread about solar.

          • Bob_Wallace

            Nuclear can lower prices or drive them below zero only when demand is very low and nuclear takes a loss rather than go offline.

            It’s a false savings. Taking that loss means that nuclear has to sell for higher when demand is up. If they are the highest accepted bidder then peak settlement prices are going to be higher.

          • Mint

            BTW, I’m fully aware of the merit order effect, and you clearly are not trying very hard to see how my arguments fit into the market.

            Take a look at your link again. You see those bidders at the end of the merit order? If they’re losing bids due to the rise of solar, then they’re going to increase their bidding price during the hours that solar isn’t available. That’s the only way to make up for lost revenue and pay the bills.

            Bids are not fixed. If you add wind or solar, the remaining generators will change their bidding strategy unless you can cut them out entirely.

            Look at the last graph: $1000-$10000 per MWh! Obviously that has nothing to do with fuel price. It’s peakers putting in insanely high bids a few times per month.

          • eveee

            Now I am sure you are spreading FUD and have lost it. Think really hard about what you just said. You just said

            “If they’re losing bids due to the rise of solar, then they’re going to increase their bidding price during the hours that solar isn’t available.”

            Exactly when is that? When the sun is not up? It isn’t 4AM when there is no load and only cheap base load is needed to fulfill demand. There are fewer and fewer hours when base load plants can take advantage of MOE because peak load is no longer all day. Just like this article says. And just like you are trying with all your cognitive might, to ignore.

          • Mint

            4AM? WTF?

            Have you been paying attention to anything? Where is the peak in ALL these net load charts, right from the beginning of the article? It’s always 7 or 8 PM.

            You (and this link) continually operate under the false premise of static prices from each generator. That isn’t true in the slightest – not in the electricity market nor any other competitive market. The lower the demand for a product, the higher you need to price it to cover manufacturing overhead.

            Fewer hours is only half the story. The other half is how much price rises during those reduced hours.

            From your own link: “Prices have indeed been rising across Australia, due to the need to replace ageing infrastructure (power poles and lines, for example); also due to rising peak demand spikes.

            To conclude this section, here is one illustration of peak prices jacking up the overall wholesale cost, the following graph is from the Zero Carbon Australia 2020 stationary energy plan:”

          • eveee

            The peak load is 7 or 8 PM after solar and wind have cut into it. Its 5PM or earlier before. You said, “If they’re losing bids due to the rise of solar, then they’re going to increase their bidding price during the hours that solar isn’t available.”

            I am trying to say, when is it that they are going to increase their bidding price during the hours when solar is not available? That can only mean after dark. They can hardly get into a bidding war over peaks at 4AM, since that is the lowest demand of the day. That leaves a very much shorted and lowered peak in the early evening. Not nearly as much time to make MOE money on. That is exactly what the article says. That is exactly what you have been trying so strenuously to avoid. FF powered plants make less money when solar generates power because they lose out on MOE windfalls at peaks.

          • Mint

            The peak load is 7 or 8 PM after solar and wind have cut into it.

            FINALLY I’m getting through to you.

            Yes, the peak net load is now 7 or 8 PM (has been for a while). So if we add more PV today, what happens to this new peak? Nothing.

            Peak shaving benefits came during the first few percent of solar penetration. Marginal benefits of PV, i.e. from 2014 onwards, are fuel savings only.

            That leaves a very much shorted and lowered peak in the early evening. Not nearly as much time to make MOE money on.

            Buddy, look at the graph in your own link. You don’t need much time to make a lot of money. A few hours per month is all that’s needed.

            Think about this from a business perspective. If you owned a $100M peaker, and are competing with other peakers for ~1000 hours per year at the end of the merit order, what is the minimum average revenue per hour you’ll settle for? Now what if your window of opportunity was cut down to 200 hours per year? Then how low would you be willing to go with your price?

          • Bob_Wallace

            I’ll butt in and answer.

            I’d raise my per MWh price high enough to cover my costs plus give me a reasonable profit.

            Of course, as that happens, someone will install storage and cut into my market even more. Thus we enter a downward spiral for gas as asking price rises and storage captures more and more of the market.

            It would probably get so bad that utilities might pay some gas plants for capacity and call on them only a few hours a year.

            Or they’d decide that it makes more sense to upgrade existing PuHS and flow battery storage, after which gas plants would take a one way trip to the scrapper.

            But that’s out 20, 30 years. Who know what magic will be invented during that interval? Not even the Shadow….

          • Mint

            And you’re absolutely right. Cost plus reasonable profit is exactly how every competitive commodity market works. Paying plants for capacity is basically an equivalent to this also.

            If storage capacity can be built for a similar price (i.e. $/kW, a different factor than the usual $/kWh metric) to natural gas capacity, then yes, it will cut into the market. I see that happening with short term batteries.

            But I don’t see storage $/kW (even PuHS) getting much lower than natural gas $/kW, so consumer savings will probably be minimal.

            There is one exception, though: Grid integrated vehicles.

            There, an EV is already being bought for another purpose, and storage payments are just a bonus. You could pay an EV owner $3 to dump 10 kWh into the grid 50 evenings a year ($0.30/kWh sounds like a lot, but it’s only $150/yr for 3kW peak handling), and no peaker could compete with that price.

            Let’s hope EVs take off.

          • Bob_Wallace

            “But I don’t see storage $/kW (even PuHS) getting much lower than natural gas $/kW, so consumer savings will probably be minimal.”

            You don’t see the cost of a tank of vanadium/whatever being cheaper than a gas plant? Include the cost of standing by for the gas plant.

            You don’t see the cost of bulldozing out a larger pool for PuHS as cheaper than a gas plant? Include the cost ….

          • Mint

            Bulldozing out a larger pool for PuHS doesn’t decrease $/kW, but rather increases it. It only decreases cost per $/kWh, and even then only if that new pool is cycled as often.

            Vanadium/whatever may get a bit cheaper than natural gas, but I don’t see it being a lot cheaper. Your figure for natural gas was $0.66/W, right? Between the battery ($0.06/Wh * 5 in the distant future?), inverter ($0.10/W in the future?), packaging, and profit, I’m guessing $0.5/W is the long term cost floor.

            But yeah, it is of course just an educated guess.

          • Bob_Wallace

            “Bulldozing out a larger pool for PuHS doesn’t decrease $/kW.”

            Yes, it reduces the overall $/kW price because fixed costs are spread over larger capacity.

            We are guessing.

            We’ll need storage. I don’t think you disagree with that. We don’t know for how long, how many days.

            Let me just set up a hypothetical. Let’s say that for 98% of the time we need storage to get us through three days. But once a year, that other 2%, we need a week of storage. Now how to fill in for those extra days?

            (BTW, I just read that UK offshore wind is 17% “firm”. Always on – baseload – whatever term you like.)

            Cheaper to install an extra turbine in a PuHS and make the reservoirs larger or install a gas plant? (We’re assuming new builds. Using existing gas plants is a different situation.)

            Cheaper to add more tanks to a flow battery facility?

            We don’t know those numbers. Could play out either way.

            Cycling isn’t the issue. Both the extra turbine/pool would get as much use as the gas plant.

            One difference is that the gas plant would need staffing/maintenance/security during 51 weeks of sitting idle. Those costs would be rolled up into the normal costs of a PuHS plant.

            I have no way of knowing what storage and gas plant costs might be 20, 30 years from now when we approach the 100% renewable grid goal. Or, more likely, after we’ve reached the 95+% renewable grid level and our gas plants start wearing out. That sort of speculation holds not a lot of interest for me. Going back 30 years and we were just starting to get personal computers, there were no cell phones, no CD or DVDs, photography and movies were film, TVs were cathode tube, and on and on and on….

          • eveee

            You don’t need to use Buddy in a demeaning way with me. Thats pointless. You are stuck like a broken record. In your endless zeal to prove that solar can’t do it all, which is true, you forgot that there are other renewables. Despite multiple requests that you observe and note the frequent synergy between wind and solar because of the diurnal land/sea breeze, you have failed to acknowledge. I don’t know if this an ego thing or whatever, but it does not harm your reputation to notice this and acknowledge it. The rest of the last response is given to exaggeration. Only the first few percent of solar penetration? When solar becomes so prevalent in California that the duck has a deep belly, then we can talk. Solars daytime output is steady at about 4G. The summer peak is at least 20G above trough. Even on not so warm days its about 30G. That means at least potential 10G just for peak generation, and possibly more out to 20G. That does not yet displace the base load. And why not displace it? We want coal plants to be displaced. Thats an even greater opportunity. What do we care if the FF peakers are no longer economic or completive? They have to compete, too, and if they don’t…. So what if we use solar instead? You are making no sense. The goal is not to keep FF peakers profitable. If you are a gas peaker and you are confronted with the hypothetical you placed, then you probably exit the business. You still have fuel costs and you cannot compete with solar on that. Your logic that solar does not get any peak shaving benefits seems to ignore the merit order effect. Every single plant chosen at the peak gets the highest price. Its true that the market for peaks is obliterated by solar. And solar can stand on its own with its low fuel cost. I fail to see a problem in that, except for the FF peakers and other base load that can no longer make up for super low night time rates by cashing in on the merit order effect during the day. Even after kicking out the other peakers during the day, solar will still on average command a higher average price until daytime rates are lower than night time most of the year. At that point, storage will become very lucrative. Finally, just because the net peak has shifted to 7 to 8 PM, (which means little because MEO is not computed by net peak, but rather the real peak demand) does not mean the daytime demand will not gather a premium. It will be less, and yes solar is doing that. As the article states, solar kills FF profits of both coal and gas because they rely on them through MEO. Thats a win.

          • Bob_Wallace

            As they run up their prices storage will start cutting them off at the knees.

            We’re in the transition, babeeee!!!!

          • eveee

            A shorter peak means higher prices for the remaining hours? Look, the peak did not just get shorter in time. It got shorter in height and time. That means less peakers are needed. Less peakers needed means less cost. The capital cost is the same even if they are never used. Only the fuel cost increases if they are used. And reserves (which might never be used) are reduced as well because the probability that they may be needed is lowered. You don’t seem to understand how this game is played. Thats why displaced file costs are not the only game in town.

  • Beth Bond

    Great article John!

  • CsabaU

    How was the predicted 2014 curve? Why should the electricity demand raise so much at 8 pm? I find in unlogical.

    • eveee

      It doesn’t. Its like carving a piece of wood. Whittle it down and what was a bump becomes a dip. What was a dip becomes a bump. Nothing increased. The shape just changed.

      • CsabaU

        Hi all, look at the curve. In 2013 there is a significant incraase at 8 pm. The postulate is that the peak will further increase significat 2014 and more slightly the years after. (If it was due to economic and population, the increase would be also in the night time. Since lighting and most electrical devices like TV are more efficient, I don’t understand why the large increase. Is the population assumed to buy more electrical gadgets and play with all of them simultanously in the evening? It is the reason I would like to see the actual 2014 curve.

        • eveee

          Thats a pretty safe bet coming from a utility. They have to make that claim to the PUC to justify more capacity and rate increases. Thats how it works for them. If demand decreases and they are stuck with expensive, long term amortized thermal PP, they are stuck with paying for them when no one buys their electricity. That is happening as a trend right now. Conservation did that. The last thing in the world they want to see is rooftop PV carving more of that out of their dwindling market. They can’t go to the PUC and say, demand dropped. Now we need a rate increase. There is no precedent for that and the public/private utility system never anticipated it. So they are stuck throwing a duck curve at people and hoping no one notices that there is a decrease in demand hiding behind it. They are pointing furiously at the head and saying, look its going up. (don’t look over there at the belly)

        • eveee

          CsabaU. I think you are right. Leave it to a utility to predict ever higher electric consumption. Good pick.

          • Bob_Wallace

            How about a picture of what is happening with US generation/demand….

    • Mint

      They’re including general demand increase from economic & population growth. The 8 pm rise is not due to solar.

    • Steven F

      It is a typical march (winter) power curve. The power increase starting at 6PM is due to people driving home, turning on the lights, TV / Radio, and cooking dinner. Many states have similar winter power curves.

      • eveee

        The cherry pick was to choose a typical march(winter) power curve where peak loads are much lower and ignore the annual peak that is much more significant in July/August. It matters much more for electric rates and capital expenditures for peak generators.

  • eveee

    John, you got that right. The duck curve had to be cherry picked. During summer heat waves there is a peak at 5PM or earlier. The duck curve will not happen during July, only in the lowest demand days of the year during Spring mostly. During the peak Summer air conditioning days, solar and wind are invaluable. As you say, the industry loses a lot of money, but ratepayers gain.

    It strikes me as an odd PR like gesture to point at a demand curve that drops and claim thats a bad thing. Bad for whom? A very subtle way for utilities to hint that they want more, not less demand. (while pretending they love conservation)

    • Mint

      The peak during solar hours is only slightly lower than the evening peak, even during the summer. There’s plenty of solar now in California, and net load (total load minus solar/wind) is in the evening now.

      For example, here’s yesterday’s total load and net load:

      (Source: http://content.caiso.com/green/renewrpt/20140721_DailyRenewablesWatch.pdf)

      Subtract existing solar, and 8PM is where the remaining demand’s peak is. More PV isn’t going to change that, and only saves fuel costs. Only batteries or pumped hydro will reduce the peak fossil fuel needed, but of course they cost as well.

      • Bob_Wallace

        Or Wyoming wind….

        • Mint

          Wyoming generates 88% of its electricity from coal:

          The more California uses Wyoming wind, the less Wyoming can use for itself.

          While wind can, on average, appear to give you a nice complement to solar (I assume this is why you suggested it), the evening rise isn’t reliable.

          • Bob_Wallace

            I’m not sure what Wyoming’s coal use has to do with anything. Wyoming has some excellent wind resources and they are starting to develop it.

            Additionally, work is underway to run a HVDC line from Wyoming’s wind areas to the Pacific Intertie and Intermountain Intertie. That will make Wyoming wind available throughout the West.

          • Larry

            Excellent reply. I’m afraid “Mint” somehow thinks all the wind in Wyoming is currently being used to generate electricity. Needs to get out from under that rock.

          • eveee

            Electricity knows no borders.

      • eveee

        That’s correct. Actually, wind and solar dovetail nicely as the graph shows. Local patterns provide that. Its diurnal. If you go back several years, you will find that the same pattern appears and solar plus wind provide the same peak hour, but the peak is lower. One must always consider the effects of a combination of sources. So more PV and wind can lower the peak, but it happens at the same time. Eventually the peak is really low, maybe even as low as the tail, and the belly is way lower.

        • Mint

          Wind does not reliably give you that diurnal curve. Look at the state-wide wind production on this day:
          And then two days later:
          (FYI, I found this randomly. I’ll do a more thorough data analysis later)

          In two days, total CA wind production at 7PM went from 3820MW down to 357MW. As a result of this and demand differences, thermal generation at that time went up from 9544MW on Tues to 15395MW on Thurs of the same week.

          I do consider the effects of a combination of sources. That’s why I keep quoting Budischak and NREL in my discussions with you. They show that combined output can reach some pretty low levels at times.

      • eveee

        The very graph you chose shows the diurnal wind variation matching up nicely with solar. Doozy. Trying to ignore the graph you presented? Not very convincing. I have been watching it. I didn’t believe it. But there it was. Day after day. Week after week. Like clockwork. Please stop cherry picking days it doesn’t happen. I know that. Its how non random it is that counts. That graph is a typical CA summer day.

        • Mint

          Please stop cherry picking days it doesn’t happen. I know that. Its how non random it is that counts.

          In what universe do grids work like this? Name a single US grid with target reliability less than 99%.

          These “cherry picked” anomalies are the only thing that matter for capacity. They are far from irrelevant.

    • Steven F

      Iit is not cherry picked. It is a typical winter power curve fore many places including California. The graph in the article is for march. In the winter wind power is highly intermittent. In California wind power is mainly a summer power source. Solar in the deserts is good most days of the year.

      • eveee

        I meant the rest of the year looks nothing like that. Take a look at today. There is no belly. The daytime demand peak is 5PM. What counts the most is the peak demand for the year. It necessitates installation of expensive generation and transmission capacity. All for a few days per year. The daily peaks are no big deal for the utility. The peak in Spring is even lower than the peak in winter. Its so confusing to discuss the load curve, its easier to show it in a 3d graph so you can see daily and seasonal cycles on the same graph. Like this:


    • Human 256

      No, the Duck’s head is the non A/C peak load. It is called the “Shoulder” in power-speak. During the A/C season, it is true that the A/C load will mean that the ramp up before the Shoulder will not be as rapid.

  • brad

    Well, that is well written and clear. Thanks.

  • Matt

    First go to TOD billing since it get people to move load to where it is cheapest, money talks!
    Yes do the thermal storage on large AC, it is a very cheap approach and allows for smaller AC.
    Then the easiest and simplest solution the the duck technical problem is to give every PV owner enough batteries to store 4 hours of the PV output. Then the bottom of the valley and top of humps would go away.

    If utilities don’t want the consumer having the batteries, then they can install them at their substations.

    • CitizenK

      Batteries are not the best solution to energy storage, they are expensive, often environmentally unfriendly and have short life-spans.

      Other methods:

      Forms of energy storage – Wikipedia

      • RobS

        They may be but if they have already been manufactured for an existing EV and you have a fully charged 60kwh battery which you only run down to 50kwh on your drive home with no intention to drive until the following morning then drawing a few kwh to take your evening demand from your car instead of the grid is just a cherry on top.

      • bink

        CitizenK, batteries at substation expensive? please provide example. Short life span, please provide example

      • Mint

        For the specific purpose of addressing the duck curve, they’ll probably be the best solution in the coming years. Several modern battery chemistries can have 10,000 cycle life if operated properly. And once we get down to around $150/kWh for raw cells (very soon), we’ll get <$1000/kW for ~5 hour storage. That's on par with gas turbine capacity and can be put anywhere, e.g. on-site at a baseload generator.

        Sure, for much longer term storage (24h+), batteries are unlikely to get cheap enough to fit the bill without a revolutionary discovery. But the topic of discussion is the duck curve.

        • eveee

          Backward. The amount of needed storage has dropped. The amount of annual peaks has dropped. The amount of daily peak hours has dropped. All that means less opportunity for storage, not more. Thats why pumped storage is losing out in Germany. You can’t buy and store cheap storage at night and sell it into cheap market during the day. The number of peak hours useful for making a profit for storage has dropped, not increased. Solar is an ideal peaker. It matches summertime air conditioning loads, the biggest load all year long, from coast to coast. You are making a problem that does not exist, then solving it.

          • Mint

            The amount of daily peak hours has dropped. All that means less opportunity for storage, not more.

            That’s not how opportunity for storage works.

            First of all, NG plants in good condition have zero chance of being replaced by storage, regardless of renewable penetration. So those are irrelevant.

            Storage becomes an option only when older peakers need investment for refurbishing. If they foresee enough of hours of demand, refurb will be worth it. If they’ll only get a few, then they’ll be shut down. Only when the latter happens will storage opportunity arise.

            Finally, battery storage only has a chance of being cost competitive if the peaks are short and frequent. That’s what the duck curve provides. A long peak increases capital cost of storage, and it loses out to peakers.

          • Bob_Wallace

            Gee, so the reports of batteries now being installed and taking business away from peaking plants are what?

            I guess we’d have to say false since you’ve declared that impossible.

            And I guess it’s also impossible for gas price to rise along with storage prices dropping….

          • Mint

            Where did I declare that impossible, smartass? Do you not see that two posts above I talk about batteries being a good solution to the duck curve?

            I don’t think you understood this reply. eveee said solar is REDUCING opportunity for storage. I was telling him how it makes storage MORE competitive.

            Those reports fit in precisely with my narrative. A few peakers used under certain circumstances get retired each year, and now that we have the duck curve, we only need 4-hour batteries for storage to work (making them more cost competitive).

            But reasonably priced storage can’t replace all functions of peakers, e.g. seasonal peaks lasting weeks, and good condition peakers won’t be replaced (by storage or anything else) for ages.


          • Bob_Wallace

            Avoid the name calling.

            It makes my delete finger itchy.

            Same for the all caps.

            Win the point with facts and skillful composition.

            Flow batteries with large tank volume, PuHS with large reservoirs and synthetic fuel made from renewable electricity can’t cover “seasonal peaks lasting weeks”?

            I agree that we will likely continue to use gas turbines for a long time. We’ve got them and as long as we use them only a few hours a year they will last until fusion is only 20 years away.

            And they’re so inexpensive to install that they might be the very deep storage solution for many decades. We could tolerate a tiny amount of fossil fuel to run deep backup, or we could run them with biogas/syngas.

            But we might find that as we add flow batteries (if that’s what we use) it might make more sense to add some more storage tanks and retire peakers rather than paying them capacity rates (either directly or through very high peaking prices).

            Which is going to have a higher ‘stand by’ cost – a gas turbine or a tank of cheap chemicals? A gas turbine or a pool of water?

          • Human 256

            Flow batteries have high W-h ratings but not high Watt ratings. That rapid ramp requires a high Watt rating.

            There is some doubt if combustion gas turbine peakers will be able to meet the ramping requirements. That is why the utilities are worried.

          • Human 256

            The head of the Duck is the Winter peak.

          • eveee

            Oh you are so confused. The head of the duck is the evening peak. It happens all the time, but during the summer, the air conditioning daytime load exceeds it. The whole duck curve thing is an artificial construct, subtracting the variable renewables (only) from the load curve. No other form of generation or load is scrutinized this way. So if some other form of generation varies, either suddenly or unexpectedly, or randomly, its not discussed. It really only applies to solar, because the curves are always daily. Unfortunately, it rarely includes wind. If it did, in places that near the coast, it would show a diurnal pattern that is opposite the solar pattern and balances it. All this means particularly little without reference to exact locations and patterns. It also doesn’t make much sense if the solar and or other generation sources are not designed to do anything in particular relative to the load. Such has been the case with existing renewables, because initially, and in legacy fashion, they have been used for fuel savings. Without any effort to make them provide ELCC, they don’t display much.
            The Duck curve is however a clever way for utilities to distract customers from the fact that solar lowers peak demand. Thats a benefit.

    • CitizenK

      PS, I agree about TOD billing though, employers should be encouraged to let employees charge their cars at work too, since that is when electricity peaks and demand wanes.

Back to Top ↑