Batteries for Home and Business Storage: Transforming the Customer Side of the Meter
Introduction: Energy Storage, the Electric Grid, and a Low-Carbon Future
If anyone could instantly generate electric battery buzz, it is Tesla CEO Elon Musk, and that is exactly what he did with his watershed announcement on April 30, 2015. In presenting to the public Tesla’s plans for conquering the home and utility battery business, Musk stated, “[o]ur goal here is to fundamentally change the way the world uses energy.… We’re talking at the terawatt scale. The goal is complete transformation of the entire energy infrastructure of the world.”1 It was not April Fool’s Day, and Musk was not being hyperbolic. According to expert analysis presented in subsequent media coverage, this was not just another product announcement from the latest popular flavor of a Silicon Valley company, it was an achievable vision. And at the heart of Musk’s vision of a zero carbon energy future was the electric battery.2 This transformative change would allow a home, a business, or perhaps a whole utility to leap toward the zero carbon economy. The full story behind this “big battery announcement”—and what it means for a clean, more distributed grid leading us on a path off fossil fuels—is complex, perhaps enigmatic, but clearly captivating.
Why are batteries gaining so much attention in regard to the potential for a low-carbon future? First, as Amory Lovins points out, “Electricity has become the connective tissue of the information age.…” It is “clean, efficient, precise, and flexible, ensuring that major infrastructure systems including communication, building industry, and even transportation will continue to shift to electricity as an energy supply source of choice.”3 While electricity continues to become the energy supply source of choice, historically its use has been limited by the fact that it could not be readily stored. Electric energy supplied to the electric grid must be constantly balanced with electric energy consumed. Our alternating current (AC) power grid must be maintained in constant equilibrium—with the electric current changing its direction of flow back and forth at a controlled frequency of 60 cycles per second (50 cycles per second in Europe). The storage problem is amplified given that the system must respond to generation or load changes in thousandths of a second rather than minutes or hours.4 Further complicating the challenge is the fact that, unlike traditional power plants, our most feasible low-carbon generation options, wind and solar power, are intermittent resources. The electrical output of these renewable resources is not directly controllable and the output can rise or fall dramatically, literally as the wind blows or clouds appear and again disappear.
Another appeal of storage stems from the growing recognition that severe weather events are increasing due to climate change and they pose challenges to a centralized rather than a more distributed grid. Batteries increasingly offer an alternative for backup power or to support local microgrids when the centralized grid is unavailable.
With this scene set, the electric battery charges forward: it’s a distributed electric resource that is as clean as the electricity it stores and that has the potential to rapidly transform both our homes and businesses as well as the electric power grid itself. The electric battery and the storage it provides offers our society a critical tool for achieving a true low-carbon economy. It might, in fact, be our best hope for quickly transitioning off of fossil fuels and avoiding the worst fates of climate change. Understanding how we achieve cost-effective carbon reduction for our electric system requires a deeper dive into the opportunities provided by batteries. Batteries offer a potentially complex array of services for our electric grid. In fact, in their Electricity Storage Handbook, the Department of Energy (DOE) and the Electric Power Research Institute (EPRI) describe a full 18 different storage applications and services.5 In this chapter, we discuss the battery storage applications for the home and business, and in Chapter 6, we discuss grid-level storage applications.
Tesla’s Big Battery Announcement
Musk, in his announcement, offered a two-prong approach to the electric storage challenge. First, he led with an enticingly attractive and increasingly affordable battery for the home and business. The Tesla Powerwall (see Figure 5.1) is a rechargeable lithium-ion battery for the home or business that was originally offered in 7-kWh and 10-kWh options.6
These “behind the customer meter” distributed energy resources could provide the customer with backup power when the local electric grid is down or give the consumer the option to manage either home solar generation or off-peak electricity to generate value from energy arbitrage. According to Tesla, the batteries could provide continuous power of 2 kW, peak power of 3.3 kW, operate between −20°C (−4°F) and 43°C (110°F), and result in a 92 percent round-trip efficiency in regard to the energy stored and the energy available for future consumption. At the price of $3,500 ($3,000 for the smaller 7-kWh version) before installation, including a 10-year warranty, suddenly many began to believe that the common electric battery might be the missing piece to the low-carbon puzzle.7 Full purchase and installation cost, including the cost of an inverter, will likely be double the reported list price for the battery.
Figure 5.1 Tesla Powerwall home energy storage system. (Courtesy of Kevin B. Jones.)
The second prong of the Tesla strategy was represented by the announcement of the Tesla PowerPack. The PowerPack, similar to the Powerwall, is a rechargeable lithium-ion battery intended for utility-scale applications beginning with a 100-kWh version, costing approximately $25,000, and that could be stacked to the gigawatt level.8 The PowerPacks can be grouped to scale from 500 kWh to over 10 mWh and can be utilized for two-hour or four-hour net power discharge.9 Nothing was necessarily breathtakingly new about this technology; instead it was the breadth and future affordability of the vision that attracted both consumers and the media. Musk postulated the outer bounds of this strategy, noting that the world would need 2 billion PowerPacks to be completely carbon neutral.10 This worldwide vision would require what some might term an incredibly ambitious level of zero carbon-generating resources to first produce this energy, which would in turn be stored by the PowerPacks. In the third quarter of 2016, Tesla announced the PowerPack 2.0, with increased energy density compared to the earlier version, and noted that it had already begun shipping it to projects, including the 52-MWh solar plus storage project with Kauai Island Utility Cooperative (discussed in Chapter 8).11 We discuss the PowerPack and the opportunities to use electric batteries for electric grid storage in Chapter 6.
Sonnen: German Engineering for Behind-the-Meter Storage
While Tesla Energy may have gained the headlines in the United States, no discussion of distributed battery storage would properly reflect the current environment without examining the progress of Sonnen, the German storage company that has aggressively entered the U.S. market and is the residential storage leader. According to Boris von Bormann, the company’s then chief executive (von Bormann later left Sonnen for their new competitor Mercedes-Benz Energy Americas, LLC), “Elon Musk has the name.… Yes, we are the leader.”12 For the U.S. market, Sonnen is pricing its sonnenBatterie Eco4 system starting at $9,950, which includes a 4-kWh, 3.0-kW lithium-ion battery, an Outback Power inverter, and intelligent control software. Sonnen rates the Sony-manufactured batteries their system utilizes for 10,000 cycles and warranties them for 10 years; thus, this is a significantly greater cycle life than the Tesla units. It offers the sonnenBatterie Eco in 2-kWh increments, with the Eco6 stepping up to 4 kW and the Eco12 reaching 8 kW (see Figure 5.2).
The Eco16 can integrate with 8 kW of solar and store 16 kWh of solar energy. It comes ready to plug into your solar system at a cost of $22,800. Three of the Eco units can be stacked together for a maximum of 48 kWh and can work in backup, max solar self-consumption, and off-grid modes. The Eco12 and above units provide premium home backup service potential, albeit at a premium price compared to a home generator. It is almost an apples and oranges comparison evaluating the costs of the Sonnen and Tesla products. While Sonnen has a significantly higher list price, it comes prewired at the factory with an integrated inverter and has a significantly longer cycle life than the Tesla Powerwall.
In mid-2016, Sonnen launched the sonnenBatterie Eco Compact (see Figure 5.3). The streamlined unit is designed to increase a home’s grid-tied solar consumption and retails for $5,950, 40 percent of the cost of Sonnen’s comparable product.
This all-in-one residential solar battery solution provides 4 kWh of energy storage and includes the standard Eco unit’s cycle rating and warranty, but it cannot provide off-grid backup service.13 According to Brett Simon, energy storage analyst at GTM Research, the Sonnen Eco Compact “is competitive with other offerings in the U.S. residential market today” at $1,500 per kWh prior to installation when taking into account its longer cycle life.14
Figure 5.2 Inside the Sonnen Eco8 home energy storage system. (Courtesy of Kevin B. Jones.)
In early 2016, Sonnen shipped its 10,000th battery, unveiled a new North American headquarters in Los Angeles, and introduced the sonnenBatterie Pro line for commercial and industrial customers, ranging from 24 kWh to 96 kWh, which it planned to begin producing by year-end at its production facility in San Jose, California. The company also claims that its systems meet all the new tariff obligations in Hawaii and that their integrated system will provide Hawaiian customers financial returns in as little as 6.5 years thanks to Hawaii’s relatively high electric rates and favorable solar resource.15 According to von Bormann, “we’re looking at the whole ecosystem for renewables, plus storage and electric vehicles and the smart home—where all of the infrastructure talks to each other and is optimized for each other, where you can really create an efficiency package that allows you to self-supply your energy and smartly manage it with control technology.… I think this is really where [the electricity sector] is going in the years to come, and we’ll see significant breakthroughs in the five- to seven-year range.”16
Batteries for the Home and Business: End-Use Storage Opportunities
While companies like Tesla and Sonnen are delivering exciting new products, what services can these battery storage systems provide for customers? In considering the opportunities provided by batteries for home or small business uses, our discussion will center on coupling these battery units with distributed solar photovoltaic (PV). While distributed wind or microhydro present similar opportunities, solar PV is and will continue to be the most ubiquitous distributed renewable resource. At the home and business level, opportunities are primarily for battery storage on the customer side of the meter (also known as “behind the meter”). As with net-metered solar, these customers are located on the utility’s distribution system and the behind-the-meter batteries will often be invisible to the utility. In other words, the utility’s meter will only record what the customer is consuming on a net basis and will not likely have the data to confirm whether the customer is offsetting some of its consumption with battery or solar power. When it comes to battery storage for the home, safety and simplicity are vitally important; thus, lithium-ion and lead acid batteries are both viable technologies in the near term and are certainly familiar to the consumer.17 It is also important to understand that a battery storage system requires more than just the battery itself to effectively interact with our AC grid. In addition to the battery, the system would also require a monitoring and control system and a power conversion system. The monitoring and control system would ensure safety and optimize performance of the system, while the power conversion system utilizes bidirectional inverters to convert DC power from the battery to the AC grid.18 As a result, battery storage for the home and business presents a number of different end-use opportunities, which can be directed by the customer with little utility involvement.
Figure 5.3 Sonnen Eco Compact. (Courtesy of Kevin B. Jones.)
From the utility perspective, all that the customer needs in order to enable the electric battery is a solitary smart meter (one per customer account) and some well-structured utility rates.19 Smart meters are increasingly being deployed to utility customers and, as of 2015, there were more than 65 million smart meters in place nationwide, covering half of all U.S. homes.20 Smart meters replace old analog technology with a computer chip, storage, and two-way communication capabilities between the utility and the customer meter, which allows the utility to implement more dynamic utility rate and pricing options.21 With a smart meter and smart rates in place, behind-the-meter batteries in tandem with the customer’s solar panels can serve a variety of useful functions supporting a low-carbon future.
Dynamic Pricing and Solar Energy Arbitrage
With a smart meter and smart pricing in place, the battery can become an effective tool for energy arbitrage and electric bill management.22 “Buy low and sell high” is the mantra of the market arbitragers that encapsulates the theory behind this approach. While electric customers by default are often charged a flat customer rate no matter whether it is morning, noon, or night, actual electric costs vary by time of day based on basic supply and demand. The real electric costs are determined by both the current usage of customers on the system and the resources that are available to serve that customer load in real time. The continued rollout of smarter technology on the electric grid is helping utilities and rate setters move toward greater harmony between real electric costs and the prices that customers pay. More and more customers, either by choice or by mandate, are increasingly facing prices that vary based on the time of day, also known as dynamic prices. These time-varying or dynamic prices may be as simple as separate peak and off-peak pricing blocks, with higher peak prices during the daytime (particularly in late afternoon hours) when electric usage tends to be higher, and lower off-peak prices during the night when household and business activity tend to quiet down. Generally, these fixed tier pricing plans are called time-of-use rates. These time-of-use rates are intended to send customers a price signal to be more energy efficient during peak periods by shifting as much use as possible to the lower-cost off-peak periods, thereby reducing peak demand on the electric system and the customer’s electric bill.
With the advent and widespread deployment of smart meters and associated communications technology, these pricing plans have increased in sophistication, leading to the addition of a dynamic pricing option known as critical peak pricing.
Under critical peak pricing (see Figure 5.4), a peak and off-peak time-of-use rate is supplemented by a significantly higher critical peak price during a finite number of hours (e.g., during afternoon hours for the 10 hottest days of the summer). Critical peak prices are often set from 3 to 10 times higher than standard rates.23 Studies have shown that these critical peak prices elicit meaningful customer reductions in electric usage, known as demand response.
Figure 5.4 Illustration of critical peak pricing. (Courtesy of Kevin B. Jones.)
While historically, dynamic pricing was intended for customers to modify their energy usage patterns, the continued growth of distributed energy resources (DERs) behind the customer meter—such as solar panels and battery storage—increasingly allows customers to take advantage of their full suite of DERs by also selling energy back to the grid when price signals are sufficiently high. While flat customer electric rates (along with flat net metering credits) provide no financial incentive for storing behind-the-meter energy for a future time period, time-of-use and critical peak pricing plans are a different story altogether. With peak and off-peak price differentials sufficiently high, customers will be able to cost-effectively store solar energy during lower price periods in order to utilize it during higher price periods, thereby reducing their utility bill and providing a revenue stream to support the cost of additional battery storage. According to Europe’s battery industry trade group, small- and medium-sized users (like a typical household or business) with battery energy storage as a component of their smart systems will see financial benefit from this time shift for self-consumption. In practice, these uses would typically require batteries to be available for one to six hours for one cycle a day.24 A cycle is considered one charge and discharge of the battery.
As more and more renewables are flowing into the grid, time-of-use rates and time periods are shifting to adjust and make for their most efficient use. In states such as California, which lead in both renewable energy and dynamic pricing, time-of-use pricing plans are adapting to the new realities of abundant renewables on the grid. They are modifying their dynamic pricing plans by shortening peak daytime hours and focusing them more acutely on the 4:00 p.m. to 9:00 p.m. peak hours. This is because the record-setting levels of solar PV now on the California grid are knocking down the midday net loads and creating more of a price valley for real electric costs rather than the midday peaks that the energy system has historically yielded (see Figure 5.5). The new realities of distributed solar production now offer midday opportunities for strategic electric sales.
Figure 5.5 As more solar is added to the grid over time, the net load in the middle of the day continues to decline, creating both opportunities and challenges for the grid. (Courtesy of CAISO.)
On the flip side, late in the afternoon is becoming an increasingly critical peak demand time, as the sun begins to go down and solar generation declines at the same time that people return home, turn on lights, and power up computers, televisions, and appliances such as air conditioning. This change is creating new opportunities for solar energy arbitrage by providing price incentives to sell stored solar energy after the sun has begun to fade in the western sky. The combination of growth of renewables and the development of home and office battery storage increasingly allows the electric grid to transition to a low-carbon reality. Home and office batteries will be able to store solar energy earlier in the day in order to more cost effectively serve higher price peak demand late in the day when solar generation is dissipating. Electric cars will also increasingly be able to take advantage of lower midday electric demand and prices, and buy energy off the grid affordably to support the evening commute. The electric battery utilizing today’s grid technology and pricing plans will increasingly allow the consumer to buy energy for low prices off the grid (charging batteries for home, office, and transportation) and then sell it back at higher rates to the grid, thereby mitigating peak afternoon demand, all the while encouraging a market transition to a lower carbon energy system.
Another option for batteries to generate savings for customers as we transition to a lower carbon economy is by reducing demand charges. What’s a demand charge? While utilities collect much of their revenue from their customers through hourly volume-based energy charges (cents per kWh), they also often have in place what are known as demand charges for their larger commercial customers. Demand charges are meant to send a price signal to customers of the cost of the customer’s peak hourly demand on the electric system. When a customer places a peak demand on the system, in order to reliably serve the customer, the utility not only must instantaneously serve that customer’s load, but the utility may have to plan in advance to construct poles, wires, transformers, and generating plants that will be in place for many years. As a result, utilities often have monthly demand charges ($ per kW) that lock in place the peak demand for a seasonal period (often utilizing both summer and winter seasons) based on the highest averaged kW demand for a 15-minute interval. For example, if a commercial customer were to set its peak usage at 200 kW on July 15, then the demand charge would be applied to the 200-kW peak usage for a six-month period going forward (often referred to as a demand ratchet) until a new peak demand was set. Once again, as new distributed energy resources, such as solar and electric batteries, are integrated into the system, the game can change.
While historically the only choice facing utilities was to reduce customer demand (demand-side management) or build new delivery and supply resources to serve growing peak demand, with the customer paying a higher monthly demand charge for their increased peak usage, now storage is increasingly becoming a useful and meaningful option. Consider the California example described earlier: growing solar resources are creating opportunities to store low-carbon energy during midday valleys in order to later inject the energy into the system to offset the same customer’s peak demand. Using the electric battery for peak demand management is offering revenue opportunities to further support battery storage deployment. A Goldman Sachs analysis estimated a five-year payback for a battery storage system for a hypothetical commercial customer in California based on revenues from time-of-use and demand charge reduction alone.25
While demand charges have historically been mostly utilized for commercial and industrial customers, some utilities have begun to explore implementing demand charges for residential customers, particularly those with net-metered solar PV systems. As noted earlier, the growth of solar PV has been pushing peak residential demand periods later into the day. While solar customers may offset their kWh usage on a monthly basis, they still likely have significant peak demand on the system, particularly when solar drops off later in the afternoon. As a result, some utilities would like to send these customers a price signal related to their peak demand usage and thus have proposed residential demand charges. While residential demand rates have often been opposed by the solar industry, they create new revenue opportunities for the storage industry. Additional time-sensitive rates for residential customers, including demand charges consistent with the Goldman Sachs analysis for commercial customers, would strengthen the business case for residential battery storage. However, it should be noted that particularly for smaller commercial and residential customers, there is significant concern regarding demand charges.26 Residential customers have much more diversity in their usage, and with the advent of smart meters, pricing for these customers can be more targeted toward well-defined peak and off-peak periods, such as with critical peak pricing plans. According to Jim Lazar of the Regulatory Assistance Project, time-varying rates are more equitable, reduce bill volatility, and are better understood by customers.27
Most residential and commercial customers who have installed solar PV have installed and operated their systems under their respective state’s net metering rules. Net metering, in its simplest form, allows a customer to benefit from behind-the-meter solar PV through billing the customer for only their net energy use on a monthly basis. During some hours of the month, the gross solar generation will be less than the customer’s gross consumption, while during other hours the solar energy generated will exceed the customer’s consumption and can result in net energy being injected into the utility’s grid. Generally speaking, under net metering, a utility compiles all of the monthly generation and load together and generates a net monthly bill. Forty-one states plus the District of Columbia and three U.S territories have some form of net metering. Beyond the basic similarities of state net metering policies, there is a multitude of differences.28 In the vast majority of states, net metering rules make the customer indifferent as to whether they actually consume the solar energy generated by their system. In other states, the customer may be compensated less for energy they export to the grid. In some states, there may be a modest economic incentive to utilize battery storage in minimizing the export of electricity to the grid over some time in order to maximize the financial benefit of the solar energy the consumer generates. While some people may find some personal principled reason for solar PV self-consumption, most state net metering laws today make the consumer content with allowing the grid to manage its excess solar generation rather than invest in additional battery infrastructure. Over time, if states trend toward reducing the incentive for electricity exported to the grid, the economics of using the battery to manage home consumption will improve, and products such as Sonnen’s Eco Compact battery are specifically designed to provide this service.
Storage as a Backup Energy Resource
As dynamic pricing and demand charge revenue opportunities support a stronger business case for home and office battery storage, customers will also increasingly be attracted to batteries for their use as a backup energy resource. Today, batteries play an important supplemental power role on a daily basis for off-grid electric uses. The lead acid battery has proven to be a reliable and cost-effective choice for customers who do not have convenient access to an existing electric grid. For off-grid systems, battery storage provides sufficient stored energy to allow energy-conscious consumers to provide modern conveniences. In states with net metering programs, solar customers who do not face expensive line extension bills in order to be tied to the grid, usually find that using the grid as their “battery”—to both take excess generation from the solar system and provide supplemental power when there isn’t sufficient sun—is the most cost-effective alternative. Thus, including backup battery storage with grid-tied systems has not recently been either cost effective or popular.
The challenge with the battery is the limited total energy available to the home or office prior to a lengthy recharge. For example, one 7-kWh Tesla Powerwall unit would provide only from two to eight hours of backup power for an average home (Tesla removed all references to its 10-kWh battery from its Web site soon after introduction and instead is relying on its 7-kWh battery, which is better optimized for daily consumption rather than backup storage). Furthermore, the Powerwall is only rated for 2 kW of continuous use, and while that will likely power the household lights, refrigerator, Internet, and television for a few hours, it is not designed to power major appliances at the same time. Your air conditioner (1.5 kW), electric range (1.2–2.1 kW), microwave (1.3 kW), toaster (1.5 kW), electric heat pump (2.5 kW), or electric clothes dryer (5.4 kW) would overpower its capacity or drain its battery more quickly.29 While multiple batteries can be stacked to provide additional power, this option very quickly becomes cost prohibitive. According to analysis by Goldman Sachs, the cost of a Tesla Powerwall unit sized to provide 48 hours of backup power would be over three times as much as a conventional diesel generator. A home or office battery performing bill management services can provide some backup power, but electric batteries still face financial barriers as reliable sources of backup power, particularly when compared to the cost and performance of home generators.30 According to Greentech Media, a 2016 survey by Enphase (a microinverter company) of 566 homeowners in California, Hawaii, Massachusetts, and New York who have or are looking to get solar systems found some interesting results. While 20 percent of these homeowners already had a backup generator, of the remaining households, 50 percent said they were interested in a backup generator. When probed further, these respondents both wanted multiple days of backup power and wanted to pay less than $10,000, posing a service and cost challenge for existing battery companies.31 Unlike generators, backup battery systems offer other sources of revenues from managing time-of-use rates and optimizing net metering tariffs, and have a clearer future path for cost reductions than fossil-fueled combustion technology.
With battery economics improving—considering both declining costs and increasing revenue generation opportunities—batteries for home and office backup service are generating newfound interest. Batteries can serve effectively as a backup resource, since they can react instantaneously and provide excellent power quality. Sonnen’s Eco12 battery with 8 kW of capacity begins to provide a meaningful backup opportunity for most residential homes, but at a retail cost of $18,750. Apparently realizing the limitations of its initial Powerwall product, Tesla announced a reconfigured Powerwall 2 in late 2016 that was more like the Sonnen products, including a bigger, more powerful battery (14 kWh, 5 kW continuous load) and integrated inverter, both housed in a more boxy cabinet (weighing 269 lb) rather than the sleek design of the earlier battery-only unit, for a reported price of $5,500 prior to installation.32 In addition, unlike fossil-fueled generators, electric batteries do not need regular oil changes, monthly cycling to charge the starter battery, produce little noise and no fumes, have lower running costs, and can remain in attractive cabinets stored in the garage or basement—even while operating.
Behind-the-Meter Pricing Plans and Services
As we have outlined above, electric utilities across the country—with California leading the way—have advanced time-of-use pricing and other rate plans that are beginning to form the economic basis for battery storage. Time-of-use pricing plans, like those offered by utilities in California, are beginning to send price signals that improve the business case for electric batteries, particularly in regions that also host significant amounts of distributed solar. We will examine case studies of how four utilities have responded differently to the distributed generation and energy storage challenges their systems are facing.
First, we will look at the Salt River Project’s implementation of a demand charge for solar customers. Perhaps nowhere has a pricing plan received so much opposition from the solar industry as the Salt River Project’s (SRP) Customer Generation Price Plan, which incited one leading solar company to sue SRP in U.S. District Court. We will take a more in-depth look at the SRP plan, since the workings of demand charges are an important, yet complex opportunity for technologies such as battery storage. Next, we will consider how another leading public power utility, the Sacramento Municipal Utility District (SMUD), is experimenting with batteries, solar PV, and time-of-use rates. Then we will look at a very different pilot at Green Mountain Power where this Vermont investor-owned utility is offering customers Tesla Powerwall batteries for a monthly fee, with no upfront cost. Finally, we will explore how the battery company STEM is the leading behind-the-meter supplier to Southern California Edison’s (SCE) storage procurement.
Case Study: SRP—Innovative Demand Rates for Solar and Storage?
The Salt River Project, headquartered in Tempe, Arizona, was formed in 1903, even before Arizona became a state. Today, SRP is one of the nation’s largest public power utilities, with approximately 1 million customers. SRP’s board of directors, which manage the utility and approve its rates without the need for state regulatory approval, are elected by the landowners in its service territory; this arrangement stands out from most other public power utilities, which are controlled by the customers or their representatives. SRP is considered an early adopter and industry leader of time-of-use rates, including innovative electric prepay service and electric vehicle charging rates.33
In 2014, SRP sent shockwaves through the Arizona solar industry when it proposed to radically restructure how it charged customers who have small-scale distributed generation, particularly rooftop solar PV customers. SRP, like many utilities that allow small-scale distributed generation through net metering tariffs, has rates that recover a substantial portion of utility fixed costs in per kWh usage charges. According to an SRP consultant, when “actual usage for any group of customers is significantly different than estimated usage for the class as a whole, as would be the case when customers install DG units, then the utility rates no long recover the full cost of service for those customers.”34 According to SRP’s analysis, 73 percent of its costs are fixed, and while solar customers significantly reduced their bills, they did not proportionately reduce their consumption of peak demand.35 SRP’s response was to propose a Customer Generation Price Plan (Rate E-27), which would raise fixed charges to $32.44 per month (compared to $20 for the time-of-use plan), and would include a time-of-use price plan with the lowest rates of any such price plan (about 50 percent lower than other options it offered). SRP would maintain net metering while significantly reducing net metering’s financial incentives. SRP’s proposal uniquely differed from other utility reactions to net metering by including a new demand charge (based on the 30-minute interval during on-peak hours when the home used the maximum amount of electricity) for residential customers. Peak hours during the summer period are defined as weekdays from 1:00 p.m. to 8:00 p.m. and during the winter are weekdays from 5:00 a.m. to 9:00 a.m. and 5:00 p.m. to 9:00 p.m. According to a report by one of SRP’s consultants, the plan is an improvement over other recent proposals by utilities because “[i]nclusion of the demand charges allows DG customers to tailor rooftop solar designs in ways to increase their savings, while simultaneously increasing the value of such systems to SRP. DG customers can also reduce their levels of instantaneous demand (through, for example, demand interlocks and/or storage), which can yield both significant additional bill savings, and additional value to the utility.”36 As ultimately approved, the SRP plan grandfathered the 15,000 existing solar customers, allowing them to keep their existing rates for up to 20 years.37
SRP’s plan faced significant opposition with a number of constituencies, who argued that the utility’s plan was unfair to solar customers. According to the newspaper The Arizona Republic, approval by the utilities board of directors followed “a series of crowded, contentions public hearings” that pitted “the utility’s solar customers, solar-panel representatives and environmentalist against SRP executives.”38 Following the board approval, Thad Kurowski, director of policy and electricity markets for the rooftop solar company SolarCity, stated, “SolarCity will have no choice but to challenge the decision in the courts.” According to Kurowski, SRP was “eliminating the ability to go solar in SRP service territory and doing it in a way that could not be justified.”39 On March 2, 2015, SolarCity sued SRP in U.S. District Court in Phoenix, alleging that the SRP rate was an anticompetitive act by a monopoly.40 SolarCity claimed that SRP’s anticompetitive rate plan caused solar installations to fall by 95 percent. In reality, according to industry reports, the impact was not quite so dramatic, as solar installations in the second half of 2015 fell by 75 percent compared to recent quarterly data.41
According to SRP treasurer Steve Hulet, the fundamental problem was with the utility’s pricing structure and not with rooftop solar, and that new price signals were a necessary change that could still benefit distributed solar generation. According to Hulet: “SRP has not simply adopted an increased fixed cost. The demand charge is a tool and a price signal, not only for the customer but the industry.”42 SRP chief financial executive Aiden McSheffrey noted that, “our new plan sends a price signal that [incentivizes] more efficient installations by the solar industry and behavior by the customer that maximizes the value of their solar systems.”43 According to Hulet, if new solar customers match the profile of existing solar customers, their bills will now only drop from $170 per month to $120 per month, but if they respond efficiently to the price signal, they could save more than $100 per month.44 As examples, Hulet suggested “the rooftop solar industry could put their solar units up west-facing to help more with peak demand,” while net-metered solar is typically installed to maximize annual kWh production rather than maximizing coincidence with peak demand. Hulet further explained that customers “can adopt new technology, whether it’s load controllers or smart thermostats or battery technology, and change their behavior to respond to those price signals.”45
Ravi Manghani, an energy storage analyst with the independent third-party energy consulting firm GTM Research, agreed with SRP’s logic and noted, “the Salt River Project ruling could open new opportunities,” especially involving battery storage.46 Manghani explained that one-third to one-half of commercial and industrial customer power costs are from demand charges and thus, the most economic use of battery storage is in these applications; residential demand charges and related reforms may work similarly to boost home energy storage.47 Solar Grid Storage CEO Tom Leyden agreed that “solar plus storage responds to the demand charge technically, whether or not it works economically.”48
Under rate plans such as SRP’s, solar-plus-storage may be economically superior to solar-only scenarios; however, it is yet uncertain whether the battery storage industry will be a cost-effective solution prior to the industry achieving further economies of scale in production. Joyce Mclaren, with the National Renewable Energy Laboratory, further cautioned that “an average household can’t do the analysis it’s going to take in order to figure out how I can make this economical for me.”49 She believes that “a household is going to have to re-learn how they’re going to manage their energy usage to respond to the cost signal.”50
The Rocky Mountain Institute (RMI) didn’t take an official position on the fairness of the policy in its report, “The Economics of Demand Flexibility,” but concluded, “by using three simple technologies to control three major loads during peak periods, the customer can reduce their peak demand without any real sacrifice in comfort or convenience.”51 RMI’s analysis demonstrated that “while demand charges may be bad news for solar installers in the short term, the good news is that simple, cheap technologies available today can reduce them by 60%” and, in the case of SRP, “this takes solar PV from out of the money and puts it back in the money.”52 The technologies modeled by RMI were smart thermostats and timers for EV charging and water heating, but RMI also notes that some solar installers are also “bundling a battery with new PV systems, which can also minimize demand charges, but at a significant cost premium relative to less-expensive demand flexibility solutions.”53
Not all solar installers share the outrage at SRP’s new policy. American Solar and Roofing, based in Scottsdale, Arizona, is focused on finding a solution that can make SRP’s new rate plan work with solar installations. American Solar and Roofing CEO Joy Seitz stated, “if Arizona residents want to go solar, we are committed to finding a solution.”54 According to Seitz, her company’s solution will both reduce afternoon peak loads and seek to avoid demand charges; this solution sounds remarkably similar to alternatives suggested by SRP executives. The company’s solution “involves aiming solar panels to the west, to capture the maximum late-afternoon sun.”55 The company will also “equip the homes with lithium ion batteries that can charge during midday when solar production peaks, and dispatch power in late afternoons” to offset peak demand.56 Coordinating this technology “is a controlling device made by German manufacturer SMA, which will allow the system to be programmed around SRP’s rates.”57 The total cost for the system is $30,000 for a 7-kW solar array with batteries. It will be a “$7,000 to $8,000 premium over the same sized array” without the batteries and controller. American Solar estimates that “the higher price means it will take an estimated 14 years to pay off the system with lower electricity bills” and, according to Seitz, “we foresee pricing coming down drastically.”58 SRP spokesman Scott Harrelson said that “the new price plan was designed to accommodate new technologies or consumer behaviors” and “battery storage is one of those technologies.”59
Case Study: SMUD Pioneering Smart Rates and Storage
Another utility that has made headlines for its pioneering work with battery storage is the Sacramento Municipal Utility District (SMUD). SMUD is the sixth largest community-owned electric utility in the United States and has provided public power throughout Sacramento since 1946. SMUD has been a longtime industry leader in promoting energy efficiency and renewable resources and has been conducting pilot programs to explore the benefits of combining residential solar PV and battery storage systems.60 While some California investor-owned utilities (IOUs) have attempted to block homeowners from installing batteries along with their solar panels, SMUD has been working to test how solar and storage can work together to provide benefits to the grid.61
SMUD has partnered with 2500 R Group, LLC (a joint venture between Pacific Housing, Inc. and Sunverge Energy, Inc.) to demonstrate a residential smart energy project in a residential housing development at 2500 R Street in midtown Sacramento. This demonstration project is located within 34 newly constructed single-family homes that were designed to be net zero energy. Each of the homes contains its own Sunverge Solar Integration System (SIS, see Figure 5.6).
The Sunverge integrated energy management solution used for this project consisted of three components:62
As an integrated system, the Sunverge system not only generates its own solar energy, but also stores it in the batteries, allowing the energy to be used when it is most critical. After the technology was installed on the customer side of the meter, the challenge was to figure out how to get the customer to use its technology to work with the utility smart meter-equipped grid and the SMUD demand response and dynamic pricing programs.63
Figure 5.6 Inside the Sunverge Solar Integration System. (Photo courtesy of Sunverge Energy, Inc.)
One of the biggest challenges stemmed from the fact that Sacramento customers’ demand peaks between 4:00 p.m. and 7:00 p.m., as residential customers begin coming home and turning on air conditioning and other appliances, just as the levels of solar generation begin to substantially fall off. Patrick McCoy, SMUD’s solar program planner, noted that SMUD is looking at the potential to shift the actual use of the solar energy from midday to late afternoon, when demand peaks.64 One of SMUD’s most effective tools has been its critical peak pricing (CPP) plan that it implemented under its Smart Pricing Options Pilot. The plan charges customers extra high prices ($0.75 per kWh during the pilot compared to on-peak rates of $0.28 per kWh for non-holiday weekdays between 4:00 p.m. and 7:00 p.m.) during peak hours for a limited number of critical peak days (up to 12 event days, generally the hottest summer days, between June and September, called by SMUD).65 Of the 34 households, 10 signed up for SMUD’s voluntary time-of-use rate with the critical peak pricing plan. Sunverge and SMUD collected data from the SIS system and smart meters in varying intervals. The data collected revealed that critical peak pricing customers were able to significantly lower demand and even inject stored solar energy into the grid at times. During nine high-demand days (called CPP event days) that were studied, Sunverge calculated that participants on critical peak pricing saved $445 over just those nine days.66 Sunverge incorporates smart thermostats with other whole home energy data and, according to the CEO Ken Munson, “there’s more value to unbundle when you get into the very specific nuances of taking these distributed energy resources and optimizing them in concert with the grid.”67 According to SMUD, their demonstration project would address the following use cases for the technology in order to measure the product’s viability:
According to Lupe Jiminez, SMUD’s Smart Grid research and development senior project manager, these customers on average saw a “1.35 kW saving stream daily and an additional 1.31 kW saving on our critical peak days”68, for a total average customer demand savings of 2.66 kW and a peak savings of 4.38 kW.69 In addition to demonstrating the SIS success in peak load shifting, analysis conducted for SMUD concluded that the system was successful within design parameters in demonstrating the ability to smooth out and firm PV output, successfully respond to control signals for regulation service, provide backup power capability during a simulated outage, and maintain power quality.
Today, the Sunverge technology is relatively straightforward, although times are changing. Sunverge Energy’s CEO Ken Munson doesn’t want his technology thought of as a “battery in a box” (or, more appropriately, “battery in a closet”). Instead, he prefers to think of his “Solar Integration System” as an “energy manager for the solar PV equipped home” and one that is seen as an asset rather than a threat to the utilities.70 According to Munson, “stuffing lithium-ion or flow or any kind of battery in a box and putting it in as a simple backup device is not that exciting. But putting a cloud layer with real-time energy services on top, and being able to aggregate and control a fleet of devices on the grid in near real-time—that is something special.”71 On February 9, 2016, Sunverge Energy announced the next generation of its Solar Integration System (SIS) and, according to Munson, the “newest [SIS] gives utilities and solar providers more options for linking multiple systems into a virtual power plant to shift peak load and provide other benefits to the homeowner and the grid.”72 All new SIS models offer an extensive array of grid services currently available with energy storage systems, including backup power mode, PV self-consumption mode, time-of-use bill optimization mode, grid nonexport mode, and peak shifting mode. These services maximize the efficient use of renewable or conventional energy and significantly increase the economic value of energy storage.73
SMUD’s Plans for Charging Forward?
SMUD’s plans for charging forward with battery storage technology are best analyzed in its reaction to California’s Assembly Bill 2514, Energy Storage Systems, which required California’s IOUs to implement mandatory storage procurement targets and required publicly owned utilities (POUs) such as SMUD to consider them and report back to the legislature. AB 2514 became law on September 10, 2010 when signed by then-governor Arnold Schwarzenegger, and on September 4, 2014, SMUD’s board of directors completed their assessment, passing a board resolution determining, “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.”74 Supporting the board’s action was the “SMUD AB 2514 Storage Procurement Report,” which was developed as part of SMUD’s regular integrated resource planning (IRP) process. The SMUD report made the following recommendations: (1) do not establish an energy storage procurement target for SMUD; (2) continue investing in energy storage technology assessment, demonstrations, and pilots; (3) develop staff expertise in customer services to provide assistance to customers considering installation of energy storage systems; (4) continue exploring potential development of the Iowa Hill pumped hydro project; and (5) monitor ongoing developments with energy storage procurement by the IOUs in California.75 Central to SMUD’s recommendation was that storage technologies (including battery storage) were not cost effective at this time, with the exception of large pumped hydrostorage (which we discuss more in Chapter 7).76 SMUD did not suggest abandoning battery storage technology and noted that lithium-ion batteries and compressed air energy storage were projected to have the most significant cost reductions over the following five years.77 SMUD also committed to continue demonstrations and pilots similar to the 2500 R Street project with Sunverge and committed to training its staff to be “trusted advisors” in helping their customers make third-party energy storage choices.78 SMUD’s history of clean energy leadership and openness to and experience with distributed energy resource solutions, and leading dynamic pricing options, send a clear message that this progressive customer-focused utility does not believe that the electric battery is yet poised to disrupt the current utility model from either the customer or utility side of the meter. Under AB 2514, SMUD must reevaluate this determination at least every three years.
However, SMUD might be reconsidering the economics of batteries, as evidenced by a recent decision. On February 5, 2016, SMUD announced in a press release that its board of directors agreed to not build the Iowa Hill pumped-storage project due to cost and financial risks.79 Pumped storage, discussed further in Chapter Seven, is a storage technology where lower-cost off-peak electricity is used to pump water into an upper reservoir, which can later be released during higher demand times to spin a turbine and generate electricity during this more valuable time. SMUD’s engineering contractor had provided a construction cost estimate of $1.45 billion.80 It was SMUD’s determination that an investment this size would significantly limit the choices SMUD has “with regard to future distributed generation technologies and significantly constrain SMUD’s future capital investments.”81 SMUD thus concluded that the project was not financially feasible and that with “recent advances in other energy storage technologies, it is likely there will be more economical alternatives” in the future.82 SMUD noted that the electric utility business is moving away from large, central power plants and that “technology for storing electricity in lithium-ion batteries has advanced at a surprising rate recently and could become economic on a larger scale in the next decade.”83 SMUD also noted that it is exploring new transmission alternatives as an option.84 SMUD’s actions continue to provide significant data and experience for utilizing batteries for behind the customer meter storage opportunities.
Case Study: GMP—Innovative Fees for Battery Services?
Green Mountain Power provides a stark contrast to SRP’s and SMUD’s price-signaling approaches to encourage appropriate smarter integration of battery storage with distributed generation of solar power. Green Mountain Power, which is an early promoter of the Tesla Powerwall, is the largest and only investor-owned utility in Vermont, serving over 260,000 customers (formed from the 2012 merger of Central Vermont Public Service Corporation and Green Mountain Power). GMP promotes the Tesla Powerwall on the homepage of their Web site and claims to be the first utility to “offer the Tesla Powerwall to customers.”85 The GMP pilot will utilize the 7-kWh model and the utility has ordered an initial 500 units. GMP will pair the Tesla technology with a SolarEdge bidirectional inverter to allow homes’ alternating current electric system to interact with the battery’s direct current, and the unit comes with a 10-year warranty. Tesla offers to take back and recycle the battery at the end of its useful life.
GMP is offering three service options for its customers. The first option is direct sale of the battery to the customer, where the customer maintains full access to the battery, and will cost the customer $6,501. While Tesla’s announced price was $3,000 for the 7-kWh model, the GMP price includes the cost of the bidirectional inverter, sales tax, and a 20 percent GMP markup. The customer will be responsible for installation and maintenance costs.
GMP’s second service option is a direct sale of the battery, but then the customer shares access to the battery with the utility. Under this option, the customer will again be responsible for paying $6,501 as well as installation costs, but the customer will be paid a monthly bill credit of $31.76 for sharing access with the utility. The customer is also required to maintain communications with the Powerwall or it may be subject to additional fees. GMP calculated the bill credit as the value of the reduction in wholesale electric and regional transmission cost savings from the utility having control of the battery during peak demand times. GMP calculated these savings based on its estimate of success in controlling the batteries and offering the energy into the regional competitive wholesale markets to reduce charges and generate additional revenues. Under this option, GMP would aggregate the batteries throughout its service territory and offer them into the markets run by ISO-NE, which is the New England electric grid operator.
Under the third and final service plan, GMP installs, owns, and maintains the Powerwall, and the customer pays no upfront fees. GMP and the customer would share access to the battery, with GMP being allowed to store energy during off-peak periods and dispatch back onto the grid for on-peak periods. GMP calculated the monthly cost of the battery to the utility to be $86 per month ($2.84 per day) and then analyzed the 10-year value stream from the battery based on its estimate of reduced RTO peak demand charges from the capacity market, and regional transmission charges as $1.69 per day. The net result is a $37.50 per month ($1.25 per day) cost to the customer. Customers who sign up for this plan will commit to remain on the rate rider for 10 years, which is the useful life of the battery, and must maintain reliable communications with the system.
For Powerwalls controlled by GMP, the utility will have the power to control the charging and discharging cycles and will reduce utility costs by discharging the batteries during high-market price periods and at times of peak load. The utility will attempt to limit its use of the batteries during periods that are expected to be high outage periods (e.g., during expected storms), and the customer will then have access to the batteries for backup power in the event of an outage. The duration of the power available from the battery will be highly dependent on the load that it is serving, but the utility estimated that a single Powerwall could provide as many as six hours of backup power (as discussed previously, this would not include the use of multiple large appliances).
The ultimate value of the Powerwall to the customer is difficult to gauge, but since GMP has not been a leader in offering cutting-edge time-of-use rates like SMUD or SRP, even for those who own the battery outright, the primary value for the customer appears to be backup power. While GMP notes that its typical winter outage lasts about 2.5 hours, well within the Powerwall’s range, those who live in northern climates or are concerned about future severe weather events tend to be less concerned with the typical short outage and more concerned with the multiday outages that occur regularly (if not frequently) within the GMP service area, which is regularly subjected to heavy snowfalls and intense winter storms. These multiday outages are not served well by the limited capacity of the Powerwall. Concerning payback periods, it would take more than 17 years for a straight payback to return the battery’s cost to the customer under the program’s second option. The third option—GMP’s initial ownership—could be considered as an interest-free 14-year installment plan from the utility to the customer.86 A GMP residential customer who used 600 kWh per month would have a bill of approximately $100, and the monthly fee for the GMP rate rider of $37.50 would equate to a 37 percent increase in the monthly bill for the benefit of having some backup power covering typical, shorter outages. At the end of the day, if the customer value proposition is limited to backup power, then it is going to be tough to compete with home generators from companies such as Generac, Cummins, and Briggs and Stratton, which can power most of the appliances in a home with virtually limitless (as long as you have the fuels on hand) amounts of backup power for $5,000 or less.87
Case Study: Stem Is the Behind-the-Meter Leader in SCE’s Storage Procurement
Beginning with the California Legislature’s passage of AB 2514, which called for a statewide energy storage mandate to enable a market transformation of these technologies, the California Public Utility Commission (CPUC) finally approved the rules on October 17, 2013. The CPUC rules required the state’s three IOUs to procure a staggering 1.3 gigawatts of energy storage by 2020. Under the proposed procurement targets, SCE would procure 90 megawatts, beginning in 2014, and ramp up biannually with the additional procurement targets set at 120 megawatts in 2016, 160 megawatts in 2018, and 210 megawatts in 2020, for 580 megawatts of storage capacity.88 Through the 2016 procurement, Stem has been the leading behind-the-meter provider for SCE.89 The vast majority of this capacity was awarded in SCE’s Local Capacity Requirement (LCR) request for offers, which required SCE to procure at least 50 megawatts of energy storage. In the end, SCE procured approximately five times that much, with Stem winning 85 megawatts of contracts in the West Los Angeles Basin.90 The significance of this procurement cannot be understated, as this was the first time that behind-the-meter storage has gone head to head in such a procurement, and startup battery storage company Stem was the biggest behind-the-meter storage provider winner. In delivering on its commitment to provide 85 megawatts of storage, Stem must identify customers in specific geographies for its behind-the-meter energy storage system, with all 85 megawatts deployed by 2021.91 Stem already has experience in California with demand response pilots and is installing its LIBs and control systems in 68 Extended Stay America’s California hotels.92 Stem’s current products range from 18 kilowatts to 54 kilowatts and are offered through a no-money-down leasing program, with costs recovered through a monthly fee paid by the customer, while the customer sees utility bill reductions by avoiding demand charges.93
In January of 2016, SCE and other California utilities announced the first winners from the Demand Response Auction Mechanism (DRAM), which is the “state’s first big attempt to bring distributed energy resources into service for the grid.”94 Potential resources range from smart thermostats to behind-the-meter batteries.95 While SCE procured just over 20 megawatts of resources, Stem was awarded 100 kW.96 The DRAM has stringent requirements, which include requiring the resources to deliver the promised amount of demand reduction for up to four hours per day, and up to three consecutive days during peak demand times.97 According to Greentech Media, Stem’s marketing director noted that these were only “small, pilot scale” numbers and that the goal was “making sure for our customers that we can help them get the first opportunity to participate in these new programs, and offer some additional value streams.”98 Stem expects larger participation in the 2017 DRAM. According to Stem director of policy Ted Ko, “this business model, stacking values and participating in wholesale markets, DRAM or other markets, is the core for Stem. Wherever we go to deploy, we will be looking for these kinds of opportunities.”99 Stem has also recently been successful in other markets. Stem was one of ten winners of an auction to supply demand response to Con Edison, where Stem will install up to 857 kW of battery storage in New York City by 2018.100 Currently most of Stem’s battery installations work off of grid electricity, although with its agreement with SunPower, it is hoping to increase opportunities for solar and storage.101
The case studies we examined—the Salt River Project, Sacramento Municipal Utility District, Green Mountain Power, and Stem—demonstrate that there are changing utility policies that can complement the declining costs of solar generation and battery storage. While these cases have largely focused on opportunities for residential customers, parties are also focusing on opportunities in the commercial and industrial markets. Furthermore, regions such as Germany, Hawaii, Puerto Rico, and even California, which have relatively high electricity prices and opportunities for meaningful time-of-use rates, are segments that might find an expedited and even smoother path toward large-scale adoption of batteries.
Each of the leading companies we have studied is targeting a different approach that enhances the business case for battery storage. Each example, in its own way, also highlights the challenges that must be overcome in order to make batteries a ubiquitous component of a low-carbon energy system. Continued reductions in both solar and battery costs will play a major role in overcoming these challenges and in bringing the electric battery closer to being a leading cost-effective choice for carbon reduction for the home and small business.
