Meet the battery-powered home

YOU HAVE to hand it to Elon Musk for his ability to change the game and capture the headlines. The billionaire entrepreneur made a fortune from the sale of PayPal, then went on to challenge luxury carmakers with the Tesla Model S, and to shock the satellite-launcher establishment with SpaceX’s cut-price Falcon 9 rockets. Mr Musk now wants to disrupt the electricity business—by changing the way people consume the stuff, preferably by generating and storing their own, and kissing the grid goodbye.

On April 30th, Tesla unveiled a line of lithium-ion storage batteries (similar to those used in its electric cars) that let homeowners and businesses save surplus electricity produced by rooftop photovoltaic panels during the day for use at night. Around a 1,000 journalists and analysts cheered Mr Musk’s announcement. Only the late Steve Jobs could generate such an enthusiastic response for so mundane a trade announcement. Such is the reality-field distortion of the Tesla brand and the company’s charismatic leader.

Tesla’s residential battery pack, called the Powerwall, comes in two sizes: a 7 kilowatt-hour (kWh) lithium-ion system priced at $3,000 and a 10 kWh unit for $3,500. Both are housed in similar wall-mounted enclosures that can be ordered in red, white or blue as well as grey or black. Neither battery pack includes an inverter/rectifier for turning the unit’s 400-volt direct current into 120-volt alternating current, and vice versa. Customers will also have to pay certified technicians to install the battery pack on a wall and wire it into the building’s main electricity panel—no job for home handymen. Installation, wiring and inverter can easily double the price.

Tesla reckons up to nine Powerwalls can be hooked together for a total of 90kWh. Beyond that, customers will be able to order (later this year) a commercial version of the battery system, called Powerpack. This will have a capacity of 100kWh, and cost a reputed $25,000. Powerpacks can be plugged together to provide up to ten megawatt-hours (MWh) of storage. Amazon is installing 4.8MWh of the units in one of its data centres in northern California. Target, a national retail chain, has a similar trial underway.

Given all the buzz surrounding Tesla’s announcement, readers would be forgiven for thinking residential energy storage was something new. If anything, Tesla is a latecomer to the game. At least a dozen other companies in America sell similar battery packs. JuiceBox Energy is typical of the new generation of storage-battery builders. Based in Silicon Valley, but with engineers drawn from the motor industry in Detroit, JuiceBox’s 8.6kWh battery pack uses hundreds of “18650” cylindrical lithium-ion cells found most often in laptop computers. Tesla chose the same rechargeable 18650 cells—so-called because they are roughly 18.6mm (three-quarters of an inch) in diameter and 65mm long—for its electric cars, and will use essentially similar cells in its energy storage packs.

While still small, the battery-storage market is increasing in line with the growth of solar installations. Last year, $128m worth of battery packs were bought in America, says GTM Research of Boston, Massachusetts. The vast majority, however, were acquired by utilities.

The problem power companies face, especially in the American southwest, is that they are inundated with renewable energy (solar and wind) during the off-peak afternoon, when the last thing they need is yet more electricity—least of all, expensive renewable energy. Meanwhile, they get precious little help from renewables during periods of peak demand in the evening and early morning.

In the absence of adequate means to store it, much of this surplus green energy has had to be dumped on the spot market below cost. That is where battery storage systems come into play: they provide utilities with a handy way of “load shifting” the surplus from daytime periods of over-supply to times when demand is at its greatest.

Load shifting is being touted for residential users as well as utilities. For homeowners with their own solar generators, the preferred way to load shift has been to sell surplus juice to the local utility during the day and buy electricity back from the grid when needed at night.

That is becoming uneconomic, as the wholesale price utilities buy electricity from home generators falls and the retail price they sell it back to homeowners rises. Meanwhile, state regulators are beginning to let local utilities reduce the number of off-peak hours in the day, making it harder for customers with solar-generators to use the grid for storage. With so-called “net metering” on the way out, battery storage is becoming increasingly popular.

Even without rooftop solar, homeowners are being encouraged to arbitrage the difference between peak and off-peak electricity rates using residential battery packs. The idea is to buy cheap electricity from the grid during the night to charge the batteries, and run the household on battery power when the local utility's rates are high. In California there is a 19 cent difference between off-peak power bought in the middle of the night and peak power in the evening.

Be warned, though. For one thing, the amount of storage capacity needed to run a household “off the grid” for the rest of the day can prove pricey. The average American home needs 3-4 kilowatts (kW) of power to operate its heating, cooling, appliances and lighting. The JuiceBox, for instance, can deliver 4.3kW of continuous power. So, one stand-alone unit costing $9,900 ought to suffice for most households. The Tesla battery pack, by contrast, has a continuous power rating of just 2kW. In this case, homeowners expecting to use an electric oven to cook supper while running a washing machine, all on battery power, would be out of luck. At least a pair of Powerwalls, costing $7,000 for the units alone, would be needed to do the job.

Also, there would be no investment tax credit for installing a stand-alone battery packs. The 30% solar investment tax credit—the American government’s main policy measure for encouraging people to install solar panels on their roofs—applies only if batteries are part of a complete solar-generation setup. Besides, the 30% incentive for homeowners is to disappear in 2017. Commercial installers will be a little better off, with their 30% tax credit falling to 10%, rather than zero.

The average American home consumes 11,000kWh of electricity a year, so a 19 cent difference between peak and off-peak rates implies an attractive annual saving of $2,000 from load shifting. Unfortunately, that ignores the lifetime cost of the battery system, not to mention the cost of charging the batteries every night. A typical battery pack for an average-sized home could cost around $13,000, including installation and inverter/rectifier. In most cases, this would be financed using a 5% bank loan over ten years. Add the cost of charging, maintenance, inflation, interest and system degradation, and the total could easily exceed $2,000 a year.

In California, load shifting might just about break even, but elsewhere in America people installing load-shifting battery packs would probably be out of pocket. Of course, homeowners who already have solar panels on the roof would benefit substantially from adding battery packs to their systems, especially as the inverter/rectifier would have already been paid for. With enough solar capacity, they would then be free to load shift their own green electricity from afternoon to evening, free of the grid.

The one surprise to come out of the recent Tesla announcement was the price of its battery packs per kilowatt-hour. At $350, that was about half of what industry veterans anticipated, says Neil Maguire, chief executive of JuiceBox. The commercial version is expected to sell for $250 per kWh. That is a real bargain, given today’s battery prices.

Judging from the company’s patents, Tesla has clearly done a lot of work on making the 18650 cells it buys from Panasonic for its electric cars cheaper and safer than commodity versions on the market. Even so, they still cost Tesla around $200 per kWh, reckons Mohamed Alamgir, director of research at battery-maker LG Chem Power in Troy, Michigan. Packaging the individual 18650 cells into modules with control circuitry and cooling where necessary just about doubles the cost. As such, Tesla is set to lose money on even its residential battery packs, let alone its industrial ones.

Why on earth would Tesla do that? Because it does not expect to go on losing money indefinitely. By forward-pricing its storage batteries, it is banking on the cost per kilowatt-hour falling further down the learning curve and crossing the break-even point sooner rather than later. Indeed, by creating extra demand for all the 18650 cells that are to be produced by Panasonic at the “gigafactory” the two companies are building jointly in Nevada, Tesla is seeking to ensure that the plant will be reasonably well loaded when manufacturing starts in late 2016. That is essential if economies of scale are to be realised.

The $5 billion gigafactory rising from the desert floor near Reno is quite possibly the biggest gamble Mr Musk has taken so far. The plant will be capable of producing enough battery packs each year to supply power for no fewer than 500,000 electric vehicles. That is a staggering number of electric vehicles for one manufacturer to turn out annually.

This year, Tesla expects to deliver 55,000 of its current Model S cars. It will introduce a companion model, a sports-utility vehicle known as the Model X, this coming autumn. Even if the two luxury electric vehicles (which use the same underpinnings, battery packs and drivelines) do not cannibalise one another’s sales too much, their combined volume is unlikely to absorb more than a fraction of the gigafactory’s output.

This is where Tesla’s proposed Model 3 comes in. It is to be a smaller, more affordable electric car built in large numbers for mainstream buyers. The aim is to get the Model 3 into production by the end of 2017. However, to keep the car’s sticker price under $35,000, Tesla’s already low battery costs have to come down still further. That means making sure the gigafactory really is fully loaded, if the claimed economies of scale are to be captured. The introduction of storage batteries for the residential and industrial market will help considerably. So will any other battery application Mr Musk's fertile mind can think of.