September 11, 2017 Categories: California Clean Energy Climate Change Low-Carbon Fuel

The Power of Cheap Batteries, or, How I Learned to Stop Worrying and Love Fast Chargers

A few weeks ago, there was a fascinating, multi-threaded twitter conversation among energy policy wonks about how lithium-ion batteries have become cheap, the pace of innovation and how policy impacts technology deployment. It was a great illustration of the good, thoughtful side of Twitter. One tweet particularly stuck with me:

That seems like a lot of air conditioners, but is it enough to be a problem? Electricity grid operators prefer not to have large loads popping on and off as vehicles fill their batteries. Proposed higher-power chargers could exacerbate the problem; charging company EVGo is currently exploring a 350 kilowatt (kW) model which could give future generations of electric vehicles enough juice for over 200 miles of range in less than 10 minutes. Tesla apparently aims even higher:

Should We Be Worried?


350 kW chargers are great for plug-in vehicle owners who want a quick top-off of clean energy, but too many vehicles charging at once could cause spikes in demand. At Musk’s “children’s toy” levels of 350 kW, 20,000 EVs charging at once would lead to about 7 Gigawatts (GW) of electric demand. Compare this to 30-40 GW typical peak demand for the grid which covers most of California. 20,000 EVs plugging into fast chargers at once is an unlikely scenario, since most people use lower-voltage chargers at home or work for most of their charging, but if this did happen, if could cause grid issues unless managed properly.  

Luckily, managing power demand is a problem for which there are many solutions. One is to schedule charging in a way that spreads out the demand; in many cases that’s what grid operators plan to do. Since most vehicles spend most of their time parked at home or work, where they could be plugged in, the grid operator could spread available capacity between plugged in cars, knowing that they have several hours to accommodate them all. As our grid gets more advanced, electric vehicles or vehicle chargers could communicate with grid controllers to efficiently schedule charging – even helping utilize surplus wind or solar energy that comes when there isn’t enough demand to use it all.

Not all charging can be scheduled, however. If someone is on a road trip, they won’t be inclined to wait around until the grid operator decides it’s convenient for them to get power. They’re also going to want a rapid re-charge, which means high voltage chargers. For this need, the question is, how do you smooth out the peaks and valleys of time-sensitive charging? Smoothing out peaks and valleys in demand sounds like a job for….

It wears glasses so you can't tell that by day, it powers an iPad mini.

The same technology which powers most EVs can also support their charging stations. By building a battery into chargers, vehicles can be recharged very rapidly without putting an excessive burden on the grid. The battery can charge overnight, when the grid has capacity and when clean wind energy is plentiful, or at midday, when solar power peaks. During the day, it can pull from the grid when doing so does not cause problems, but dip into the battery during busy times.

At What Cost?


Batteries are getting cheaper all the time, but they’re not free, and you may need to put enough battery capacity on the charger to fill several cars before it has time to re-charge, so the critical question is: how much will these batteries cost?

The answer: probably not all that much.

Consider a hypothetical 350 kW charger located in a public place. This charger can transfer enough energy to drive an EV for 200 miles in less than 10 minutes, so it’s basically the EV equivalent of a gas pump. Let’s assume that people use it more or less like they use gas pumps right now: for most stations, the busiest times are weekdays during commute hours, with moderate traffic during working hours and very little activity at night.

Tuesday-Thursday refueling activity at a typical urban/suburban fueling station. Monday and Friday show a similar profile. Source: Gas Technology Institute (2008)

Let’s assume that during its busiest hour, around 4pm, the charger is in near-continuous use: 10 minutes of charging with one idle minute as the next car pulls up. If we scale the remaining hours of the day to match the activity in the graph, our charger would average just over 177 kW across the day, or about half of its maximum power. Let’s say our goal is to add enough batteries to allow the charger to meet all of its needs by drawing a steady 180 kW from the grid, eliminating all the day’s peaks and valleys. A 1.1 megawatt-hour battery, enough to fill 17 base-model Tesla Model 3’s, gives enough storage with a bit of wiggle room to spare. It stores enough energy during the night that it can ride out a busy day, including an hour of near-continuous operation in the afternoon without running dry.

Tesla is currently reporting battery costs slightly under $200/kWh; which means insulating the grid from the charger’s variable demand it could add up to $220,000 in cost. That’s a lot of money, but there are three things to consider: First, high-power chargers, like the 350 kW model we’re thinking about, are not going to be cheap. It takes very robust equipment to handle that much power. There aren’t credible source for the actual cost of one of these things, since they’re only in pilot projects at the moment, but most researchers I’ve talked to agree that it could be over $1 million per charger for the first several years. Adding $220,000 to the cost per charger for batteries may be tolerable. Second, the cost of adding batteries trades off against the cost of building a higher-power connection to the grid; you can make up some or all of the battery costs if your transformers, conduit and power management systems don’t need to be as powerful. Grid operators can find the optimal balance between battery capacity and a robust grid connection. If our charger  draws 200 kW instead of 180 kW, the amount of battery required drops from 1100 to 860 kWh. A 250 kW grid connection drops needed battery capacity all the way to 310 kWh. You can also use batteries to allow the charger to reduce its demand at times when the grid is under the most strain. If we gave our charger on the 200 kW connection to the grid 1.2 MWh of batteries, it could drop its demand on the grid by half from 5 to 9 PM, reducing the need to run fossil-fueled “peaker” plants. Batteries offer charging providers and grid operators the flexibility to find the balance of storage and power best suited for local conditions.

Third, and maybe most importantly, the battery offers value beyond smoothing demand to the grid. Batteries can absorb excess power when brief fluctuations make it available and supply short jolts when needed to ensure that grid conditions stay within tolerances. Recent tests have demonstrated that batteries can serve this purpose, which previously would require keeping power plants running at idle, or other costly measures. A charging station operator could choose to commit part of their battery capacity to providing ancillary services during part of the day, providing another revenue stream to offset their costs.  

Finally, it’s important to remember that these costs are probably the worst-case scenario, for the situations where scheduling or simply relying on existing resources isn’t an option.

Cheap Batteries Solve A Potential Problem of Cheap Batteries

Much has been written about the potential of EV’s to reduce emissions of climate pollutants like CO2, improve air quality and help support a renewable electricity grid. The rapid decline in battery costs has unlocked revolutionary advances in clean transportation. All of these benefits depend on being able to fuel EVs in a convenient, grid-smart fashion. Simply deploying chargers and hoping for the best leaves the potential for transient spikes in demand if many EV owners happen to plug into high-power chargers at once. There are many tools available to charging station providers and grid operators to manage this risk, including cheap batteries. A moderate amount of battery capacity can dramatically reduce a charger’s demand on the local grid and even help stabilize the local grid by providing ancillary services. Widely available low-power charging at home and work can serve the vast majority of vehicle charging needs, and can help integrate variable renewable energy sources when managed with smart grid applications. Far from signaling a new threat to grid stability, batteries and the vehicles they power could make tomorrow’s grid even more reliable and less expensive than it is today.

Join the Conversation