The (Battery) Acid Test

Dominion coupled solar rooftops at Randoph-Macon College with experimental battery storage.

Dominion coupled solar rooftop at Randolph Macon College with experimental battery storage.

Dominion Virginia Power is running experiments to see how well battery storage works with solar electricity. The verdict so far: mixed. 

Last year Austin-based ViZn Energy Systems had developed a zinc-iron redox battery system, the product of eight years of work, which it touted as safe, efficient, durable and flexible. The battery, the company stated, could absorb excess electricity when the sun was at its peak, inject power into the grid when production faded, and smooth over short, spiky fluctuations in voltage as output varied with cloud cover.

When Dominion Virginia Power opened a 50 KW roof-top solar installation at Randolph-Macon University in Hanover County in April 2015, it installed one of ViZn’s 48 kW “flow” batteries along with a smaller, 7 kW “wet cell” battery to test in real-world conditions.

“We wanted to evaluate and demonstrate some different battery technologies, explains Brett Crable, director of new technology and energy conservation. A big question was how the batteries would stand up to continual charging and re-charging.

As it turned out, the flow battery did not meet expectations from a reliability standpoint, says Crable. When the lease on the battery system expired, Dominion chose not to renew it.

As Dominion prepares to integrate increasing volumes of intermittent solar and wind power into its distribution system, it is looking at battery storage as one tool for dealing with the inevitable fluctuations of voltage and power. PJM Interconnection, the regional transmission organization of which Dominion is a part, says the high-voltage transmission system can handle up to 30% wind and solar without making it vulnerable to blackouts. But Dominion wants more hands-on experience with its lower-voltage distribution lines (which connect substations to homes and businesses) before committing to battery storage on a large scale.

Unlike the batteries people buy for home use, one size does not fit all in an industrial application. Batteries have varying characteristics depending upon the chemical composition and internal structure. Some batteries are designed for power (releasing electricity faster) and others for energy (total electricity stored). Some hold up better than others from charging and re-charging. Some are toxic, others non-toxic.

“As technology evolves, you look at power, energy, durability and safety,” says Crable. “Depending upon the application,  you’re looking for different characteristics.”

In addition to its small-scale deployment at Randolph-Macon, Dominion is testing a lithium-ion battery at an experimental solar facility in Kitty Hawk, N.C.

Among the dozens of battery deployments around the country, the Randolph-Macon project is one of only two facilities in Virginia using battery storage at the moment, according to the  U.S. Department of Energy Global Energy Storage Database. The other is a 25 kW zinc bromine flow battery at a Visa data center in Loudoun County.

Electro-chemical energy storage projects in northeastern quadrant of U.S.

Electro-chemical energy storage projects in northeastern quadrant of U.S.

While Crable sees battery storage as useful for narrow applications, he doesn’t regard it as practical for residential and small business customers who fantasize about going off the grid.

Say you’re an individual homeowner with a house of average size and energy consumption, he says. You’ll need about 5 kW capacity of solar to meet your needs during normal weather conditions. To keep the juice flowing on nights and cloudy days, you’ll need to augment the solar panels with a battery. If you want enough power to keep on the lights for four days — enough to sustain you through a long winter storm — that would be a very large battery. Then you’d need enough solar panels to fully recharge that battery while also meeting your household needs….which would require even more solar panels You probably don’t have enough rooftop to hold them all.

“Battery storage is complementary,” Crable says. “But as a replacement for the power grid, it is not economic or cost-competitive.”

The grid is evolving and modernizing, he says. “Solar will be part of that, and battery storage will continue to evolve and play an important role.” But it’s too early to say whether Dominion will deploy battery storage on a large scale in the future. “We always seek to learn.”

There are currently no comments highlighted.

14 responses to “The (Battery) Acid Test

  1. It does occur to me – that once a battery comes along that will work for a typical home – what would keep more and more people from installing their own solar/battery set-up and forget about Dominion and net-metering except when they just need to top off with more from the grid? And if Dominion wants to up the price for “availability” then that’s going to drive the ROI even higher for even more robust home systems.

    And if Dominion pushs too far – people will buy standby generators themselves…

  2. LtG, I with respect to cost, suspect not in my lifetime … I hope so in yours…

  3. @JohnB – probably neither of our lifetimes!

  4. re: ” As Dominion prepares to integrate increasing volumes of intermittent solar and wind power into its distribution system, it is looking at battery storage as one tool for dealing with the inevitable fluctuations of voltage and power. PJM Interconnection, the regional transmission organization of which Dominion is a part, says the high-voltage transmission system can handle up to 30% wind and solar without making it vulnerable to blackouts.”

    still don’t understand the “logic” of seeing battery storage as a way to smooth out the variability of solar but not natural gas for the same role.

    a standby generator with a piped connection to natural gas -can and does operate in an autonomous mode with no human operator needed.

    It comes on based on a sensor that detects the presence or absence of electricity on line –

    the same generator – once in generating mode – senses how much demand there is and adjusts it’s output to meet that demand.

    for instance, if someone is just running the TV and the lights it runs at a different rate than if the TV, lights and the Heat Pump comes on – that generator then increases it output to compensate.

    In the very same way -a much bigger version – a utility-scale gas turbine could be sensing “demand” on the grid that would exceed what solar was generating – an supplement it – and essentially vary it’s output based on what the line demand is , as it varies… “intermittently”

    If solar alone can meet the demand then the utility turbine idles. if demand starts to exceed what solar can provide then the turbines fires up but runs at a low state -just enough to supply what is needed.

    If solar goes completely away -rain or night -then the utility gas turbine runs at a high state – enough to supply demand…

    when morning comes – and the sun comes up and solar starts ramping up – the turbine starts to ramp down.

    the “battery” is – essentially the natural gas – in the pipeline.

    it “turns on” when it is needed and it sits inert when not – it works a LOT LIKE a “battery” does.

    the beauty of this is that you conserve gas – you don’t burn any more than you need.. and you “harvest” solar WHEN it IS available – as opposed to viewing solar as “intermittent” or “variable” or “undependable” – it becomes the fuel of choice – WHEN it IS “available” and you use backup when it is not.

    it “works” EXACTLY the same way a residential stand-by generator works – you’d use solar when it is available – you use the stand-by generator when it is not.

    the primary difference is that the systems either is full on or full off as opposed to running in dual mode – modulating.. per fuel availability.

    this is a simple problem for computers and software… it’s not even rocket science.

  5. How about 2020?
    …But the utility industry is on a timetable, he added. Many analysts, including those who put together Oncor’s Brattle report, expect that many storage technologies will hit widespread cost-effectiveness on the grid somewhere around 2020, which could spark massive proliferation of the resource.

    Dominion should be looking at control systems, not so much the batteries according to this ….
    “After a push to change those rules failed in the Texas legislature last year, Oncor directed its attention back to its grid optimization efforts while it prepares for the next legislative session in 2017. This past summer, it unveiled a microgrid at one of its operating facilities near Lancaster, Texas, combining solar, storage and a diesel microturbine in an attempt to eliminate outages at the facility.
    Through that microgrid and other storage pilot projects, Oncor found that ‘the control systems are much more the issue than the batteries themselves,’”

  6. re: control systems

    here’s a question.

    WHERE should the control systems be?

    at the end point where the use is occurring?

    or somewhere in between the end use and where the power is generated – distributed?

    let me give an example.

    right now – I have a water heater – with a box the utility controls.

    I just installed a smart thermostat – that I control but also the utility can.. and I’m not sure exactly what they can do or not.. and at this point I’m not sure they know exactly – what they want to do – either – a little scary.. and to be honest – I’m not sure I want them deciding how warm or cold the house should be – or them basing their decision on things more important to them than me…. I think I’d rather the relationship be based on the cost of the power even if it varies rather than them choosing for everyone… different folks will make different decisions… based on a variety of factors unique to each user – not like a herd that all do the same.

    • An interesting discussion! You say, “still don’t understand the “logic” of seeing battery storage as a way to smooth out the variability of solar but not natural gas for the same role,” and then describe the way it could work. Yes, the gas turbine could function to fill the valleys and shave the peaks exactly as you describe. There’s a big caveat, however, having to do with the design of big natural gas combined-cycle (NGCC) generators, which gain a lot of efficiency through that “combined cycle” stuff that requires much longer lead times for starting up and shutting down. But beyond that, the way the grid operates today, the generator that wants to sell electricity to the grid must be capable of responding to instructions from the grid’s system operator, i.e., the unit must be “dispatchable,” to make the most money in the wholesale energy market.

      You’ve got to stop thinking about solar generators and backup gas generators as discrete, localized pairs of equipment. Think of the 12-state-wide PJM grid as ALL the solar and ALL the gas-fired generators in PJM, backing each other up, along with all the other generators in PJM thrown in for good measure. It would waste a lot of potential efficiency if two geographically-adjacent generators, one gas and one solar, were operated together by one commercial owner to achieve a localized demand target and PJM only saw the sum of their outputs, rather than each component separately. PJM matches up all the generation across its 12 state area to find the very next-cheapest-to-run generator within that area. PJM will run that next cheapest unit, which may be cheaper to operate than your gas unit, and you should be glad it did because over time your bill for energy taken from the grid will be lower, and/or your payment for energy you deliver to the grid will be higher. As the solar output declines each afternoon, PJM doesn’t care whether it cranks up a gas unit in NC or a similar unit in OH to pick up the load.

      The PJM system operator will ask the next-cheapest-to-run generator across the entire 12-state area to turn on as the solar output declines; but if you are your own system operator and your “system” has only solar and one gas unit, that gas unit is what you run when you have to. In contrast, PJM has a provisional generation dispatch chart all figured out for each day, minute by minute, based on the load forecast and weather forecast and every other input variable affecting loads, with generators in economic order (gas-fired or not) ready to ramp up just when needed, and the actual dispatch is adjusted every few seconds by computer-assisted operators. PJM can achieve lowest overall cost of operation with its hundreds of generators far more efficiently than a single commercial customer with 2 or three generators can. But you are correct, the principles are the same, and the generators that back up the solar output do, in fact, work similarly to the way batteries would work.

      As for batteries versus natural gas, as Jim points out, batteries just aren’t there yet. Look at Lithium Ion batteries for example: despite Mr. Musk’s boasting, Li+ batteries use expensive materials including scarce cobalt and are difficult to construct (any internal defects make them potentially unstable, even explosive), and their useful life cycle (number of recharges) is OK but not great, after which you don’t reuse the materials but throw them away. But it’s getting closer, that day when safe, stable, long life, cheap-to-make batteries are available, and if the cost curve for new-technology batteries drops anything like as fast as the cost of solar photovoltaic cells has dropped recently, that will indeed revolutionize local grid operations. And so I applaud Dominion experimenting and getting involved to gain some early experience with how to integrate batteries into the operation of their system.

    • LG, you ask, “WHERE should the control systems be? at the end point where the use is occurring? or somewhere in between the end use and where the power is generated – distributed?”

      The goal in running an electric system is to match supply with load (let’s leave storage such as batteries out of this for the moment to keep it simple).

      If your entire system is contained within your own property — your own microgrid, so to speak — you run just enough generation to supply your load. Solar runs itself, so you have your natural gas unit set up to run automatically to supplement that. Except, the gas unit takes a while to get up and running, so you have to crank it up in advance to get it ready (as “spinning reserve”) to take on part of the load when needed. And, the gas unit is a fixed increment of kW and you only need a portion of that; the rest goes to waste. Your controls are right there, between your generation and your main switchboard serving your load.

      Unless, that is, you can sell that excess to the grid. Or if you have no excess and your load continues to increase, you must buy from the grid or cease some of your demand. Now, to make that possible, you need a two-way interconnection, with a meter or two, and disconnect switches for safety. Controls remain the same.

      But here you are, operating your own micro-grid on a net basis, when you could turn over control over the generation piece to a distant system operator and sell your generation on a gross basis, and at the same time buy ALL of your demand from the grid on a gross basis. That way, you would get the same efficiencies as the entire grid, and not have to control anything (except, turn on or shut down the generator when instructed to do so). All the system controls will be out there on the grid.

      Unfortunately, in Virginia that’s illegal. You can sell your generators’ gross output to PJM, yes; but you cannot buy directly from PJM, you have to deal with the locally franchised utility, DVP. Now DVP (or in your case, a co-op customer of DVP’s) is located within PJM and buys much of its power from PJM but you won’t see the PJM wholesale price; instead, you pay an averaged annual price (with markup) approved by the SCC and based on all of Dominion’s and your co-op’s sources, plus separate charges for the delivery system and billing. At one time you still had the legal option to buy from competing retail suppliers of electricity, and DVP and your co-op had to deliver the competitor’s electricity over their wires; but that state of affairs, called “retail access,” was largely repealed by the GA.

      So we have a hybrid arrangement here. You can sell your generator’s output to the grid. You have to buy from the local franchised utility. The only alternative is keep the output of your generators for yourself, if it’s lower cost than the price to buy from the utility. Unfortunately, most of the time, for conventional generation, it won’t be lower cost; but solar is cheap electricity so it makes sense to keep that for yourself if you can build solar generation on site.

      So now you have to have controls over your solar setup on-site, but controls over your net demand for power from the grid are take care of by PJM, or their intermediaries, DVP and your co-op. This is inefficient, and burdensome, but if you want the cheapest electricity, that’s what you do.

      When should you run that gas generation of yours? Well of course, we’re assuming the sun has gone down, and if your retail supplier can beat the cost to you to generate after dark for yourself, you should buy instead of generate. There’s also the cost of controlling all this on-site yourself; it’s worth something to let the retail supplier handle it all. It might help if, for example, that retail supplier would sell electricity to you under a special “solar backup” rate schedule devised to give you a price even closer to the grid wholesale price of electricity during those backup hours of the day; but the retail suppliers in Virginia have little incentive to offer that, these days.

      As for controlling all this, as long as you are talking about the grid, you are talking about PJM, and its master control center is in Pennsylvania. DVP effectively runs its operations center outside of Richmond as a satellite of the PJM control center but also as the center for all DVP distribution switching. Your co-op has its own control center, also working closely with the centers run by DVP, ODEC and PJM. They match generation to load as directed by PJM’s dispatch, and handle the switching on their parts of the grid. If you ALSO want to do your own local controlling of your own generation, fine. It will be very inefficient to operate in this manner; but the resulting price of the electricity you generate for yourself may, in some hours, be lower than the retail price from your local utility. Hopefully it will be lower over enough hours to be worth all the equipment cost and operating hassle; but DVP is betting that it won’t be.

      • Acbar – good commentary. I do think what is missing from your discussion is the duty to serve. A public utility generally has the duty to serve all potential customers within the franchise area. Some are cheap. Some are average. But some are expensive to serve. Also, some won’t use enough commodity to recover the fixed costs.

        If a utility has a duty to serve, allowing customers within the franchise area to purchase electricity from third parties, unless the third parties are “guilt suppliers” that charge higher prices because they are green, which, in turn, depresses demand, the utility risks loosing the higher-spending, lower-cost customers. Uneconomic bypass. Either rates have to go up for the higher-cost and/or lower-volume customers, or we have theoretical death spiral. I’m not a big fan of Dominion, but it needs to be treated fairly.

  7. good comments, informative, and much appreciated

    but some additional observations:

    combined-cycle gas plants now come in hybrid versions so they can ramp up quick as peakers then transition to more efficient combined cycle.

    ” A good example is Sloe Centrale, a 2 x 430
    MW F-class single shaft plant in the
    Netherlands, where 30-minute start-up
    times ….

    High availability and reliability power
    plants, such as combined cycle units, are
    required in order to compensate for
    fluctuating renewables.”

  8. I think it would be financially “tough” for residential to do their own power right now but it’s not keeping commercial, municipal or even Va Tech from doing their own on-site power…

    Here’s Bedford:


    Bedford Town Council approves solar farm project

    Next summer, the town of Bedford could be seeing a brand new solar farm installed on town property — a first for any municipality in Virginia.

    The Bedford Town Council voted unanimously to approve the solar farm, which would be located on 20 acres on Draper Road, adjacent to the old town landfill.

    The developer, O2 emc, based out of North Carolina will develop the utility scale solar farm in partnership with the Bedford electric Department in the town.

    “We have been receiving proposals for over two years,” Wagner said the town received about 10 proposals. “At some point we mutually found each other. We selected them based on their performance and qualifications as well as passion for the project and lowest price.”

    That price comes in at the tune of 6.19 cents per kilowatt hour.

    The town is going to acquire a power purchase agreement or a PPA with O2 — a contract between the two parties where the developer generates the electricity and the buyer purchases it.

    It will be a three megawatt solar farm and will capture energy from the sun and turn it into electricity, which will be exported onto the town’s electric distribution system. All of the power generated will be put on the grid and can be used by local homes and business..

    Stephens said the solar farm is an “attractive” use of the land, which has been affected by the landfill for a number of years.

    “The town is happy to find a productive use for that land over a long time period,” he said. “It won’t create any noise or smell and doesn’t require services of water or sewer. It is one of the best possible uses of the land that’s been sitting there pretty much unused.”

    The rows of solar panels will be around 15-feet high at most and would track sunrays and run north to south and follow the sun east to west generating the highest amount of power from the lowest amount of space, Stephens said.”

    this is going on all over Virginia now from the Eastern Shore to Chesapeake to Louisa!

    the question is – how much will those other options cost – and will that cost be so high so as to wipe out the cost advantages of harvesting solar.

    well – if the grid price for electricity is going to be higher to compensate for more local site solar – all that is going to do is drive people further towards more even more local site changes to further reduce their costs of higher priced grid electricity. It will do the opposite of driving them away from solar.

    It will include any/all available options – more solar, higher efficiency replacement equipment, on-demand water heaters, etc gas-powered backup generators – which by the way – fire up autonomously, in about 2 minutes..or less when the grid drops.

    • LG, I agree, there are quite a few who are doing all this in spite of the obstacles. Large institutions and governmental entities are the best able: they are most likely to be located on a “campus” of contiguous land, with sufficient facilities to merit hiring an energy manager who can focus on managing all the complications of electrical generating on site, switching and buying/selling with the grid, operating and maintaining all the equipment, forecasting daily operating schedules, and staffing it all. This is not a task for the harried homeowner in his ‘spare’ time.

      But if a homeowner has the time and wants to try, it’s possible. As TomH reminds us, there are companies out there that will come in and do a turnkey solar project in your back yard if you agree to lease the equipment and purchase the power output from them. The State’s utility regulators are fearful of them making an undue profit from these deals, and Dominion views them as a threat to its exclusive franchise to sell electricity at retail; but for the busy homeowner it’s a partial solution to the systems management issue. As for the “undue profit” earned by these solar PPA providers — presumably the homeowner wouldn’t sign a solar PPA deal unless it saved him something on his electric bill relative to DVP’s charges, plus, his family gets solar-power bragging rights.

      You make an insightful statement: “if the grid price for electricity is going to be higher to compensate for more local site solar – all that is going to do is drive people further towards more even more local site changes to further reduce their costs of higher priced grid electricity.” Yes, but! Look carefully at what you just said. The “grid price for electricity” is not what the retail customer pays; he must buy from a middle-man: DVP or a co-op; they in turn pay the grid price for electricity, then tacks on a charge to distribute that power. If the grid price becomes “higher to compensate for more local site solar” that is going to be as a result of the reduced efficiency of the electric grid operating primarily as backup to solar and wind rather than as the primary source of power around-the-clock. But that’s not a bad thing. First, you have to acknowledge that there’s a huge offsetting benefit to everyone built into the solar tradeoff; the cheap solar comes with higher-priced backup but the combo is still cheaper than the alternative would be: less solar and more low-cost 24-hour baseload generation. Second, there’s the improved efficiency of new fast-cycling gas turbines and homeowner load management (e.g., grid-controlled water heaters), as you point out. Third, there’s the prospect, sadly unrealized to date, of decent battery technology to time-shift some of that solar power into after-dark hours. Add it all up and it means there will be competitive pressure on the likes of DVP to pass through the lowest grid price it can for all that solar-backup power, using all the tricks at its disposal such as high-efficiency cycling and batteries, rather than have that institution or business or homeowner install its own backup generators. And I think DVP will win that competition because, ultimately, the grid can provide that backup power more efficiently than a bunch of smaller, stand-alone fossil-fueled units can, even with middle-men involved. All those small backup units pose the greatest real threat of electric supply inefficiency.

      Batteries are a bit of a wild card because their efficiency potentially isn’t a function of their size, but we have a long way to go before we fully understand how cheap, scalable batteries will change the grid. I think it’s likely that batteries — like solar generation itself — will prove to be efficient and cost-effective both for distributed installations and for massive central-station-type installations.

      Visions of the future! Hope I see some of it.

  9. good conversation.

    Let me revise my statement – ” “if the grid price for electricity is going to be higher to compensate for more local site solar”

    if the price of off-site electricity that a homeowner would have to purchase to supplement whatever their local site could not provide…..

    the homeowner won’t care where that power comes from or who is supplying it .. they wonj’t care if Dominion is in a war with other competitors or what the SCC does or not – only what the bottom line cost is at wire coming into their house.

    I equate this to modern-day behaviors with regard to to other things – for instance, people USED to have loyalty to come brand of gas – no more – you go for the lowest price at a reputable retailer…

    ditto – Toyota versus Kia/Hyundai ..

    electricity is a commodity – in the minds of consumers… they don’t understand the truly complex nature of it – from neither a technical nor a business perspective.

    people have changed in other ways.. they don’t like paying a lot of dollars to fill up their car and they don’t like gawd-awful monthly electric bills and it’s more than just money – it’s the idea that it’s wasteful and bad for the environment…

    I just think the momentum is in that direction and the technology evolving rapidly towards not only solar but turn-key – autonomous systems where the homeowner installs it – and it operates without intervention while providing hourly/daily status reports just to let the owner know the system is operating as designed.

    I think all the ruminations going on with Dominion – is going to be OBE by the advancement of technology and apparently the many loopholes that already exist for other businesses to get around some of Dominions would-be regulatory-assisted blockades over the longer run.

  10. here’s another one:

    35-megawatt solar farm proposed for farmland in Norge

    …….. p to 35 megawatts in Norge, according to a letter sent by the developer to homeowners around the site.

    The solar facility is proposed to be built on a 223-acre site…. power up to 7,000 homes in the area

    you start multiple projects like this across the state – and they’re going to add up to a sizeable deployment of solar… far, far more than Dominion was talking about doing……..

Leave a Reply