California public utilities vote no on energy storage

In 2002 California set itself an ambitious renewable energy target – a third of its electricity from renewables by 2020 – and according to Figure 1 of the California Public Utilities Commission second quarter 2014 report it’s going to meet it:

Whether it will meet it, however, remains an open question because the more intermittent renewable energy California adds the more difficult it becomes to integrate it with the grid. This problem in fact became evident fairly early on, and in an attempt to solve it California in 2010 passed legislation (AB 2514) to encourage its publicly owned utilities to install energy storage – batteries, thermal, flywheels, CAES, pumped hydro (not exceeding 50MW), whatever worked – requesting them to develop viable and cost-effective plans and submit end-2016 energy storage capacity targets by October 1, 2014. This deadline has now passed, the ballots are in and the submissions from 29 of 31 California publicly owned utilities (two of the links provided don’t work) have been published here (h/t Mark Miller for the link).

And how much energy storage did the 29 utilities commit to?

27.6 MW, equal to roughly 0.1% of their combined peak load. Enough to keep the lights on for maybe a few seconds:

  • Two utilities (Los Angeles Department of Water and Power and Redding Public Utilities) committed to new storage projects totaling 27.6 MW.
  • Two utilities (Riverside Public Utilities and Vernon Gas and Electric) set targets of zero MW.
  • One utility (Glendale Water and Power) adopted its existing 1.5 MW of installed thermal storage as its target.
  • The remaining 24 utilities declined to set a target, almost all on the grounds that energy storage is cost-ineffective and/or technologically immature.

Most of the submissions consist of copies of Board Resolutions with little or no technical backup. The one from Burbank Water and Power is typical:

A target for Burbank Water and Power to procure energy storage is not appropriate at this time due to lack of fully developed, cost-effective energy storage opportunities.

Although others are a little more specific on costs, such as Roseville Electric:

Staff concluded current storage technologies are 2 to 10 times higher cost than energy from Roseville’s existing portfolio of resources.

and Turlock Irrigation District:

The results of the staff study concluded that energy storage systems are currently not cost effective and in most cases increased cost by millions each year.

Sacramento Municipal Utility District, however, goes into detail. Here’s an excerpt from SMUD’s submission:

Since 2008, SMUD has invested over $30 million dollars in internally and externally funded research to understand and prepare SMUD and its customers for eventual deployment and utilization of energy storage. Staff has been conducting various field demonstrations, studies, and assessments of different storage technologies, used for different applications ranging from transmission scale to distribution scale to customer scale systems. On technical issues, this body of work has assessed technology performance including such factors as efficiency, reliability, and durability. On economic issues, this body of work has assessed capital costs, installation costs, operation costs, value, and cost effectiveness. Additionally through this body of work, staff has assessed grid integration issues and strategies for interconnecting, aggregating, visualizing and controlling storage systems from grid planning and operations perspectives. Based upon this body of research, staff finds the storage applications examined are not cost effective at this time, with the exception of large scale pumped hydro storage.

Which helps SMUD not at all because large-scale pumped hydro is the one storage option California doesn’t allow.

So here we have a near-unanimous vote of no confidence in energy storage from utility professionals whose job it is to supply reliable power to consumers and who understand the realities of the electricity market. From it we can conclude:

    1.  That California is not going to get any meaningful amount of energy storage capacity before 2020.
    2.  That in all likelihood no one else is going to get any either. The economics just aren’t there (pumped hydro excluded).
    3.  That lack of energy storage capacity will continue to limit the grid penetration of non-dispatchable renewable generation for the foreseeable future.

Key excerpts from the submissions of all 29 utilities are listed below for reference. I’ve left the “Whereases”, “Now Therefores” and “Be It Resolveds” out to improve readability:

Alameda Power & Water: (finds) that energy storage systems are not currently viable or cost-effective for Alameda Municipal Power and (recommends) that procurement of energy storage systems be deferred until further justified.

City of Anaheim: ES technologies are relatively new and expensive not to mention vary in maturation and … are space intensive and difficult to site in an urban setting such as Anaheim and … do not assist Anaheim with reducing peak demand as Anaheim is already resourced to meet such demands.

Azusa Light and Water: (C)ommercially available energy storage technologies, although technically feasible, are not cost-effective at present.

(City of Banning – inoperative link.)

Biggs Municipal Utilities: (C)ommercially available energy storage systems are not currently viable and cost effective for the city at this time, and the City is not adopting procurement targets at this time.

Burbank Water & Power: (A) target for Burbank Water and Power (BWP) to procure energy storage is not appropriate at this time due to lack of fully developed, cost-effective energy storage opportunities.

City of Cerritos: The application of utility-owned and operated energy storage technology to serve the City’s electric utility customers over the next three years is more costly than the value of benefits.

City of Colton: Energy storage systems at this time are not economically viable due to their high cost.

Glendale Water & Power: It is recommended that the City Council approve Glendale Water & Power’s current installed energy storage capacity of 1.5 MW as GWP’s energy storage procurement target for the purpose of compliance with California Assembly Bill No. 2514 (2010).

(Healdsburg Electric Department: inoperative link.)

Lodi Electric Utility: has determined that the economics behind energy storage deem it not cost-effective at this time.

City of Lompoc: it is not cost-effective for the City to develop ES procurement targets at this time.

Merced Irrigation District: (S)ince MID does not purchase renewable energy… does not have direct access to, nor does it produce renewable energy for use in its electric retail system separate and apart from the power it purchases …. is not aware of any cost-effective technologies and/or applications that have been identified for the District’s operations, and since energy storage systems are generally intended to be tied to a renewable generation source, there is no purpose for implementing a procurement target for energy storage systems for use within MID’s electric retail service at this time.

Los Angeles Department of Water & Power: There are two projects that are deemed eligible energy storage systems namely, a generation connected storage with a net incremental capacity of 21 MW and an incentivized customer connected storage with a raised peak demand shift of 3MW, and LADWP will primarily rely on these two projects to fulfill its 2016 procurement targets totaling 24MW.

Modesto Irrigation District: has not identified reliability or operational needs that could be met only with energy storage and that would require the adoption of mandatory energy storage procurement targets …. (and therefore adopts) a policy that it is currently not appropriate for the District to adopt formal energy storage procurement targets.

City of Needles: determines that establishing a target for the City of Needles and the Needles Public Utility Authority to procure energy storage systems is not appropriate due to the absence of a clear and present need for energy storage systems.

City of Palo Alto: (A) target for the city of Palo Alto utilities to procure energy storage systems is not appropriate due to lack of cost-effective options.

Pasadena Water & Power: It is not appropriate at this time to establish procurement targets for energy storage systems to be procured by Pasadena Water and Power due to lack of cost-effective viable options.

Pittsburg Power Company (Island Energy): Staff recommends that the Board adopt no energy storage systems for Island Energy since energy storage systems are not feasible or cost-effective for Island Energy’s current operation.

Port of Oakland: Energy storage systems are not viable nor cost effective for the Port at this time.

City of Rancho Cucamonga: The City decline(s) establishing a procurement target for energy storage pursuant to AB 2514 due to the lack of cost- effective energy storage options

Redding Electric Utility: While not legislatively mandated to do so, in June 2012, the Council approved the expansion of REU’s (Thermal Expansion Storage) Program. The TES Program expansion provides for approximately 2 MW of PLS that would be in addition to REU’s existing TES/PLS already procured and installed from 2005 through May 2012 (1.3 MW). In addition, the Program expansion includes contract provisions for 2016-2017 allowing the City to procure additional TES/PLS systems up to 0.8 MW should REU operating conditions warrant more TES/PLS capability. In summary, REU’s energy storage targets for 2016 and 2020 are 3.6 MW and 4.4 MW respectively.

Riverside Public Utilities: Recommend(s) that the City Council adopt an Energy Storage Procurement Target per Assembly Bill 2514 of zero megawatt at this time as none of the viable applications of energy storage technologies/solutions that may benefit RPU are cost effective.

Roseville Electric: Staff concluded current storage technologies are 2 to 10 times higher cost than energy from Roseville’s existing portfolio of resources. In addition, the extent of application benefits remains unproven. Therefore energy storage procurement targets are not appropriate at this time.

City & County of San Francisco: concludes that it is not cost-effective for the SFPUC to adopt an electric storage procurement target at this time.

Sacramento Municipal Utility District: determines that the adoption of energy storage procurement targets is not appropriate at this time due to the lack of viable and cost effective energy storage options prior to the target dates set forth in Assembly Bill 2514.

City of Santa Clara: With the exception of pumped hydroelectric power, very little commercially available energy storage is currently cost-effective.

Truckee-Donner Public Utilities District: finds that energy storage systems are not currently viable and cost effective for this District, and the District is currently not adopting procurement targets.

Turlock Irrigation District: The results of the staff study concluded that energy storage systems are currently not cost effective and in most cases increased TID cost by millions each year.

Vernon Gas & Electric: Vernon Gas & Electric staff recommends that the city adopt energy storage procurement targets of zero megawatt hours by December 31, 2016, and December 31, 2020, because energy storage is not cost- effective, and, therefore, not appropriate for the City and City customers.

City of Victorville: (A) review of existing energy storage technologies did not identify an application that would be cost-effective at this time; however, it is anticipated that energy storage systems will become more commercially tested and cost-competitive with other resources with the passage of time and improvements in technology.

This entry was posted in Energy, Political commentary and tagged , , . Bookmark the permalink.

56 Responses to California public utilities vote no on energy storage

  1. Dave Rutledge says:

    Hi Roger,

    A timely update. I am rooting for Ares rail storage

    As I am sure you are aware, unlike other states and countries, California renewables accounting is by consumption rather than production. So our 33% renewables will not be comparable to Germany’s because of substantial imports. At CAISO, in-state renewables are 17% ytd compared with 15% last year. The entire gain is solar. We will see how much sun the grid can take.


    • Euan Mearns says:

      ARES storage – several hundred MW for about 10 hours – may work with solar. This is basically the same principle as pumped storage. So the choice is between building reservoirs, building railways or building nukes.

      • Glen Mcmillian says:

        Storage is the toughest nut in the renewables industry by far and away and there just may not be any way to ever have both cheap and dependable storage on the producers end.

        But the rail storage idea seems to me to be simple and easily deployable in technical terms. There are certainly plenty of places in the US at least where a rail storage system could be built and such a system would not take long at all to build once permitted and financed.

        My guess is that we are going to be getting storage defacto at the consumer end of the grid in the form of smart appliances and thermal storage of heat and cool within the foreseeable future. There is plenty of room in a typical mcmansion for a hundred gallon water heater with two or three times the usual amount of insulation which means the plumber can install defacto storage of wind and solar adequate for a day or two so far as hot water is concerned.

        A good sized pile of crushed stone or even dry soil in the crawl space (itself properly insulated) of a new house can be used to store heat energy adequate for a few days which will take care of the intermittency of wind power in winter for that particular ( NEW ) house.

        And for what it is worth I expect battery powered automobile sales to take off like a rocket within a decade as consumers overcome their reluctance to buy them based on fear of reliability and resale value.Battery prices are coming down by half every five years or so and at that rate in ten years the cost of a truly dedicated electric car may be hardly any higher than that of a conventional car…and less after considering the reduced maintenance.

        I have no opinion about electric cars being used to help trim peak loads but they will certainly be useful in making good use of off peak wind and thus lower the demand for oil substantially if they are numerous enough.

        In the end the price of fossil fuels due to inflation and taxes and depreciation is eventually going to be high enough to force the use of renewable power on the basis of fuel costs alone whenever possible.

        This does of course imply that we will still have to have two thirds or more enough conventional capacity to handle peak loads- but that is not an engineering problem. It is a political and financial problem. That capacity will be there once the regulatory authorities rearrange the rates paid to justify having it AVAILABLE rather than actually generating most of the time.

        We have plenty of other heavy duty infrastructure that is used only intermittently and think nothing of it. Major highways are lightly used most of the day except at so called ” rush hours”. A lot of large conventional retailers run their stores at a loss almost the whole year and make it up plus their annual profit between Thanksgiving and Christmas.

        I have a heavy duty pickup truck sitting in the driveway that costs me a substantial amount of money for insurance and taxes that is seldom driven more than once or twice a week and then not very far. I drive a smaller one or a very small car whenever I can. But that larger four by four goes places on the farm and hauls loads the smaller truck can’t and it goes when the roads are iced up.

        So I have resigned myself to the three or four dollars it costs me every day it just sits there in insurance premiums and taxes..

        Rust and depletion never sleep.Inflation only catnaps. Renewables plus backup conventional generation will eventually be cheaper than conventional generation alone.

        Now of course I am one of the people who do believe we are looking at runaway climate change if we don’t cut back the co2 in a big way.So I believe in pushing as hard as possible on the renewables front NOW.

    • Hi Dave:

      Eight miles of track, 7.8% grade, 21,000 tons of concrete. Hope the brakes work. 😉

      The problem with ARES is going to be finding suitable terrain. There aren’t too many places on Earth where you get a gradual 8 mile long slope with an elevation change of 3,300 feet and nothing in the way.

      • Hi Roger,

        You may be right. ARES is starting up near the Tehachapi wind farms. We will see how they do.


        • I can’t find anywhere around Tehachapi with a 3,300ft drop in eight miles either. I guess I must be missing something.

          • Euan Mearns says:

            I bet the solar guys get really pissed off when they discover that half of their power goes to storage and they need to double up on array size.

            Is there any data on round trip efficiency for the trains. What they want to do is to couple this idea to death trains from Wyoming. Use less energy pumping train up the hill, load it with coal, and capitalise on a little tectonic potential energy. Green coal power 😉

          • Willem Post says:

            Better yet,
            We could have solar energy move the trains up the High Plains which have plenty of wind, load them with coal and oil, unfurl the sails and not only coast downhill, but have the wind push as well.

          • Dunno about the solar going uphill or the wind pushing downhill, but a loaded 15,000-ton coal train in Gillette, Wyoming, elevation 4,500ft, has 1.35^11 foot pounds of potential energy relative to a destination power plant at sea level, equivalent to ~50,000 MWh of electricity if I’ve done my sums right.

  2. Hugh Sharman says:

    Great analysis Roger! Shocking for a storage buff like me but I have suspected for years and from my own experience that much froth must disappear from this business before it can be taken seriously!

  3. Willem Post says:


    That is a very timely post.

    As a result, I added this section to this article.

    RE and its Impending Adverse Impact on the California Grid: Without domestic balancing and generating capacity, and without domestic, economically viable energy storage, and without extensive connections to other grids that have EXCESS balancing and generating capacity, the adverse impact of increasing variable, intermittent RE on a domestic grid will become unmanageable. Here is an example:

    The more variable, intermittent RE California adds, the more difficult it becomes to integrate it with the grid. This problem became evident early on, and in an attempt to solve it California, in 2010, passed legislation to encourage its publicly owned utilities to install energy storage – batteries, thermal, flywheels, CAES, pumped hydro (not exceeding 50MW), whatever worked – requesting them to develop viable and cost-effective plans and submit their end 2016 energy storage capacity targets by October 1, 2014. The submissions from 29 of 31 California publicly owned utilities have been published here

  4. Mark Miller says:


    In sunny CA we would like to leverage the plug-in vehicles we hope to have on the roads soon to stabilize the grid a bit. A meeting will be held on the subject shortly to see how we are doing:

    “California Energy Commission staff will lead a public workshop with the California Independent System Operator (California ISO), California Public Utilities Commission (CPUC), and California Air Resources Board (CARB) to review the progress of research called for in the California Vehicle-Grid Integration Roadmap as part of the Governor’s Zero-Emission Vehicle Action Plan. Stakeholder discussions on future solicitations in this topic area will also take place. Commissioners may attend the workshop.”

    We have been supporting various R&D efforts over the years on this subject as you can see from our current projects:

    It will be interesting to see if UCSD goes for one of the grants to modify their charging stations to allow for the flow of electrons from car batteries back to the grid. How all the new capital costs are going to be allocated is going to be difficult given that “our UCSD on-campus charging stations really do charge $.49 per kwh!” per arttoscience’s recent post on charging costs-

    • Hi Mark:

      Thanks for your input (and again for the link).

      Energy storage plans are often heavy into vehicle-grid integration but It’s never been clear to me how this is going to work in practice. The main problems as I see them aren’t so much the intricacies of charging and discharging millions of EVs, which are difficult enough, but a) who is going to pay for all these EVs and b) where is the electricity needed to power them going to come from?

      Personally I like hybrids.

      You say “We have been supporting various R&D efforts”. Does the “we” mean you work with the California Energy Commission?

      • Mark Miller says:

        Hi Roger,

        I drive so few miles these days, vs my corporate and consulting days, that my older diesel vehicle works just fine for me. If I was still commuting 30 miles a day one way I would be evaluating the newer Honda Accord hybrid, one of the newer clean diesels, and a plug in EV that also has a standard FF engine (ie Chevy Volt).

        My existing PG&E service is 200 AMP so I wouldn’t need to worry about having to upgrade. Unfortunately, I don’t have a spare 220/240 volt line in a location that would permit reasonable charging times at home.

        A couple of years ago my wife nixed my plan to change out our electric dryer for a propane fueled one (which would of helped reduce our monthly kWh usage). She may be willing to reconsider switching out our dryer given the recent changes to our electric rate schedule which will lead to an 18% increase in our monthly bill for electrical energy usage in the winter months- analysis based on last year’s actual usage data for December with our TOU E-7 Net meter rate schedule.

        I have been expecting increased cost allocations to our electric bill for a few years as our rate structures were unsustainable given that PG&E’s costs for securing RE generation and the transmission infrastructure to get the utility scale RE to the grid has to be paid for by someone. My SWAG for the reallocation of PG&E’s costs to my bill were off as I only expected 5 to 10% increases in our Average kWh costs. No I do not work for the CEC, but I have spent a lot of time over the last 8 years trying to understand how we are planning to decarbonize. Living really close to Sacramento has allowed me to attend many of the public meetings held by the CEC and CARB.

        These days I try to keep up with our strategies and the tactical plans via being on the distribution list(s) from the CEC. PG&E was kind enough to send me a paper copy of their 2012 Rate Design Window filing which is the documentation I used to come up with my SWAG of the cost allocation for my electrical energy usage.

        It will be interesting to see if some of our energy efficiency upgrades will provide enough EE improvements to offset some of the increased costs for our usage. I tried to include some form requirements to go along with the functional improvements we have undertaken. It takes a LONG time to re-glaze 20 big windows that have 8 panes per window. Thank goodness our contractor who installed some new siding on one exterior wall let me observe where our window frames could use some insulation and air gap closing. Our 1883 house needed exterior paint and some friend contractors shared their expertise with me on ways to caulk up area’s that required some creative gap filling. An ecologist associate told me that the spiders and various creatures who used to live in the gaps in the siding at the edge of the walls and windows would be just fine as they would find some other place to catch their pray/dinner.

        My 70 and 80 year old in-laws are finding it a bit painful dealing with the recent electrical energy cost increases as they are on fixed incomes and their 1970’s vintage home is all electric. We have been collecting up wood from our parcel over the last few years and splitting it for their use. Last time I checked the back part of our parcel I noticed yet another dead tree (a large pine this time). Over the last 4 years we have had 3 Oaks die- one was actually blown apart via a lightning strike that burned down the trunk into the roots.

        • Hi Mark:

          The average household consumption of electricity in the US in 2012 was 940 kWh/month. The average in California was 573 kWh/month. My consumption of grid electricity in November was 31 kWh, for which I paid the princely sum of $2.35.

          But it would have been a different story without the solar panels on my roof. My usage of grid electricity would have shot up into the hundreds of kWh range, I would have been paying a much higher rate and I would have had a bill of over $100.

          And the solar panels were unsubsidized too.

          Goes to show that solar can work in small doses in the right place. And living a mile high at latitude 20N is a good place to be.

          I know the Sacramento area quite well. I butted heads with legislators there when I was working in Salton Sea geothermal back in the 1980s, my ex-daughter-in-law is from Roseville and my son went to UOP.

          I have spent a lot of time over the last 8 years trying to understand how we are planning to decarbonize. If you ever figure it out please let me know. 😉

      • Willem Post says:

        The Toyota PriusV, 44 MPG, EPA combined, is used as taxis in Paris where gas last year was $9/gal. No plugging in.

        The Toyota plug-ins are not selling that well.

        • There are two reasons plug-in EVs aren’t selling well. One is that they cost about twice as much as conventional vehicles, the other is “range anxiety”. We can expect that costs will decrease in the future, making EVs more affordable, and that ranges will increase too, lessening range anxiety. But watch out when the government starts discharging your EV battery back into the grid. You may not make it to work next day.

    • Batteries aren’t going to ever work:

      Nor will using them to store power:
      Peterson, S.B.; Whitacre, J.F.; Apt, J. (2010). “The Economics of Using PHEV Battery Packs for Grid Storage.” Journal of Power Sources (195:8); pp. 2377–2384.

      The results suggest that vehicle owners are not likely to receive sufficient incentives from electricity arbitrage to motivate large-scale use of car batteries for grid energy storage.

      without energy storage, the grid is too unstable:

  5. Willem Post says:

    “So I believe in pushing as hard as possible on the renewables front NOW.”

    Please read this article

  6. Hugh Sharman says:


    Terribly sorry but you have this wrong.

    California (CAISO) remains set to meet its GW+ targets. My colleagues in California have pointed out the errors in your report but probably the best summary of the current situation is by Union Bank of Switzerland at

    • Euan Mearns says:

      Hugh, Its bad form to turn up and say we got it wrong and to then not say how. The report you link to has 268 MW of storage projects and while I’ve only scanned it I don’t see any mention of MWh let alone TWh of storage. The main table also has 1698 MW of new gas fired power. This a kind of tells you how California is going to balance load does it not?

    • Er no, Hugh, I haven’t got it wrong.

      I was talking about publicly-owned utilities (POUs), as the title and text make clear. You are talking about investor owned utilities (IOUs).

      Assembly Bill 2514 allowed the POUs to decline to set storage targets, which almost all of them did. The IOUs had their combined 1.325 GW storage target set for them. They didn’t have the option of declining.

      Even if this target can be met 1.325GW of storage would still work out to only a few percent of combined California IOU peak load. And since AB 2514 doesn’t specify a GWh target the IOUs would be able to meet the letter, if not the spirit of the Act by delivering 1.325GW for only a few minutes.

      • Chris Nelder says:

        A bit more detail on the SCE contracts:–storage-and-efficiency-to-meet-electricity-needs_100017103/#.VFz_G_CRQ5s.twitter
        “Nobody quintuples the order of something they don’t like,” notes Browning. “That’s a great leading indicator of the cost and performance of storage.”
        “SCE decided to purchase 250 MW of energy storage because it felt it had a higher value than any other generation asset (including natural gas, wind and solar).”

        Note too that the world’s largest merchant solar power system (i.e., no PPA) just went online…in Chile:

        • California IOUs actually took a rather dim view of AB2514, as the following excerpt from SCE’s October 2013 filing with the CPUC attests:

          4.4.2. Parties’ Comments

          The proposed targets have been met with mixed responses. SCE warns that the targets are very aggressive and will come at a high cost to California ratepayers, especially if they are poorly designed and the pathway is too rigid. SCE advises that the Commission should remain flexible by periodically revisiting the targets and the pace of procurement. Similarly, PG&E argues that the energy storage procurement targets should be shifted, so that less is required in 2014 and 2016, and correspondingly more is required in 2018 and 2020. PG&E believes that, consistent with its experience in the RPS Program, the cost of storage projects will decrease as storage technologies evolve.

          SDG&E puts forth various arguments in opposition to the proposed procurement targets. First, SDG&E contends that the timeline and level of the targets are arbitrary. As support, it maintains that technical analysis from this proceeding does not justify the proposed level of procurement targets. Second, SDG&E argues that any procurement targets should be related to a specific need or solve a specific problem. SDG&E contends that there has been no examination as to what level of distribution level, transmission level and customer level energy storage would be beneficial to each utility or local area within a utility’s service area. Finally, SDG&E maintains that if procurement targets are adopted, these targets should not be in place until 2020 because energy storage systems are not mature enough to have specific interim targets before then.

          Other parties also maintain that it is premature to mandate targets. MEA expressed concern regarding the lack of data related to performance and cost-effectiveness on all identified use cases and warns about pursuing “storage for storage’s sake.” Similarly, DRA states that the Commission should not adopt targets without further analysis of whether storage is the only option to service grid functions. Pilot Power shares the same views but suggests that the numbers be revisited in two years to see if they are justified.

          Other parties maintain that it is inappropriate to set targets at this time because there is no demonstrated need for additional resources. CalWEA notes that other Commission proceedings have not identified a need for integration until 2020 and concludes that “[if] there is no need, by definition it is not possible to cost-effectively satisfy that need with additional (un-needed) resources.” CEERT proposes that procurement targets should not be established until energy storage technology eligibility and cost effectiveness have been determined. Jack Ellis asserts that procurement targets are unnecessary and that developers and sponsors of storage projects should be free to develop projects that they wish since cost effectiveness analyses suggest benefit/cost ratios greater than 1 for a variety of applications. AReM believes that the picking and choosing of winners could be more effectively managed through competitive markets rather than government mandates.

          • Chris Nelder says:

            Sure, sure. Nobody likes having targets imposed on them. But SCE’s decision to contract 5x what they were required to appears to have been their voluntary move. Regulatory rhetoric is one thing; money on the table is another entirely.

          • “money on the table is another entirely.”

            Indeed it would be if it was SCE’s money, but it seems that it won’t be.


            Both SCE and SDG&E are authorized to include the costs of the procurement authorized today through the Cost Allocation Mechanism, consistent with its established rules, and/or other applicable procurement cost allocation processes.

            The Cost Allocation Mechanism, or CAM, is designed to ensure that the costs of new resources procured to ensure local or system reliability are shared equally among all utility distribution customers, regardless of their generation provider.

        • Euan Mearns says:

          Chris, Roger has a solar array at his home in Mexico made viable by the fact that electricity prices there for the wealthy are very high. Energy storage is of course a good idea. The best and biggest stores are of course FF and U – Th 😉 It does not surprise me that some companies are investing in storage. It does surprise me that more are not doing so. But would you agree that it makes the cost of new renewable electricity even more expensive. We are heading for system with lots of new renewables + storage + inter-connectors + 100% backup.

          Solar in sunny climes may work – if proven to have large positive ERoEI. But where I live it doesn’t – and yet it is being mindlessly rolled out. The storage issue with solar is an order of magnitude different to wind. If you have reliable sunshine you really only need storage for a 24 hour cycle. The challenge with wind is I believe impossible to bridge at reasonable cost. The things that works with both muscle (GW) and stamina (h) is pumped hydro. But the scale of deployment required to bridge long wind lulls is not achievable in the UK.

          As far as I can see none of yours or Rogers links discuss hours. All the stats are GW – delivery potential. Not specifying durability suggests to me that none of these schemes are serious.

          Coire Glas in Scotland is massive, but still only has a 50 hour storage capacity.

          • Mark Miller says:


            Over the last 6 years SMUD has spent 30 million on various energy storage projects. Starting on page 95 of their recent board meeting package their technical experts summarize how their various efforts have worked out (actual performance vs original plan).

          • Euan: The storage commitment was part of a package deal SCE/SDG&E made with the California Public Utilities Commission to replace the lost generation from the recently-closed San Onofre nuclear plant, and I’m sure a lot of behind-the-scenes wheeling and dealing went on before the final targets were agreed on. But SCE/SDG&E got basically what they wanted out of the deal – a lot of new gas-fired capacity – and committing to more storage capacity than they wanted to commit to may well have been the price they had to pay to get it.

            I think the SMUD document Mark links to gives a much better idea of what the economics of energy storage really are. SMUD spent $30 million evaluating six different storage technologies relative to nine different load-following applications and the only one they found to be economic was large-scale pumped hydro.

          • Chris Nelder says:

            Scotland’s another matter entirely…

            But staying with the topic at hand… It’s important to note that the POUs only serve about a quarter of the state. The other three-quarters are served by the big three IOUs, and they have to fulfill just a part of the new storage mandate, which was wisely constructed to encourage storage across a full range of market sectors.

            Of course the costs will be allocated across the customer base regardless of generation provider. How else would it be done in a (sort-of) “deregulated” market? Or if you’re suggesting that one of the big three utilities should not be buying 250 MW of storage capacity at this point…well I guess we’ll just have to disagree about that.

            I assume storage projects are so frequently listed in terms of MW capacity because their load factors aren’t always known up front…or maybe it’s just easier to work with those numbers. It would be better perhaps to state them in terms of Ah (but not MWh). But I don’t see anything nefarious about that.

            If the point is that the POUs aren’t signing up for storage just yet…ah well. No big deal. They’re a small part of the picture and they’ll come around when costs are lower. Perfectly fine for the IOUs to do the heavy lifting and testing in order to start integrating storage on the grid. Remember, this was always intended to be just the initial pilot phase of a very long-term program, not a one-shot ultimate referendum on the entire concept of grid power storage.

            If the point is to suggest that non-trivial storage projects aren’t happening in California: Sorry, that’s simply not true.

            If the point is to suggest that storage generally is dumb, or that it doesn’t work economically as a complement to high wind and solar generation, sorry, well that doesn’t appear to be the case either. Not just in Calfornia…check out Texas:

            If the point is that renewables plus storage costs more over the long term than simply remaining with the fossil fuel and nuclear incumbency, I don’t think that’s proved at all; the recent cost trends are that fossil and nuclear generation costs continue to rise while renewable costs continue to fall. And surely the examples of Texas and California speak for themselves. Though the situation is different for every region so that’s a difficult thing to generalize about.

            If the point is that wind can’t do much without a lot of storage, how is it that Scottish wind turbines produced 982,842 megawatt-hours of electricity in October — 126% of Scotland’s residential electrical needs? One would think they’d be overflowing with storage capacity at that level.

            But this thread has drifted quite a bit and I’m no longer sure what the point is. Anyway I didn’t come here to debate energy transition in general. I was just trying to offer a little clarity on the California storage question.

          • Euan Mearns says:

            Chris, one of my recurring themes is horses for courses. Advocacy often leads to uniform blanket sort of proposal where you either have to be in favour of renewables or not. I believe that the point of Roger’s post is to show that where companies were given a choice, most chose not to invest in storage. Cheap, scalable, environmentally benign grid-scale storage is of course a good thing. But it does not yet exist and personally I doubt it ever will. But if renewable hobbiests want to install a battery pack or some such good luck to them. Authorities mandating companies to install storage is drifting towards a command economy and presumes that said authorities have a deep and sure grasp of a myriad issues surrounding climate change and society’s ability to run on something other than fossil fuel.

            I believe Scotland should have a lot more pumped hydro. But when you look at the scale of the challenge and the environmental impact and how futile it is then you begin to question if it is such a good idea. The natural sites for pumped hydro in Scotland are along the Great Glen – Loch Ness et al. But the trouble is when you start pumping the river leaving the loch may run dry and when you generate you will have a flood. If you have time you should have a read:


            Scotland is absorbing a lot of wind power at present because it is attached to a much larger country. That larger country pays for most of the subsidies that would bankrupt Scotland alone and absorbs the surplus power when it exists. Increasingly, surplus power is being spilled. Consumers are paying top $ to wind companies to spill power.

            Wind works in Denmark because it is balanced off Scandinavian hydro. You seem to be anti-nuclear – well wait and see what happens to Sweden’s ability to balance Danish wind if they close down their nukes and have to use their hydro for base load. And there is wide spread opposition in Norway and Sweden to extending the hydro facilities because of the environmental impact.

            Wind also works in Portugal because it too also has a large hydro park. Luis has never managed to get his head around the horses for courses argument. I’m not against wind in Portugal if the Portuguese don’t mind every ridge covered in wind mills and expensive electricity. Portugal had a problem in any case affording to pay for imported FF. The UK has 1.6 GW of hydro that only has capacity to run at about 10 to 20% load. It is a pin prick compared with UK peak demand over 55 GW.

            The Great Glen is also intensively used for recreation since the famous Caledonian Canal runs through it. Built at a time when engineers were engineers. And our mountains are also used intensively for recreation and our limited “wilderness” is increasingly being destroyed by wind turbines. We have an important mountaineering tourist industry that is put at risk by turbines subsidised by the mountaineers for some of the time to generate power that is spilled and for a lot of the time to produce nothing at all.

            Solar in S California is perhaps different.

          • Chris Nelder says:


            Cheap, scalable, environmentally benign grid-scale storage is of course a good thing. But it does not yet exist and personally I doubt it ever will.

            I’m going to guess you didn’t look at that link I provided regarding the storage purchase offer proposed in Texas by Oncor.

            And there are more large storage projects coming – Li-ion and CASE:

            I wouldn’t expect the first tranche of grid-scale storage projects to be cheap (although it seems that these are already competitive, or are expected to be within 2-3 years). Like any new energy technology, some learning and scaling will bring costs down over a period of years. Anyone who doubts that grid-scale storage married with renewables is going to be highly competitive with conventional power by 2018-2020 simply isn’t paying attention.

          • Euan Mearns says:

            Chris, the Lion battery scheme is a kiddies toy. The CAES scheme in Utah a different beast. 60 GWh for $1.5 billion is comparable to the Coire Glas pumped storage scheme at 30 GWh for £800 million. I reckoned that Coire Glas would be better used diurnally with nuclear. Trying to use it to store wind is futile IMO. And the same will apply to your Utah CAES scheme – where the dimensions are awesome. It may work sensibly with solar. What is the round trip efficiency of CAES?

          • Chris Nelder says:

            Sorry, haven’t looked into the efficiency of that particular CAES project.

            Two more 20 MW (7.8 MWh) Li-P battery projects were announced today in Chicago:

            Like other smallish projects, those will be used primarily for frequency regulation.

            But the aforementioned Oncor offer is for 5 GW (!) of battery storage across Texas.

            Storage isn’t just one thing, and next to nothing can be said about it generally. There are lots of different technologies, applications, needs, and markets (it’s about a lot more than just backing up renewables) and there is a great deal of ferment happening across the sector. I think one of the most interesting players is Germany-based Younicos. Definitely one to keep an eye on. For CAES, have a look at what LightSail is working on.

  7. Paolo Pulicani says:


    Tech: Power: ask price bid price
    MW $/MWh $/MWh

    WIND 769 51.8 57.4
    FV 890 106 87

    Contract duration: 20 years

  8. Pingback: Roger Andrews: California public utilities vote no on energy storage | Peak Energy & Resources, Climate Change, and the Preservation of Knowledge

  9. Pingback: AWED Energy & Environmental Newsletter: November 10, 2014 - Master Resource

  10. Chris & Euan: Getting a little cramped up above so I’m starting afresh down here:

    Chris, you said earlier: “I’m going to guess you (Euan) didn’t look at that link I provided regarding the storage purchase offer proposed in Texas by Oncor.”

    Well, I did.

    “The Dallas-based transmission company is proposing the installation of 5,000 megawatts of batteries not just in its service area but across Texas’ entire grid. Statewide, Oncor estimates a total price tag of $5.2 billion.”

    That works out to $1,040/kW installed, which is more than the $917/kW EIA estimates a new natural gas combined cycle plant – which actually generates electricity instead of just moving it around – would cost. Is that economic? I guess it depends on local factors.

    Another statement in the link; “Oncor believes it can make the economics work” suggests that the $5.2 billion cost is a guesstimate. Back in May another article featuring Oncor (link below) noted that battery storage was “still prohibitively expensive on a utility scale”, with CEO Bob Shaphard agreeing that “The math isn’t there yet”.

    And at the time Oncor wasn’t sure that battery storage technology would work even on the small-scale:

    “Oncor is investing $1 million to install six storage batteries across its Dallas grid to test how well the technology can function as back-up power supply during short outages.”

    Clearly something has happened since May to convince Oncor that gigawatt-scale battery storage is worth pursuing. But if it’s as promising as Oncor claims, why isn’t everybody doing it?

    Maybe because Oncor is a power distributor, not a power generator?

    • Chris Nelder says:

      Roger: It’s definitely an interesting development that a transmission company like Oncor wants to make this move – making storage part of the grid, versus treating it like generation. That makes more sense from a technical standpoint, actually. And it will be a big deal for the grid power storage market if its growth happens via monopoly rate base versus competitive “merchant” providers.

      Oncor has researched the cost trends and talked to the battery manufacturers and determined that this will be a good bet for them. “Everyone assumed the price point was five to six years out. We’re getting indications from everyone we’ve talked to they can get us to that price by 2018.”

      Obviously, a $5 billion project will take some advance planning…

      Why isn’t everybody doing it? Um…because it’s a new thing to do? Because Texas is a fairly unique situation on several counts?

      • Chris:

        The question of how a generator’s perspective might differ from a distributor’s is an intriguing one but not something I can comment on intelligently. And Texas is of course a unique situation. Everywhere is.

        But now I’m going to reverse my position entirely and assume that battery storage is indeed on the brink of commercial feasibility. What happens then? Well, the first thing we find is that we have oodles of solar and wind power but no batteries to store it in:

        Wind plus solar installed capacity 2011: 310,183 MW, 2013: 459,544 MW (BP data)
        Installed battery storage capacity 2011: 370 MW, 2013: 734 MW

        (battery numbers from

        It’s difficult to estimate how much storage capacity is needed to smooth out wind and solar fluctuations, but according to some crude assumptions I made it’s around 50% of installed wind and solar capacity. If so then we have a current storage capacity deficit of around 200,000 MW with the gap widening all the time as more wind and solar goes in. Can we make up this deficit? Tough, I would say. We would probably do better to mandate that all future wind and solar projects come with storage attached and let the deficit take care of itself. That would guarantee the rapid expansion of battery and other storage technologies provided the added costs and complexities didn’t kill wind and solar in the process.

        • Chris Nelder says:

          Roger: How much storage you need depends entirely on your set of assumptions. And it’s a complex set of assumptions about the grid power mix, the prices, the demand from electricity trade partners, the weather, the shape of the loads, the technologies, etc. etc. etc.

          For example: You can have oodles of solar and wind power, without needing batteries, and without having to dump any of it, if you have an adequately large and interconnected grid. Exporting power is obviously the first choice. Both Scotland and Denmark export their excess wind power, so how much storage do they technically need? (Does your 50% estimate hold true for Scotland, today?) To simplify it somewhat (leaving aside all the other tricky questions): Only when an entire system of interconnected countries (e.g., Europe) has no more capacity to absorb excess RE might storage become important.

          It’s all theoretical at this point, since no one is getting close to having the real-time data to know exactly how much storage is needed. And of course every grid and country is different.

          For another example: Craig Morris (the best English-language pundit on Germany’s Energiewende in my opinion) has done a whole series of blog posts recently on a new Fraunhofer study which did a scenario analysis to try to answer the question: How much storage would be needed to support a 100% renewable grid? Here’s a good place to start:

          Here, we see that this 100 percent supply of renewable electricity requires roughly 15 percent storage. Nonetheless, I do not see from the background paper how we get to the 612.4 TWh consumed in a year; this figure on the visualization does not tally with the background paper.

          The background paper does not say how much battery storage is required but does state that 13 GW of power-to-gas capacity is assumed.

          It thus seems that the background paper is not exhaustive. To find the missing information about batteries, I therefore went over to another interactive chart that might interest you – the “scenario map” (once again, only in German). There, we find that 55 GW of batteries are assumed, along with 13.5 MW of pumped hydro and compressed air storage. The efficiency of the batteries is assumed to be 85 percent, compared to 75 percent for pumped hydro and compressed air.

          You can click back through the whole series of blog posts starting here:

          But it’s important to understand the assumptions used in the scenario. This Q&A with the lead author of the study elaborates on that:

          To wit:

          We therefore end up with 55 kWh of battery capacity hooked up to the amount of PV assumed to be installed in households, producing a 55 GW fleet of batteries as a part of the scenario. The need for storage depends on the use of other flexibility options (such as load management and grid expansion), so other 100 percent renewable outcomes are certainly imaginable with far less battery capacity in light of the relatively low capacity factor.

          My gut (based on a variety of similar studies), says that you don’t begin to need storage until RE gets to about half the total supply for an entire interconnection. And then you can cut down on the need for storage at least a half a dozen different ways…

          But there are many ways to think about it. In December 2012 I wrote up a study on the PJM Interconnection which found that overbuilding wind & solar generation capacity and having a more interconnected grid could virtually do away with the need for storage on a 100% renewable grid.

          About a year ago I also did a little thing on California’s storage mandate:

          But in many ways it’s more a business and political question than a technical question…another piece I wrote on that:

          There are lots of interesting studies out there that have tried to figure out the best mix, including cost-optimization studies. Look around. Here’s a good one by NREL, focusing on an 80% renewables scenario:

          The short answer is: It’s not a back-of-the-envelope calculation. It’s very complicated. It varies from place to place. And nowhere have we gotten to the point where we have real-world data on what the technical limits might be. I think Hawaii is probably the closest thing to a real-world example so far, and they’re only beginning to deploy any real storage, which is mostly distributed anyway. When we do get to the point where we discover technical limits, there will be many different ways of overcoming those limits, and we’ll have to sort out what’s best for each unique circumstance.

  11. Pingback: Recent Energy And Environmental News – November 10th 2014 | PA Pundits - International

  12. Hugh Sharman just sent me an article which links to this post and presents the Investor-Owned Utility side of the story. I link to it here in the interests of balance. (Pun intended) 😉

  13. Mark Miller says:


    Thanks to Eddy, our 3 year old Golden/Chow, I had to refile a bunch of paper this morning. I put my copy of “Table B-7: Average Residential Electricity Rates (2010 Cents/kWh), Updated Forecast” back into my 33%RES folder.

    The CA legislature updated the 20%RES to the current 33%RES a few years ago. Other than increasing the % of RE as part of the standard the legislature corrected a perceived fairness problem (the public utilities were not mandated to meet a RES in the 20%RES). The updated legislation corrected this perceived unfairness and required both the public and private electrical services providers in the state to meet the 33%RES.

    Thought you might like a copy of the table as it includes a few public utilities forecasted residential rates as well as the big three private electric utilities rates (the table is on page 33):

  14. Chris Nelder says:

    Those interested in seeing a cost-optimized model of how storage might play a role on the California grid might want to check out this recent study by NREL, GE and others:

    It shows three different scenarios with varying amounts of RE, gas generation, storage, etc. and concludes:

    With this portfolio, the California electric sector can reduce GHG emissions by more than 50% below 2012 levels in 2030:
    ⁻ With minimal rate impact
    ⁻ Without compromising reliability
    ⁻ With minimal curtailment of renewable energy
    ⁻ With a stable gas fleet that is dispatched with minimum cycling

    • Euan Mearns says:

      Chris, I had a quick look at the presentation. I wish I had more time. My quick impression is this:

      1) California is big, 2) California has mountains and a lot of hydro, 3) California sits atop a subduction zone that transfers magmatic heat towards the surface – has geothermal (watch out for earth quakes and volcanos) 4) California is sunny.

      This is like southern Germany that has a lot of these things. California is optimally placed for renewables and good luck – the wealth of California and the things I mention above are probably not a coincidence. But lets see you do that in London, Liverpool or Glasgow.

      Also noted that wind does not play a big role in CA. Solar correlates with demand, wind is stochastic.

    • I took a slightly more detailed look, and my attention was drawn to this graph, which according to the caption shows the “baseline” case for the years 2030.

      It shows that there is indeed very little requirement for storage (highlighted in black to make it easier to see). But there are reasons for this. First is the low overall level of intermittent wind and solar penetration (looking at the graph I would guess 10-15%). Second is the very large amount of gas generation (looking at the graph I would guess over 50%). With this much load-following capacity the question is why you would need storage at all.

      The graph is also quite similar to the status quo in Germany, where they are already having to export surplus solar power because the grid can’t absorb it.

      • Euan Mearns says:

        Roger, thanks for posting the chart – it saved me doing it 🙂 CSP, wind and storage are all but irrelevant – why would you bother? I don’t know how many times I will have to say this, but here goes, solar PV in sunny climates has potential, solar correlated with AC demand. Not sure what the QF bit of CHP-QF means? And if the objective here is to reduce CO2, why use gas for base load instead of nukes? – Chris?

        • Chris Nelder says:

          I am baffled why you guys would focus in on the “baseline scenario” and conclude that “CSP, wind and storage are all but irrelevant” when the whole point of that study was to explore the “target” and “advanced” scenarios, in which they become a much larger part of the mix. Do you just hear what you want to hear?

          I’m also surprised you would ask why gas instead of nukes, when the whole starting point of this thread was how California intends to deal with the shutdown of the San Onofre nuclear plant. ?? If you don’t know why that plant was shut down, I wrote a piece on that last year:

          I think I’m done here.

          • Euan Mearns says:

            Bye then Chris, it was good to have you around for a while at least. I had a quick look at your nuclear piece, which didn’t seem unbiased to me. But looking at that nuke right on the shore north of San Diego I’d be crapping myself. Just as well California doesn’t get earth quakes – eh?

            40 years of operational life! Same in the UK, way beyond design life. And the problems are mainly in the steam generation plant, not in the thermal reactor – same in the UK. Lets see how off shore wind compares.

  15. Jay Alt says:

    California, that’s the place you oughta be. Today’s storage options may not make utilities jump for joy there. Yet. But who is it ‘forcing’ Texas’ biggest fossil fuel utility to ask for $2 billion in storage to stabilize the TX grid and save consumers money?

Comments are closed.