Clean Power

Published on February 19th, 2016 | by Glenn Meyers


SolarCity To Deploy Tesla Energy Batteries For 52 MWh Of Evening Electricity Storage On Kaua’i, Hawaiʻi

February 19th, 2016 by  

According to Pacific Business News, Tesla Energy batteries will be used for a SolarCity solar farm and energy storage system which is being developed for Kauai Island Utility Cooperative (KIUC) in Kaua’i, Hawaiʻi.

The Tesla Energy batteries will supply a 52 MWh utility-scale energy storage system in order to help KIUC meet evening peak demand, which typically occurs between 5:00 pm and 10:00 pm. The storage system will be located at a 12 MW solar farm.

Tesla storage teslaenergy_utility1_750_422

Artist’s impression of Tesla’s utility-scale storage systems. Image via Tesla Motors.

Of its commitment to solar energy, KIUC president and CEO David Bissell has said, “No other utility in the US has a higher percentage of large-scale solar on its grid than KIUC.”

It is hoped the solar farm and storage capacity will help KIUC further reduce fossil fuel use. SolarCity said it would charge the utility 14.5 cents per kilowatt-hour for power from the batteries in a 20-year arrangement. News of the solar storage deal was reported in September 2015.

This project awaits approval from the state’s regulator, the Public Utilities’ Commission. SolarCity and KIUC asked last year for permission to accelerate development, so that it could begin by April this year, when existing ITC (investment tax credit) rates were due to drop.

As much as 95% of daytime demand on Kauai is met by solar at present but KIUC has targeted extending this number into the evening hours. When all is said and done, SolarCity’s grid-connected project will provide the 52 MWh utility-scale battery facility, built to distribute up to 13 MW of solar power.

“That a small co-op on Kaua‘i can become a world and national leader in energy transformation in such a brief time is something all of our members can be proud of and celebrate,” said Bissell.

Expect many others to be tracking the progress of this development.

Check out our new 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , , , , , ,

About the Author

is a writer, producer, and director. Meyers was editor and site director of Green Building Elements, a contributing writer for CleanTechnica, and is founder of Green Streets MediaTrain, a communications connection and eLearning hub. As an independent producer, he's been involved in the development, production and distribution of television and distance learning programs for both the education industry and corporate sector. He also is an avid gardener and loves sustainable innovation.

  • Anthony Gaona

    I feel the TESLA Powerwall changes the game and makes economic arguments against solar obsolete. Do you think there is one thing that Elon Musk and his boys haven’t thought of? We’re dealing with some pretty intelligent human beings here. Has anyone heard of Powur? Solar City has carefully selected Powur to be the category creating direct sales company representing SolarCity products and services nationwide. They will officially launch Powur when the TESLA battery is rolled out for residential use. Text me at 210-288-4266 for more info on Powur!

  • Eric Gold

    By the way, I am curious why Hawaii does not have a higher integration of wind power along with the PV. According to wikipedia very little additional wind has been added in the last 3 years.

    Perhaps naively, I have always presumed that a smart mix of wind and PV will lead to small storage requirements.

  • newnodm

    Solarcity got whacked in Hawaii when net metering was removed. I expect part of the low price is to move decidedly into commercial/utility.

    • Bristolboy

      Yes, I suspect as a “demonstration” type project they may be willing to accept a lower profit margin or even worse, to get into the market.

  • jburt56

    Quantity has a quality all its own.

  • vensonata

    Start your calculators, gentlemen. This number 14.5 cents kwh answers many questions and ends some extremely biased speculation. Solar City has given us the final answer to the question, “how much does solar with storage cost per kwh?” Now we need to figure out the formula that leads to such a figure. If solar by itself is selling for 12 cents kwh, then how on earth can the battery costs be added to that and come out with 2.5 cents kwh additional? No battery in the world can do that at present. So let us say that the PV is “really costing 8cents kwh. (PV prices in southern U.S. at utility scale can be as low as 6 cents) now we are left with 6.5 cents kwh for the battery. That is still amazing. Consider that somewhere in that 6.5 cents a profit of at least 1.5 cents needs to be made. So it seems they are estimating actual levelized cost of storage at 5 cents kwh. And that means that the Powerblock must have 5000 cycles over its lifetime. And apparently this is a 20 year agreement. It may mean Solar City expects those batteries to actually last 20 years or more. Now possibly they have built in a replacement at year 10 or 15, expecting batteries to have fallen drastically in price by that time, but I cannot determine this from the information given.

    • Ronald Brakels

      The sun can set as late as around 7:20 in Hawaii, so some of that 14.5 cent electricity from 5:00-10:00 pm could be coming directly from solar panels, particularly if they are west facing or tracking. So that could be helping keep their costs down.

      • vensonata

        This is for evee as well. I think they are saying that this 12 mw PV array exclusively powers this 52 mw battery. That is, all of the energy goes into the battery during the day, none of it goes directly to the grid. By evening you can continue to provide up to 13 mw to the grid for about 4 hours through the stored energy. This energy which is combined PV and battery costs 14.5 cents per kwh. That is also what Newnodm seems to be saying. The other two PV arrays are devoted to supplying energy directly to the grid as it generates it in real time. They charge about 12 cents kwh for this. But the mark up must be substantial.

        But Lord I could be wrong, because it just sounds too good to be true.

        • bink

          power is a commodity. capacity service is not. there is a capacity payment in this contract

        • bink

          a battery plus storage contract would have a energy component and a capacity (service) component to it. they don’t mix the two.

        • eveee

          Take a look at the GTM article from last year. Its a PPA. Than means the price is a purchased electricity price. And the article makes it clear that Kauai, which is already 95% solar during the day, is going for the night time load with storage. Game over for diesel.

          “In 2016, SolarCity will construct a 17-megawatt photovoltaic solar array on 50 acres adjacent to an existing power plant owned by Kauai Island Utility Cooperative (KIUC). The installation will include a 52-megawatt-hour battery system. SolarCity will sell power from the project to KIUC under a 20-year power-purchase agreement.”

          “Under the agreement, KIUC will purchase power for 14.5 cents per kilowatt-hour — considerably less than the utility currently pays for comparable diesel generator capacity.”

          Thats all you need to know. Is a PPA. Thats the price.

          Its not too good to be true. The solar electricity is much less than 10c/kwhr. We know utility solar in California is under $1.50/watt. PPAs are going for 4c/khwr in Austin Texas. Whats so hard to believe about that?

          Eight hours a day in Hawaii on many days, solar is already supplying 95% of daytime load. The evening load is smaller, and for a shorter period of time, perhaps 4 hours. So the blended price is one third solar plus storage, two thirds solar directly to load. At that rate, storage could be over 15c/kwhr and still get the PPA bid at 14.5c and turn a profit. The early AM load is negligible.

          • neroden

            See my earlier comment: I believe that with reasonable but optimistic estimates of the solar price (4c/kwh if unsubsidized, 3c/kwh if it’s getting a big federal subsidy), and pessimistic estimates of the percent of energy run through the battery, the implied cost of storage is either 21c/kwh or 23c/kwh.

            That’s a cap on the cost; it might be lower. This is significantly lower than Lazard’s estimates for “peaking plant” applications and significantly lower than Lazard’s estimates for “PV integration” applications. This is cheaper than zinc and competitive with onshore gas peaker plants.

            I suspect Lazard’s analysis is faulty due to its financing cost assumptions, which are way too high.

          • eveee

            Yes. There are some problems with valuation and identifying paybacks. Sonia Aggarwal has a nice paper about Market reform, for one.

            Lazards seems to have made an attempt to guesstimate some of those issues, but SC has clearly found a way to make value for storage. its a bit surprising that it is so low.

            There is a whole bunch of info from utility dive if you are interested. Some questions can be put to rest there.
            If memory serves, both the solar and storage benefited from a ITC.

        • eveee

          You may be right about it only supplying at night from storage.

          “”We’re really excited about [this project] because it’s effectively, as far as we know, the first utility-scale solar-plus-storage project done on a purely commercial basis that is affecting the shift of the vast portion of PV production to the nighttime peak,” Bob Rudd, SolarCity’s director of energy storage project development, told the audience at a recent storage conference. ”

          You tell me. Idk. The vast portion?

          . Kauai has a solar surplus. But that would be truly amazing. I mean, why add solar when its a surplus? That would mean the cost of solar from excess is so low, that the cost is almost entirely storage. Even in the article its not entirely clear how the intend to operate it. I mean, even if they said it would be operated one way, that doesn’t mean they could not change it. See what you can figure out from the article. Its pretty good.

    • eveee

      The levelized cost of solar is not 5c/kwhr. The system cost is skewed because the bulk of storage is only a few evening hours, while solar provides the largest load during the day. The early AM hours have negligible load. Since most of the larger daytime load is provided directly by solar, it dilutes the expense of storage in the calculation.
      If the rest of the US is any example, large scale utility solar is as low as 4c/kwhr, but probably at least as low as 6 to 8c. Daily cycled storage for a PowerWall is about 12c/kwhr, but PowerPack would be lower, probably only 8 to 10c/kwhr. Indeed, there is room for profit in there.

    • Otis11

      …ah, I believe there’s a fundamental misunderstanding here. Unless I’m mistaken, that 14.5 c/kWh is the price differential. They are charging the grid 14.5 c to take one kWh and place it back on the grid at a later time – aka, that does not include the cost of the electricity itself.

      Simply, the cost of storage is 14.5 c/kWh.

    • bink

      Power only. Read the article its says “power” and then references (2) other, solar arrays that KIUC owns having no battery storage that come in under the 0.145 cents price here.

      • vensonata

        See my recent post just above.

    • eveee

      Here is GTM on Hawaii solar plus storage for 24/7 power. Its essentially a power purchase agreement. And it seriously undercuts diesel.

      “In 2016, SolarCity will construct a 17-megawatt photovoltaic solar array on 50 acres adjacent to an existing power plant owned by Kauai Island Utility Cooperative (KIUC). The installation will include a 52-megawatt-hour battery system. SolarCity will sell power from the project to KIUC under a 20-year power-purchase agreement(PPA). A PPA covers the complete cost of power production.

      “The most interesting thing about this project is that it’s firm solar power,” said Peter Rive, founder and CTO of SolarCity. “That’s a new and important class of utility-scale solar power system. The batteries will store all the energy produced by the solar array, and then dispatch it to the grid as needed.”

      Under the agreement, KIUC will purchase power for 14.5 cents per kilowatt-hour — considerably less than the utility currently pays for comparable diesel generator capacity. Furthermore, KIUC noted that this is “only slightly more than the cost of energy from KIUC’s two existing 12-megawatt solar arrays, whose output is available only during the day.”

    • neroden

      I think you may be overestimating the “real” solar price. Hawaii is practically optimal for solar power. If utility scale PV in the southern US costs 6 cents/kwh, assume that in Hawaii it costs 5 cents/kwh (it might be as low as 4 cents).

      So then if we assume it all goes through the battery, the Powerblock battery costs between 9.5 cents/kwh and 10.5 cents/kwh. They expect the levelized cost to be 9 cents/kwh or below.

      Since it’s a 20 year agreement I’d assume they expect 20 years of service (anything else would be *highly speculative* and not investment-great) which means 7300 daily cycles or more.

      This is still implausibly low! Lazard’s LCOE was estimating levelized costs at utility scale to be 25 cents or more. 9 cents is way, WAY lower.

      Ronald is almost certainly correct that some of this electricity is coming directly from the solar panels, in the early evening.

      Sooooo, suppose that *half* of the electricity is direct from the solar panels (this is the maximum plausible amount given the 5-10 PM delivery times and the time of sunset) and half of it goes through the battery. The half direct from the solar panels costs 4 cents / kwh, and the half which goes through the batteries costs 18.5 cents / kwh (so the average comes out right). Then the batteries by themselves cost 14.5 cents / kwh.

      Which is *STILL* far far less than Lazard’s estimates, so it’s great news.

      Either SolarCity has lost money on this deal, or the batteries are way way way cheaper than Lazard’s estimates.

      Assuming SolarCity isn’t losing their shirt…. We know the batteries cost less than 14.5 cents / kWh, which is a record low. This is equivalent to $145/MWh, if you want to compare it to Lazard’s numbers. It’s cheaper than a diesel reciprocating engine, and cheaper than anything on Lazard’s entire Levelized Cost of Storage analysis.

      • vensonata

        That is a thoughtful analysis. It is the way I would have done it (in fact that is the way I do it in reality since I am off grid with solar and battery) however…The numbers 12mw and 52 mwh just seen to close to perfect symmetry to think there would be much juice left over to pump into the grid during the day. Notice that there are two other 12 mw arrays providing daytime power. So it would make sense that about 50% of that would be required from 5-10 pm. And now this is what I think is really happening. KIUC feeds the battery for free from their PV array and buys it back for 14.5 cents kwh. That means the actual cost to them is really say 5 cents for their contribution plus another 14.5 cents for the battery costs. Total 19-21 cents kwh. Still good, and we find out that Solar city can sell stored kwh for 14.5 cents and still make a profit. Which leads me to believe their actual final costs are about 11 cents kwh. This time I am more confident that I am closing in on how it really works.

        • neroden

          Thanks. I don’t quite understand what you’re suggesting for the business model. I was doing an ultra-pessimistic assessment just to get a sense of what’s going on. Even with a pessimistic assessment, they seem to be cheaper than the Lazard estimates, which is great. I look forward to more projects and the next Lazard report…

          • vensonata .

            I was very doubtful about Lazard’s familiarity with storage. It was a first report and I am glad they are paying attention but I think they realized that almost no one knows very much about batteries and battery makers frequently seemed to be involved in some kind of lying contest. I find it bizarre that this particular technology is so opaque to the consumer.

        • Eric Gold

          I never would have thought of that twist LOL
          It puts an upper bound cost at about 20 cents a kWh for battery sourced juice and say 5 cents a kWh for direct sourcing.

          If direct and battery sourced are in 1:1 consumption ratios the wholesale cost averages 12.5 cents a kWh.
          If direct and battery sourced are 1:2 then the average wholesale cost is 10 cents a kWh.

          Goodbye diesel, and good riddance!

          By the way, pvwatts estimates solar collection for Kauii at 1.5 kWh/watt*a. Not that great, I presume due to clouds.

      • Peter Voight

        Something doesn’t add up. Income is very low.

        52MW*$0.145 = $7,500/day.

        Cost may look nearly plausible if everything is levelised over 24hrs, but hourly income is small.

        Ignoring solar costs, storage costs

        Powerpack is said to be $250/kwh

        Total = $13million.
        To do that, assemble at least 6million cells into 520 enclosures. Repeat all enclosure and related hardware costs 520 times.
        Efficiency will be at best 80%.

        Power output is only 13MW, so can’t earn faster than $1,185/hour
        It would be better to spend the capital on Subway franchises.

        • eveee

          Low income? Well $7,500 a day doesn’t sound like much. Annually, thats 2.737 million. If the initial storage cost really is 13 million, without interest, it pays back in 4.75 years. Thats assuming the solar energy costs nothing, which is not true. But the energy cost could be as low as 4c/kwhr with the ITC. We need to be careful here to distinguish cost/kwhr/cycle we are calculating vs initial storage capacity in cost/kwhr.

          Why would efficiency be at best 80%? Inverters are above 95%, and battery round trip is 0.92. That all assumes efficiencies at maximum power and current. But why would anyone do that while grid tied? There is no good reason since its programmable.

          For the battery, efficiency is VI/(V-IR)I or 1/(1-I2R)

          Pardon the 2 not be superscript for squared.

          Thats one way. Round trip squares that yet again with the efficiency dominating term becoming I4R, a very steep function. Even at 0.707 max charge/discharge, loss falls to one fourth of that, making efficiency about 98%.

          • vensonata

            Agreed with all you said above. However, take note of my recent revelation just above. Kuai feeds their PV into the battery and pays 14.5 cents kwh to take it out. The feed in is not counted in the 14.5 cent number. Their total cost is PV at say 5 cents plus the 14.5 cents they pay to Solarcity and the final number is 19.5 cents for solar and storage combined. They sell that to the customer at probably about 22 cents kwh or more from 5pm to 11 pm. That, I think, is how it works.

          • vensonata

            Kuai rates are actually 34 cents kwh. so that utility makes a nice return on their 19.5 cents kwh costs.

          • Eric Gold

            The utility website residential price sheet says $10 a month for a connection charge and 29 cents a kWh

          • eveee

            I am not sure that the feed in is not counted in the 14.5c number. I just don’t know. If you can find out for sure, please let me know. Its really hard to get a super clear, unambiguous take on this. One might even have to go to the PPA. It looks like the utility dive article is saying all of the solar goes into the battery and its a PPA. How on earth could Tesla possibly do a PPA, which is a sale price for electricity, don’t care how you do it, if it doesn’t take into account both solar and storage? I don’t know. It looks like they sell some solar in the day, and some at night. They are talking about it being on demand. That sure sounds like 24/7 to me. I read that they have certain minimum demand required in the PPA deal.
            I could buy your numbers, too. They make sense. I wish I had better verification.

          • vensonata .

            I have a gut feeling that I got it right. I would prefer to find out I was wrong and that the price is combined PV and storage, but I fear that 14.5 cents sounds like the real number just for the battery. However, personally I would jump at that number compared to most lead acid. From my point of view, all of this sounds cheap as heck.

          • vensonata

            And now I think I am wrong again! I now think eveee is right and the price is PV and battery combined…for 14.5 cents. I realized that it is SolarCity that owns the Panels and battery. Formerly I though KUIC owned the Panels and solar city owned the battery. So I am delighted to be wrong (possibly) once again and that battery prices are much better than I had formerly stated.

          • Eric Gold

            I can show you the arithmetic if you are interested, but it works out to all of the 12MW PV production from the day goes into the 52 MWh battery for later use when the sun is down.

          • eveee

            Go for it, thanks. Thats what I read, but that calculation would buttress it. The energy has to come from somewhere. What puzzles me is that it could come from any of the solar powering 95% of Kauai’s load. Perhaps, SC packaged the deal as a power solution. It had to be a business model so it makes sense that they controlled the whole solution.
            I mean they did package it as a PPA, and thats the traditional way to do that, as an generation source, but this time with a twist. If all this turns out real, its a revolutionary step.

          • Eric Gold

            Why is my cat hanging out at your place ?

            Ah.. back to the topic: I first want to clarify that I am only saying that it takes a day of average Hawaiian sunshine to fill up a 52 MWh battery from a 12 MW array.

            PVWatts reports that a watt of installed PV collect 1.5 kWh of energy a year, so 4.109 Wh a day. 12 watts then collects
            12*4.109 = 49 Wh a day. 12 MW then collect 49 MWh a day, or a few percent less than the size of the battery.

            As aside #1, the Kaui utility reports annual purchases of the different fuels each year. I fail to reconcile those numbers with 95% PV contribution during the day.

            Aside #2, I *suspect* that we are missing the important money question here, insofar as the larger question is how to run a large island on almost all clean energy. I posit that most days are handled by a smart mixture of wind+PV+hydro+battery, and only a small number of days require fossil fuel backup. So how does a company offer a profitable back-up service that is only used say 1-5% of the days of the year ? What is a low fixed cost, low capitalization backup ?

          • eveee

            Nice calculations. I like my kitty. #1 is interesting. Should be watched.
            #2 Methinks its not 1-5% days of the year, but rather most evenings. If this facility can chip away at expensive evening demand, it winds. Judging from the prices cited for diesel vs solar plus battery, its a winner. On Kauai, amazingly, solar plus storage seems poised to kick out the remaining FF, or at least such a large percentage, that GG emissions will drop dramatically and costs too. A win win.

          • Peter Voight

            Yes, but it’s $7.500 day or $315/hr. That’s the revenue, not income or profit. A large and long term investment to earn pocket change.

            If only it were as simple as I^2*R
            80% is a good figure, even if MPPT losses are ignored.
            Charging and discharging are separate cycles or mixed.

            It’s not like a small home system. Distribution and control,
            cabling and cooling. It’s not so easy to dismiss restive losses, because low resistance conductors cost money. And it’s not all DC so power factor correction and all that goes with it,

            If the maximum output is not used, then power can’t be delivered in the required time, so earnings would fall again.

          • neroden

            For future reference the way to notate superscripts for squaring is x^2

          • eveee

            Thanks. Someone showed a nice link to all of them. Cant remember where.
            So I 2 R . Ausgezeichnet

  • Dona Paula

    I was trying to calculate the total revenues for the BESS alone considering they are charging 14.5 cents per kilowatt-hour (Solar + BESS). If the cost of BESS portion is around 9 cents/unit, 52 MWh, 80% DOD, 1 cycle a day, 20 years and 250 days of sunshine in hawaii. Total units generated 208,000,000. Total revenue 9 X total units = approx. 19 million Dollars.
    So 19 mill/52,000 units = $365/Kwh for LCOE. I am assuming that this price includes maintenance (corrective and preventive) for 20 years and includes the PCS and BMS.

    • eveee

      How do you separate out the revenues for storage? How do you know how much energy is provided by solar directly to load, and then how much energy is provided by storage? I am not sure what you mean by unit. Also, how much of that 52MWhr is used daily? The cost of Tesla PowerPack has been stated at $250/kwhr.

      • bink

        This is load shifting application. All the energy production is being shifted to the 5 p.m. – 10 p.m. demand period

        • eveee

          What does that have to do with your false comment about Tesla batteries? Tesla has two battery types. Learn to read yourself. Load shifting is a daily occurrence. The daily cycle battery chemistry is different from the weekly.

          • bink

            get a life. I recognize you are ignorant as hell so no responding to you anymore. And this is a specific grid application not a hypothetical scenario

  • onesecond

    I want to put pv panels on top of that storage farm.

  • Harry Johnson

    If 95% of daytime demand is met by solar, then will more PV be added for evening use? This is an excellent test case for the US market and all the islands and remote communities around the world.
    SolarCity should set up a turnkey program that allows these communities to get off expensive imported dirty diesel power.

    • Bryan

      The rest of the US market doesn’t have Hawaii’s high diesel fuel powered rates so it won’t fly anytime soon in the rest of the US.

      • Bristolboy

        It might be sooner than you think, once it is proven to work in Hawaii there will be lower risks elsewhere.

        It is also not that much more expensive than evening wholesale prices in places like Italy, Greece etc where costs can be around $100/MWh.

      • Damon Wright

        I would think that other island nations and remote communities currently relying a lot on diesel would be comparable, even if the rest of the US market isn’t.

        Much of Micronesia will probably follow in Hawaii’s footsteps as soon as this is proven to work. Hawaii is a good guinea pig for them.

        • Bryan

          Hawaiians have the income to support such an investment. There are few island nations whose inhabitants can afford this expensive technology.

          • blueseasons

            How long have you lived here? I don’t think you know anything about incomes over here. And please don’t lump all “Hawaiians” as you call them into one category. The other island, other than Kauai have much lower electricity rates. It’s not likely this project could even work out financially on Oahu, Maui or Big Island.

  • Mary Grikas

    Is the 14.5 cents for the storage, or does it include generation? In other words, does KIUC pay first for the PV solar generation, then an additional 14.5 cents for the power from the batteries?

    • newnodm

      Both solar and storage for 14.5

      • eveee

        Yes. More info.

        “The cost of energy from the facility sold by SolarCity will undercut “the current costs of conventional generation”, KIUC said, continuing that it will be “only slightly more” expensive than the daytime electricity generated by the co-operative’s two existing 12MW utility-scale PV plants.

        “KIUC has been investigating energy storage options for more than two years and price has always been the biggest challenge. This is a breakthrough project on technology and on price that enables us to move solar energy to the peak demand hours in the evening and reduce the amount of fossil fuel we’re using,” KIUC’s president and CEO David Bissell said.”

        • Frank

          I remember reading that back in the good old days when oil prices were high, electricity costs ranged between 38 and 46 cents a kwh depending on island. I think the average last year was 28 cents for all of Hawaii. Now that’s retail, but this still sounds like a winner.

          • newnodm

            But even with costly electricity and gasoline, the average Hawaiians energy bill is not great. The have very little HVAC, and their low latitude means good sunlight year round. (Except when I visit Kauai, and then it rains for a week.)

          • Sreehari Variar

            lol, you should go to california sometime. They will be very thankful if you can make it rain.

          • blueseasons

            Sorry. I have lived here for 24 years. KIUC has the highest rates in the world still, and we have very high electric bills. Average consumption in kWh is low due to a lack of a need for heating. Less than 5% of single family homes have air conditioning, so for the majority, that’s not a big expense either. If you believe paying a hundred dollars a month for lights, entertainment and cooking is “not great” , you are mistaken.And on top of that you have a healthy propane bill just to heat water. If you have electric water heating, your consumption is much higher and so is your bill.

          • blueseasons

            The other islands that have a power grid have lower cost electricity than Kauai. Kauai is still .48 a kWh and that’s not going to change. Electricity independent of fossil fuel and inexpensive electricity are two different things. KIUC has concentrated on the former.

        • Otis11

          Mmm… I’m not convinced that’s storage and solar… at 14.5 c/kWh it could still buy electricity during mid-day and undercut evening peaker plants. That, and the cost of storage is unlikely to be 9.5 c/kWh even after a 30% tax credit within the next few months.

          Possible, yes, but unlikely (IMHO)

          Edit: However, if I’m wrong, that’s terrible news for peaker-plants world-wide, good for baseload plants, and very good for wind/solar combo (could profitably cycle twice/day to meet morning peak with wind and evening peak with solar to lower effective capital cost of the batteries – as well as boost the value of energy during solar/wind production periods)

          • bink

            Its not, its only for the energy or power. The other (2) projects that were referenced are solar only with no battery and their costs come in slightly under the Solar City install for the solar array.

            especially when the article above says “power”

          • Bob_Wallace

            If it’s any help, EOS Energy Systems has said that they could purchase low priced, off grid electricity from the grid and sell it back for about 12 cents per kWh. That’s an all-in price.

          • Otis11

            Mmm… is that today? And that claim also isn’t terribly clear – it doesn’t necessarily state that the 12c includes the cost of the power bought. (The key word to look for is ‘at’ – Them saying they are selling it ‘at 12c/kwh’. Saying they can sell it ‘for’ 12c/kwh technically has the legal loophole that they can later claim their service is 12c/kwh excluding the cost of electricity. English is terrible that way…)

            Also, even if they are able to do that, they could be ‘purchasing’ power at near-zero prices. That’s simply not going to happen (profitably) with solar – the only reason things hit near-zero prices are when there’s a production credit (wind) or a continuity advantage (Large thermal/nuclear).

            I would believe storage costs in the 12-15c/kwh range. Possibly even getting down to 10c/kwh. But getting 10c/kwh storage and 4.5c/kwh solar would be a new record for both categories. Unlikely scenario at best…

          • Bob_Wallace

            Their web site has changed from the one I last looked at. On it they included all costs including the purchase of electricity.

            I think solar will easily go below 4.5c/kWh. Three cents or less is where I think it’s heading. Same for onshore wind. Wind is almost there (without subsidies).

          • Otis11

            Oh sure… I wouldn’t be surprised to see wind get to 1.8c/kwh and solar down below 2.5c/kwh at point of production… (Those are likely on the low side, but not unattainable.)

            The problem is I don’t see that happening today. Even 4.5c/kwh would be a record today. As would 10c/kwh storage. These will undoubtably happen (after correcting for inflation – and probably even before the correction), and likely soon, but I just don’t see it as an immediate thing that anyone sane would be advertising or writing in contracts at this point…

          • bink

            EOS is a different system made for capacity shifting lithium is a lot more expensive. Look at that rendering of the Tesla install that is a very expensive install with HVAC even if you had 100 kWh dc cost at 1 cycle per day that, that works.

          • eveee

            Read the GTM article then, a second source. It makes it crystal clear that the project provides power after dark and that the project is being paid for as a power purchase agreement at 14.5c/kwhr. That means its supplying power during the day at solar rates for kwhrs, not just at night. The effective cost of kwhrs at night could be greater than 14.5c/kwhr solar plus storage. In other words, storage is only a small part of the power delivered. Most is direct solar.


            ““The most interesting thing about this project is that it’s firm solar power,” said Peter Rive, founder and CTO of SolarCity. “That’s a new and important class of utility-scale solar power system. The batteries will store all the energy produced by the solar array, and then dispatch it to the grid as needed.”

            Under the agreement, KIUC will purchase power for 14.5 cents per kilowatt-hour — considerably less than the utility currently pays for comparable diesel generator capacity.”

            Somehow there is a disconnect on the description of the solar, because the GTM article says its a 17MW solar array. See what you think.


          • Otis11

            Golly… so much misinformation about this out there. Hard to tell what’s correct.

            So, from my best assessment, they’re installation 17MW of solar, 52 MWh of battery and the battery charge/discharge rate of 13MW. That means they will be selling some of that power at time of generation as they simply can’t store it all at the rate produced – so the average cost will be 14.5c/kWh where about 60% goes solar + battery while 40% is sold straight solar.

            Am I getting close? Haha.
            (Also a chance I completely misunderstood or got terrible information from the various sources… but these seem to be the most often quoted numbers)

            Regardless though – it does seem to suggest that the price does include the cost of storage and energy. If true, I guess it’s not the first time Tesla has shocked me… (I might need to recalibrate my expectations…)

          • eveee

            Yes. Thats the idea. One idea. Those are my prognostications, educated guesses, etc. But they are guided by the simple notion that much of the energy provided is direct solar. A fraction is from storage. And I don’t know any more than the articles say. In fact, if there is too much solar, the battery charging costs are zilch.

            Here is one from utility dive, usually a good source.

            From the horses mouth so to speak.

            “The power generated by the PV cells will be used solely to charge a 13 MW battery array capable of providing 52 MWh to customers of Kauai Island Utility Cooperative (KIUC), the island’s sole electricity provider. That means the solar cells will charge the batteries during the height of the day, and the batteries will discharge the stored power to customers during the evening peak between 5 p.m. and 10 p.m.”

            “Anyone that’s been out to Kauai will notice that they have a lot of solar on the island and really don’t have any appetite at all for solar at midday,” Rudd said. “If anything, they were already in a bit of a curtailment state during certain days. So, they love solar, they want more because it’s cheaper than what they otherwise would realize, but they don’t need it during the day.”

            They put a different blush on the rose. To understand what was done, one might have to read the PPA itself. Kauai already has penny of solar, so the solar source is cheap. Utility dive makes it sound like the problem SC solved was excess daytime solar(Kauai is already at 95% solar on sunny days) being used to charge the batteries during the day and discharge at night. One problem solved. But Kauai is also wanting to replace unreliable lead acid units.

            Interestingly, there is some hydro in there.



          • Otis11

            Yeah, they way they word it, it’s still not apparent that they are contracting the total power average of the solar+storage at 14.5 c/kwh… but that’s my best guess at understanding it.

            BTW – them filling up when energy is ‘almost zero’ during the middle of the day is shooting themselves in the foot since they are also producers… They have to cover the average cost of solar and the average cost of storage no matter where the price goes.

          • eveee

            Yes. So they are well motivated to sell power in early evening if there are higher rates at night. I think there are some business reasons why the details are not clear. But there are some practical ones, too. Demand varies. Notice the contract stipulates a minimum purchase.

        • bink

          the other two have no storage so the price is only for the power (or) energy

          And I quote:

          KIUC said, continuing that it will be “only slightly more” expensive than the daytime electricity generated by the co-operative’s two existing 12MW utility-scale PV plants.

      • Mary Grikas

        Thanks for clarifying. That’s great!

      • Bryan

        I wonder if this is why solar city has been posting losses every quarter lately ?

        • Bob_Wallace

          Have you looked at a financial statement from them?

          I wonder if they might be losing money in the same way Tesla is. Investing in growth.

          • neroden

            Solar City is a *financing* company.

            Yes, they are investing in growth. But I’m not sure that’s it.

            How much money SolarCity makes is mostly about the difference between the effective interest rate paid by the customers, and the rate which SolarCity has to pay on its “green bonds”.

            At the moment, SolarCity’s interest rate paid out for those “green bonds” is quite high. I’m not sure how big the spread is, but it may not be very large.

            I have trouble analyzing finance companies which is why I’ve stayed away from SolarCity. Manufacturing is easier for me to analyze.

      • bink

        You are incorrect. go back up top and read the article. The power (or) energy is 0.14.5 cents kWh.

        The article references 2 other projects that have no battery storage attached to them that come in at a slightly lower cost of energy than on this project

        • newnodm

          No, I’m not incorrect

          • bink

            Ummm, yes you are and it is obvious that you are incorrect because you don’t provide numbers or a methodology for a base calculation.

            The article stated 0.145 cents for power (that’s it in a nutshell) it said nothing about a capacity payment.

            The difference is energy is a commodity and capacity is (a service) not.

            I can guarantee the contract had a capacity payment clause, one which they were not willing to publicize

          • newnodm

            I emailed the media relation person at KIUC (which is a co-op/nonprofit) and ask if there were capacity payments in addition to the PPA. Jim Kelly of KIUC responded “No”.

            Your turn.

          • bink

            cut and paste the email and post it. this is peak power there is no way you are going to tell me they left that much money on the table. avg US peak price is 0.145 cents. Hawaii is three to four times that

          • newnodm

            You are welcome to email KIUC yourself. I got a response in about ten minutes on a sunday.

  • Ross

    Will the farm have all its power sent to the batteries for release during the evening peak?

    • newnodm


  • Bristolboy

    The price for delivering this power is much less than I expected, and probably compares very well to peaker plants on such an island. I assume this is so cheap due it’s size, the low-risk US investment climate and presumably high irradiance. That said, where such islands lead the rest of the world will follow so it will be good to see how long it is before such technology breaks to larger markets.

    • Bryan

      The rest of the world won’t follow because the rest of the world’s power isn’t as expensive as hawaii’s power.

      • eveee

        Most of the world won’t just yet. Except for Australia. And Germany perhaps. There is rural India, too.

        • Ronald Brakels

          We won’t follow with on-grid storage in Australia. There is not enough difference between the average high and average low wholesale electricity prive over 24 hours to pay for it currently. However, we could definitely install home and business energy storage here as the opportunity for arbitage between the low solar feed-in tariff and the high retail electricity price here is consistantly high. Unfortunately this will take longer than many hoped as recent excitment about low cost home energy storage has turned out to just be hype.

          • eveee

            But what if solar plus storage was offered to micro grids at 14.5c/kwhr? Wouldn’t that beat the pants off grid electricity in Australia? I read Australian developers are considering that idea, even encouraged by the utility. Except for the utility encouragement part, that seems to be what is happening in Hawaii.

          • Ronald Brakels

            Sorry, I should have been clearer. Storage in Australia will go before the meter. That is in a home or business or micro grid before the meter that connects it to the grid. New rooftop solar solar with a 5.3 centaroo feed-in tariff I would save about 23.5 centaroos or 16 US cents per kilowatt-hour of stored electricity in my state. So storage at 14.5 US cents a kilowatt-hour will pay for itself. Unfortunately we are not there yet since the Powerwall has turned out to be a bit pony. .

          • neroden

            “a bit pony”? As in vaporware? The Powerwall will be available at cheap prices in a couple of years, but yeah, it’s not *really* available now, thanks to low production rates.

          • Peter Voight

            I don’t know about ‘vapourware’ but the battery is sold as part of 5kWp solar bundles costing around AUD16,000. The same without the battery can be bought for 5,500-6,500, putting the battery+inverter at $10,000.
            The consummer magazine Choice, calculated payback of 20 years, using rates similar to what I pay.

            Specs are not what was expected. 6.4kWh, 60% retention at 10 years etc.

            There isn’t enough to be saved, except for those who are paying exceptional tariffs. But, plans can include natural gas, so don’t expect the lowest price for both. Solar ownership is still well in the minority.

            I don’t have solar. I pay 0.23c/kwh peak, 0.11 off peak.
            Typical bills can be found at Ausgrid’s site.

            I doubt that batteries will become much cheaper, and certainly not soon. The Powerwall weighs 95kg, perhaps there are10kWhr of cells, so 45kg. That leaves 50kg of supporting materials. If prices fall, they will not be in proportion to cell price. At least not for Tesla.

          • eveee

            Do you have a source for the 60% retention in 10 years?
            Its not in any official document. I do find it in a disreputable source, tho.

            Many DC-AC inverters have higher efficiency. Like 97.5%.

            The specs really have not changed. They are stated in a different way. The output is now given minus the efficiency.

            7kwhr x 0.92 = 6.44kWhr

            Before it was stated as the energy storage and efficiency separately.

            Also, the efficiency is at the rated power of 3.3kW, not 2kW
            Since the efficiency is round trip, the loss is squared if both charge and discharge rates are reduced. There really is no reason to charge or fully discharge the storage quickly and at full rate when connected to the grid. The battery efficiency is controlled by I squared R. Another non linear expression. That means at as little reduction as half power, the loss falls off rapidly.

            I read the article. The paybacks are based on assumptions that change the amounts substantially. This abc source shows one situation that changes the payback to about 10 years. If its not fully used, you don’t get payback as fast. No surprise there.


            Its unlikely 50kg of supporting materials unless you mean not structurally. There is a DC-DC converter, BMS, and cooling system.

          • Peter Voight

            Yes, I know, there are all sorts of claims and numbers, but all figures I have stated come from Tesla’s documents.
            The retention figures are in the warranty published 23-Jan-2016.
            at least 85% at 2 years or 2MWh aggregate discharge
            at least 72% at 5 years or 9MWh
            at least 60% at 10 years or 19MWh
            60% is 3.84kWh
            Yes, efficiency is the discharge efficiency, so 7kwh means 6.4kwh useable output. DC to AC at 97.5% would bring 92.5% to 90%. Useable output to load is 5.76kWh

            There is a DC/Dc converter in the path from the 400V bus to the cells. More information is needed to find round trip, but for sure, input is greater than 7kWh, so DC bus round trip will be less than 92.5%.
            Efficiency is stated at 2kw. From the operators’ manual of your link
            Round-trip efficiency;
            “Values provided for 25 °C, 2 kW charge/discharge power, 400 V to 450 V DC bus.”
            Same as the retention warranty, 2kW.
            There is also a separate spec sheet that says the same.

            The power output was raised as a result of remarks about not being able to run a kettle and such. I expect that 3.3KW was seen as being more useful.

            For the same output voltage, current at 3.3kW will be 1.65 times that at 2kW. That is quite a difference.
            Assuming simple resistance, I^2R losses at 3.3kW will increase by 2.72 times over 2kW.

            The ABC/Choice article does show payback less than 20 years, but only if all solar energy is used. The battery can output only 6.4kWh. Equivalent grid cost is 6.4kWh*$0.218 = $1.40/day.
            Payback time is >20 years for the battery. That’s the cost and return for adding the battery to solar.

            That’s right. 50kg of non-cell materials. If the battery’s design demands cooling fans, heat exchangers, whatever, they are costs that do not scale with cell price.

          • eveee

            Not sure what you mean by usable output to load. According to the given numbers. 0.9 x 7kwhr = 6.3kwhr.

            Why is more information for round trip needed? Its given. 92%. So no, I don’t see how its less. Its given for a system including the DC-DC, not for the battery pack itself. Otherwise there would have to be a separate citation for the DC-DC efficiency.

            The warranty figures look poor. But its a warranty, not the actual performance. I could say the warranty for solar panels is 10 years. But they last 25 years at expected degradation rates. So it goes. The warranty on some cars is 50 or 60,000 miles. It doesn’t really tell much except thats what the warranty is. Same for the 2kW rating on the warranty. I don’t like it, either, but there it is. It should be higher for all of them.

            Did you miss the part at the end of the ABC article that says it could be nearer 10 year payback?

            My point about the other materials in the ESS is they are not all structural. And I don’t know the weight of the cells. Even if you look at the weight of a Tesla car pack, its not the weight of the cells. Thats where you need to start. I doubt the weight has much to do with cost. And you are right, those costs don’t necessarily scale the same as the battery. But I doubt they have much cost content compared to the battery. In typical battery packs, only about 5% or so might be BMS, etc. DC-DC converters are cheap. A DC-AC converter could be more expensive and drops in cost precipitously as kw capacity rises.

          • Peter Voight

            Figures are given by Tesla at the 400V bus. DC to AC add losses.
            If you say the battery is 7kW, but 92.5% is 6.4Kw, then input additional to 7kWn , must be provided during charging.
            Overall less than the stated 92.5%.

            The warranty is what it is.

            I did not miss the ABC calculation. Battery adds the storage feature that earns $1.40/day. To have that feature costs $10000, so >20 years.payback.

            The home powerwall weight 95kg. But, there are around 45kg or cells ( each 45g or so) totaling 45kg. The remainder (50kg) is what is needed to make those cells into a battery.

          • Ronald Brakels

            No… no… pony doesn’t mean vaporware. It’s short for pony and trap. And as a result, a certain company’s reputation has gone a bit pear shaped here.

            Peter Voight has exhaustively covered why we are feeling disillusioned in his comments.

      • Bristolboy

        I wasn’t implying it will happen straight away, but in time it may. Island nations (like Hawaii) were the first where normal solar and wind were economic – now such areas are spreading around the world.

Back to Top ↑